0.1tesis 2023 03 20

Instituto Tecnológico y de Estudios Superiores de Monterrey
Campus Monterrey
School of Engineering and Sciences
Object Manipulation by Collaborative Autonomous Underwater

Vehicles Driven by a Model-free Second-order Sliding Mode Controller
with Finite-time Convergence
A dissertation presented by
Josué González García

Submitted to the
in partial fulfillment of the requirements for the degree of
Doctor of Philosophy
In
Engineering Science
Monterrey Nuevo León, June 1st, 2023

Instituto Tecnológico y de Estudios Superiores de Monterrey
Campus Monterrey
The committee members, hereby, certify that have read the dissertation presented
by Josué González García and that it is fully adequate in scope and quality as a partial
requirement for the degree of Doctor of Philosophy in Engineering Science.
_______________________
Dr. Alfonso Gómez Espinosa
Tecnológico de Monterrey
Principal Advisor
_______________________
Dr. Luis Govinda García Valdovinos
Center for Engineering and Industrial Development
Energy Division
Committee Member or Co-advisor
_______________________
Dr. Tomás Salgado Jiménez
Center for Engineering and Industrial Development
Energy Division
Committee Member
_______________________
Dr. Enrique Cuan Urquizo
Committee Member
_______________________
Dr. Jesús Arturo Escobedo Cabello
Committee Member
_______________________
Dr. Rubén Morales Menéndez
Dean of Graduate Studies
ii
Declaration of Authorship
I, Josué González García, declare that this dissertation titled, Object Manipulation
by Collaborative Autonomous Underwater Vehicles Driven by a Model-free Second-order
Sliding Mode Controller with Finite-time Convergence, and the work presented in it are
my own. I confirm that:
• This work was done wholly or mainly while in candidature for a research degree at
this University.
• Where any part of this dissertation has previously been submitted for a degree or
any other qualification at this University or any other institution, this has been
clearly stated.
• Where I have consulted the published work of others, this is always clearly
attributed.
• Where I have quoted from the work of others, the source is always given. With the
exception of such quotations, this dissertation is entirely my own work.
• I have acknowledged all main sources of help.
• Where the dissertation is based on work done by myself jointly with others, I have
made clear exactly what was done by others and what I have contributed myself.
___________________________
@2023 by Josué González García

All rights reserved
iii
Dedication
Thanks for all your unconditional confidence, support, patience, and

encouragement. You were my main motivation for pushing through this work.
iv
Acknowledgements
I would like to express my deepest gratitude to Tecnológico de Monterrey for the

support on tuition and CONACyT the support for living.
v
Object Manipulation by Collaborative Autonomous Underwater Vehicles Driven by a
Model-free Second-order Sliding Mode Controller with Finite-time Convergence
By
Abstract
The use of Autonomous Underwater Vehicles (AUV) has expanded in recent years
to include them in inspection, maintenance, and repair missions. These missions require
the automation of tasks such as autonomous navigation and station-keeping, which are
pretty challenging due to the complexity of the vehicle itself and the underwater
environment. Traditional control strategies have been used to deal with those problems
before, but their performance is limited since they do not consider non-linearities, external
disturbances, or model uncertainties. Non-traditional controllers have been explored too,
but have shortcomings such as being model-based, parameter-dependant, or so.
Additionally, the nature of some objects implies the involvement of multiple vehicles to
manipulate them, making the manipulation mission even harder. Collaboration of AUVs
is a strong challenge since underwater communications are limited or null. An advanced
control strategy dealing with the trajectory tracking and the station-keeping problem would
be desirable for simplicity and robustness. If this controller could also help coordinate
multiple vehicles without requiring them to communicate with each other, it would make
collaborative manipulation tasks reachable. For this purpose, a model-free high-order
Sliding Mode Controller is presented in this project. Unlike others control approaches, the
proposed control strategy achieves finite-time convergence to a practical zero error in a
time-base that the user can arbitrarily define. This characteristic will be used to coordinate
the navigation of two AUVs to manipulate an object underwater. The performance of the
proposed controller was evaluated by numerical simulations and experiments in a semi-
Olympic swimming pool and compared with classic and state-of-the-art control strategies
regarding trajectory tracking and station-keeping problems. Results have shown that the
proposed controller can achieve finite-time convergence of the tracking errors to a
practical zero value in the predefined time-base. Results also demonstrated that the
convergence time could be arbitrarily selected by the user and achieved by the controller
resulting in, as far as the author knows, the first controller with this characteristic for AUVs
navigation. The simulations and experiments also showed that the proposed controller
outperformed classical and state-of-the-art controllers regarding the Root Mean Square
of the tracking errors and energy consumption. Finally, the proposed model-free high-
order SMC coordinated two BlueROV2 vehicles in a collaborative manipulation mission
without vehicle-to-vehicle communication. Numerical simulation results confirmed that the
proposed controller could manage the vehicles to follow a predefined coordinated
trajectory to approach, grip, transport, and release an object without communicating with
each other.
vi
List of Figures
Figure 1. Autonomous Underwater Vehicle (AUV) technologies for localization and navigation. ......................................... 6
Figure 2. Dead-Reckoning drift effect. ................................................................................................................................ 7
Figure 3. Vertical deflection. ............................................................................................................................................... 8
Figure 4. AUV equipped with two side-scan Sound Navigation and Ranging (SONARs). ................................................... 9
Figure 5. Forward-Looking SONAR (FLS) placed at vertical and horizontal orientations. .................................................. 10
Figure 6. Acoustic localization systems: (a) Long Baseline, (b) Short Baseline, (c) Ultra-Short Base Line. ....................... 11
Figure 7. The schematic diagram for Short Baseline (SBL) system. ................................................................................. 12
Figure 8. Range-only/Single-beacon positioning of a fixed target from a moving vehicle. ................................................. 14
Figure 9. Geomagnetic map features. .............................................................................................................................. 16
Figure 10. Terrain-Referenced Navigation. ....................................................................................................................... 17
Figure 11. Optical localization systems based on active landmarks. (a) AUV following an array of active markers, (b) AUV
locating an entrance by an arrangement of active markers. ............................................................................................................. 18
Figure 12. Simultaneous Location and Mapping (SLAM) of (a) an AUV equipped with a sensor to map its environment and
(b) digital reconstruction of the environment. ................................................................................................................................... 20
Figure 13. Loosely Coupled (LC) vs. Tightly Coupled (TC) sensor fusion schemes. ......................................................... 22
Figure 14. Sensor fusion alternatives for AUV positioning. ............................................................................................... 24
Figure 15. Waypoint navigation representation. ................................................................................................................ 28
Figure 16. Path following navigation representation. ........................................................................................................ 29
Figure 17. Trajectory tracking navigation representation. ................................................................................................. 29
Figure 18. Parallel navigation of AUVs. ............................................................................................................................ 33
Figure 19. Leader-follower navigation of AUVs................................................................................................................. 33
Figure 20. Collaborative mission of AUVs. a) Survey missions. b) Intervention missions. ................................................. 35
Figure 21. Reference frames for underwater vehicles....................................................................................................... 39
Figure 22. Thruster configuration for the BlueROV2. A) Top view. B) Front view. ............................................................. 42
Figure 23. Model-free second-order sliding mode control with finite-time convergence complete block diagram............... 48
Figure 24. BlueROV2 simulator. Simulink block diagram. ................................................................................................. 49
Figure 25. Parameterized desired spiral trajectory of the Unmanned Underwater Vehicle (UUV) BlueROV2. ................... 50
Figure 26. Results with 𝑡𝑏 = 5 𝑠. (a) Trajectory tracking. (b) Position tracking errors. ...................................................... 53
Figure 27.The simulated three-dimensional trajectory of the BlueROV2 with 𝑡𝑏 = 5 𝑠. .................................................... 53
Figure 28. Control signal for each thruster. Simulation with 𝑡𝑏 = 5 𝑠. .............................................................................. 54
Figure 31. Error convergence for different controllers. (a) Surge error. (b) Sway error. (c) Heave error. (d) Yaw error. ..... 56
Figure 32. Results for the finite-time convergence controller subject to ocean currents and 𝑡𝑏 = 5 𝑠. (a) Trajectory tracking.
(b) Position tracking errors. .............................................................................................................................................................. 57
Figure 33.Experimental set-up at Tecnologico de Monterrey, Campus Querétaro. (a) Ground control station at a side of the
semi-Olympic pool. (b) BlueROV2 deployed into the water. (c) Autonomous trajectory tracking mission. ......................................... 58
Figure 34. BlueROV2 hardware configuration .................................................................................................................. 59
Figure 35. BlueROV2® software configuration. ................................................................................................................ 59
Figure 36. Controlled trajectories with the PID controller. (Left) Depth. (Right) Heading. .................................................. 61
Figure 37. Controlled trajectories with the PID controller. ................................................................................................. 61
Figure 38. Controlled trajectories with the asymptotic model-free 2nd-order SMC. (Left) Depth. (Right) Heading. ............ 62
Figure 39. Best controlled trajectories with the asymptotic model-free 2nd-order SMC. .................................................... 62
Figure 40. Controlled trajectory with the model-free high-order SMC with finite-time convergence in a 5 s predefined-time.
........................................................................................................................................................................................................ 63
Figure 41. Tracking error for depth and heading control. .................................................................................................. 63
Figure 42. Time base generator results for 𝑡𝑏 = 5 s. ......................................................................................................... 64
Figure 43. Time-varying gain α(t) result for 𝑡𝑏 = 5 s. ......................................................................................................... 64
Figure 44. Controlled trajectories with the model-free high-order SMC with finite-time convergence in predefined-time. (Left)
Depth. (Right) Heading. ................................................................................................................................................................... 64
Figure 45. Velocities estimations by Euler and Exact differentiators. ................................................................................ 65
Figure 46. RMSE comparison for the trajectory tracking with the different controllers. (Left) Depth. (Right) Heading. ....... 65
Figure 47. Control coefficients on the thrusters for the different control approaches. ........................................................ 66
Figure 48. Simulation results for finite-time trajectory tracking and station-keeping in the 𝑥, 𝑦, 𝑧 positions and 𝜓 orientation.
External disturbances were introduced in the interval 10 𝑠 ≤ 𝑡 ≤ 18 𝑠 as ocean currents (𝜈𝑜𝑐). ....................................................... 69
Figure 49. Experimental set-up. (A) Semi-Olympic swimming pool. (B) Control station. (C) BlueROV2 deployment. (D)
Station-keeping task execution. ....................................................................................................................................................... 70
Figure 50. Experimental set-up, hardware configuration. .................................................................................................. 70
Figure 51. Additional thruster for external disturbance. A) Front view diagram. B) Implementation. .................................. 71
Figure 52. Experimental set-up, software configuration. ................................................................................................... 71
Figure 53. Depth (left) and tracking error (right) results of the control test. No external disturbances were introduced. ..... 72
Figure 54. Experimental results of the control test for depth station-keeping. Control signal 𝜏𝑧 (left) and thruster coefficients
𝑢5,𝑢6 (right). No external disturbances were introduced. ................................................................................................................. 72
Figure 55. Results of the open-loop experiments. Depth of the vehicle (upper left) with an external disturbance of ~50 𝑁
(upper right) introduced at 𝑡 = 13 𝑠. Depth of the vehicle (lower left) with an external disturbance of ~ − 40 𝑁 (lower right) introduced
at 𝑡 = 13 𝑠. ....................................................................................................................................................................................... 73
vii
Figure 56. Experimental results for depth station-keeping with an external disturbance of ~10 𝑁 introduced in the interval
13 𝑠 ≤ 𝑡 ≤ 25 𝑠. Depth (left) and tracking error (right). ...................................................................................................................... 74
13 𝑠 ≤ 𝑡 ≤ 25 𝑠. Control signal 𝜏𝑧 (left) and thruster coefficients 𝑢5,𝑢6 (right). .................................................................................. 74
Figure 58. Additional thruster coefficient of −0.25 in the interval 13 𝑠 ≤ 𝑡 ≤ 25 𝑠. .............................................................. 75
Figure 61. Additional thruster coefficient of −0.50 in the interval 13 𝑠 ≤ 𝑡 ≤ 25 𝑠. .............................................................. 76
Figure 62. Time-varying additional thruster coefficient in the interval 13 𝑠 ≤ 𝑡 ≤ 25 𝑠. Minimum value −0.25 and maximum
value 0.25. ....................................................................................................................................................................................... 77
Figure 63. Experimental results for depth station-keeping with a time-varying external disturbance introduced in the interval
Figure 64. Experimental results for depth station-keeping with a time-varying external disturbance introduced in the interval
Figure 65. Cooperative manipulation task. Trajectory design for both vehicles. ................................................................ 83
Figure 66. MFSOSMC x,y, and z trajectory tracking (no external disturbances). BlueROV2 #1 (left). BlueROV2 #2 (right).
........................................................................................................................................................................................................ 86
Figure 67. MFSOSMC ϕ,θ, and ψ trajectory tracking (no external disturbances). Left: BlueROV2 #1. Right: BlueROV2 #2.
........................................................................................................................................................................................................ 86
Figure 68. BlueROV2 #1 velocity tracking with the MFSOSMC (no external disturbances). .............................................. 87
Figure 69.BlueROV2 #2 velocity tracking with the MFSOSMC (no external disturbances). ............................................... 87
Figure 70. UUV1 and UUV2 trajectory tracking the MFSOSMC in the collaborative manipulation of an object (no external
disturbances). (a) Isometric view. (b) Top view. ............................................................................................................................... 88
Figure 71. Tracking error for all the controllers (no external disturbances). (a) 𝑥 − 𝑎𝑥𝑖𝑠. (b) 𝑦 − 𝑎𝑥𝑖𝑠. (c) 𝑧 − 𝑎𝑥𝑖𝑠. ........... 89
Figure 72. Tracking error for all the controllers (no external disturbances). Left: 𝜙 orientation. Right: 𝜓 orientation. ......... 90
Figure 73. Performance indicators for all the controllers (no external disturbances). Left: Euclidean distance in the 𝑥 − 𝑎𝑥𝑖𝑠.
Right: relative heading orientation. ................................................................................................................................................... 90
Figure 74. Tracking error for all the controllers (with external disturbances). (a) 𝑥 − 𝑎𝑥𝑖𝑠. (b) 𝑦 − 𝑎𝑥𝑖𝑠. (c) 𝑧 − 𝑎𝑥𝑖𝑠. ......... 91
Figure 75. Tracking error for all the controllers (with external disturbances). Left: 𝜙 orientation. Right): 𝜓 orientation. ...... 92
Figure 76. Performance indicators for all the controllers (with external disturbances). Left: Euclidean distance in the 𝑥 −
𝑎𝑥𝑖𝑠. Right: relative heading orientation. .......................................................................................................................................... 92
Figure 77.Control signals for UUV1 by all the controllers (with external disturbances). ..................................................... 93
viii
List of Tables
Table 1. Commercial Inertial Measurement Unit (IMU)-Attitude and Heading Reference System (AHRS) systems. 9
Table 2. Commercial acoustic positioning systems. 12
Table 3. SLAM representation methods. 20
Table 4. Technologies for AUV localization and navigation summary. 24
Table 5. Commercial acoustic modems. 26
Table 6. SNAME notation for underwater vehicles. 39
Table 7. BlueROV2 parameters. 42
Table 8. Gain matrices for classic controllers and the proposed model-free second-order SMC with finite-time convergence.
52
Table 9. Controller parameters for simulation of the desired spiral trajectory. 52
Table 10. Root Mean Square (RMS) values for thrusters control coefficients. 56
Table 11. Experimental results for the vertical thruster coefficients in the station-keeping phase. 78
Table 12. Experimental results for the 𝜏𝑧 control signal in the station-keeping phase. 79
Table 13. Parameters for the controllers used in the collaborative manipulation simulations 85
ix
Contents
Abstract ................................................................................................................vi
List of Figures ...................................................................................................... vii
List of Tables ........................................................................................................ix
Chapter 1: Introduction ......................................................................................... 1
1.1 Motivation ................................................................................................. 1
1.2 Problem statement and context ................................................................ 1
1.3 Research question ................................................................................... 2
1.3.1 Hypothesis ......................................................................................... 2
1.4 Solution overview ..................................................................................... 2
1.4.1 Objectives .......................................................................................... 2
1.4.2 Proposal ............................................................................................. 3
1.4.3 Methodology ....................................................................................... 3
1.5 Main contributions .................................................................................... 4
1.6 Thesis structure ........................................................................................ 4
Chapter 2: Literature review ................................................................................. 5
2.1 Introduction .............................................................................................. 5
2.2 Underwater vehicle localization ................................................................ 7
2.2.1 Dead-reckoning and inertial navigation .............................................. 7
2.2.2 Acoustic navigation ............................................................................ 9
2.2.3 Geophysical navigation .................................................................... 15
2.2.4 Optical navigation ............................................................................. 18
2.2.5 Simultaneous location and mapping ................................................. 19
2.2.6 Sensor fusion ................................................................................... 22
2.2.7 Localization and navigation overview ............................................... 24
2.3 Underwater communication .................................................................... 25
2.3.1 Underwater communication overview ............................................... 28
2.4 Autonomous navigation of AUVs ............................................................ 28
2.5 The Station-keeping problem ................................................................. 32
2.6 Collaborative AUVs ................................................................................ 32
2.6.1 Collaborative navigation ................................................................... 32
2.6.2 Collaborative AUVs missions ........................................................... 35
2.6.3 Collaborative AUVs overview ........................................................... 38
Chapter 3: Unmanned Underwater Vehicles modeling ....................................... 39
x
3.1 Kinematics and hydrodynamics .............................................................. 39
3.2 BlueROV2 underwater vehicle ............................................................... 42
Chapter 4: Controller design ............................................................................... 44
4.1 Model-Free High Order Sliding Mode Controller .................................... 44
4.2 Time Parametrization of 𝜶 Gain ............................................................. 45
4.3 Stability Analysis .................................................................................... 45
4.4 Further Considerations ........................................................................... 47
4.4.1 Reference frame transformation ....................................................... 47
4.4.2 Saturation Constraints ...................................................................... 47
4.4.3 Exact differentiator ........................................................................... 48
Chapter 5: Controller validation for trajectory tracking ........................................ 49
5.1 Numerical simulator................................................................................ 49
5.1.1 Time-parameterized trajectory definition .......................................... 50
5.1.2 Comparing to classic controllers....................................................... 51
5.1.3 Results and discussion..................................................................... 52
5.2 Experimental validation .......................................................................... 57
5.2.1 Experimental setup........................................................................... 57
5.2.2 Time-parameterized trajectory definition .......................................... 60
5.2.3 Comparing with other controllers ...................................................... 60
5.2.4 Results and discussion ..................................................................... 61
5.3 Chapter conclusions ............................................................................... 66
Chapter 6: Controller validation for station-keeping ............................................ 68
6.1. Numerical simulator ................................................................................. 68
6.2. Experimental validation ............................................................................ 69
6.2.1. Experimental setup ............................................................................ 69
6.2.2. Results and discussion ...................................................................... 71
6.3. Chapter conclusions ................................................................................. 79
Chapter 7: Coordinated manipulation task .......................................................... 81
7.1. Coordinated manipulation task design ..................................................... 81
7.2. Comparing with other controllers .............................................................. 83
7.2.1. Finite-time second-order SMC (FTSOSMC) ...................................... 83
7.2.2. Non-singular Terminal Sliding Mode Control ..................................... 84
7.3. Numerical simulations .............................................................................. 85
7.3.1. Simulations without external disturbances ......................................... 85
xi
7.3.2. Simulations with external disturbances .............................................. 90
7.4. Chapter conclusions ................................................................................. 93
General conclusions ........................................................................................... 95
Future work ......................................................................................................... 95
Abbreviations and acronyms .............................................................................. 96
Bibliography ........................................................................................................ 98
Published papers .............................................................................................. 112
Autonomous underwater vehicles: Localization, navigation, and communication
for collaborative missions ........................................................................................ 112
Model-free high order sliding mode control with finite-time tracking for unmanned
underwater vehicles ................................................................................................. 113
Experimental Validation of a Model-Free High-Order Sliding Mode Controller
with Finite-Time Convergence for Trajectory Tracking of Autonomous Underwater
Vehicles ................................................................................................................... 114
Model-Free High-Order Sliding Mode Controller for Station-Keeping of an
Autonomous Underwater Vehicle in Manipulation Task: Simulations and Experimental
Validation ................................................................................................................. 115
Finite-Time Controller for Coordinated Navigation of Unmanned Underwater
Vehicles in a Collaborative Manipulation Task ........................................................ 116
Resume ............................................................................................................ 117
xii
Chapter 1: Introduction
1.1 Motivation
Over the years, many Unmanned Underwater Vehicles (UUVs) have been
designed to accomplish a wide range of applications in the scientific, commercial, and
military areas. UUVs have permitted us to deepen our knowledge of the oceans and
seafloors. Their use in tasks such as structural inspection [1], [2], environmental risks
detection [3], and mapping underwater structures [4], among others, increases the safety
of the mission and the reliability of their results [5] and reduces the operational costs
considerably [6]. In addition, UUVs have been widely used as an alternative to extremely
costly, time-consuming, and risky human underwater operations [7]. They can perform
some tasks that humans cannot do due to the challenging underwater environment. Most
of the tasks performed by these vehicles are for data-gathering applications [8]. However,
there is an increasing interest in their use in inspection, maintenance, and repair
operations [9] that require manipulation and interaction with objects in the underwater
environment. Regardless these vehicles are remotely operated (ROVs) or autonomous
(AUVs), certain operations, such as high-precision navigation and station-keeping,
require some form of autonomy from the vehicle. Autonomy is difficult to achieve due to
the vehicle's high non-linearity and the underwater environment's uncertainties [10].
Recent research has addressed the complete automation of such tasks [11]–[14].
However, the industry requires the accomplishment of underwater tasks whose
complexity involves the collaboration of two or more vehicles, which means new
approaches must be developed to deal with the problems of multi-vehicle systems.
1.2 Problem statement and context
Underwater missions are pretty challenging. Assembling an AUV with the proper
technology to overcome localization, navigation, and communication shortcomings is
difficult for researchers who must limit their proposals to numerical simulations. It is even
more complex when collaborative AUVs refer to. As a result, most of the work reported
for AUVs is meant for surveillance tasks. Intervention missions reported mainly involve
carrying an object underwater, and the efforts have remained at numerical simulations.
Moreover, some of them correspond to teleoperated vehicles since it is a task hard to
automate. The vehicle’s coordinated navigation and station-keeping are two main
problems to work on to achieve collaborative intervention missions.
Vehicle navigation can be autonomously performed using waypoint tracking, path
following, and trajectory tracking [15]. Contrary to waypoint tracking and path following,
in the trajectory tracking problem, the vehicle must follow a time-parametrized path,
meaning it must reach a certain point in a given time [16]. This attribute makes trajectory
tracking suitable for collaborative AUVs missions, particularly when the communication is
limited or null. Nevertheless, the integration of time and space restrictions makes it the
most complex autonomous navigation method. An advanced control strategy must be
designed to overcome that complexity. Most of the controllers found in the literature have
either asymptotic or exponential error convergence; however, a finite-time convergence
controller is highly desirable due to the time-parametrization of the AUVs trajectories. An
1
advanced control algorithm that allows the user to select the convergence time of the
tracking errors will have some advantages for the vehicle's navigation, such as higher
robustness, faster response, and higher accuracy, among others [17]. It will also make
the design and implementation of a collaborative mission easier by allowing coordination
of the movement of the vehicles.
Once the vehicle reaches its reference, it should maintain its position and
orientation during manipulation. It is known as the station-keeping problem [18], and it is
challenging to overcome unknown external disturbances caused by the underwater
environment or the manipulation task itself. Station-keeping is critical to the success of a
mission as it allows the operator or autonomous manipulator to have better control of the
operation and reduces the risks of collisions that could damage either the vehicle, the
manipulator, or the object.
Researchers usually consider a control algorithm for navigation and another for the
station-keeping problem. However, additional advantages, such as reducing the
algorithm complexity, computational costs, and the latency of the controller, will be
brought off if the same controller could manage the trajectory tracking station-keeping
problem.
Considering all this, the challenge in this research is designing a collaborative
manipulation task for two AUVs, focusing on developing an advanced control algorithm
to solve the trajectory tracking and station-keeping problems that could be evaluated in
actual underwater vehicles.
1.3 Research question
This research aims to answer the question:
Is it possible to set up a team of underwater vehicles to autonomously approach,

grab, transport, and release an object in an underwater environment?
1.3.1 Hypothesis
Based on the state-of-the-art, the following hypotheses are proposed:
• A single controller can manage AUVs' trajectory tracking and station-
keeping problems.
• A finite-time controller with a user-defined convergence time can improve
the vehicle’s accuracy, robustness, and energy consumption performance.
• A coordinated autonomous navigation scheme can manage a collaborative
manipulation mission without vehicle-to-vehicle communication.
• A finite-time controller with a user-defined convergence time can coordinate
multi-vehicle navigation in a collaborative manipulation mission.
1.4 Solution overview
1.4.1 Objectives
The research presented in this work focuses on designing and developing a
collaborative manipulation mission for autonomous underwater vehicles. To fulfill such an
aim, the main objective of the research is:
2
• Design and implement a two-vehicle collaborative manipulation mission to
autonomously approach, grab, transport, and release an object.
The following specific objectives are considered to support the main objective:
• Explore the state-of-the-art regarding localization, navigation,
communication, and collaboration of underwater vehicles.
• Implement an advanced control algorithm for trajectory tracking and station-
keeping in AUVs.
• Evaluate the control algorithm designed by performing experiments in
actual underwater vehicles.
• Design and implement a collaborative manipulation mission considering the
localization, navigation, and communication restrictions.
• Evaluate the collaborative manipulation mission by numerical simulations
and comparations with state-of-the-art finite-time controllers.
This research does not address the manipulator design or object identification and
grabbing problems, which are considered solved.
1.4.2 Proposal
A collaborative manipulation task between two AUVs is proposed in this work. The
proposed scheme for the mission does not consider vehicle-to-vehicle communication.
Instead, it uses a model-free second-order sliding mode control with finite-time
convergence to coordinate the two vehicles in the task. The designed task is to approach
an object, grab it, move it to a desired final position for its release, and return the AUVs
to a defined home position. The proposed controller converges to a practical zero error in
a user-defined time, guaranteeing that the vehicles follow a time-parameterized trajectory.
This proposal is evaluated and validated by numerical simulations and experiments in a
semi-Olympic swimming pool. The performance of the proposed controller is also
compared with traditional controllers, non-traditional asymptotic controllers, and two
state-of-the-art finite-time controllers.
1.4.3 Methodology
The scientific method is proposed beside the following methodology:
• State-of-the-art research: An extensive literature review regarding

underwater localization and communication systems, control algorithms for
AUV navigation, Station-keeping, and AUV collaboration is necessary to get
familiar with the problem of autonomous collaborative manipulation.
• Vehicle hydrodynamical modeling: The proposal considers a model-free
control approach; however, assembling the BlueROV2 hydrodynamical
model is required for simulation.
• Numerical simulator: A numerical simulator for the BlueROV2 underwater
vehicle will be programmed in a MATLAB/Simulink environment.
• Control strategy: An advanced controller for AUV navigation and station-
keeping will be integrated, considering a model-free second-order sliding
mode controller with finite-time convergence. Its main feature will be that
the user can arbitrarily set the convergence time.
• Trajectory tracking validation: The controller will be evaluated through
numerical simulations and experiments in a semi-Olympic swimming pool
3
considering a single BlueROV2 underwater vehicle. The controller
performance will be compared with traditional and non-traditional
controllers.
• Station-keeping validation: The controller will be evaluated through
numerical simulations and experiments in a semi-Olympic swimming pool
considering a single BlueROV2 underwater vehicle. The controller
performance will be compared with traditional and non-traditional
controllers.
• Collaborative manipulation: A collaborative manipulation mission will be
designed for two BlueROV2 vehicles. Vehicle-to-vehicle communication is
not considered; thus, exploiting its arbitrarily user-defined convergence time
feature, the control strategy developed is used to coordinate the navigation
of the vehicles during the mission.
• Evaluation: the collaborative manipulation mission will be evaluated by
numerical simulations and compared with two state-of-the-art finite-time
controllers.
1.5 Main contributions
This work expects to contribute to underwater robotics research, particularly to

advanced control strategies, collaborative AUV intervention missions, and coordinated
navigation. The expected contributions are:
• A literature review on autonomous underwater vehicles localization, navigation,
communication, and collaborative missions.
• A finite-time model-free second-order SMC for trajectory tracking and station-
keeping of AUVs. Its main feature is that the user can arbitrarily set the controller’s
convergence time.
• A numerical simulator for the BlueROV2 underwater vehicle.
• Experimental validation of control algorithms for AUV navigation.
• Experimental validation of control algorithms for AUV station-keeping.
• A collaborative manipulation task design where no vehicle-to-vehicle
communications are needed.
Previous contributions are also expected to be published as research papers in

SCOPUS-indexed Journals.
1.6 Thesis structure
The rest of the document is structured as follows:

• Chapter 2: Literature review
• Chapter 3: Unmanned Underwater Vehicles modeling
4
• Chapter 4: Controller design
• Chapter 5: Controller validation for trajectory tracking
• Chapter 6: Controller validation for station-keeping
• Chapter 7: Coordinated manipulation task
• General conclusions
Chapter 2: Literature review

The problems of localization, communication, and navigation must be addressed
before establishing a collaborative scheme for AUVs. An extensive literature review was
made as this research's first step. This section discusses the different approaches for
AUVs localization, communication, autonomous navigation, and collaborative schemes.
The main results of this Chapter were published as a review paper [10]:
J. González-García, A. Gómez-Espinosa, E. Cuan-Urquizo, L. G. García-

Valdovinos, T. Salgado-Jiménez, and J. A. Escobedo Cabello, “Autonomous
underwater vehicles: Localization, navigation, and communication for
collaborative missions,” Appl. Sci., vol. 10, no. 4, 2020, doi:
10.3390/app10041256
2.1 Introduction
Navigation and localization are two of the most critical challenges for underwater
robotics [19]. They continue to be problems to solve for many applications, such as
collaborative missions. Traditional methods for underwater localization of AUVs include
Dead-Reckoning (DR) and Inertial Navigation Systems (INS) [20]. DR and INS are some
of the earliest established methods to locate an AUV [21]. These systems rely on
measurements of the water speed and the vehicle’s velocities and accelerations that,
upon integration, lead to the AUV position. They are suitable for long-range missions and
have the advantage of being passive methods —they do not need to send or receive
signals from external systems— resulting in a solution immune to interferences. The main
problem with these methods is the accuracy drift.
In the accuracy drift, the position error grows over time due to different factors,
such as the ocean currents and the sensors' accuracy, which cannot sense the
displacements produced by external forces or the effects of the Earth’s gravity. Using
geophysical maps to match the sensors' measurements is an alternative to dealing with
the accuracy drifts of the inertial systems. This method, known as Geophysical Navigation
(GN) [22], allows for accomplishing longer missions, maintaining a relatively low position
error. However, there is a need to have the geophysical maps available before the
mission, which is one of the main disadvantages of GN. The high computational cost of
comparing and matching the map with the sensor data must also be considered.
Acoustic ranging systems have been another common alternative for AUV
navigation [23]. These systems can be implemented using acoustic transponders to
locate an AUV in either global or relative coordinates. However, most of them require
complex infrastructure, and the cost of such deployments could be higher than other
methods. Researchers have recently explored new alternatives for AUV localization and
navigation.
5
Optical technologies have become very popular for robots and vehicles in land or
air environments [19] but face harsh conditions in underwater environments that have
delayed the development of such technologies for AUVs. When the underwater conditions
permit proper light propagation and detection, visual-based systems can improve the
accuracy of the position estimations and reach higher data rates than acoustic systems.
Finally, recent advancements in sensor fusion schemes and algorithms contribute
to developing hybrid navigation systems, taking advantage of different solutions to
overcome their weaknesses. A sensor fusion module improves the AUV state estimation
by processing and merging the available sensor data [24]. Some common sensors are
INS, Doppler Velocity Loggers (DVL), and depth sensors. Recently, the INS
measurements are also being integrated with acoustic/vision-based systems to produce
a solution that, beyond reducing the accuracy drifts of the INS, will have high positioning
accuracy in short-ranges. All these technologies are summarized in Figure 1, including
the main sensors used and the approaches taken in each case.
Figure 1. Autonomous Underwater Vehicle (AUV) technologies for localization and navigation.
Other challenges must be addressed besides the problem of self-localization and

navigation to implement a collaborative team of AUVs. Since there is a need for sharing
information between vehicles, communication is a significant concern. The amount and
size of the messages will depend on the collaborative scheme used, the number of
vehicles, and the communication system capabilities. Acoustic-based communication
performs better than light-based communication in range but not in data rates. It also
suffers from small bandwidth, high latency, and unreliability [25]. Radio-based
communication has had few practical underwater applications despite its notable merits
in the terrestrial wireless network field [26].
The collaborative navigation scheme is also a mandatory issue to be considered.
The underwater environment is complex for single-vehicle navigation, and now multiple
vehicles are requested to navigate among each other. A proper formation has to ensure
safe navigation for every single vehicle. Since there is no need to interact with the
6
environment, survey missions are simpler to implement and have been performed
successfully for different applications, such as mapping or object searching and tracking.
Intervention missions are usually more challenging than survey missions due to the
manipulators' or actuators' complexity. In either case, since an experimental setup is
difficult to reach, many efforts are being tested in simulation environments. All these
topics are analyzed in the following sections.
2.2 Underwater vehicle localization
This section summarizes the underwater localization methods found in the

literature, their different approaches, how they estimate the vehicle’s position, their
advantages and shortcomings, and the results obtained in simulated and real
environments. Finally, an analysis of all this information is presented.
2.2.1 Dead-reckoning and inertial navigation
The simplest method to obtain a position for a moving vehicle is by integrating its
velocity in time. This method, known as DR [21], requires knowing the vehicle's velocity
and direction, usually accomplished with a compass and a water speed sensor. The main
problem in DR is related to the presence of an ocean current, as illustrated in Figure 2,
because it will add a velocity component to the vehicle that the speed sensor does not
detect. Then, the method's accuracy will be strongly affected, especially when the vehicle
navigates at a low velocity.
Figure 2. Dead-Reckoning drift effect.
Inertial sensors can improve the navigation accuracy and reliability of DR methods.
The INS consists of three mutually-orthogonal accelerometers aligned to a gyroscopic
reference frame. The measured accelerations are integrated to obtain the vehicle' velocity
and position. The fact that inertial navigation is self-contained—it neither emits nor
receives any external signal—is one of its most significant strengths, making it a stealthy
navigation solution, immune to interference or jamming [27]. However, the error on the
pose estimations is known to increase over time and depends on the sensors' accuracy.
7
Mathematically, the total acceleration is denoted as 𝑟̈ , and can be expressed as follows
[28]
𝑟̈ = 𝑎 + 𝑔, (1)
where 𝑎 is the acceleration calculated by the INS and 𝑔 is the gravitational acceleration.
Since the accelerometers do not sense gravity, the vehicle's position obtained by
integrating the acceleration measurements will result in an error. Gyroscopic drifts are
also a source of error that can result in significant misalignments between the sensor
frame and the Earth-fixed reference frame, causing navigation errors that grow over time.
A Global Positioning System (GPS) is a common method to correct these errors.
However, to correct the error accumulated by the INS, the vehicle must go to the surface
to obtain a new GPS location at regular intervals, which can waste time and resources.
Integration of INS and GPS data can also be complex since those systems are based on
entirely different principles.
Even if the instruments were perfect, the estimations of an INS would result in an
error [22]. The gyroscopic reference frame is aligned to an ellipsoid Earth reference
model. The reference ellipsoid conforms roughly to the shape of the earth, and in
particular, to the mean sea level. If the mass of the Earth were homogeneously distributed
within the ellipsoid, the gravity vector would be normal to the reference ellipsoid surface.
However, due to the Earth's mass's inhomogeneous distribution, the gravity vector can
have significant components tangential to the reference ellipsoid surface (known as
vertical deflections), as shown in Figure 3. Since an INS cannot distinguish between the
tangential components of earth gravity and the horizontal acceleration of the vehicle,
these gravity disturbances cause errors in the INS velocity and position estimations.
Figure 3. Vertical deflection.
The latest advances in Micro Electro Mechanical Systems (MEMS) inertial sensors
profoundly affect the current availability of Inertial Measurement Units (IMUs). IMUs have
become attractive for many applications where size, weight, power, and cost are key
considerations [29]. This set of sensors can be used to implement an Attitude and
Heading Reference System (AHRS) or an INS. Some MEMS-based systems
commercially available are shown in Table 1. Nevertheless, despite technological
8
developments in inertial sensors, INS are underway to reduce the accuracy drift at a few
meters of uncertainty over one hour of unaided inertial navigation [30]. Damping
techniques, using water speed measurements, are used to control velocity and position
errors caused by uncorrected vertical deflection and inertial sensor errors [31]. However,
this is at the cost of introducing an additional error source (the water-speed/ground-speed
difference caused by ocean currents). Another alternative to reduce these effects is using
maps of vertical deflection compensation values as a function of latitude and longitude to
compensate for the measured accelerations.
Table 1. Commercial Inertial Measurement Unit (IMU)-Attitude and Heading Reference System
(AHRS) systems.
Heading Pitch and Roll Data

Manufacturer Product Name Accuracy/ Accuracy/ Rate Depth Rate (m)
Resolution Resolution (Hz)
Impact Subsea ISM3D [32] ±0.5°/0.1° ±0.07°/0.01° 250 1000–6000
Seascape
Seascape Subsea ±[32]0.5°/0.01° ±0.5°/0.01° 400 750
UW9XIMU-01 [33]
Inertial Labs AHRS-10P [34] ±0.6°/0.01° ±0.08°/0.01° 200 600
SBG Systems Ellipse2-N [35] ±1.0°/- ±0.1°/- 200 -
TMI-Orion DSPRH [36] ±0.5°/0.1° ±0.5°/0.1° 100 500–2000
VectorNav VN-100 [37] ±2.0°/0.05° ±1.0°/0.05° 400 -
XSENS MTi-600 [38] ±1.0°/- ±0.2°/- 400 -
2.2.2 Acoustic navigation
Acoustic-based signals propagate better underwater than others, such as radio
and electromagnetic, and can reach considerable distances allowing AUV navigation
using acoustic transponders. Some navigation methods based on acoustic signals are
Sound Navigation and Ranging (SONAR) and acoustic ranging.
SONAR
There exist different methods to employ a SONAR for AUV navigation. Two basic
configurations are the side-scan SONAR [23] and the Forward-Looking SONAR (FLS)
[39]. Both are used to detect objects which can be seabed changes, rocks, other vehicles,
and even marine species. When an AUV is in operation, it must be able to detect these
objects to update its navigation trajectory and avoid collisions, known as obstacle
avoidance.
For the side-scan SONAR, the transducer device laterally scans when attached to
the AUV, as illustrated in Figure 4. A series of acoustic pings are transmitted and received
back. The ping travel time and the speed of sound in water are used to determine the
existence of features located perpendicular to the direction of motion.
9
Figure 4. AUV equipped with two side-scan Sound Navigation and Ranging (SONARs).
The FLS uses a searchlight approach, steering the sonar beam scanning forward
of the vessel and streaming soundings continuously. FLS can be placed at different
locations on the vehicle, as shown in Figure 5, to ensure that the AUV can detect
obstacles from different directions.
Figure 5. Forward-Looking SONAR (FLS) placed at vertical and horizontal orientations.
Two-dimensional images can be produced to survey the ocean and its features.
While indicating what exists on the sea or seafloor, these images do not contain relative
or global localization information.
Traditional obstacle avoidance planning methods include potential field, Bandler
and Kohout (BK) products, particle swarm optimization, fuzzy controller, etc. Galarza et
al. [40] designed an obstacle avoidance algorithm for an AUV. The obstacle detection
system disposes of a SONAR, and its use guarantees the safety of the AUV while
navigating. Obstacle avoidance is performed based on a fuzzy reactive architecture for
different forward speeds of the vehicle. The algorithm was validated under a
computational simulation environment running in MATLAB. During the simulated route,
the vehicle remained at a minimum distance of 5 m from the obstacles, reducing its
forward reference speed of 1 m/s to values between 0.02 m/s and 0.4 m/s; thus, safe
navigation around obstacles was achieved without losing the trajectory of navigation and
reaching all the waypoints. Braginsky et al. [41] proposed an obstacle avoidance
methodology based on data collected from two FLS placed in horizontal and vertical
orientations. FLS data is processed to provide obstacle detection information in the 𝑥𝑧
and 𝑥𝑦 planes, respectively. For horizontal obstacle avoidance, the authors used a two-
10
layer algorithm. The first process of the algorithm is based on BK products of fuzzy
relation as a preplanning method; the second process is a reactive approach based on
potential field and edge detection methods. If the horizontal approach fails to find a path
to safely avoid the obstacle, a reactive vertical approach is activated. The SONAR used
in experimentation has a detection range of up to 137 m and is operated at 450 kHz.
During the test, the mission definition for the AUV was to move from a starting point to a
target point. Despite the maximum range of the FLS, decisions were made when an
obstacle was within 40 m of the AUV.
Lin et al. [42] implemented a Recurrent Neural Network (RNN) with Convolution
(CRNN) for underwater obstacle avoidance. Offline training and testing were adopted to
modify the neural network parameters of the AUV autonomous obstacle avoidance
learning system, so self-learning is applied to the collision avoidance planning. Combining
this learning system with FLS simulation data enables online autonomous obstacle
avoidance planning in an unknown environment. Simulation results showed that the
planning success rate was 98% and 99% for the proposed CRNN algorithms; meanwhile,
it was 88% and 96% for the RNN algorithms. The authors concluded that the CRNN
obstacle avoidance planner has the advantages of short training time, simple network
structure, better generalization performance, and reliability than an RNN planner.
Acoustic ranging
In acoustic ranging positioning systems, AUVs are equipped with an acoustic
transmitter that establishes communication with a set of hydrophones. The distance
between the AUV and the hydrophones can be calculated through the propagation time
of the acoustic signal by knowing the propagation velocity of sound underwater. Then, a
location for the AUV concerning the set of hydrophones can be obtained by geometric
methods. One of the differences between acoustic systems is the arrangement of the
hydrophones. In Long BaseLine (LBL) systems, hydrophones are fixed within a structure
or any other known underwater point of reference (landmark) [43]. The length of the
baseline can be up to hundreds of meters. In Short BaseLine (SBL) and Ultra Short
BaseLine (USBL) systems, hydrophones are placed on buoys, vehicles at the surface, or
even on a second AUV. SBL systems measure a relative position between the reference
sound source and the receiving array. In USBL, the relative location from the hydrophone
to the moving target is calculated by measuring the phase differences between acoustic
elements [44]. The baseline length is measured in meters for SBL and decimeters for
USBL systems. In either arrangement, hydrophones are generally located by GPS. In
Figure 6, all three configurations for acoustic localization systems are illustrated.
(a) (b) (c)
11
Figure 6. Acoustic localization systems: (a) Long Baseline, (b) Short Baseline, (c) Ultra-Short Base Line.
A schematic diagram of an SBL system is represented in Figure 7. Three

hydrophones, represented by 𝐻1, 𝐻2, and 𝐻3, are located at the points 𝑂, 𝑁, and 𝑀, at
the origin of the reference frame and along the 𝑥 and 𝑦 axes, respectively. The distance
from the detected vehicle to the 𝑖 hydrophone is called oblique distance, which is denoted
by 𝐷𝑖 , with 𝑖 = 1, 2, 3.
Figure 7. The schematic diagram for Short Baseline (SBL) system.
The vehicle receives a signal from a hydrophone (𝐻1) and sends a reply received
by all the hydrophones (𝐻1, 𝐻2, 𝑎𝑛𝑑 𝐻3); then, the signal run time is measured. The
propagation time of the acoustic signal from the transmitter in the vehicle to the
hydrophone base (𝑇𝑖 ) is used to obtain the oblique distance with the equation [44]:
𝐷𝑖 = 𝑉 · 𝑇𝑖 , (2)
where 𝑉 is the nominal speed for underwater acoustic signals and is used as 𝑉 =
1435 𝑚/𝑠. The location of the vehicle’s transmitter (𝑃) with coordinates 𝑋𝑝 , 𝑌𝑝 , and 𝑍𝑝 ,
can be calculated using a traditional SBL model as follows
𝑃 = (𝑋𝑝 , 𝑌𝑝 , 𝑍𝑝 ), (3
)
𝐷12 − 𝐷22 + 𝑁 2 (4
𝑋𝑝 = ,
4𝑁 )
𝐷12 − 𝐷32 + 𝑀2 (5
𝑌𝑝 = ,
4𝑀 )
𝑍𝑃
1⁄ 1⁄ 1⁄
2 2 2 2
{[𝐷12 − 𝑋𝑝2 + 𝑌𝑝2 ] 2
+ [𝐷22 − (𝑋𝑝 − 𝑁) + 𝑌𝑝2 ] − [𝐷32 − 𝑋𝑝2 + (𝑌𝑝 − 𝑀) ] } (6
)
= .
3
An error exists between the measured position and the actual position of the
transmitter. Among other factors, it is caused by not considering the sound velocity
12
variations produced by changes in the underwater environment conditions such as depth,
temperature, density, and salinity. The accuracy of an acoustic positioning system will
depend on different factors, such as the distance and operational depth range, the
number and availability of hydrophones, and the operational frequency. A few commercial
baseline acoustic systems and accuracy specifications are shown in Table 2.
Table 2. Commercial acoustic positioning systems.
Operating Depth Range

Name Type Accuracy Range (m)
(m)
EvoLogics S2C R LBL [45] LBL Up to 0.15 200–6000
GeoTag seabed positioning system
LBL Up to 0.20 500
[46]
µPAP acoustic positioning [47] USBL Not specified 4000
SUBSONUS [48] USBL 0.1–5 1000
1% of distance range (1 m for a 100 m
UNDERWATER GPS [49] SBL/USBL 100
operating range)
Although acoustic systems have been used in the past, they are still used as the
primary localization and navigation system for AUVs or teams of AUVs and Unmanned
Surface Vehicles (USVs). Batista et al. [50] worked on a filter for combining LBL and
USBL systems to estimate underwater vehicle position, linear velocity, and attitude. This
filter considers an underwater vehicle moving in a scenario where a set of fixed landmarks
is installed in an LBL configuration. The vehicle is equipped with a USBL acoustic
positioning system. The filter performs well even in the presence of sensor noise in a
simulated environment. The resulting solution ensures a quick convergence of the
estimation error to zero for all initial conditions. However, it could not be a practical
solution for some cases since it requires a complex infrastructure.
Coordinated navigation of surface and underwater vehicles is proposed by
Vasilijević et al. [51]. The proposed scheme has the purpose of serving as a first-
responder monitoring team on environmental disasters in oceans. The USV is connected
to a ground station via Wi-Fi for control and monitoring; meanwhile, acoustic
communication sends instructions to the AUV and retrieves information from it. A GPS is
mounted on the USV to locate the vehicles obtaining a position on geographic
coordinates. Once the USV gets a location, a USBL system is used to get a relative
location of the AUV regarding the surface vehicle. An algorithm is run to convert them to
a global position so the control station can know where both vehicles are. This allows the
precise localization of pollution or any other problems found by the vehicles and is
intended to help to plan a rapid response. Limitations on the USBL system are not a
problem in this scenario as long as the USV and AUV remain in close range for
communication. Sarda et al. [52] used a digital USBL system for AUV recovery. The AUV
was equipped with a receiver array of four transducers, and a transponder array was
mounted on a USV, serving as a recovery station. The system proposed is not only
capable of estimating the distance between the AUV and the recovery location, but it is
also able to measure horizontal and vertical bearings. The system has an update period
of 3 s and has an accuracy of less than a meter. Its main limitation is the sensing range,
AUV must be 25 m within the target localization, or the system measurements are
13
considered erroneous. Field experiments showed a success rate of 37.5% in recovering
the AUV.
Range-only—also known as Single-beacon—localization is another alternative to
traditional acoustic localization systems that has gained attention in recent years. The
concept of range-only/single-beacon positioning can be divided into two groups
depending on the way they are used [53]: as a navigational aid for a moving vehicle or as
localization of a stationary or moving target. All these methods use a set of ranges
between a target and different static nodes, known as anchor nodes. Typically, these
ranges can be obtained using the time of flight given the speed of sound in water. Then,
the unknown underwater target position problem can be solved using trilateration, where
in general, three or more points are needed in 2D dimensions and at least four points in
3D scenarios.
A method for target positioning from a moving vehicle—which periodically
measures the range to the underwater target—is represented in Figure 8.
Figure 8. Range-only/Single-beacon positioning of a fixed target from a moving vehicle.
The underwater target position 𝑃𝑡 is calculated using the moving vehicle positions
𝑃𝑖 and the ranges measured between the moving vehicle and the target 𝑟̅𝑖 expressed as
𝑟̅𝑖 = ‖𝑃𝑡 − 𝑃𝑖 ‖ + 𝑤𝑖 , (7)
where 𝑤 i is a zero mean Gaussian measurement error. Different methods are used to
solve the system and find the target position through ranges: linearization of the function
to find a closed-form least squares solution, and an iterative minimization algorithm
minimizing a cost function related to the maximum likelihood estimate.
Bayat et al. [54] presented an AUV localization system that relied on the
computation of the ranges between the vehicle and one or more underwater beacons, the
location of which may be unknown. The system aimed to compute an AUV position
estimate in real-time and construct a map composed of the beacons locations
estimations. Experiments were performed with three autonomous marine vehicles
following three different trajectories. Minimum-energy estimation, projection filters, and
multiple-model estimation techniques were used as observers to compare the results. A
combination of those estimators produced the best results regarding the error in the
trajectory followed by the AUV, reduced from tens of meters to some meters in the test's
first three minutes. Villacrosa et al. [55] presented a solution to range-only localization
using a Sum of Gaussian (SoG) filter. Two variations of the SoG filter were proposed and
tested in real experiments, where an AUV performed an autonomous localization and
homing maneuver. The results in all experiments showed that the AUV was able to home
14
with an error smaller than four meters. A vision-based algorithm corroborated the results.
Masmitja et al. [53] developed a range-only underwater target localization system where
a wave glider performed as a moving LBL in simulations and real sea tests. The study
aimed to determine the best path and its characteristics, such as the number of points,
radius, and offset, to obtain the desired target localization performance. Results showed
that with a minimum number of 12 points, a radius greater than 400 m, and an offset as
low as possible, the Root-Mean-Square Error (RMSE) could be minor than 4 m.
Zhang et al. [56] presented a new method to solve problems of LBL systems, such
as communication synchronization among hydrophones. The system considers a
Strapdown Inertial Navigation System (SINS) and the formation of a matrix of several
virtual hydrophones. A single sound source is placed at the bottom of the sea and sends
periodic signals; meanwhile, a single hydrophone is installed on the AUV. In the AUV
navigation trajectory, four selected recent positions of the AUV are regarded as four
virtual hydrophones of the LBL matrix, constituting a virtual LBL matrix window.
Simulation results indicate that the proposed method can effectively compensate for the
position error of SINS. Thus, the positioning accuracy can be confined to 2 m.
2.2.3 Geophysical navigation
To avoid the problem of INS drifts and the cost of infrastructure for underwater
acoustic systems, GN is a favorable alternative. These approaches match the sensors'
measurements with geophysical parameters such as bathymetry, magnetic field, and
gravitational anomaly contained in a map. Navigation technology based on GN can
correct the INS error over time [57] without bringing the AUV to the surface. The
navigation algorithm estimates navigation errors and sends them to the vehicle navigation
system to correct its position. This method allows the vehicle to maintain required position
accuracy by providing continuous corrections without needing external sensors, such as
GPS. The main limitations of GN are the need for a map available before the mission and
the computational complexity of searching for a map correlation with sensor estimations.
Conversely, the key advantage of GN over other technologies is the extensive operating
range when in use. Given a map, GN provides bounded localization error with accuracies
dependent on the DR navigation, the map resolution, and the sensitivity of the
geophysical parameter to change vehicle state [55].
GN matching algorithms are classified into two different broad categories: batch
methods and sequential methods [27]. The main algorithms for those methods have been
TERCOM and Iterated Closest Contour Point (ICCP) [58] for batch methods; SITAN,
Beijing university of aeronautics and astronautics Inertial Terrain-Aided Navigation
(BITAN), and BITAN-II [31], [59], [60] for sequential methods. TERCOM correlates active
range sensor observations with a digitized elevation database of terrain. Meanwhile, the
essence of SITAN is the acquisition mode and tracking mode, which are state-estimation
problem based on an Extended Kalman Filter (EKF) after the non-linear system state and
observed equations are linearized. Particle Filters (PF) and Bayesian estimators are also
algorithms used in sequential methods.
Gravity navigation
For an INS, the effects of a change in the local gravitational field are
indistinguishable from the vehicle accelerations. This is problematic since the Earth’s
gravitational field is far from uniform. One alternative is complementing the INS with
gravity navigation. A gravity sensor—gravimeter or gradiometer—measures gravity and
15
its gradient at the AUV location, while an INS estimates the vehicle’s position. A
gravimeter measures gravity anomaly or the deviation in the magnitude of the gravity
vector relative to a nominal earth model. A gradiometer is a pair of accelerometers with
parallel input axes on a fixed baseline that measures gravity gradients or the rate of
change of gravity regarding linear displacement [30]. The difference in the
accelerometer’s output excludes the linear vehicle acceleration but contains the
gravitation gradient across the baseline. Based on the sensor's position and
measurements, the database searches for the best gravity and gradient fit. Then the
optimal matching position will be used to correct the position error of the INS. Han et al.
[61] proposed a matching algorithm for gravity-aided navigation, combining an ICCP
algorithm with a Point Mass Filter (PMF) algorithm. The algorithm involved a two-step
matching process. First, given a large initial position error, the PMF based on the vehicle
position variable can obtain an instructional position in real-time. Then, the ICCP
algorithm can be employed for further matching. A numerical simulation was performed
with a 12 h sailing period to verify the validation of the proposed matching algorithm. The
speed of the underwater vehicle was set to 10 nmi/h. Simulation tests indicated that
compared with the conventional ICCP algorithm, the proposed algorithm could achieve
better results regarding latitude and longitude positioning errors, which were reduced up
to 56% and up to 65% compared to the INS standalone.
Geomagnetic Navigation
Geomagnetic Navigation relies on magnetic sensors, and its essence is the Fitting
of Two Point Sets (FTPS) process, where a marine geomagnetic map is used for
matching [27]. Many features in a geomagnetic field can be matched [62], such as the
intensity of the total field 𝐹, the horizontal component 𝐻, the north component 𝑋, the east
component 𝑌, the vertical component 𝑍, the declination 𝐷, the inclination 𝐼, the
geomagnetic gradient, and more. These features are illustrated in Figure 9.
Figure 9. Geomagnetic map features.
Zhao et al. [63] studied two matching algorithms, TERCOM and ICCP, used in
geomagnetic matching navigation. An experiment was designed to test the accuracy of
the underwater navigation system, using a Differential GPS (DGPS) receiver to provide
the vehicle's exact position. In the results, matching positioning errors in the 𝑥 or 𝑦
directions were less than 100 m. Authors conclude that both TERCOM and ICCP can
achieve credible geomagnetic navigation, with the difference that ICCP can provide a
16
real-time positioning solution and TERCOM cannot. Ren et al. [20] presented a new
algorithm to solve FTPS in geomagnetic localization. The algorithm was an improved
version of the ICCP algorithm based on Meng’s algorithm. Simulation results showed that
the ICCP-Meng algorithm performed better than the original ICCP algorithm in dealing
with geomagnetic-matching localization. Wang et al. [64] presented a new method based
on integrating TERCOM, a K-means clustering algorithm and an INS. An experiment was
implemented to evaluate the accuracy and stability of the method proposed. INS and
DGPS were set on the surveying vessel. The positioning result from a DGPS was used
to verify the accuracy of this new method. After completing the experiment, the error of
the new method was minor than 50 m; meanwhile, the traditional method showed a seven
times higher error.
Bathymetric navigation
Isobaths are a straightforward use of bathymetric maps for AUV. An isobath is an
imaginary curve connecting all points with the same depth below the surface. A controller
[65] can be designed for an AUV to follow an isobath whit only low-level localization
equipment—such as an echo sounder—and ensures that it never leaves a pre-defined
area. Terrain-Referenced Navigation (TRN), Terrain-Aided Navigation (TAN), and
Terrain-Based Navigation (TBN) are all similar approaches for GN [66]. These systems
estimate the errors in a primary navigation system—such as an INS—and the terrain
database to provide a highly accurate position estimate relative to the digital terrain
database. TBN operates by correlating the terrain profile overflow with the terrain
information stored in the terrain database. A basic measurement equation [67] for TBN is
given by
𝑦 = ℎ(𝑥) + 𝑒, (8)
where ℎ(·) is the terrain elevation function, 𝑥 is the vehicle location, 𝑦 is the measured
terrain height, and 𝑒 is the measurement noise. An example of terrain correlation in one
dimension for a single altimeter measurement is represented in Figure 10.
Figure 10. Terrain-Referenced Navigation.
Zhao et al. [68] worked on a TAN algorithm that combined TERCOM and PF.
Experiments were performed to compare the proposed algorithm with the BITAN II
algorithm. Results showed that the North and East position error remained below 100 m
for the new algorithm, and the mean error was less than half compared with the BITAN-II
17
algorithm. Based on those results, the authors concluded that their system was more
reliable, with higher positioning precision and better stability than the one used for
comparison.
Salavasidis et al. [69] proposed a low-complexity PF-based TAN algorithm for a
long-range, long-endurance, deep-rated AUV. The algorithm's potential was investigated
by testing its performance using field data from three deep (up to 3700 m) and long-range
(up to 195 km in 77 h) missions performed in the Southern Ocean. The authors compared
TAN results to position estimates through DR and USBL measurements. Results showed
that TAN holds the potential to extend underwater missions to hundreds of kilometers
without the need for surfacing to re-initialize the estimation process. The RMSE of the
TAN algorithm was up to 7 times lower when compared with the DR measurements, and
the absolute water-depth difference was reduced up to 66% when compared with USBL
measurements. Meduna et al. [70] proposed a TRN system for vehicles with low-grade
navigation sensors to improve the navigation capabilities of simple DR systems. The
algorithm uses an 8-dimensional particle filter for estimating critical motion sensor errors
observed in the vehicle. Field trials were performed on an AUV with DR navigational
accuracy of 5%–25% of Distance Traveled (DT). The ability of TRN to provide 5–10 m
navigational precision and an online return-to-site capability was demonstrated in those
field trials.
2.2.4 Optical navigation
Optical technologies are a suitable option for providing information about the
environment. Optical systems can be implemented with a camera or an array of optical
sensors. Despite the poor transmission of light through water, which results in a limited
range for imaging systems [71], different algorithms and techniques are being studied. In
Figure 11, two examples of optical systems are shown; where the AUV must detect and
follow active landmarks within a structure (a) or identify a pattern made with active marks
to navigate through it (b).
(a) (b)
Figure 11. Optical localization systems based on active landmarks. (a) AUV following an array of active
markers, (b) AUV locating an entrance by an arrangement of active markers.
An optical detector array sensor system was presented for AUV navigation by Eren
et al. [72]. The performance of the developed optical detector array was evaluated for its
capability to estimate the position, orientation, and forward velocity of AUVs regarding a
light source fixed underwater. Numerical simulations based on a spectral angle mapper
18
algorithm showed that a hemispherical frame design with a 5 × 5 photo-detector array
was sufficient to generate the desired position and orientation feedback to the AUV. The
detection accuracy was 0.2 m in translation (surge, sway, and heave) and 10° in
orientation (pitch and yaw). Some optical or artificial vision systems have been applied to
AUVs for different purposes, such as docking and recovery. Zhong et al. [73] developed
an artificial vision system that detects a set of lamps around the desired docking location
for an AUV. The AUV uses a binocular localization method to locate the docking platform
and navigates to reach it. Navigation lamps were mounted at the entrance of the docking
station as active beacons. Three standard underwater green lights were symmetrically
positioned on the docking model around the center of the three lights. A ship model
experiment has been conducted in a laboratory to evaluate the algorithm's feasibility. The
test results demonstrated that the average localization error was approximately 5 cm, and
the average relative location error was around 2% in the 3.6 m range. A similar approach
was proposed by Liu et al. [74]. The authors presented a vision-based framework for
automatically recovering an AUV with another AUV in shallow water. The proposed
framework contains a detection phase for the robust detection of underwater landmarks
mounted on the docking station in shallow water and a pose-estimation phase for
estimating the pose between AUVs and underwater landmarks. During ground
experiments, they observed that the mean position and orientation errors were 1.823°
and 6.306 mm, respectively, in the absence of noise, and 2.770° and 9.818 mm,
respectively, in the presence of strong noise. Field experiments were performed to
recover a sub-AUV by a mother vessel in a lake using the proposed framework and
showed that the algorithm outperformed the state-of-the-art method regarding the
localization error.
Although these systems showed a response with high accuracy, pre-installed
infrastructure is needed to implement them. An alternative approach is using a camera or
set of cameras to identify features in the environment or targets for the AUV mission.
Monroy et al. [75] developed a micro AUV with an artificial vision system that allows it to
follow an object by color. A Hue Saturation Value (HSV) filter was implemented in the
artificial vision system. A non-linear proportional-derivative controller was programmed in
the vehicle to stabilize the heave and surge movements. An intervention AUV addresses
a search and recovery problem in [76]. The problem consisted of finding and recovering
a flight data recorder. The mission was compounded by two stages: survey and
intervention. As the system was tested on a water tank, the survey stage consisted of a
pre-defined trajectory of the AUV. This trajectory guarantees that images taken by the
AUV cameras cover the complete bottom of the tank. Once the survey is done, the flight
data recorder is identified on the images by applying an HSV histogram and then located;
so the intervention stage can occur. Even though these techniques are pretty popular on
land and air robots, working with cameras has several restrictions underwater. Before the
mission, it must know what the robot is looking for; the robot must be pointed to an object
of potential interest, and HSV boundaries must be manually selected until it is well
detected. It also has the inconvenience that colors are not the same underwater as above
water because illumination strongly affects them.
2.2.5 Simultaneous location and mapping
Simultaneous Location and Mapping (SLAM) is a technique that consists of a
mobile robot, such as an AUV, being placed at an unknown location in an unknown
19
environment and making it able to build a consistent map of the environment and
determinate its location within this map [77]. In Figure 12, a SLAM solution is represented
where an AUV is equipped with a sensor to explore the environment and create a digital
reconstruction. Color codes can be used to represent information such as the distance
between the vehicle and obstacles.
(a) (b)
Figure 12. Simultaneous Location and Mapping (SLAM) of (a) an AUV equipped with a sensor to map its
environment and (b) digital reconstruction of the environment.
There are different SLAM representation methods used to reconstruct the

environment. Each one has its shortcomings and advantages. Choosing the best one
depends on the application desired, which can be inspection, navigation, or interaction
with objects. The principal representation methods are listed in Table 3.
Table 3. SLAM representation methods.
Method Type Description Applications

Models the environment as a set of landmarks
Localization and mapping
landmark-based maps 2D/3D extracted from features as points, lines, corners,
[78].
etc.
Discretizes the environment in cells and assigns Exploring and mapping
Occupancy grid maps 2D
each cell's occupancy probability. [79].
Describes the 3-D geometry by a large Obstacle avoidance and
Raw Dense Representations 3D
unstructured set of points or polygons. visualization [80].
Boundary and Spatial-Partitioning Generates representations of boundaries, Obstacle avoidance and
3D
Dense Representations surfaces, and volumes. manipulation [81].
Underwater SLAM can be categorized as acoustic-based and vision-based [39].

The perception of optical devices is constrained by poor visibility and noise produced by
sunlight in shallow waters. Moreover, they can provide high frequencies and high
resolution for a lower cost than an acoustic system. On the other hand, a high-definition
FLS can offer a promising alternative for working under challenging conditions.
In [4], Hernández et al. presented a framework to allow an AUV to explore unknown
environments and create a 3D map simultaneously with an acoustic system. The
framework comprises two main functional pipelines. The first allows the AUV to create an
acoustic map online while planning collision-free paths. The second pipeline builds a
photo-realistic 3D model using the gathered image data. This framework was tested in
20
several sea missions and the results validated its capabilities. Palomer et al. [82] used a
multi-beam echo-sounder to produce high-consistency underwater maps. Since there is
no general method to evaluate the consistency of a map, the authors computed the
consistency-based error [83] and proposed a 3D statistic method named #Cells. The
statistic method counted the number of cells each bathymetric map occupies within the
same 3D grid. If a map occupies fewer cells, it is probably because its point clouds are
more densely packed due to better registration. The algorithm was tested using two real-
world datasets. Three surfaces were created for different navigation methods: DR, USBL,
and the proposed algorithm. Regarding the number of occupied cells, the proposed
method occupied 5.76% fewer cells than the DR model and 7.24% less than the USBL
model.
Gomez-Ojeda et al. [84] implemented a visual-based SLAM algorithm. The authors
compared a stereo Point and Line SLAM (PL-SLAM) with ORB-SLAM, point-only, and
line-only systems. Results showed superior performance of the PL-SLAM approach
relative to ORB-SLAM in accuracy and robustness in most dataset sequences. The mean
translational error was minor for PL-SLAM in 55% of the sequences, and the mean
rotational error was minor in 73% of the cases. Nevertheless, that work was not tested
for underwater applications. After that, Wang et al. [85] proposed a method to improve
the accuracy of vision-based localization systems in feature-poor underwater
environments using the PL-SLAM algorithm for localization. Three experiments were
performed, including walking along the wall of a pool, in a linear and irregular route. The
experimental results showed that the algorithm was highly robust in underwater low-
texture environments due to the inclusion of line segments. At the same time, the
algorithm achieved a high accuracy of location effectively. The attitude error—computed
as shown in Equation 9 —was 0.1489 m, representing the 2.98% DT.
Attitude error = √(𝑒𝑟𝑟𝑜𝑟_𝑥)2 + (𝑒𝑟𝑟𝑜𝑟_𝑦)2 + (𝑒𝑟𝑟𝑜𝑟_𝑧)2 (9)
The authors conclude it can be implemented in AUVs' future navigation and path
planning. Ferrera et al. [86] proposed what they considered the first underwater dataset
to study underwater localization methods from low-cost sensors aiming to explore the
capabilities of visual-based SLAM in real and challenging environments. The dataset has
been recorded in a harbor. It provides several sequences with synchronized
measurements from a monocular camera, a MEMS-IMU, and a Pressure Sensor (PS). A
20 frames per second (fps), 600 × 512 px monochromatic camera and a 200 Hz IMU were
among the sensors used in the dataset acquisition. As a benchmark, the authors ran
experiments using state-of-the-art monocular SLAM algorithms and then compared ORB-
SLAM, Semi-direct Visual Odometry (SVO), and Direct Sparse Odometry (DSO). Results
showed an absolute translation error between 24–52 cm, 24–67 cm, and 2–56 cm for
each applied method, highlighting the potential of vision-based localization methods for
underwater environments. With the same idea, Joshi et al. [87] formed their datasets from
an underwater sensor suite—equipped with a 100 Hz IMU and a 15 fps, 1600 × 1200 px
stereo camera—operated by a diver, an underwater sensor suite mounted on a diver
propulsion vehicle, and an AUV. Experiments were conducted for each dataset
considering the following combinations: monocular; monocular with IMU; stereo; and
stereo with IMU, based on the modes supported by each Visual Odometry (VO) or Visual
Inertial Odometry (VIO) algorithm. Despite often failing to track the complete trajectory,
21
DSO and SVO had the best reconstructions for the tracked parts. As expected, stereo
performed better than monocular. The results confirmed that incorporating IMU
measurements leads to higher performance than the pure VO packages.
2.2.6 Sensor fusion
The main inconvenience of an INS is that the position and orientation accuracy
drifts over time. The system must correct its error by periodically comparing its position
estimation with a fixed location measured from additional sensors—such as a GPS—to
keep it under the limits expected for safe AUV navigation. To overcome this, the INS can
be fused with other sensors. Two main schemes for sensor fusion are Loosely Coupled
(LC) and Tightly Coupled (TC). The basic difference is the data shared by the sensors. In
an LC scheme, a solution for the position or orientation of the AUV is obtained for each
sensor individually and then blended using a filter—such as a Kalman Filter (KF)—to get
a more accurate and reliable solution. In a TC scheme, raw measurements of the sensors
are processed directly on the filter to overcome problems such as poor signal quality or
limited coverage thanks to the filter’s capabilities to predict the vehicle's pose. In this case,
a more robust filter is needed, so variants of the KF are commonly used [88]—such as an
EKF or Unscented Kalman Filter (UKF)—. Filter selection is essential to get a better
solution for the vehicle’s pose, and besides the sensor fusion approach adopted,
accuracy, numerical efficiency, and computational complexity must be considered. LC
and TC schemes are represented in Figure 13, with velocity estimation from an INS and
a DVL as an example.
Figure 13. Loosely Coupled (LC) vs. Tightly Coupled (TC) sensor fusion schemes.
Most of the sensor fusion systems for AUV navigation are those of an INS aided
by a DVL; typically, the fusion is under an LC scheme [89]–[91] with a linear filter.
However, when DVL measurements are limited, an LC algorithm leaves the INS to work
alone. This produces an accumulative error that gets bigger with time.
Liu et al. [92] explored a TC scheme as an alternative to that problem. The authors'
approach includes depth updates given by a depth sensor among raw measurements
from the DVL to help the INS and avoid the drift caused by the limited measurements on
22
the LC approach. Different trajectories were simulated for an AUV, including a straight-
line trajectory for 1,800 s with a constant velocity. In simulations, the INS, DVL, and PS
update frequencies were 200 Hz, 1 Hz, and 1 Hz, respectively. For the 𝑥, 𝑦, and 𝑧 axes,
a gyro drift of 0.01°/h and a 100 μg accelerometer bias were introduced as INS errors;
0.002 m/s as a constant DVL error, and 0.05 m as a constant PS error. The results
showed a cumulative error of 1000 m at the end of the trajectory for the LC approach and
only 10 m in the TC case. The same disadvantages of the LC fusion of an INS/DVL
system were addressed by Tal et al. [93]. The authors designed a navigation system
based on a 150 Hz INS aided with a 1 Hz DVL, a 0.5 Hz magnetometer, and a 0.25 Hz
PS under an Extended Loosely Coupled (ELC) approach within an EKF. They focused
their work on exploring cases where only partial measurements of the DVL were available
and used external information to complete the velocity calculation of the vehicle. Different
trajectories of a vehicle were simulated to test their system. Results showed better
performance by the ELC scheme, improving by up to 38% of the RMSE compared to the
standalone INS and up to 12% with a TC scheme.
Another approach is the fusion of INS with acoustic systems. Zhang et al.
investigated using an AUV positioning method based on a SINS and an LBL under a TC
algorithm [94], [95]. The authors were looking to solve position error accumulation of
AUVs. They compared the TC and LC approaches by simulating an AUV trajectory under
different conditions, such as changing the number of hydrophones available. Test results
demonstrated that the system proposed in this work is more reliable than the LC approach
since the error on the trajectory—mainly when approaching or leaving the hydrophones
array—was up to 50% lower.
Artificial vision is also being fused with INS to improve its performance. Manzanilla
et al. [96] addressed autonomous navigation for AUVs. They used artificial vision fused
with an IMU on an LC algorithm. Parallel Tracking and Mapping (PTAM) was implemented
to localize the vehicle from a visual map, using a single camera—15 fps, 640 × 480 px.
Then, an EKF was used to fuse the visual information with data from an IMU to recover
the map scale and improve the pose estimation. In this work, fully autonomous trajectory
tracking was successfully achieved and compared using standalone PTAM and the
sensor fusion approach. Results showed that the trajectory followed by the vehicle using
sensor fusion has a maximum error below 20 cm while the standalone PTAM drifts up to
60 cm.
The EKF is the most widely used nonlinear filtering approach in TC schemes. EKF
is based on a simple linear approximation to the nonlinear equations. However, there are
too many unknown disturbance factors underwater, and they cannot be established in
suitable mathematical models in the kinematic equation. Other alternatives to a traditional
EKF have been explored. Li et al. [97] proposed a multi-model EKF integrated navigation
algorithm. It was designed to solve the harsh underwater environment problems. Based
on the probabilistic data association theory, this algorithm was compared with standard
EKF in a lake trial using an AUV equipped with an IMU, an AHRS, and an LBL system
with four acoustic beacons. Results showed better performance by the multi-model EKF
since the error between actual positions and estimations was less than 12 m. The
algorithm showed to overcome disturbances that produced peaks of over 400 m on
traditional EKF estimations. Chen et al. [98] worked on another alternative to an EKF for
TC SINS/LBL navigation systems. Instead of applying an EKF, they used a Near-Real-
23
Time (NRT) Bayesian framework. They compared the NRT framework with EKF
approaches with an accurate and poor initialization. Results showed better performance
by the NRT solution with an 80% reduction of the measurement residuals with a poor
accurate yaw error initialization.
The main alternatives for sensor fusion based on an INS are summarized in Figure
14.
Figure 14. Sensor fusion alternatives for AUV positioning.
2.2.7 Localization and navigation overview

General conclusions regarding sensor performance for non-traditional AUV
navigation and localization technologies are shown in Table 4.
Table 4. Technologies for AUV localization and navigation summary.
Navigation Information
Approaches Accuracy Range Results
Technology Available
Depending on the
From 5 m up to
distance from obstacles,
Distance from hundreds of Experimental in real
SONAR from 5–10 cm to more
obstacles. meters from conditions.
than a meter (10–120
obstacles.
cm).
Acoustic Depending on the
distance from the
Acoustic range Up to tens of
hydrophone array and Experimental in real
(LBL, SBL, Position meters from the
the frequency, from conditions.
USBL). array.
some centimeters up to
tens of meters.
24
Gravity, Meters. Depending on Kilometers from Simulation,
Geophysical geomagnetic, Position the map resolution and the initial Experimental under
TAN, TRN, TBN filter applied. position. controlled conditions.
Simulation,
Position and Up to 20 cm for position 1–20 m from
Light sensors. Experimental under
orientation and 10° for orientation. markers.
Optical controlled conditions.
relative to a
Up to 1 cm for position 1–20 m from Experimental in real
Cameras target.
and 3° for orientation. markers. conditions.
Up to tens of
Position and Experimental in real
Acoustic meters from
orientation From some centimeters conditions.
targets.
SLAM relative to the up to more than a
Simulations,
mapped meter. 1–10 m from
Cameras Experimental under
environment. targets.
controlled conditions.
Depending on the
approach and filter
Position, Kilometers from
applied, the Simulations,
Sensor fusion ELC, LC, TC. orientation, and the initial
accumulative error can Experimental.
velocity. position.
be reduced up to some
meters (5–20)
After the literature review, it can be considered that acoustic-based technologies

are still a reliable alternative for AUV localization and navigation; although they require
more infrastructure than others. Future work must consider the possibility of including
them in collaborative AUV teams. Acoustic systems must overcome low updated rates
and limited accuracy (at long ranges) to avoid collisions in AUV formations, especially
when navigating a few meters of each other.
On the other hand, visual-based localization technologies—including SLAM—have
gained attention in recent years. Unlike acoustic methods, these technologies can
estimate both position and orientation. They also reach a higher accuracy which is critical
for the collaborative navigation of AUVs. Thus, it is an exciting and reliable option for
some particular tasks under specific environments. Nevertheless, most of them are on an
early level of readiness since they have been tested only in very controlled environments.
It seems complicated for visual-based systems to overcome challenging underwater
conditions. Moreover, it is hard for researches to find the proper conditions to test their
visual-based and visual-SLAM algorithms in real underwater setups. Some datasets are
being collected to deal with that, such as the AQUALOC dataset [99], which is dedicated
to developing SLAM methods for underwater vehicles navigating near the seabed. The
Autonomous Field Robotics Laboratory (AFRB) [100] has some datasets for the same
purpose.
2.3 Underwater communication
The rapid attenuation of higher frequency signals and the unstructured nature of
the undersea environment make it difficult to establish a radio communication system for
AUVs. For those reasons, wireless transmission of signals underwater—especially for
distances longer than 100 m—relies almost only on acoustic waves [26], [101].
Underwater acoustic communication using acoustic modems transforms a digital
25
message into a sound that can be transmitted underwater. The performance of these
systems changes dramatically depending on the application and the range of operation
[102]. The main factors in choosing an underwater acoustic modem are:
• Application: Consider the type and length of the message (Command and control
messages, voice messages, image streaming), frequency of operation, and operating
depth.
• Cost: Depending on the complexity and performance, from some hundred up to
$50,000 (USD).
• Size: Usually cylindrical, with lengths from 10 cm to 50 cm.
• Bandwidth: Acoustic modems can perform underwater communication at up to some
kb/s. The length of the message and time limitations must be considered as well.
• Range: The range of operation for the vehicle’s communication impacts the cost of
the system. Considerer than a more extended range will increase the latency and
power consumption of the system. Acoustic modems are suitable for short distances
up to tens of km.
• Power consumption: Depending on the range and modulation, the power
consumption is in the range of 0.1 W to 1 W in receiving mode and 10 W to 100 W in
transmission mode.
Table 5 contains a few options of acoustic modems commercially available.
Table 5. Commercial acoustic modems.
Max Bit Rate Frequency Band

Name Range (m)
(bps) (kHz)
Teledyne Benthos ATM-925 [103] 360 2000–6000 9–27
WHOI Micromodem [104] 5400 3000 16–21
Linkquest UWM 1000 [105] 7000 350 27–45
Evologics S2C R 48/78 [106] 31,200 1000 48–78
Sercel MATS 3G 34 kHz [107] 24,600 5000 30–39
L3 Oceania GPM-300 [108] 1200 45,000 Not specified
Tritech Micron Data Modem [109] 40 500 20–28
Bluerobotics Water Linked M64 Acoustic Modem [110] 64 200 100–200
The working principles of underwater acoustic communication can be described as

follows [111]: First, an electrical transmitter converts information into an electrical signal.
Second, after digital processing by an encoder, the transducer converts the electrical
signal into an acoustic signal. Third, the acoustic signal propagates through the water
medium and propagates the information to the receiving transducer. In this case, the
acoustic signal is converted into an electrical signal. Finally, after the decoder deciphers
the digital signal, the electrical receiver converts the information to audio, text, or picture.
Acoustic communications face many challenges, such as small bandwidth, low
data rate, high latency, and ambient noise [112]. These shortcomings might provoke a
latency of several seconds or even more than a minute. Considering these, Yang et al.
[113] analyzed formation control protocols for multiple underwater vehicles in the
presence of communication flaws and uncertainties. The error Port-Hamiltonian model
about the desired trajectory was introduced. Then, with relative information constraints or
uncertainties, the formation control law was achieved by solving specific limitations of the
26
linear matrix inequality problem. Abad et al. [114] introduced a communication scheme
between the AUVs and a unique representation of the overall vehicle state that limits
message size. Every reported position and path plan is encoded using a grid encoding
scheme to limit the data sent. The authors implemented a decentralized model predictive
control algorithm—centralized methods are typical for swarms of AUVs—to control teams
of AUVs that optimizes vehicle control inputs to account for the limitations of operating in
an underwater environment. They simulated their proposal and showed the effectiveness
of their approach in a Mine/Countermine mission. Another way to deal with acoustic
communication issues was presented by Hallin et al. [115]. They proposed that enabling
the AUVs to anticipate acoustic messages would improve their ability to complete
missions. They outlined an approach to AUV message anticipation in AUVish-BBM, an
acoustic communications language for AUVs [116], based on a University of Idaho-
developed paradigm called Language-Centered Intelligence (LCI). They demonstrated a
new application of LCI in cooperative AUV operations and argued that message
anticipation could be effectively deployed to correct message errors. The structure,
content, and context of individual messages of AUVish-BBM, together with its associated
communication protocol, supply a systematic framework that can be utilized to anticipate
messages expressed by AUVs performing collaborative missions.
The absence of an underwater communication standard has been a problem for
collaborative teams of AUVs. In 2014, Potter et al. presented the JANUS underwater
communication standard [117], a basic and robust tool for underwater communications
designed and tested by the NATO Centre for Maritime Research and Experimentation.
This opened the possibilities for simple integration of different robots using this standard
[118]–[121] for collaborative tasks such as underwater surveillance.
Optical technologies have been tested either stand-alone [122], [123] or as a
complement to an acoustic system [101] to improve the performance of underwater
communication. Laser submarine communication has advantages such as a high bit rate,
higher security, and broad bandwidth. Blue-green light (whose wavelength is 470–580
nm) penetrates water better, and its energy attenuation is less than any other wavelength
light [124]. Thus, researchers have explored optical underwater communication systems
based on blue-green light to allow an underwater vehicle to receive a message from an
aerial/spatial system at any depth despite its actual speed, course, and distance from the
transmitter. Wiener et al. [122] sought a system to deliver a message from a satellite to a
submarine, avoiding the need for the submarine to navigate close to the surface to
retrieve the message, as happened with the radio-frequency systems used at the time.
The authors stated that blue light has the potential to accomplish the result expected in
the future. However, their research only briefly represented what could be expected when
working in such a complex environment. Puschell et al. [123] first demonstrated two-way
laser communication between a submarine vehicle and an aircraft. They concluded that
a blue-green laser communication system could, someday, reach operational
requirements. Sangeetha et al. [125] conducted experiments to study the optical
communication between an underwater body and a space platform using a red laser with
a 635 nm wavelength. Results showed that the performance of the red-light system was
lower than expected for a blue-light system in terms of the attenuation coefficient
observed. Corsini et al. [126] worked on an optical wireless communication system where
both transmitter and receiver were underwater. The optical signal with a 470 nm
27
wavelength was obtained by modulating two LED arrays and received by an avalanche
photodiode module. The error-free transmission was achieved in the three configurations
under test (6.25 Mbit/s, 12.5 Mbit/s, and 58 Mbit/s) through 2.5 m of clean water.
Although some authors, such as Wiener and Puschell, claimed that laser
communication systems for underwater vehicles could be possible, recent studies
showed that technology is still limited. Laser-based systems cannot reach a target with
more than a few tens of meters in depth under ideal conditions [127]. Thus, Farr et al.
[101] developed an optical communication system that complements and integrates an
acoustic system. The result was an underwater communication system capable of
offering high data rates and low latency when within optical range, combined with long
range and robustness of acoustics when outside of optical range. Authors have
demonstrated robust multi-point, low-power omnidirectional optical communications over
ranges of 100 m at data rates up to 10 Mb/s using blue-green emitters.
2.3.1 Underwater communication overview
The nature of underwater environments makes it difficult to use communication
systems with high-frequency signals. This is due to the rapid attenuation that permits
propagation only at very short distances. Acoustic signals have a better performance but
face many challenges, such as signal interferences and small bandwidth, which results
in the need for time synchronization methods, producing a high latency in the system.
Another option is a light-based system, which offers a higher bandwidth but at
short/medium ranges. Blue/green light has better propagation underwater than any other
light. However, when the range for communication is increased, the power consumption,
weight, and volume of the equipment required also increase.
2.4 Autonomous navigation of AUVs
Autonomous navigation can be performed using waypoint tracking, path following,

and trajectory tracking [15]. The vehicle navigates through a set of pre-defined waypoints
in the waypoint tracking method. This method is easier to achieve, but it could produce
some uncertainty in the trajectory followed by the vehicle between two waypoints [128].
This navigation method is illustrated in Figure 15; note that the vehicle can take different
courses for the same waypoints in different runs.
Figure 15. Waypoint navigation representation.
The path following navigation ensures the vehicle follows the desired path to reach
its destination, which is geometrically defined in cartesian coordinates [129]. This
28
navigation method is illustrated in Figure 16; the vehicle is asked to follow a predefined
path without time restrictions.
Figure 16. Path following navigation representation.
In the trajectory tracking problem, the vehicle must follow a time-parametrized

path, meaning it must reach a certain point in a given time [16]. The integration of time
and space restrictions makes trajectory tracking the most complex of the referred
autonomous navigation methods. This navigation method is illustrated in Figure 17.
Figure 17. Trajectory tracking navigation representation.
Considering this, the highly non-linear dynamics of the AUV, and the presence of
external disturbances, new control methods are required to fulfill the extreme challenge
of trajectory tracking in AUVs.
Several control strategies have been applied in the past to achieve high precision
trajectory tracking. The Proportional Integral Derivative (PID) control is one of AUVs' most
employed control strategies due to its simple structure. They work well for systems where
it is possible to control all outputs through the inputs—meaning they are not
underactuated—and are widely used for heading, depth, and surge control [130]–[133].
However, one limitation of PID controllers is that they are tuned to deal with specific
conditions; the performance will be affected if those conditions change. In addition, PID
controllers do not consider nonlinearities, which will eventually deteriorate the controller's
performance. Other strategies are used with the PID controller to compensate for some
of the uncertainties faced in the underwater environment. Dong et al. [134] designed a
fuzzy rules-based strategy to tune a PID depth controller. Experiments were conducted
to test the controller, showing a depth control accuracy of up to 3 cm. Yang et al. [135]
adopted a cascade PID structure to deal with horizontal displacements in the heading
control. Experimental results showed better performance when compared with a
29
traditional PID controller since oscillations were reduced and even eliminated when
controlling the heading of the ROV. Zhang et al. [136] used Particle Swarm Optimization
(PSO) to optimize the parameters of a PID controller for horizontal displacements.
Simulation results showed better performance than a PID controller but were weaker than
a fuzzy controller. Another approach to deal with traditional controllers’ issues is to include
an intelligent algorithm [137], [138] in the control law to achieve an adjustment of the
controller parameters or self-tuning. This is known as adaptive control, which does not
require an exact vehicle model; only a small amount of prior knowledge suffices.
Backstepping control [139] and Sliding Mode Control (SMC) [140] are other
common control methods widely used in the trajectory tracking of AUVs. Backstepping
Control (BC) is a model-based nonlinear controller based on the Lyapunov theory [141]
and works by sectioning the dynamic system of the vehicle into subsystems. Hence, the
controller begins with a known stable system and then iterates to obtain proper controllers
for the rest of the subsystems.
Yan et al. [142] applied backstepping control to the trajectory tracking of an AUV
to a docking station. Simulations of their controller showed that all tracking errors
converge to a small neighborhood of the origin. Zhou et al. [143] proposed a solution for
underactuated AUVs tracking using a combination of a Neural Network (NN) algorithm
and a backstepping sliding mode controller. Simulations showed a smooth tracking of the
desired three-dimensional trajectory. Yan et al. [141] proposed a backstepping sliding
mode controller with fuzzy switching gain. Simulation results showed that the controller
design could guarantee that the AUV accurately tracks the desired path in the presence
of time-varying interference and effectively eliminate the chattering in the traditional SMC
method.
The SMC has been extensively applied due to its simple form, robustness, and
tolerance to model uncertainty. However, SMC has the problem of chattering. For AUV
control, the chattering causes high-frequency changes in the thrusters' speed, leading to
high energy consumption, increased friction, and thruster damage [144]. Lv et al. [145]
proposed a fault-tolerant control method integrated with thrust allocation based on the
sliding mode theory. Experimental results showed a reduction of up to 94% of the steady
error compared with a conventional SMC. García-Valdovinos et al. [146] proposed a
second-order SMC combined with a backpropagation NN control scheme for AUVs to
deal with external disturbances and parameter variations. Experimental results showed
that the controller could reduce the position error to zero robustly. The online adjustments
of the weights in the NN compensated for the variations in the vehicle hydrodynamics and
perturbations. Ramezani-al et al. [147] proposed an SMC with an adaptive gain to
eliminate the effects of external chattering and noise vulnerability. In numerical
simulations, the proposed controller drives the vehicle to the desired trajectory in a limited
time.
Despite the numerous approaches for trajectory tracking of AUVs encountered,
just a few of them have considered a finite-time convergence. A finite-time convergence
controller would have many advantages, such as faster convergence time, higher
accuracy, and better anti-disturbance capability [17]. Finite-time tracking controllers have
been widely and successfully used to control fully actuated robot arms [148], [149].
However, little research has been done on underactuated vehicles. The particular
nonlinear dynamics of AUVs have other challenges, for instance, the hydrodynamics
30
effects inherent to the vehicle’s geometry and water density. Yan et al. [150] adopted the
globally finite-time tracking control strategy and a PID-SMC model-based controller to
track a predefined trajectory for a AUV. Simulations showed that the control laws could
achieve strongly robust and preferably control performance for the horizontal trajectory
tracking control in the presence of parameter perturbation and unknown currents. Their
work was extended to three-dimensional trajectory tracking by Yu et al. [151]. Simulation
results show that the vehicle can converge to the trajectory and remain in a small
neighborhood even in the presence of ocean currents, wherein a tiny fluctuation in the
tracking errors is observed. Chu et al. [152] proposed a local recurrent NN for finite-time
trajectory tracking of ROVs with an unknown dynamical model. The authors designed a
state observer based on sliding mode to estimate those variables that cannot be
measured by the ROV and achieved finite-time convergence with the adaptative controller
proposed. Qiao et al. [153] proposed two control schemes based on sliding mode for the
tracking control of AUVs in the presence of uncertainties and disturbances: an adaptive
integral terminal sliding mode control and an adaptive fast integral terminal sliding mode
control. Each controller comprises a kinematic and a dynamic controller, where the
kinematic can achieve local finite-time convergence of the tracking error to zero. Guerrero
et al. [154] proposed an adaptative high-order SMC with finite-time convergence for an
AUV. The authors validated the proposed controller with real experiments for depth and
yaw trajectories, considering parametric uncertainties and external disturbances. Liu et
al. [155] proposed a scheme based on a non-singular terminal SMC manifold. The results
showed better convergence rate and tracking precision performance without chattering.
Liu et al. [156] presented a finite-time second-order SMC for trajectory tracking of
underwater vehicles subject to system uncertainties and unknown disturbances.
Fewer works can be found in the literature concerning cooperative underwater
manipulation tasks. Casalino et al. [157] and Manerikar et al. [158] presented a control
algorithm for UUVs' cooperative transport of large objects. This work was part of the
project MARIS, which aimed to develop cooperative control strategies for multiple UUVs.
They presented results from numerical simulations that assumed that the object had
already been grasped and that information exchange between vehicles was minimal.
Simetti et al. [159] developed a unified architecture for controlling single and cooperative
UUVs. In support of their proposal, the authors presented a complete simulation of a
manipulation task requiring minimal communication. In the simulations, the vehicles
approached the object position, then performed the grasp and used the proposed
algorithm to move the object to its destination in a defined final position.
As can be seen, the work found in the literature is usually validated through
numerical simulations since it is challenging to assemble an experimental setup with
multiple vehicles for this kind of application. Moreover, a simulated environment permits
authors to make assumptions about the hardware or software that a real system may not
fulfill. A formal hardware or software integration testing method would be necessary to
ensure a successful and safe system implementation. However, due to the limitations
present in real experiments, such as sensor resolution, measurement rate, and processor
speed, it is possible that a controller does not perform as expected when programmed in
a real system.
31
2.5 The Station-keeping problem
In manipulation operations, autonomous navigation drives the vehicle to the object

or surface on which the task will be performed. Once the vehicle reaches its position, it
should maintain its position and orientation during the manipulation, overcoming unknown
external disturbances caused by the underwater environment or the manipulation task
itself. This is known as station-keeping [18] and is critical to the success of a mission as
it allows the operator or autonomous manipulator to have better control of the operation
and reduces the risks of collisions that could damage either the vehicle, the manipulator,
or the object. Sakiyama et al. [160] proposed a disturbance observer for the station-
keeping of AUVs. Numerical simulations introduced external disturbances of 10 N in the
horizontal plane. The results showed an initial push for the vehicle of no more than 5 cm,
which is compensated after a fraction of a second. Ding et al. [18] investigated a modified
adaptive generalized super-twisting algorithm supported by an adaptive tracking
differentiator and reduced-order extended state observer for free-floating manipulation
under model uncertainties and external disturbances. The authors performed numerical
simulations to verify the feasibility and efficiency of their proposed control scheme when
sensor noise, external disturbances, and 30% of parameter uncertainties were
introduced.
The RMSE for the position was kept below 0.5 cm and 0.2° for orientation. Vu et
al. [161] designed an SMC-based station-keeping algorithm for an AUV subjected to
model uncertainties and ocean currents disturbances in the horizontal plane. In
simulations, the authors used a random value between −35% and 35% as model
uncertainties and included ocean currents in the x, y, and z axes with a mean value of
0.5 m/s. At the beginning of the simulations, the AUV position shifted with a maximum
error of 5 cm. After a couple of seconds, the vehicle returns to its desired position.
Experiments with the AUV with an unknown added mass and an external current of 1 knot
resulted in a position drift of up to 20 cm. It is undeniable that real-world experiments are
very challenging. Even control systems that show robust and exceptional performance in
numerical simulations are subject to performance drift when used in the real world.
2.6 Collaborative AUVs
Collaborative work refers to an interaction of two or more AUVs to perform a

common task: collaborative navigation, exploration, target search, and object
manipulation. A team of AUVs navigating on a specific formation can significantly expand
the applications for underwater missions, such as those requiring proximity to the seafloor
or covering a wide area for search, recovery, or reconstruction. At first, researchers
focused on how multiple vehicles could obtain data simultaneously from the same area
of interest. Nowadays, their focus has moved to the trajectory design and operation
strategies for those multi-vehicle systems [162].
2.6.1 Collaborative navigation
Groups of AUVs can work together under different navigation schemes, generally
parallel or leader-follower [111]. In a parallel formation, as illustrated in Figure 18, all
AUVs are equipped with the same systems and sensors to locate and navigate
themselves precisely and communicate with their neighbor AUVs.
32
Figure 18. Parallel navigation of AUVs.
In a leader-follower scheme, as illustrated in Figure 19, the leader AUV is equipped

with high-precision instruments. Meanwhile, follower AUVs are equipped with low-
precision equipment [163]. Communication is only required between the vehicle leader
and its followers. There is no need for the followers to communicate with each other.
Figure 19. Leader-follower navigation of AUVs.
The lower cost and the reduced communications needs make the leader-follower
scheme the main navigation control method of AUVs. Its basic principles and algorithms
are relatively mature [111]. However, unstable communications and communication
delays are still challenging and must be addressed. Since there are problems with signals
when multiple systems emit simultaneously, the co-localization of AUVs is mostly based
on time synchronization. However, time synchronization methods have shortcomings,
such as needing AUVs to go to the surface to receive the synchronization signal. As an
alternative, Zhang et al. [164] studied collaborative navigation and positioning multi-AUVs
without time synchronization. The authors established a collaborative navigation
positioning model for multi-AUVs and designed an EKF for collaborative navigation. This
design only needs the time delay of the AUV itself and does not need to consider whether
the AUV has synchronized with others. In simulated experiments, they compared the
precision of their algorithm with a prediction model and results showed that, even when
error increases over time, the precision of co-localization without time synchronization
was higher. Yan et al. [165] addressed the problems of leader-follower AUV formation
control with model uncertainties, current disturbances, and unstable communication. The
method's effectiveness is simulated by tracking a spiral helix curve path with one leader
AUV and four follower AUVs. A second-order integral AUV model with a nonlinear
33
function and current disturbances was established, considering model uncertainties and
current disturbances. The simulation results showed that leader-follower AUV formation
controllers are feasible and effective. After an adjustment period, all follower AUVs can
converge to the desired formation structure, and the formation can keep tracking the
desired path. Cui et al. [166] focused on the problem of tracking control for multi-AUV
systems and proposed an adaptive fuzzy-finite time control method. This algorithm
combines algebraic graph theory with a leader-follower architecture to describe the
system's communication. Then, the error compensation mechanism is introduced. Finally,
the application of finite time and fuzzy logic systems improves the convergence rate and
the robustness of the multi-AUV system. The effectiveness of the proposed algorithm was
illustrated by simulation. Choosing the architecture of 4 AUVs, including one leader and
three followers. The algebraic graph theory and a fuzzy logic system technique are
integrated into the distributed controllers to avoid unknown internal and external
interferences. The simulation results demonstrate the proposed algorithm's effectiveness
and the multi-AUV system's robustness with a faster convergence speed than other
algorithms. When AUVs navigate closed formations, the delay between the transmission
and reception of the acoustic signals represents a high risk. Therefore, a solution with a
response time significantly faster must be explored. Bosh J. et al. [19] developed an
algorithm for AUVs navigating on a close formation. Light markers and artificial vision are
used to estimate a target vehicle's pose at short ranges with high accuracy and execution
speed. In the experiments presented, the filtered pose estimates were updated at
approximately 16 Hz, with a standard deviation lower than 0.2 m in the distance
uncertainty between vehicles at a 6 m - 12 m range. As expected, the results showed that
the system performs adequately for vehicle separations smaller than 10 m, while the
tracking becomes intermittent for longer distances due to the challenging visibility
conditions underwater. Another alternative for cooperative navigation of AUVs is using
systems that allow the vehicles within a team to help each other with their localization.
Teck et al. [167] proposed a TBN system for cooperative AUVs. The approach consists
of an altimeter and acoustic modem equipped on each vehicle and a bathymetric terrain
map. The localization is performed via decentralized particle filtering. The vehicles in the
team are assumed to have their system time synchronized. Simple scheduling is adopted
so that each vehicle in the team broadcasts its local state information sequentially using
acoustic communication. This information includes the vehicle's current position, the filter
estimated covariance matrix, and the latest water depth measurement. When another
vehicle receives the acoustic signal, the time-of-arrival can be calculated to determine the
inter-vehicle distance. Results showed that localization performance improves as the
number of vehicles in the team increase, at least up to four, and when they are in range
for proper acoustic communication. In those conditions, the average positioning error was
in the range of a few meters. Tan et al. [168] developed a cooperative path planning for
range-only localization. The authors explored using a single-beacon vehicle for range-
only localization to support other AUVs. Specifically, they focused on cooperative path-
planning algorithms for the beacon vehicle using dynamic programming formulations.
These formulations consider and minimize the positioning errors accumulated by the
supported AUV. Implementation of the cooperative path-planning algorithms was in a
simulated environment. The simulations were conducted with different ranging aids, each
transmitted from a single beacon. The ranging aids used were a single fixed, circularly
34
moving, and cooperative beacon. Experimental results were also obtained by a field trial
was conducted near Serangoon Island, Singapore. Average error was reduced up to 19.1
m over a traveled distance of 1.5 km. De Palma et al. [169] made a similar approach. The
problem addressed by the authors consisted of designing a relative localization solution
for a networked group of vehicles measuring mutual ranges. The aim of the project was
to exploit inter-vehicle communications to enhance the range-based relative position
estimation. Vehicles are considered capable to know their own position, orientation, and
velocity regarding a common frame. Such vehicles share their own information through
their communication channel, and they can obtain measurements of their relative
Euclidean distance with respect to several other agents. The connection topology of the
agents was represented through a relative position measurement graph and a simulation,
relative to a group of 4 agents, was performed with different connection topologies. During
the simulation, z error remained in the range of ± 1 m after a time-lapse of 3000 s.
2.6.2 Collaborative AUVs missions
Surveillance and intervention are typically the kinds of missions designed for teams
of AUVs. Surveillance missions require the AUVs to detect, localize, follow, classify
targets, inspect, or explore the ocean. Meanwhile, intervention missions require the AUVs
to interact with objects within the environment. Examples of both missions are
represented in Figure 20.
Figure 20. Collaborative mission of AUVs. a) Survey missions. b) Intervention missions.
Collaborative survey missions

A good search mission needs to minimize the number of vehicles required and
maximize the efficiency of the search. Oceanic, biologic, and geologic variability of
underwater environments impacts the search performance of teams of AUVs. Baylog J.
et al. [172] applied a game theoretic approach to optimizing a search channel
characterization of the environment to address search planning in these conditions, where
the detection process is prone to false alarms. The search space is partitioned into
discrete cells where objects of interest may be found. The game theory approach seeks
to find the equilibrium solution of the game rather than the optimal solution to a fixed
objective of maximizing the value-over-cost. A sequence of searches by four search
agents over a search region was planned and simulated to demonstrate effectiveness in
achieving the game objective. Li et al. [170] proposed a sub-region collaborative search
strategy and a target-searching algorithm based on a perceptual adaptive dynamic
prediction. The reality of the local environment is obtained using the FLS of the multi-AUV
35
system. The simulation experiments verified that the algorithm proposed successfully
searched and tracked the target.
Moreover, in the case of an AUV failure, it can also ensure that other AUVs
cooperate to complete the remaining target search tasks. Algorithms for collaborative
search based on NN are being designed to overcome the variability of the environment
and the presence of obstacles. Iv et al. [171] presented a region search algorithm based
on a Glasius Bio-inspired Neural Network (GBNN), which can be used for AUVs to
perform target search tasks in underwater regions with obstacles. This algorithm divides
the search area into several discrete sub-areas, and connections are made between
adjacent neurons. AUVs and obstacles are introduced to the network as excitation
sources to avoid collisions during the search. Constructing hypothetical targets and
introducing them into the NN as stimulating excitation sources allows the AUVs to quickly
search for areas where the target is likely to exist and efficiently complete the search task.
Sun et al. [172] designed a new strategy for collaborative search with a GBNN algorithm.
A grid map is set up to represent the working environment, and NNs are constructed
where each AUV corresponds to a NN. All AUVs must share environmental information,
and each AUV is treated as a moving obstacle in the region to avoid vehicle collisions.
The simulation was conducted in MATLAB to confirm that multi-AUVs can plan
reasonable and collision-free coverage paths through the proposed algorithm and reach
full coverage on the same task area with the division of labor and cooperation. Yan et al.
[173] addressed a control problem for AUVs tracking a moving target with varying velocity.
For this algorithm, at least one AUV is assumed to be capable of obtaining information
about the target, and the communication topology graph of the vehicle is supposed to be
undirectedly connected. Simulations were made using MATLAB to demonstrate the
efficiency and effectiveness of the proposed control algorithm, considering a system with
three vehicles.
Collaborative intervention missions
There is much more in the underwater environment for AUVs beyond survey
missions. Manipulating objects, repairing structures or pipes, recovering black boxes, and
extracting samples, among other tasks, make it necessary to have a platform with the
capacity to navigate and perform them autonomously. Nowadays, these are mostly done
by crewed or remotely operated vehicles. Researchers have worked in the design and
development of such platforms in recent years, which is difficult even for a single
Intervention AUV (I-AUV) due to the complexity of the vehicle itself plus the manipulator
system. The Girona 500 I-AUV is an example of a single-vehicle platform for intervention
missions. This vehicle is used to dock autonomously into an adapted subsea panel and
perform the intervention task of turning a valve and plugging in/unplugging a connector
[174]. The same vehicle was also equipped with a three-fingered gripper and an artificial
vision system to locate and recover a black box [14]. Other projects, such as the Italian
national project MARIS [175], have been launched to produce theoretical, simulated, and
experimental results for intervention AUVs either stand-alone or for collaborative teams.
The aim of the MARIS project was the development of technologies that allow the use of
teams of AUVs for intervention missions, in particular: reliable guidance and control,
stereovision techniques for object recognition, reliable grasp, manipulation and
transportation of objects, coordination and control methods for large object grasp and
transportation, high-level mission planning techniques, underwater communication, and
36
the design and realization of prototype systems, allowing experimental demonstrations of
integrating the results from the previous objectives. The open-frame fully actuated robotic
platform R2 ROV/AUV was used for the MARIS project, and the Underwater Modular
Manipulator (UMA), none of them developed within the project. A vision system and a
gripper were designed for the autonomous execution of manipulation tasks. Tests were
performed to assess the correct integration of all the components, with a success rate of
a grasping operation of up 70% [13]. This project was an important achievement in terms
of autonomous underwater manipulation, and the theoretical studies for multi-vehicle
localization and collaborative underwater manipulation systems will be the next step to
be demonstrated in field trials. For collaborative I-AUVs, Simetti et al. [176][159]
described a novel cooperative control policy for transporting large objects in underwater
environments using two manipulator vehicles. The cooperative control algorithm
considers all mission stages: grasping, transportation, and the two vehicles' final
positioning of the shared object. The cooperative transportation of the object is carried
out to deal with the limitations of acoustic communication. This was achieved successfully
by exchanging just the tool frame velocities. A simulation was done, with the UwSim
dynamic simulator, using two vehicles with 6 degrees of freedom to test the control
algorithms. The simulation consisted of completing the following tasks: keeping away
from joint limits, keeping the manipulability measure above a certain threshold,
maintaining the horizontal attitude of the vehicles, maintaining a fixed distance between
the vehicles, reaching the desired goal position. The system managed to accomplish the
final objective of the mission successfully, by transporting the object to the desired goal
position. Conti et al. [177] proposed an innovative decentralized approach for cooperative
mobile manipulation of intervention AUVs. The control architecture deals with the
simultaneous control of the vehicles and robotic arms, and the underwater localization.
Simulations were made in MATLAB Simulink to test the potential of the system. The
cooperative mobile manipulation was performed by four AUVs placed at the four corners
of the object, and obstacles were introduced as spheres. According to the authors, results
were very encouraging because the AUV swarm keeps the formation during the
manipulation phase and the object during an avoiding phase performed due to obstacles.
Cataldi et al. [178] worked in cooperative control of underwater vehicle-manipulator
systems. The authors propose an architecture that considers most of the underwater
constraints: uncertainty in the model, low sensor bandwidth, position-only arm control,
and geometric-only object pose estimation. The simulated system, designed in MATLAB
and adapted with SimMechanics, consisted of two AUVs transporting a bar. Results on
bar position and applied forces on end effectors provided promising results on its possible
real applications. Heshmati-alamdari et al. [179] worked on a similar system. A nonlinear
model predictive control approach was proposed for a team of AUVs transporting an
object. The model must deal with the coupled dynamics between the robots and the
object. The feedback relies only on each AUV local measurement to deal with
communication issues, and no data is exchanged between the robots. A real-time
simulation, based on UwSim dynamic simulator and running on the Robot Operating
System (ROS), was performed to validate the proposed approach. The goal for the team
of AUVs was to follow a set of pre-defined waypoints while avoiding obstacles within the
workspace which was successfully achieved.
37
2.6.3 Collaborative AUVs overview
Collaborative missions are pretty challenging to implement. Most authors use
MATLAB Simulink to perform their simulations and some tools, such as the former
SimMechanics (now called Simscape Multibody). Another simulation environment
commonly used for underwater robotics is the UnderWater Simulator (UWSim) [180]. With
those tools, researchers are pushing state-of-the-art control, localization, and navigation
algorithms.
Significant issues, such as vehicle localization, must be solved to achieve AUVs
collaboration. Global positioning systems are not available in underwater environments,
so the vehicles’ localization must be addressed with acoustic or inertial methods, which
are usually slow and not as accurate as needed for collaborative automated manipulation
missions. If the inter-vehicle communication system is good enough in terms of range,
bandwidth, and rate, range-only/single-beacon can be an effective method for target
localization and collaborative navigation of teams of AUVs. Vision-based systems are
also an option that has the potential to control AUV formations without the need to rely on
inter-vehicle communications, but only if the environmental conditions are favorable for
light propagation and sensing.
Communication is another big issue when working with multiple vehicles in an
underwater environment. Radio and other signals commonly used in ground or aerial
systems cannot be used underwater. Acoustic communication is used instead. However,
it has shortcomings and challenges since it suffers from small bandwidth, propagation
delays, and unreliability [181]. Vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I)
communications are usually limited to a minimal data exchange to deal with the limitations
of the acoustic systems. A decentralized scheme based solely on vehicle sensing where
no V2V or V2I communications will permit the vehicles to make decisions autonomously
[182], enhancing their performance. In this scenario, an autonomous navigation controller
with finite-time convergence would have some advantages, such as faster convergence
time, higher accuracy, and better anti-disturbance capabilities. If the user could select the
convergence time arbitrarily, it would have some extra advantages, such as lower energy
consumption and guaranteeing that the vehicle meets and tracks the desired trajectory at
a convenient moment.
38
Chapter 3: Unmanned Underwater Vehicles modeling
This Chapter introduces the general equations for kinematics and hydrodynamics
of the UUVs. Then the BlueROV2 underwater vehicle from Blue Robotics® [183] is
presented. This vehicle is the one used for simulation and experimentation.
3.1 Kinematics and hydrodynamics
The kinematics of an underwater vehicle can be described by two reference frames

[184], as shown in Figure 21. The orthonormal axes are called 𝑥, 𝑦, and 𝑧 for the Earth-
fixed frame and 𝑥𝑏 , 𝑦𝑏 , and 𝑧𝑏 for the Body-fixed frame.
Figure 21. Reference frames for underwater vehicles.
The Society of Naval Architects and Marine Engineers (SNAME) has defined a
convention for notating an underwater vehicle's position, orientation, velocities, forces,
and moments, as shown in Table 6.
Table 6. SNAME notation for underwater vehicles.
Movement Name PositionVelocityForce/Moment

X translation Surge 𝒙 𝒖 𝑿
Y translation Sway 𝒚 𝒗 𝒀
Z translation Heave 𝒛 𝒘 𝒁
X rotation Roll 𝝓 𝒑 𝑲
Y rotation Pitch 𝜽 𝒒 𝑴
Z rotation Yaw 𝝍 𝒓 𝑵
The position and orientation of the vehicle in the Earth-fixed frame are described
as 𝜂, its velocities in the Body-fixed frame as 𝜈, and the forces and moments in the Body-
fixed frame as 𝜏
𝜂 = (𝜂1 , 𝜂2 )𝑇 = (𝑥, 𝑦, 𝑧, 𝜙, 𝜃, 𝜓)𝑇 , (10)
𝑇 𝑇
𝜈 = (𝜈1 , 𝜈2 ) = (𝑢, 𝑣, 𝑤, 𝑝, 𝑞, 𝑟) , (11)
𝜏 = (𝑋, 𝑌, 𝑍, 𝐾, 𝑀, 𝑁)𝑇 (12)
Fossen [184] describes the hydrodynamic model of underwater vehicles by
Newton–Euler Equations as
𝑀𝜈̇ + 𝐶(𝜈)𝜈 + 𝐷(𝜈)𝜈 + 𝑔(𝜂) = 𝜏 + 𝜔, (13)
39
𝜏 = 𝐵𝑡 𝑢𝑡 , (14)
where
𝑀 ∈ℝ6×6 is the inertial and added mass matrix,
𝐶 ∈ℝ6×6 is the rigid body and added mass centripetal and Coriolis matrix,
𝐷 ∈ℝ6×6 is the hydrodynamic damping matrix,
𝑔 ∈ℝ6×1 is the restitution forces vector,
𝐵𝑡 ∈ℝ6×6 is the thruster allocation matrix,
𝑢𝑡 ∈ℝ6×1 is a vector containing the force generated by the thrusters, and
𝜔 ∈ℝ6×1 represents external disturbances.
The model provided in Equation 13 holds the following properties:

Property 1. The inertia Matrix 𝑀𝜂 (𝜂) is symmetric and positive definite, i.e., 𝑀𝜂 (𝜂) =
𝑀𝜂 (𝜂)𝑇 > 0, 𝑀̇𝜂 (𝜂) = 0.
Property 2. The derivative of the inertia matrix 𝑀̇𝜂 (𝜂) and the Coriolis and centripetal
matrix 𝐶𝜂 (𝜈, 𝜂) satisfy 𝑥 𝑇 [𝑀̇𝜂 (𝜂) − 2𝐶𝜂 (𝜈, 𝜂)]𝑥 = 0, ∀𝑥 ∈ ℝ𝑛 , 𝑥 ≠ 0. Thus 𝑀̇𝜂 (𝜂) − 2𝐶𝜂 (𝜈, 𝜂)
is skew-symmetric.
Property 3. The damping matrix 𝐷𝜂 (𝜈, 𝜂) is non-symmetric and strictly positive, i.e.,
𝐷𝜂 (𝜈, 𝜂) > 0, ∀ 𝜐, 𝜂 ∈ ℝ𝑛 .
Property 4. The dynamic model of a UUV is linearly parametrizable by the product of a
regressor 𝑌(𝜂, 𝜂̇ , 𝜂̈ ) ∈ ℝ𝑛 × 𝑝 composed of known functions, and a vector 𝜃 ∈ ℝ𝑝 composed
of dynamic parameters. That is 𝑌(𝜂, 𝜂̇ , 𝜂̈ )𝜃 ∈ ℝ𝑛 .
Property 5. Boundedness of dynamic terms. For constant 𝛽𝑖 > 0
1. The inertia matrix 𝑀𝜂 (𝜂) satisfies 𝛽1 < 𝜆𝑚 (𝑀𝜂 (𝜂)) ≤ ‖𝑀𝜂 (𝜂)‖ ≤ 𝜆𝑀 (𝑀𝜂 (𝜂)) < 𝛽2
with 𝜆𝑚 and 𝜆𝑀 denoting the minimum and maximum eigenvalues of 𝑀,
respectively;
2. The Coriolis and centripetal vector 𝐶𝜂 (𝜈, 𝜂)𝜂̇ satisfies ‖𝐶𝜂 (𝜈, 𝜂)𝜂̇ ‖ ≤
𝛽3 ‖𝜈‖‖𝜂‖, ∀ 𝜈, 𝜂 ∈ ℝ𝑛 ;
3. The damping vector 𝐷𝜂 (𝜈, 𝜂)𝜂̇ satisfies ‖𝐷𝜂 (𝜈, 𝜂)𝜂̇ ‖ ≤ 𝛽4 ‖𝜈‖‖𝜂‖, ∀ 𝜈, 𝜂 ∈ ℝ𝑛 ;
4. The vector of restoring forces 𝑔(𝜂) satisfies the following ‖𝑔(𝜂)‖ < 𝛽5 .
A series of kinematic transformations can be applied to Equation 13 to express the
model of the vehicle in the Earth-fixed frame
𝜂̇ = 𝐽(𝜂2 )𝜈 ↔ 𝜈 = 𝐽−1 (𝜂2 )𝜂̇ , (15)
̇ −1
𝜂̈ = 𝐽(𝜂2 )𝜈̇ + 𝐽(𝜂2 )𝜈 ↔ 𝐽 (𝜂2 )[𝜂̈ − 𝐽(𝜂2 )𝜈], (16)
with
𝐽 (𝜂 ) 03×3
𝐽(𝜂2 ) = [ 1 2 ], (17)
03×3 𝐽2 (𝜂2 )
and
𝑐𝜃 𝑐𝜓 −𝑠𝜓 𝑐𝜙 + 𝑠𝜙 𝑠𝜃 𝑐𝜓 𝑠𝜙 𝑠𝜓 + 𝑠𝜃 𝑐𝜙 𝑐𝜓
𝐽1 (𝜂2 ) = [𝑠𝜓 𝑐𝜙 𝑐𝜙 𝑐𝜓 + 𝑠𝜙 𝑠𝜃 𝑠𝜓 −𝑐𝜙 𝑐𝜓 + 𝑠𝜃 𝑠𝜓 𝑐𝜙 ], (18)
−𝑠𝜃 𝑠𝜙 𝑐𝜃 𝑐𝜙 𝑐𝜃
1 𝑠𝜙 𝑡𝜃 𝑐𝜙 𝑡𝜃
𝐽 (𝜂 ) = [0 𝑐𝜙
2 2
−𝑠𝜙 ], (19)
𝑠𝜙 𝑐𝜙
0 𝑐𝜃 𝑐𝜃
𝜂2 = [𝜙, 𝜃, 𝜓], (20)
40
where 𝐽1 (𝜂2 ) ∈ ℝ3×3 and 𝐽2 (𝜂2 ) ∈ ℝ3×3 are the matrices relating the linear and angular
velocity components, 𝜈1 and 𝜈2 , to the Earth-fixed frame. The symbols 𝑐𝑎𝑛𝑔𝑙𝑒 , 𝑠𝑎𝑛𝑔𝑙𝑒 , and
𝑡𝑎𝑛𝑔𝑙𝑒 are abbreviations for cos (𝑎𝑛𝑔𝑙𝑒), sin (𝑎𝑛𝑔𝑙𝑒), and tan (𝑎𝑛𝑔𝑙𝑒), respectively.
Then, after some mathematical manipulation the hydrodynamic model of the
underwater vehicle can be described as follows
𝑀𝜂 (𝜂)𝜂̈ + 𝐶𝜂 (𝜈, 𝜂)𝜂̇ + 𝐷𝜂 (𝜈, 𝜂)𝜂̇ + 𝑔𝜂 (𝜂) = 𝜏𝜂 (21)
where
𝑀𝜂 (𝜂) = 𝐽−𝑇 (𝜂)𝑀𝐽−1 (𝜂), (22)
𝐶𝜂 (𝜈, 𝜂) = 𝐽 (𝜂)[𝐶(𝜈) − 𝑀𝐽 (𝜂)𝐽(̇ 𝜂)]𝐽 (𝜂),
−𝑇 −1 −1
(23)
𝐷𝜂 (𝜈, 𝜂) = 𝐽−𝑇 (𝜂)𝐷(𝜈) 𝐽−1 (𝜂), (24)
𝑔𝜂 (𝜂) = 𝐽−𝑇 (𝜂)𝑔(𝜂), (25)
−𝑇 (𝜂)𝜏.
𝜏𝜂 = 𝐽 (26)
Assumption 1. To avoid a possible singularity problem in 𝐽(𝜂), the pitch angle 𝜃 is

bounded as
|𝜃| < 𝜃𝑀 < 𝜋/2, (27)
where 𝜃𝑀 stands for the upper bound of 𝜃 and it is a known positive constant.
Assumption 2. The Jacobian transformation matrix 𝐽(𝜂) is bounded by a known positive
constant 𝐽𝑠𝑢𝑝 [153] so that
𝑠𝑢𝑝𝜂 ‖𝐽(𝜂)‖ ≤ 𝐽𝑠𝑢𝑝 . (28)
Remark 1. Well-posed Jacobian matrix. The transformation in Equation 15 is ill-posed
when 𝜃 = ±90°. A quaternion approach might be considered to overcome this singularity.
However, the vehicle is not expected to operate at 𝜃 = ±90°. In addition, the UUV is utterly stable
in roll and pitch coordinates.
Ocean currents can be included as external disturbances in the simulation by using

relative velocity—the difference between the real velocity and the velocity of the ocean
current—as described by Fossen [184]:
𝜈𝑟𝑒𝑙 = 𝜈 − 𝜈𝑜𝑐 , (29)
where 𝜈 is the vehicle velocity and 𝜈𝑜𝑐 is the ocean current velocity.
A generalized vector for an irrotational ocean current velocity is described
by
𝜈𝑜𝑐 = [𝑢𝑜𝑐 , 𝑣𝑜𝑐 , 𝑤𝑜𝑐 , 0,0,0]𝑇 , (30)
where 𝑢𝑜𝑐 is the ocean current velocity from the north, 𝑣𝑜𝑐 is the ocean current velocity
from the east, and 𝑤𝑜𝑐 is the ocean current velocity from below.
Defining 𝛼𝑜𝑐 as the angle of attack and 𝛽𝑜𝑐 as the slide slip angle, every element
of the velocity vector can be calculated as
𝑢𝑜𝑐 = 𝜈𝑜𝑐 𝑐𝑜𝑠 𝛼𝑜𝑐 𝑐𝑜𝑠 𝛽𝑜𝑐 ,
𝑣𝑜𝑐 = 𝜈𝑜𝑐 𝑠𝑖𝑛 𝛽𝑜𝑐 , and (31)
𝑤𝑜𝑐 = 𝜈𝑜𝑐 𝑠𝑖𝑛 𝛼𝑜𝑐 𝑐𝑜𝑠 𝛽𝑜𝑐 .
Then, the model given by Equation 13 can be modified as follows
𝑀𝜈̇ 𝑟𝑒𝑙 + 𝐶(𝜈𝑟𝑒𝑙 )𝜈𝑟𝑒𝑙 + 𝐷(𝜈𝑟𝑒𝑙 )𝜈𝑟𝑒𝑙 + 𝑔(𝜂) = 𝜏, (32)
41
3.2 BlueROV2 underwater vehicle
The vehicle used in this work is the BlueROV2 from Blue Robotics®. It has a
vectored configuration with six thrusters arranged, as shown in Figure 22.
Figure 22. Thruster configuration for the BlueROV2. A) Top view. B) Front view.
Matrices and vectors involved in the vehicle’s hydrodynamics contain numerous

unknown parameters, making it infeasible to estimate. Nevertheless, some assumptions
and considerations can be made because of the BlueROV2 geometry and operation,
resulting in a considerable reduction of unknown parameters. These conditions are listed
below:
• Since BlueROV2 operates at relatively low speeds (i.e., less than two m/s),
lift forces can be neglected.
• BlueROV2 is assumed to have port-starboard and fore-aft symmetry, and
the center of gravity (CG) is in the symmetry planes.
• BlueROV2 is assumed to operate below the wave-affected zone. As a
result, disturbances of waves ω on the vehicle are negligible.
• The thruster allocation of BlueROV2 does not permit active control of the
pitch orientation 𝑞. However, the motion around this axis is considered self-
regulated due to the vehicle's buoyant restoring moments, resulting in the
following reduced system
𝜈 = (𝑢, 𝑣, 𝑤, 𝑝, 0, 𝑟)𝑇 , (33)
The physical and hydrodynamic parameters of BlueROV2 are summarized in
Table 7 as identified from [185], [186].
Table 7. BlueROV2 parameters.
Parameters Symbol Value Unit

Mass m 10 kg
Buoyancy B 100.06 N
Weight W 98.1 N
Center of gravity 𝒓𝑮 = (𝒙𝒈 , 𝒚𝒈 , 𝒛𝒈 ) 𝒓𝑮 = (𝟎, 𝟎, 𝟎) m
Center of buoyancy 𝒓𝑩 = (𝒙𝒃 , 𝒚𝒃 , 𝒛𝒃 ) 𝒓𝑩 = (𝟎, 𝟎, 𝟎. 𝟎𝟐) m
Inertia moment I=diagonal(𝑰𝒙 , 𝑰𝒚 , 𝑰𝒛 ) I=diagonal(0.16, 𝟎. 𝟏𝟔, 𝟎. 𝟏𝟔) 𝒌𝒈 ∙ 𝒎𝟐
Added mass parameters 𝑿𝒖̇ -5.5 𝒌𝒈
42
𝒀𝒗̇ -12.7 𝒌𝒈
𝒁𝒘̇ -14.57 𝒌𝒈
𝒎𝟐
𝑲𝒑̇ -0.12 𝒌𝒈 ∙
𝒓𝒂𝒅
𝒎𝟐
𝑴𝒒̇ -0.12 𝒌𝒈 ∙
𝒓𝒂𝒅
𝒎𝟐
𝑵𝒓̇ -0.12 𝒌𝒈 ∙
𝒓𝒂𝒅
𝒔
𝑿𝒖 -4.03 𝑵∙
𝒎
𝒔
𝒀𝒗 -6.22 𝑵∙
𝒎
𝒔
Linear Damping parameters 𝒁𝒘 -5.18 𝑵∙
𝒎
𝒔
𝑲𝒑 -0.07 𝑵∙
𝒓𝒂𝒅
𝒔
𝑴𝒒 -0.07 𝑵∙
𝒓𝒂𝒅
𝒔
𝑵𝒓 -0.07 𝑵∙
𝒓𝒂𝒅
𝒔𝟐
𝑿𝒖|𝒖| -18.18 𝑵∙ 𝟐
𝒎
𝒔𝟐
𝒀𝒗|𝒗| -21.66 𝑵∙ 𝟐
𝒎
𝒔𝟐
𝒁𝒘|𝒘| -36.99 𝑵∙ 𝟐
Quadratic Damping parameters 𝒎
𝒔𝟐
𝑲𝒑|𝒑| -1.55 𝑵∙
𝒓𝒂𝒅𝟐
𝒔𝟐
𝑴𝒒|𝒒| -1.55 𝑵∙
𝒓𝒂𝒅𝟐
𝒔𝟐
𝑵𝒓|𝒓| -1.55 𝑵∙
𝒓𝒂𝒅𝟐
The matrices that comprise the hydrodynamical model in Equation 13 are

considered as defined in [184].
The thruster allocation matrix for the BlueROV2 vehicle is assembled as
0.7071 0.7071 −0.7071 −0.7071 0 0
−0.7071 0.7071 −0.7071 0.7071 0 0
0 0 0 0 −1 −1
𝐵𝑡 = (34)
0 0 0 0 0.115 −0.115
0 0 0 0 0 0
[−0.1773 0.1773 −0.1773 0.1773 0 0 ]
Remark 2. The thruster allocation of BlueROV2 does not allow active control of the pitch
angle 𝜃. However, the motion around this axis is self-regulated due to the vehicle’s buoyant
restoring moments.
43
Chapter 4: Controller design
A model-free second-order SMC is proposed in this work to deal with autonomous
navigation and station-keeping problems. This is a finite-time convergence controller
whose main feature is that the user can arbitrarily select the convergence time. This
Chapter introduces this controller, presents its corresponding stability analysis, and
further considerations for its implementation.
4.1 Model-Free High Order Sliding Mode Controller
According to Property 4, the Equation 21 is linearly parameterizable by the product

of a regressor 𝑌(𝜂, 𝜂̇ , 𝜂̈ ) ∈ ℝ𝑛 × 𝑝 , consisting of known nonlinear functions and a vector 𝜃 ∈
ℝ𝑝 with constant parameters. This parametrization can be rewritten in terms of a nominal
reference 𝜂̇ 𝑟 and its time derivative 𝜂̈ 𝑟 as
𝑀𝜂 (𝜂)𝜂̈ 𝑟 + 𝐶𝜂 (𝜈, 𝜂)𝜂̇ 𝑟 + 𝐷𝜂 (𝜈, 𝜂)𝜂̇ 𝑟 + 𝑔𝜂 (𝜂) = 𝑌(𝜂, 𝜂̇ , 𝜂̇ 𝑟 , 𝜂̈ 𝑟 )𝜃. (35)
Subtracting Equation 35 from both sides of Equation 21 results in the open-loop
error hydrodynamics expression:
𝑀𝜂 (𝜂)𝑆𝑟̇ + 𝐶𝜂 (𝜈, 𝜂)𝑆𝑟 + 𝐷𝜂 (𝜈, 𝜂)𝑆𝑟 = 𝜏𝜂 − 𝑌(𝜂, 𝜂̇ , 𝜂̇ 𝑟 , 𝜂̈ 𝑟 )𝜃, (36)
where 𝑆𝑟 = 𝜂̇ − 𝜂̇ 𝑟 is called the extended error.
The nominal reference 𝜂̇ 𝑟 is defined in terms of the position and velocity paths as
follows
𝑡
𝜂̇ 𝑟 = 𝜂̇ 𝑑 − 𝛼𝜂̃ + 𝑆𝑑 − 𝐾𝑖 ∫0 𝑠𝑖𝑔𝑛(𝑆𝜂 )𝑑𝜎, (37)
where 𝜂̃ = 𝜂 − 𝜂𝑑 is the tracking error of the position, 𝜂𝑑 is the desired trajectory, 𝐾𝑖 is a
diagonal positive definite 𝑛 × 𝑛 gain matrix, 𝛼 is a gain yet to be defined, 𝑠𝑖𝑔𝑛(𝑥) is the
𝑠𝑖𝑔𝑛𝑢𝑚 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 of the vector 𝑥, and
𝑆 = 𝜂̃̇ + 𝛼𝜂̃, (38)
−𝑘𝑡
𝑆𝑑 = 𝑆(𝑡0 )𝑒 , (39)
𝑆𝜂 = 𝑆 − 𝑆𝑑 , (40)
with 𝑘 > 0.
Remark 3. Since 𝑆(0)𝑒 −𝑘𝑡 serves as an input of 𝜂̃̇ + 𝛼𝜂̃, a vanishing function 𝑆𝑑 is used
to make 𝑆𝜂 = 𝑆(0) − 𝑆𝑑 = 0 for 𝑡 = 0. 𝑆𝑑 aims to improve the transient response, and does not
affect the stability or the boundedness of the controller's signals. After a brief period slightly
greater than 0, 𝑆𝜂 becomes 𝑆 and would be bonded for all time.
Assumption 3. The nominal reference 𝜂̇ 𝑟 and its derivative 𝜂̈ 𝑟 are bounded by positive
scalars 𝛽𝑖 , 𝑖 = 6, … ,9 as follows:
||𝜂̇ 𝑟 || ≤ β6 + ||α|| ||𝜂̃|| + λM (K i )I < β7 ,
||𝜂̈ 𝑟 || ≤ β6 + ||α|| ||𝜂̃̇ || < β9 . (41)
The extended error 𝑆𝑟 can be rewritten as
𝑡
𝑆𝑟 = 𝑆𝜂 + 𝐾𝑖 ∫0 𝑠𝑖𝑔𝑛(𝑆𝜂 )𝑑𝜎 , (42)
and its derivative as
𝑆𝑟̇ = 𝑆𝜂̇ + 𝐾𝑖 𝑠𝑖𝑔𝑛(𝑆𝜂 ), (43)
from which the model-free high-order SMC is obtained with the following control law:
𝜏𝜂 = −𝐾𝑑 𝑆𝑟 , (44)
where 𝐾𝑑 is a diagonal definite positive 𝑛 × 𝑛 gain matrix.
44
Remark 4. Note that the controller does not require any knowledge of the hydrodynamics
or parameters of the vehicle.
Finally, to apply the control signals in the vehicle, the following transformation must
be applied
𝜏𝜐 = 𝐽−1 (𝜂2 )𝜏𝜂 . (45)
4.2 Time Parametrization of 𝜶 Gain
When the 𝛼 gain in Equation 38 is set to a constant value, the control law given in
Equation 44 leads to a second-order SMC with asymptotic convergence, as reported in
[187]. In the controller proposed for this work, a time-varying feedback gain 𝛼(𝑡) is
proposed to achieve finite time convergence of the tracking errors to a very small value
in a directly user-defined time. According to Parra-Vega [188], the 𝛼(𝑡) gain can be
defined as:
𝜉̇ (𝑡)
𝛼0 (1−𝜉(𝑡))+𝛿 , 0 ≤ 𝑡 ≤ 𝑡𝑏
𝛼(𝑡) = { , (46)
𝛼𝑐 , 𝑡 > 𝑡𝑏
where 𝛼0 = 1 + 𝜀, 0 < 𝜀 ≪ 1, 0 < 𝛿 ≪ 1, and 𝛼𝑐 > 0.
A time base generator (TBG) 𝜉(𝑡) provides a smooth 0 to 1 transition, the duration
of which can be arbitrairy defined by the user with a time-base parameter (𝑡𝑏 ). This TBG
is given by
(𝑡−𝑡0 )3 (𝑡−𝑡0 )4 (𝑡−𝑡0 )5
𝜉(𝑡) = 10 − 15 (𝑡 + 6 (𝑡 , (47)
(𝑡𝑏 −𝑡0 )3 𝑏 −𝑡0 )4 𝑏 −𝑡0 )
5
and its derivative

(𝑡 − 𝑡0 )2 (𝑡 − 𝑡0 )3 (𝑡 − 𝑡0 )4
𝜉̇ (𝑡) = 30 − 60 + 30 , (48)
(𝑡𝑏 − 𝑡0 )3 (𝑡𝑏 − 𝑡0 )4 (𝑡𝑏 − 𝑡0 )5
gives a bell-shaped velocity profile.
For Equation 47 and Equation 48, the following conditions hold: 𝜉(𝑡𝑏 ) = 1, 𝜉(𝑡0 ) =
𝜉̇ (𝑡0 ) = 𝜉̇ (𝑡𝑏 ) = 0, where 𝑡0 represents the initial time.
Finally, the solution of the differential equation in Equation 38 is

𝜂̃(𝑡) = 𝜂̃(𝑡0 )[1 − 𝜉(𝑡) + 𝛿]𝛼0 . (49)
Equation 49 represents a family of solutions converging smoothly to a small value.
Since 𝜉(𝑡𝑏 ) = 1 when 𝑡 = 𝑡𝑏 , the solution becomes
𝜂̃(𝑡𝑏 ) = 𝜂̃(𝑡0 )𝛿 𝛼0 , (50)
when the time base is reached.
Remark 5. The time base parameter 𝑡𝑏 can be arbitrarily chosen by the user and does
not depend on the vehicle's initial conditions, parameters, or hydrodynamics.
4.3 Stability Analysis
Theorem 1. In a closed-loop system, the control law described by Equation 44 and

the model described by Equation 21 lead to
𝑀𝜂 (𝜂)𝑆𝑟̇ = −𝐾𝑑 𝑆𝑟 − 𝑌(𝜂, 𝜂̇ , 𝜂̇ 𝑟 , 𝜂̈ 𝑟 )𝜃 − 𝐶𝜂 (𝜈, 𝜂)𝑆𝑟 − 𝐷𝜂 (𝜈, 𝜂)𝑆𝑟 , (51)
where finite-time tracking is guaranteed if 𝐾𝑑 and 𝐾𝑖 are large enough for small initial error
conditions.
45
Proof of Theorem 1. The stability analysis has been divided into two parts: Part I
proves the stability of tracking errors with all the closed-loop signals bounded. Part II
proves that the velocity and position tracking errors converge to zero.
Part I. Boundedness of the closed loop trajectories.
Consider the following Lyapunov candidate function
1
𝑉 = 2 𝑆𝑟𝑇 𝑀𝜂 (𝜂)𝑆𝑟 , (52)
and its time derivative wich yields to
𝑉̇ = −𝑆𝑟𝑇 𝐾𝑑 𝑆𝑟 − 𝑆𝑟𝑇 𝐷𝜂 (𝜈, 𝜂)𝑆𝑟 − 𝑌(𝜂, 𝜂̇ , 𝜂̇ 𝑟 , 𝜂̈ 𝑟 )𝜃,
≤ −𝑆𝑟𝑇 𝐾𝑑 𝑆𝑟 − ‖𝑆𝑟 ‖‖𝑌(𝜂, 𝜂̇ , 𝜂̇ 𝑟 , 𝜂̈ 𝑟 )𝜃‖,
≤ −𝑆𝑟𝑇 𝐾𝑑 𝑆𝑟 − ‖𝑆𝑟 ‖𝜌(𝑡), (53)
≤ −𝜆𝑚 (𝐾𝑑 )‖𝑆𝑟 ‖2 + ‖𝑆𝑟 ‖𝜌(𝑡),

where the skew-symmetric property described in Property 2 was applied and the norm of
𝑌(𝜂, 𝜂̇ , 𝜂̇ 𝑟 , 𝜂̈ 𝑟 )𝜃 was replaced by an upper bound defined by a state-dependent function
𝜌(𝑡).
According to Property 5 and Assumption 3, the desired trajectories and vehicle
dynamics are bounded—there are upper bounds for 𝑀𝜂 , 𝐶𝜂 , 𝐷𝜂 , 𝑔𝜂 , 𝜂̇ 𝑟 , 𝜂̈ 𝑟 —so it can be
proven that 𝑌(𝜂, 𝜂̇ , 𝜂̇ 𝑟 , 𝜂̈ 𝑟 )𝜃 is also upper-bounded [189]. If the initial error is small enough
and 𝐾𝑑 is large enough, one can infer the negative definiteness of outside of the small
sphere 𝜀0 = {𝑆𝑟 |𝑉̇ ≥ 0} centered at the origin 𝑉̇ (𝑆𝑟 ) = 0. This boundedness in the ℒ∞
sense leads to the existence of the constant 𝜀1 > 0, so that
||𝑆𝑟̇ || ≤ 𝜀1 . (54)
At this point, the stability of the tracking errors has been proven.
Part II. Existence of the second-order sliding mode.
Consider the second-order dynamical system defined by the time derivative of the
nominal reference described in Equation 43 as
𝑆𝜂̇ = −𝐾𝑖 𝑠𝑖𝑔𝑛(𝑆𝜂 ) + 𝑆𝑟 . (55)
𝑇
Now, consider the multiplication of Equation 55 by 𝑆𝜂 :
𝑆𝜂𝑇 𝑆𝜂̇ = −𝐾𝑖 𝑆𝜂𝑇 𝑠𝑖𝑔𝑛(𝑆𝜂 ) + 𝑆𝜂𝑇 𝑆𝑟 , (56)
applying the bound given by Equation 54 and considering 𝜇 = 𝜆𝑚 (𝐾𝑖 ) − 𝜀1 , results in
𝑆𝜂𝑇 𝑆𝜂̇ ≤ −𝜆𝑚 (𝐾𝑖 )|𝑆𝜂𝑇 | + |𝑆𝜂𝑇 |||𝑆𝑟̇ ||,
≤ |𝑆𝜂𝑇 |(−𝜆𝑚 (𝐾𝑖 ) + 𝜀1 ), (57)
𝑇
≤ −𝜇|𝑆𝜂 |.
If 𝜆𝑚 (𝐾𝑖 ) > 𝜀1 , then 𝜇 > 0. This guarantees the sliding mode at 𝑆𝜂 = 0 and 𝑡𝑔 =
|𝑆𝜂 (𝑡0 )|
. Note that 𝑡𝑔 = 0 for any initial condition 𝑆𝜂 (𝑡0 ), implying the enforcement of the
𝜇
sliding mode at any time and
𝑆𝜂 = 𝜂̃̇ + 𝛼𝜂̃ = 0 ↔ 𝜂̃̇ = −𝛼𝜂̃. (58)
Equation 58 implies the convergence of the tracking errors to a very small sphere
centered in the origin
(𝜂̃, 𝜂̃̇) = 0, (59)
in a finite-time (𝑡𝑏 ), as described in Equation 49 and Equation 50.
46
Remark 6. Considering the solution in Equation 58 for 𝑡 = 𝑡𝑏 , an 𝛼0 parameter very close
to 1 and a very small 𝛿, the tracking errors are limited to a very small neighborhood, 𝜀2 , from the
origin. In practice, this may represent required accuracy or a practical zero error. For 𝑡 > 𝑡𝑏 , the
time-varying gain 𝛼(𝑡) must be reset to a constant value 𝛼𝑐 > 0. Since a sliding mode is induced
at any time, 𝜂̃(𝑡) ∈ 𝜀2 , ∀ 𝑡 > 𝑡𝑏 , and 𝜂̃̇ = −𝛼𝑐 𝜂̃(𝑡), ∀ 𝑡 > 𝑡𝑏 , 𝜂̃(𝑡) converges exponentially, which
means that the tracking errors quickly tend to zero, leading to
η → ηd , η̇ → η̇ d , ∀ t > t b . (60)
Remark 7. Tuning of parameters 𝛼0 , 𝛿: It can be observed from Equation 49 that, given
an initial error condition 𝜂̃(𝑡0 ) = 0, the error 𝜂̃(𝑡) is zero no matter the value of 𝛼0 and 𝛿, since the
vehicle is already in the desired trajectory. For the cases where 𝜂̃(𝑡0 ) ≠ 0, a simple way to tune
these parameters is by fixing either 𝛼0 around 1 (slightly greater or smaller than 1) or 𝛿 > 0 very
close to zero. Then, given the initial error condition 𝜂̃(𝑡0 ), the desired practical zero error 𝜂̃(𝑡) at
𝑡 = 𝑡𝑏 , and by fixing 𝛿 in Equation 50, 𝛼0 can be computed as
𝜂̃(𝑡𝑏 )
𝛼0 = 𝑙𝑜𝑔𝛿 ( ), (61)
𝜂̃(𝑡0 )
or, by fixing 𝛼0 , 𝛿 can be computed as follows
𝛼0 𝜂̃(𝑡𝑏 )
𝛿 = √( ). (62)
𝜂̃(𝑡0 )
Notice that the practical zero error 𝜂̃(𝑡𝑏 ) is limited to the accuracy and resolution of the
measurement sensors.
4.4 Further Considerations
4.4.1 Reference frame transformation

The control signal 𝜏𝜂 must be transformed from the Earth-fixed frame to the Body-
fixed frame to obtain the forces and moments needed in the vehicle. This is achieved by
the following transformation
𝜏 = 𝐽−1 (𝜂2 )𝜏𝜂 . (63)
Then the coefficient vector 𝑢 of the thrusters is calculated as
𝑢 = 𝐵𝑡 −1 𝐾𝑇 −1 𝜏, (64)
where 𝐾𝑇 is a diagonal matrix containing the maximum force that each thruster can
deliver.
4.4.2 Saturation Constraints
Since the control signal vector 𝑢 is limited to −1 ≤ 𝑢 ≤ 1, its computation was
subjected to saturation constraints. For the first four elements of the 𝑢 vector,
corresponding to the four horizontal thrusters 𝑇1 , 𝑇2 , 𝑇3 , and 𝑇4 , if the major absolute value
is greater than one, 𝑢1 , 𝑢2 , 𝑢3 , and 𝑢4 are computed as
𝑢𝑖
𝑢𝑖 = , (65)
max (𝑢1 , 𝑢2 , 𝑢3 , 𝑢4 )
with 𝑖 = 1, 2, 3, 4.
Now for the elements 𝑢5 and 𝑢6 , corresponding to the vertical thrusters 𝑇5 and 𝑇6 ,
their value is just limited to −1 ≤ 𝑢𝑗 ≤ 1
𝑢𝑗 = 1, 𝑖𝑓 𝑢𝑗 > 1
𝑢𝑗 = { (66)
𝑢𝑗 = −1, 𝑖𝑓 𝑢𝑗 < −1
47
with 𝑗 = 5, 6. These saturation constraints were applied for the simulations with every
controller.
4.4.3 Exact differentiator
The tracking error 𝜂̃̇ of the vehicle’s velocity is needed to calculate the control law
defined in Equation 38. The sensors of BlueROV2 do not measure velocity, so the
velocities of the vehicle must be estimated. A simple Euler differentiator can be used for
this purpose. However, this differentiator is sensitive to noise and results in an inaccurate
estimation of the vehicle velocities. This problem was addressed by programming an
exact differentiator algorithm. This algorithm is based on the design by Levant [190] and
is given by:
𝑗̇0̇ = 𝑤0 , (67)
1
𝑤0 = −𝜆1 [𝑗0 − 𝑓(𝑡)]2 𝑠𝑖𝑔𝑛(𝑗0 − 𝑓(𝑡)) + 𝑗1, (68)
𝑗̇1̇ = −𝜆2 𝑠𝑖𝑔𝑛(𝑗0 − 𝑓(𝑡)), (69)
where 𝑓(𝑡) is the original signal to be differentiated, 𝑠𝑖𝑔𝑛(𝑥) is the signum function for the
argument 𝑥, and 𝜆1 = 1.5 𝜆2 = 1.1 are constant gains. After a brief adjustment time, 𝑗0 is
considered as the original filtered signal and 𝑗1 as its derivative.
𝑗0 = 𝑓(𝑡), (70)
̇
𝑗1 = 𝑓(𝑡), (71)
Remark 8. The exact differentiator accuracy and sensitivity to noise degrade as a
function of the increase in its parameters and sampling time. However, high-order differentiations
expect a drop in performance (refer to [190]). Since the differentiator in this work considers only
the first derivative, sensitivity to noise is not considered a problem.
The complete control scheme is shown in the block diagram in Figure 23.
Figure 23. Model-free second-order sliding mode control with finite-time convergence complete
block diagram.
48
Chapter 5: Controller validation for trajectory tracking
This chapter presents the efforts made to validate the proposed controller for the
trajectory tracking problem of a single AUV. The validation took place at a numerical
simulator, where it was compared with classic control algorithms. Then, the controller was
validated through experiments in a semi-Olympic water pool. Its performance was
compared with a PID and an asymptotic 2nd-order SMC. The hardware and software
modifications to the BlueROV2 required for experiments are also presented. The results
in this Chapter were published in the full research papers [191], [192]:
J. González-García et al., “Model-free high order sliding mode control with finite-time
tracking for unmanned underwater vehicles,” Appl. Sci., vol. 11, no. 4, pp. 1–22, 2021, doi:
10.3390/app11041836.
J. González-García, A. Gómez-Espinosa, L. G. García-Valdovinos, T. Salgado-Jiménez,

E. Cuan-Urquizo, and J. A. E. Cabello, “Experimental Validation of a Model-Free High-
Order Sliding Mode Controller with Finite-Time Convergence for Trajectory Tracking of
Autonomous Underwater Vehicles,” Sensors, vol. 22, no. 2, 2022, doi: 10.3390/s22020488.
5.1 Numerical simulator
The BlueROV2 hydrodynamics—as presented in Section 3.1—was programmed

in a simulator developed in MATLAB/Simulink® to test and validate the proposed control
scheme. This simulator also considers ocean currents as external disturbances, as
described at the end of Section 3.1. This simulator was used to perform all the numerical
simulations in this work. The block diagram of the Simulink workspace is shown in Figure
24.
Figure 24. BlueROV2 simulator. Simulink block diagram.
49
5.1.1 Time-parameterized trajectory definition
A spiral trajectory was selected to control all DOFs of the BlueROV2 (𝑥, 𝑦, 𝑧, 𝜑, 𝜓)
while moving simultaneously at surge, sway, heave, and yaw. The desired positions 𝜂𝑑
and velocities 𝜂̇ d are given by the following equations
𝜂𝑑 = (𝑥𝑑 , 𝑦𝑑 , 𝑧𝑑 , 𝜑𝑑 , 𝜓𝑑 ), (72)
𝜂̇ 𝑑 = (𝑥𝑑̇ , 𝑦𝑑̇ , 𝑧𝑑̇ , 𝜑𝑑̇ , 𝜓𝑑̇ ), (73)
with
𝑥𝑑 = 𝑥𝑖 + 𝑅 𝑐𝑜𝑠(𝜔𝑡),
𝑦𝑑 = 𝑦𝑖 + 𝑅 𝑠𝑖𝑛(𝜔𝑡),
𝑡
𝑧𝑑 = 𝑧𝑖 + (𝑧𝑓 − 𝑧𝑖 ) 𝑡 ,
𝑠
𝜑𝑑 = 0,
𝜓𝑑 = 𝜓𝑖 + 𝜔𝑡,
and
𝑥̇ 𝑑 = − 𝜔 𝑅 𝑠𝑖𝑛(𝜔𝑡),
𝑦̇ 𝑑 = 𝜔 𝑅 𝑐𝑜𝑠(𝜔𝑡),
(𝑧𝑓 −𝑧𝑖 )
𝑧̇𝑑 = ,
𝑡𝑠
𝜑̇ 𝑑 = 0, and
𝜓̇𝑑 = 𝜔,
where 𝑅 is the radius for the spiral trajectory, 𝜔 is a constant, (𝑥𝑖 , 𝑦𝑖 ) is the center of the
circumference of the spiral, 𝑧𝑖 is the initial depth for the trajectory, 𝑧𝑓 is the final depth of
the trajectory, ψi is the initial yaw, and 𝑡𝑠 is the simulation time.
𝜋 𝑟𝑎𝑑
The desired spiral trajectory with 𝑅 = 1 𝑚𝑒𝑡𝑒𝑟, 𝜔 = 6 𝑠 , (𝑥𝑖 , 𝑦𝑖 ) = (0,0), 𝑧𝑖 = 0,
𝑧𝑓 = 10 𝑚𝑒𝑡𝑒𝑟𝑠, and a simulation time 𝑡𝑠 = 30 𝑠 is plotted in Figure 25.
Figure 25. Parameterized desired spiral trajectory of the Unmanned Underwater Vehicle (UUV)
BlueROV2.
50
Zhang et al. [193] performed some experiments where high ocean currents
affecting a UUV were measured. Their results were used for simulation purposes in this
work, and the following ocean currents were introduced as constants
𝑚
𝑢𝑜𝑐 = 0.75 𝑠 ,
𝑚
𝑣𝑜𝑐 = 0.25 , and
𝑠
𝑚
𝑤𝑜𝑐 = 0.25 . (74)
𝑠
5.1.2 Comparing to classic controllers
Other classic control approaches, such as PID, Feed-Back Linearization (FBL),
and Lyapunov-based controllers [184], [194], will be simulated for the trajectory tracking
of the BlueROV2 to compare the performance of the proposed Model-Free Second-Order
SMC (MFSOSMC).
PID controller
The control signal used for the PID controller is given by
𝜏𝜂 = −𝐾𝑝 𝜂̃ − 𝐾𝑖 ∫ 𝜂̃ 𝑑𝑡 − 𝐾𝑑 𝜂̃̇, (75)
where 𝐾𝑝 , 𝐾𝑖 , and 𝐾𝑑 are the proportional, integral, and derivative gain matrices,
respectively. An initial set of gain matrices was proposed, and several simulations were
performed to find the one that produces the best results for the trajectory tracking of all
DOFs. As a result, the gain matrices for this controller are defined as
𝐾𝑝 = 𝑑𝑖𝑎𝑔[140,140,140,140,0,140],
𝐾𝑖 = 𝑑𝑖𝑎𝑔[120,120,120,120,0,120], and
𝐾𝑑 = 𝑑𝑖𝑎𝑔[180,180,180,180,0,180], (76)
Feed-back linearization controller
The control signal used for the FBL controller is given by:
𝜏 = 𝑀𝑎𝑏 + 𝐶(𝜈)𝜈 + 𝐷(𝜈)𝜈 + 𝑔(𝜂), (77)
𝑏
where 𝑎 is the commanded acceleration. Here, a PID controller will be used as an
auxiliary controller to compute 𝑎𝑏 as
𝑎𝑏 = 𝜈̇ 𝑑 − 𝐾𝑝 𝜈̃ − 𝐾𝑖 ∫ 𝜈̃ 𝑑𝑡 − 𝐾𝑑 𝜈̃̇, (78)
where
𝜈̃ = 𝜈 − 𝜈𝑑 , (79)
𝐾𝑝 , 𝐾𝑖 , and 𝐾𝑑 will take the same values as in Equation 76.
Lyapunov-based controller
The control signal used for the Lyapunov-based controller is given by:
𝜏 = 𝑀𝜈̇𝑟 + 𝐶𝜈𝑟 + 𝐷𝜈𝑟 + 𝑔(𝜂) − 𝐾𝑑 𝑆, (80)
where
𝑆 = 𝜈 − 𝜈𝑟 , (81)
−1
𝜈𝑟 = J 𝜂̇ 𝑟 , (82)
−1 ̇
𝜈̇𝑟 = 𝐽 [𝜂̈ 𝑟 − 𝐽𝜈𝑟 ], and (83)
𝜂̇ 𝑟 = 𝜂̇ 𝑑 − 𝛼𝜂̃, (84)
where K and 𝛼 are gain matrices which, after several iterations to find a suitable
performance of the controller, were defined with the following values
𝐾𝑑 = 𝑑𝑖𝑎𝑔[80,80,80,80,0,80], and
𝛼 = [0.80,0.80,0.80,0.80,0,0.80]. (85)
51
The values for the gain matrices used in the simulations of this work are
summarized in Table 8.
Table 8. Gain matrices for classic controllers and the proposed model-free second-order SMC with
finite-time convergence.
Controller 𝑲𝒑 𝑲𝒊 𝑲𝒅 𝜶 𝜿
PID 140 120 180 - -
Feed-back linearization 140 120 180 - -
Lyapunov-based - - 80 0.80 -
Model-free second-order SMC with finite-time convergence - 5 800 𝜶(𝒕) 5
5.1.3 Results and discussion

A simulation was performed so that a computational model of the BlueROV2
follows the spiral trajectory described before. The initial position for the vehicle was set to
𝑥=2 𝑚𝑒𝑡𝑒𝑟𝑠, 𝑦=−1 𝑚𝑒𝑡𝑒𝑟, and 𝑧=2 𝑚𝑒𝑡𝑒𝑟𝑠 so it started slightly away from the desired
trajectory. The effects of ocean currents were neglected for this simulation. The
parameters and gains for the proposed MFSOSMC with finite-time convergence are
shown in Table 9.
Table 9. Controller parameters for simulation of the desired spiral trajectory.
Parameter Value Parameter Value

𝒕𝟎 0 𝜹 0.001
𝒕𝒃 5 𝜿 5
𝛂𝟎 1.01 𝑲𝒊 𝒅𝒊𝒂𝒈[𝟎, 𝟓, 𝟓, 𝟓, 𝟓, 𝟓]
𝛂𝒄 20 𝑲𝒅 𝒅𝒊𝒂𝒈[𝟖𝟎𝟎, 𝟖𝟎𝟎, 𝟖𝟎𝟎, 𝟖𝟎𝟎, 𝟎, 𝟖𝟎𝟎]
Results for trajectory tracking and errors of the proposed controller are presented
in Figure 26. The controlled trajectory smoothly converged with the desired trajectory at
the time-base proposed without overshoots. After that, the vehicle followed the desired
trajectory without deviations. The errors decreased over time to become zero at the
designated 𝑡𝑏 except for the pitch, which, as established before, was not actuated, and
the resulting increase in its error—smaller than 0.5°—was caused by the vehicle’s
dynamics while moving forward.
52
Figure 26. Results with 𝑡𝑏 = 5 𝑠. (a) Trajectory tracking. (b) Position tracking errors.
A three-dimensional trajectory of the BlueROV2 simulation and the control

trajectory can be seen in Figure 27.
Figure 27.The simulated three-dimensional trajectory of the BlueROV2 with 𝑡𝑏 = 5 𝑠.
The control coefficients for each of the BlueROV2 thrusters are shown in Figure
28. These control coefficients were −1 ≤ 𝑢 ≤ 1, where |𝑢| = 1 represents that the
controller demands 100% of the thruster force. The maximum power was demanded at
the start of the simulation, and after reaching the time-base, the control signals remained
constant, and no chattering was observed.
53
Figure 28. Control signal for each thruster. Simulation with 𝑡𝑏 = 5 𝑠.
Since 𝑡𝑏 can be tuned arbitrarily for the user, further simulations were performed
for the proposed controller under the same gains and parameters but for different time-
bases. Simulation results for trajectory tracking and errors are shown in Figure 29 and
Figure 30 for a time-base of 3 and 7 seconds, respectively. In both cases, the vehicle met
the desired trajectory at the designated time-base 𝑡𝑏 .
54
The controllers presented and described in Section 5.1.2 were simulated to

compare their error convergence to zero and the control coefficient computed for the
BlueROV2 thrusters. As can be seen in Figure 31, the surge and sway error of the
proposed MFSOSMC converges to zero at the desired time. The rest of the controllers
converged at different times. In PID and FBL, there was no stationary error, but an
oscillation in its value was observed through the simulated time. All controllers converged
to zero for the heave error, despite the oscillations resulting in the PID and the FBL.
Finally, the FBL exhibited some troubles converging to zero during the first 20 seconds
for the yaw orientation but finally did it. The performance of the proposed MFSOSMC with
finite-time convergence surpassed the rest of the controllers, even when FBL and
Lyapunov-based were model-based.
55
Figure 31. Error convergence for different controllers. (a) Surge error. (b) Sway error. (c) Heave error. (d)
Yaw error.
The Root Mean Square (RMS) value of the control coefficients computed for the
thrusters was obtained to evaluate the performance of the proposed controller regarding
energy consumption from the thrusters. The results are shown in Table 10. The mean
column of the table represents the mean value of the coefficients considering all six
thrusters. As can be seen in the table, the average thruster force demanded by the
proposed MFSOSMC was in the range of 0.1004 and 0.1296—1.0 being the maximum—
this was up to 50% less than the forces demanded by a traditional PID or FBL controller
even when it converged faster. The Lyapunov-based controller converged to the
trajectory approximately at 7 s, and its mean value for the control coefficient of the
thrusters was 23% bigger than the 2nd Order SMC with t 𝑏 = 7 𝑠.
Table 10. Root Mean Square (RMS) values for thrusters control coefficients.
Controller Thruster 1 Thruster 2 Thruster 3 Thruster 4 Thruster 5 Thruster 6 Mean

PID 0.2078 0.1946 0.2010 0.2105 0.1311 0.2041 0.1915
Feed-back linearization 0.1676 0.1876 0.1909 0.1693 0.2851 0.2760 0.2127
Lyapunov-based 0.1278 0.1500 0.1380 0.1192 0.0678 0.1358 0.1231
Model-free 2nd Order SMC with 𝑡𝑏 = 3 𝑠 0.1449 0.1400 0.1333 0.1378 0.0801 0.1418 0.1296
56
The effects of the ocean currents described in Equation 74 were included to test
the controller response to external disturbances. Results for trajectory tracking and errors
of a simulation with the proposed MFSOSMC with 𝑡𝑏 = 5 𝑠 are shown in Figure 32.
Despite the high ocean currents introduced, the BlueROV2 still smoothly converged to
the desired trajectory at the designated 𝑡𝑏 , there were not overshoots at all, and the
vehicle followed the designed trajectory with an error of zero after the time-base was
reached.
Figure 32. Results for the finite-time convergence controller subject to ocean currents and 𝑡𝑏 = 5 𝑠. (a)
Trajectory tracking. (b) Position tracking errors.
5.2 Experimental validation
As concluded in the literature review, the state-of-the-art regarding advanced

control algorithms for AUVs is highly limited to numerical simulations. Therefore,
validating the proposed controller through experimentation is a strong differentiator of this
research. This Section presents those experiments, their setup, and the results achieved.
5.2.1 Experimental setup
The proposed MFSOSMC with finite-time convergence was tested in different
experiments, and its performance was compared with traditional PID and an asymptotic
second-order SMC. The BlueROV2 was modified in its hardware and software to follow
predefined depth and yaw trajectories autonomously. The experimental setup can be
observed in Figure 33.
57
Figure 33.Experimental set-up at Tecnologico de Monterrey, Campus Querétaro. (a) Ground control
station at a side of the semi-Olympic pool. (b) BlueROV2 deployed into the water. (c) Autonomous
trajectory tracking mission.
The experiments were performed in the Tecnologico de Monterrey Campus

Queretaro semi-Olympic swimming pool. All the runs were made in the same robot—
BlueROV2—using three full-charged batteries—one for every control scheme—from 9:00
a.m. to 12:00 p.m. This intends to perform the experiments under the same conditions.
An OMENO15 Laptop with an Intel CORE i7 processor running UBUNTU 16 as the
operative system was used as the ground control station. No external disturbances were
introduced to the experiments.
BlueROV2 hardware
Inside the BlueROV2, a Raspberry Pi® 3 (RPi) acts as the processor, and it runs
the control algorithms and manages the different sensors. The Rpi runs Lubuntu as an
operative system and is connected to a control station through a tether. A BAR-30 high-
resolution pressure sensor from BlueRobotics® is connected to the robot through an I2C
port in the RPi to measure the depth 𝑧 of the vehicle. A smart sensor BNO-055 from
Bosh® is connected to a serial port of the Rpi and is used to estimate the orientations
𝜑, 𝜃, 𝜓. The thruster’s speeds are controlled by Pulse Wide Modulation (PWM) signals
from the RPi, which goes through a set of 30 A Electronic Speed Controllers (ESC).
Finally, a 14.8 V, 18A Ah battery supplies energy for the robot. A diagram of the hardware
configuration in the BlueROV2 used for the experiments is shown in Figure 34.
58
Figure 34. BlueROV2 hardware configuration
BlueROV2 software
In its Kinetic version, ROS was used to program the algorithms to manage the
sensors, thruster’s speed, and control schemes. Five nodes were used in total, as shown
in Figure 35.
Figure 35. BlueROV2® software configuration.
The IMU sensor node (1) manages the BNO-055 sensor, and it can obtain a new
set of 𝜑, 𝜃, and 𝜓 orientations at a 100 Hz rate. The Pressure Sensor node (2) manages
the Bar-30 sensor and obtains a new 𝑧 position at a 10 Hz rate. Both nodes publish the
estimated positions to be used as inputs to the Control Algorithm node (3). In this node,
the different control algorithms were programmed along with a user interface to allow the
user to select between the PID control, the asymptotic 2nd-Order SMC, and the
MFSOSMC with finite-time convergence. The user can also modify the controller
parameters and gains from this interface. The desired position is generated as defined by
the user throughout the user interface—either as a set point or as a time-variant—and
compared with the actual position estimated by the sensors. Then, the selected control
law is applied, and a thruster’s coefficients vector is computed and published. The
59
Thruster Management node (4) reads the thruster’s coefficients vector and converts it to
the proper PWM signal for each thruster to be generated by the RPi peripherals. The
Manual Control node (5) is used before the experiment to drive the BlueROV2 to its initial
position.
5.2.2 Time-parameterized trajectory definition
A sinusoidal trajectory was chosen for the vehicle to track both depth (𝑧𝑑 ), and yaw
(𝜓𝑑 ). Those trajectories are given by
𝑧𝑑 = 𝑧𝑖 + 𝐴𝑧 sin(𝜔𝑡), and
𝜓𝑑 = 𝜓𝑖 + 𝐴𝜓 sin (𝜔𝑡). (86)
where 𝑧𝑖 and 𝜓𝑖 are the initial desired position or offset, 𝐴𝑧 and 𝐴𝜓 are the desired
amplitudes, and 𝜔 is the frequency for the signal. Then, the desired velocities 𝑧̇𝑑 and 𝜓̇𝑑
are
𝑧̇𝑑 = −𝜔𝐴𝑧 cos(𝜔𝑡), and
𝜓̇𝑑 = −𝜔𝐴𝜓 (𝜔𝑡). (87)
5.2.3 Comparing with other controllers
Experiments were carried out using three different control laws for the trajectory
tracking of the robot. The aim was to compare the performance of the proposed algorithm
with other common controllers used for this purpose.
PID controller
The control law used for this controller is given by
𝜏𝜂 = −𝑘𝑝 𝜂̃ − 𝑘𝑖 ∫ 𝜂̃ 𝑑𝑡 − 𝑘𝑑 𝜂̃̇, (88)
where 𝑘𝑝 , 𝑘𝑖 , and 𝑘𝑑 are the proportional, integral, and derivative gain matrices,
respectively. For the experiments in this work, the gain matrices that gave the best results
are
𝑘𝑝 = 𝑑𝑖𝑎𝑔[0,0,100,100,0,100],
𝑘𝑖 = 𝑑𝑖𝑎𝑔[0,0,1,1,0,1],
𝑘𝑑 = 𝑑𝑖𝑎𝑔[0,0,80,10,0,80] (89)
Model-free second-order SMC with asymptotic convergence
The control law used for this controller is given by
𝜏𝜂 = − 𝑘𝑑 𝑆𝑟 , (90)
with
𝑡
𝑆𝑟 = 𝑆𝜂 + 𝐾𝑖 ∫0 𝑠𝑖𝑔𝑛(𝑆𝜂 )𝑑𝜎 ,
𝑆𝜂 = 𝑆 − 𝑆𝑑 ,
𝑆𝑑 = 𝑆(𝑡0 )𝑒 −𝜅𝑡 , and
𝑆 = 𝜂̃̇ + 𝛼𝜂̃, (91)
where 𝛼 > 0, 𝜅 > 0, 𝑘𝑖 , and 𝑘𝑑 are constant gains, 𝑡0 is the initial time, and 𝑠𝑖𝑔𝑛(𝑆𝜂 ) is
the sign function of 𝑆𝜂 . For the experiments in this work, the gain set that gave the best
results is
𝛼 = 10, 𝜅 = 5,
𝑘𝑖 = 𝑑𝑖𝑎𝑔[0,0,0.05,0.005,0,0.001], and
𝑘𝑑 = 𝑑𝑖𝑎𝑔[0,0,100,0.1,0,0.1] (92)
60
5.2.4 Results and discussion
The control law used for the proposed MFSOSMC controller with finite-time
convergence in these experiments is given by Equation 90. The 𝛼 gain in Equation 91 is
substituted by the time-variant gain 𝛼(𝑡) as defined in Equation 46 to achieve finite-time
convergence. The parameters used were 𝛼𝑐 = 10 for depth control, 𝛼𝑐 = 5 for heading
control, and, in both cases, 𝛼 0 = 1.005 and 𝛿 = 0.001. The rest of the gain parameters
(𝜅, 𝑘𝑖 , and 𝑘𝑑 ) stays as defined in Equation 92. The BlueROV2 was programmed with
the desired depth and heading trajectories described in Equation 86 and the desired
velocities described in Equation 87. The desired roll angle was set to 𝜙 = 0°, the initial
desired positions were set to 𝑧𝑖 = 0.50 m and 𝜓 = 270°, the desired amplitudes were 𝐴𝑧 =
1
0.25 m and 𝐴𝜓 = 30°, the period was set to 𝑇 = 𝜔 = 20 s. Results for different runs with
the PID controller implementation are shown in Figure 36.
Figure 36. Controlled trajectories with the PID controller. (Left) Depth. (Right) Heading.
The best result for the PID controller is shown in Figure 37. A significant error can
be observed in the depth trajectory control. The heading trajectory control error is
significantly minor, but there is an overshoot in its transitory response.
Figure 37. Controlled trajectories with the PID controller.
The same experiment was performed by applying the second-order SMC with
asymptotic convergence. Results for trajectory tracking of 𝑧 and 𝜓 in different runs of the
experiment are shown in Figure 38. The graph in Figure 39 includes the best performance
61
of the asymptotic 2nd-order SMC, there is a faster response compared with the PID
controller, and the error once the robot meets the trajectory is also smaller.
Figure 38. Controlled trajectories with the asymptotic model-free 2nd-order SMC. (Left) Depth. (Right)
Heading.
Figure 39. Best controlled trajectories with the asymptotic model-free 2nd-order SMC.
Results for the trajectory tracking by the MFSOSMC with finite-time convergence
in a predefined time-base of 5 s are shown in Figure 40. Note that, at the beginning of the
controlled trajectory, the BlueROV2 approaches slowly to the desired trajectory and then
accelerates to reduce the error to practically zero in the selected time-base of 5 s. After
the time-base is reached, the controller maintains the robot in the desired trajectory with
a minimum RMSE (< 1 cm and <2.3°), as is shown in Figure 41. Those errors can be
attributed to hardware limitations such as sensor accuracy, resolution, measurement
rates, processor speed, and communication baud rates, which limit the control cycle
frequency.
62
Figure 40. Controlled trajectory with the model-free high-order SMC with finite-time convergence in a 5 s
predefined-time.
Figure 41. Tracking error for depth and heading control.
The resulting TBG for this experiment is shown in Figure 42. As described in
Equation 47 and Equation 48, there is a smooth transition from zero to one completed
precisely at the given time-base. The resulting time-varying gain 𝛼(𝑡) is shown in Figure
43, as described in Equation 46. Several runs of the same experiment were performed
with the proposed MFSOSMC with finite-time convergence defining different time-bases.
The results of the controlled depth and yaw trajectories are shown in Figure 44. It can be
observed that the controller drives the robot to the desired trajectory to match it in the
desired time-base for all cases.
63
Figure 42. Time base generator results for 𝑡𝑏 = 5 s.
Figure 43. Time-varying gain α(t) result for 𝑡𝑏 = 5 s.
Figure 44. Controlled trajectories with the model-free high-order SMC with finite-time convergence in
predefined-time. (Left) Depth. (Right) Heading.
Velocities estimations are another source of noise and, therefore, errors in the
trajectory tracking problem for the SMC. The exact differentiator described in Section
4.4.3 was used for the asymptotic and finite-time SMC experiments to reduce these
errors. The results of the 𝑧̇ and 𝜓̇ estimations for a finite-time controlled trajectory are
64
shown in Figure 45. As can be observed, the uncertainty in the estimations was reduced
significantly using the exact differentiator.
Figure 45. Velocities estimations by Euler and Exact differentiators.
The RMSE was computed as a performance indicator of the different controllers

applied. As the error varies during the convergence, the RMSE was calculated after the
robot was in the desired trajectory. For the finite-time controller, the robot is considered
in the trajectory when 𝑡 > 𝑡𝑏. For the PID and the asymptotic SMC controllers, the robot
is considered in trajectory once the controlled trajectory intersects with the desired
trajectory. Results for the different control approaches are shown in Figure 46.
Figure 46. RMSE comparison for the trajectory tracking with the different controllers. (Left) Depth. (Right)
Heading.
In the depth trajectory tracking, the mean RMSE —considering all the runs of the
experiment —is 4.07 cm for the PID control (PID), 1.68 cm for the asymptotic 2nd-order
SMC (SMC), and the best was the high-order SMC with finite-time convergence (SMC-
FT) with a value of 1.03 cm. For the yaw trajectory tracking, the mean RMSE was 6.29°
for the PID control, 4.58° for the asymptotic 2nd-order SMC, and the highest was the high-
order SMC with finite-time convergence with a value of 2.72°. Another performance
indicator is the energy consumption in the thrusters. A control coefficient of +1.0
represents the maximum thrust in one direction, and a control coefficient of −1.0
represents the maximum thrust in the opposite direction. In both cases, the power
consumption is approximately 625 W. A higher magnitude of this coefficient means more
65
energy consumption from the corresponding thruster. The control coefficient for the
different thrusters as a result of the experiments shown in Figure 37, Figure 39, and Figure
40 are shown in Figure 47.
Figure 47. Control coefficients on the thrusters for the different control approaches.
The resulting control coefficients for thrusters 1, 2, 3, and 4 were identical in

magnitude. The same occurred for thrusters 5 and 6. The RMS values of the control
coefficients were computed as a performance indicator as follows
1
𝑢𝑅𝑀𝑆 = √𝑁 ∑𝑁 2
𝑖=1 𝑢𝑖 , (93)
where N is the number of samples.
The RMS value of the control coefficient was 0.155 for the PID controller, 0.206 for
the asymptotic SMC, and 0.101 for the finite-time SMC with 𝑡𝑏 = 5 𝑠, considering the six
thrusters. The finite-time controller could drive the vehicle to the desired trajectory with
35 % less energy demanded than the PID controller, even when they reach the desired
trajectory simultaneously. The asymptotic SMC was faster in its approach to the desired
trajectory, resulting in an energy demand twice as much as the finite-time controller's
demand.
As can be observed in the different performance indicators, the finite-time SMC
outperforms the asymptotic SMC, even when they were programmed with the same
parameters and gains (except for 𝛼(𝑡)). This is a clear example of the advantages of
using a finite-time convergence solution. Performing a tuning on the PID or asymptotic
SMC to achieve a faster or slower convergence will also modify the controller's
performance in the trajectory tracking. This does not happen with the finite-time controller,
where a simple change in the 𝑡𝑏 parameter will lead to a faster or slower convergence
maintaining its performance and robustness. The user benefits from this feature to obtain
the best controller performance according to the task's demands.
5.3 Chapter conclusions
Finite-time trajectory tracking of underactuated AUVs is a considerable challenge.

Their nonlinear dynamics and the challenging underwater environment conditions make
it more complex for the controller to achieve a stable and robust performance as they do
in other systems and environments. The model-free second-order SMC with finite-time
convergence synthesized in Chapter 4 was implemented for an underactuated nonlinear
66
AUV to track a desired trajectory. Such a controller has some advantages compared with
other finite-time model-based first-order SMC reported in the literature since it neither
requires the dynamic model structure nor the knowledge of any hydrodynamical
parameter from the vehicle. In addition, this controller does not present the chattering
effect common in other conventional SMC. The TBG was included in the controller by the
parametrization of the 𝛼 gain at the sliding surface, making it time-varying instead of
constant.
Results of the simulations have shown that the controller managed the BlueROV2
to smoothly converge to the desired trajectory at a specific time without overshoots and
then follow it with a practical zero tracking error. Other classic controllers, such as PID
and FBL, could not perform similarly. Since the time-base can be tuned arbitrarily for the
user, simulations were repeated with different time-bases, and the results remained
consistent and satisfactory. The proposed solution's performance was also evaluated
regarding the energy consumption demanded by the different controllers. It was shown
that the proposed MFSOSMC demands up to 50% less power from the thrusters than
other controllers simulated and did not present the chattering effect common in some
SMC controllers. Finally, the vehicle was subject to high ocean currents. Simulation
results showed that the proposed controller could still meet the desired trajectory at the
designated time, smoothly and without overshoots, and then maintain the vehicle in the
desired trajectory when other classic controllers could not.
The proposed MFSOSMC was also tested and validated through experimentation
in a semi-Olympic swimming pool using a BlueROV2 robot. This robot was modified and
instrumented to follow time-parametrized trajectories at depth and orientation. Results
showed that the MFSOSMC with finite-time convergence could drive the robot to the
desired trajectory in a predefined time and maintain it there with a small error for all the
experiment runs. The controller's performance was then compared with a PID controller
and an asymptotic 2nd-order SMC in the RMSE. The mean depth error after the total of
the experiments was 75% smaller for the finite-time convergence controller compared
with the PID control.
Additionally, it was 38% smaller than the asymptotic 2nd-order SMC, which kept
the same parameters and gains of the finite-time controller except for the 𝛼 gain.
Regarding the yaw trajectory tracking, the same indicator was 57% and 41% smaller for
the finite-time controller than the PID and asymptotic SMC, respectively. Although it was
expected that the finite-time controller would outrun the PID control performance, the fact
that it outperformed its asymptotic equivalent demonstrates the advantages of a finite-
time convergence controller in the AUV navigation problem. Another advantage of using
this finite-time controller is the smaller energy consumption. The RMS thruster coefficient
was 35% smaller when compared with the PID solution, with both controllers converging
simultaneously. It was also half of the RMS coefficient computed for the asymptotic SMC.
This controller will be handy for collaborative AUVs navigation considering scenarios
where the communication between AUVs is limited or does not exist.
67
Chapter 6: Controller validation for station-keeping
This Chapter contains the results of the numerical simulations and the experiments
performed to validate the proposed MFSOSMC controller for the station-keeping problem.
In both scenarios, there are two test phases: autonomous navigation to the desired
position and station-keeping for the BlueROV2 position. The discussion in this Chapter
focuses on the station-keeping phase of the tests. The results in this Chapter were
published in the full research paper [195]:
J. González-García, A. Gómez-Espinosa, L. G. García-Valdovinos, T. Salgado-Jiménez,

E. Cuan-Urquizo, and J. A. E. Cabello, “Model-Free High-Order Sliding Mode Controller
for Station-Keeping of an Autonomous Underwater Vehicle in Manipulation Task:
Simulations and Experimental Validation,” Sensors, vol. 22, no. 12, p. 4347, 2022, doi:
10.3390/s22124347.
6.1. Numerical simulator
For initial validation of the proposed controller, numerical simulations were

performed in the simulator described in Section 5.1. The controller parameters used in
this work are the same as those used to validate the trajectory tracking problem. The
sampling time for these simulations was set as a variable with a maximum value of 0.01
s.
The simulations consisted of the autonomous navigation of the BlueROV2 to reach
the desired 𝜂𝑑 position in a predefined time-base, arbitrarily set to 𝑡𝑏 = 8 𝑠. After the
vehicle is in the reference, the station-keeping phase of the simulation begins. In the
𝑚 m
station-keeping phase, the external disturbances 𝑢𝑜𝑐 = 0.75 𝑠 , 𝑣𝑜𝑐 = 0.25 s , and 𝑤𝑜𝑐 =
m
0.25 s were introduced at 𝑡 = 10 𝑠 as ocean currents. Then they were removed at 𝑡 =
18 𝑠. The simulation results are shown in Figure 48 for the 𝑥, 𝑦, 𝑧 positions and 𝜓
orientation. It can be observed that the controller can keep the BlueROV2 in reference for
all positions and orientations. Contrary to the simulation results reported in [18], [160],
[161], there is no initial push on the positions when the disturbance is introduced, and the
tracking error remains zero for all simulations. This demonstrates the robustness of the
controller to overcome external disturbances quickly and effectively.
68
Figure 48. Simulation results for finite-time trajectory tracking and station-keeping in the 𝑥, 𝑦, 𝑧 positions
and 𝜓 orientation. External disturbances were introduced in the interval 10 𝑠 ≤ 𝑡 ≤ 18 𝑠 as ocean currents
(𝜈𝑜𝑐 ).
6.2. Experimental validation
Experiments in a semi-Olympic swimming pool were performed to evaluate the

proposed control scheme. An external disturbance was introduced to the system; then,
the vehicle was asked to remain at a reference position in depth. The external disturbance
was kept unknown to the vehicle controller. This Section contains the experimental setup,
results, and discussion of the experiments performed.
6.2.1. Experimental setup
The BlueROV2 platform used in the experiments shown in Section 3.2 was
modified in its hardware and software to introduce an unknown external disturbance
directly in its z-axis and to perform the station-keeping experiments. The experiments
were performed in a semi-Olympic swimming pool at the Tecnologico de Monterrey,
Campus Queretaro. The location and experimental setup are shown in Figure 49.
69
Figure 49. Experimental set-up. (A) Semi-Olympic swimming pool. (B) Control station. (C)
BlueROV2 deployment. (D) Station-keeping task execution.
BlueROV2 hardware
The hardware configuration used in the experiments is shown in Figure 50. A
laptop with an Intel® CORE i7 processor and UBUNTU 16 as the operating system was
used as the control station. A tethered cable and a Fathom-X interface from BlueRobotics
were used to gain remote access to the RPi onboard the BlueROV2. This RPi is the
vehicle’s processor that runs the control algorithm and manages the sensors and
actuators. It runs Lubuntu as the operational system. A Bar-30 high-resolution pressure
sensor by BlueRobotics was used to estimate the depth of the vehicle. It is wired to the
Rpi via an I2C interface. The thrusters' velocities are controlled by PWM signals sent from
the RPi to a set of 30A ESC. Finally, a 14.8 V, 18A Ah battery provides power for all the
electronics.
Figure 50. Experimental set-up, hardware configuration.
An additional thruster was added to the BlueROV2 configuration to act as an

unknown external disturbance. This thruster was placed along the z-axis of the vehicle,
as shown in Figure 51.
70
Figure 51. Additional thruster for external disturbance. A) Front view diagram. B) Implementation.
BlueROV2 software
The software was implemented using ROS. The kinetic version of ROS was
installed on the RPi, and five nodes were programmed in Python to manage the system.
A pressure sensor (1) node manages the Bar-30 sensor and estimates the depth of the
robot. Then it publishes the 𝑧 position of the vehicle at a 100 Hz rate. A control algorithm
(2) node reads the 𝑧 position, runs an exact differentiator to estimate the velocity 𝑧̇ ,
contains the controller parameters, executes the control algorithm, and provides a simple
user interface. This node works at a 100 Hz frequency and publishes the thruster’s
coefficient vector 𝑢 to the thruster management (3) node, which generates the PWM
signals to control the thrusters. The external disturbance (4) node introduces the unknown
external disturbance into the system. It does not share the magnitude or direction of the
external disturbance with the controller. Finally, a manual control (5) node was
programmed to manually move the vehicle to its initial position. This software
configuration is shown in Figure 52.
Figure 52. Experimental set-up, software configuration.
6.2.2. Results and discussion

The proposed controller was validated by a series of experiments conducted in a
semi-Olympic swimming pool. The controller parameters used in these experiments were
the same as those used in the experimental validation for the trajectory tracking problem.
71
The aim is to demonstrate that the controller can maintain the 𝑧 position of the vehicle in
the presence of unknown external disturbances without adjusting its parameters. The
sampling time for the control algorithm was set at 0.01 s.
The experiments consisted of the autonomous navigation of the BlueROV2 to
reach the desired 𝑧 position in a predefined time-base, arbitrarily set to 𝑡𝑏 = 8 𝑠. After the
vehicle was in the reference, the station-keeping phase of the experiments began. An
experiment was conducted to compare the controller's performance in the station-keeping
phase without external disturbances. This experiment is referred to as the control test.
The depth position and tracking error results of the control test are shown in Figure 53.
Figure 53. Depth (left) and tracking error (right) results of the control test. No external disturbances were
introduced.
The control signal 𝜏𝑧 and the coefficients of the vertical thrusters 𝑢5 and 𝑢6 are
shown in Figure 54.
Figure 54. Experimental results of the control test for depth station-keeping. Control signal 𝜏𝑧 (left) and
thruster coefficients 𝑢5 ,𝑢6 (right). No external disturbances were introduced.
An external disturbance—unknown to the controller—was introduced in the

following experiments by varying the additional thruster coefficient. The coefficient range
of [-1,1] corresponds to a force between −40 to +50 N along the 𝑧 − 𝑎𝑥𝑖𝑠. Some open-
loop station-keeping tests were performed to demonstrate the effects of the proposed
disturbance. The results of these tests are shown in Figure 55. The additional thruster
disturbance drives the BlueROV2 to the bottom or top of the pool in less than 5 s.
72
Figure 55. Results of the open-loop experiments. Depth of the vehicle (upper left) with an external
disturbance of ~50 𝑁 (upper right) introduced at 𝑡 = 13 𝑠. Depth of the vehicle (lower left) with an external
disturbance of ~ − 40 𝑁 (lower right) introduced at 𝑡 = 13 𝑠.
A series of experiments were performed for station-keeping in the depth of the

BlueROV2. The results for the 𝑧 position and tracking error of the experiment with an
external disturbance of about −10 N (with an additional thruster coefficient of −0.25) are
shown in Figure 56. The resulting RMSE was 1.26 cm, which is quite similar to the RMSE
from the control test. This means the controller can maintain its performance in the
presence of unknown external disturbances of the same magnitude as in the simulations
in related work [160].
73
Figure 56. Experimental results for depth station-keeping with an external disturbance of ~10 𝑁
introduced in the interval 13 𝑠 ≤ 𝑡 ≤ 25 𝑠. Depth (left) and tracking error (right).
The control signal 𝜏𝑧 and the coefficients of the vertical thrusters 𝑢5 and 𝑢6 are
shown in Figure 57. The control signal increased its mean value by 86% compared to the
control test. This higher demand on the controller is evidence of the significance of the
introduced external disturbance.
introduced in the interval 13 𝑠 ≤ 𝑡 ≤ 25 𝑠. Control signal 𝜏𝑧 (left) and thruster coefficients 𝑢5 ,𝑢6 (right).
The coefficient for the additional thruster is shown in Figure 58. The disturbance is
introduced at 𝑡 = 13 𝑠 and maintained until 𝑡 = 25 𝑠.
74
Figure 58. Additional thruster coefficient of −0.25 in the interval 13 𝑠 ≤ 𝑡 ≤ 25 𝑠.
The results for the 𝑧 position and tracking error of the experiment with an external
disturbance of about −20 N (with an external thruster coefficient of −0.50) are shown in
Figure 59. Note that this external disturbance force is twice that of the simulations in [160].
The tracking error increases during the period when the external disturbance is
introduced. The RMSE is 2.4 cm for the station-keeping phase of the experiment, which
is twice as high as the RMSE of the control test. Nevertheless, this indicator is still low
compared to the experimental results reported in [161].
As can be seen in Figure 60, the control signal 𝜏𝑧 and thus the coefficients 𝑢5 and
𝑢6 increase their mean value. This is due to the controller requesting more power from
the thrusters to overcome the unknown external disturbance trying to push the vehicle out
of its reference.
75
The control signal in this experiment is almost five times larger than the same
signal in the control test. The thruster coefficients are almost three times as large as the
thruster coefficients in the control test. The coefficient of the additional thruster for this
experiment is shown in Figure 61.
Figure 61. Additional thruster coefficient of −0.50 in the interval 13 𝑠 ≤ 𝑡 ≤ 25 𝑠.
Some experiments were performed with time-varying external disturbances, such

as the one shown in Figure 62.
76
Figure 62. Time-varying additional thruster coefficient in the interval 13 𝑠 ≤ 𝑡 ≤ 25 𝑠. Minimum value
−0.25 and maximum value 0.25.
In these experiments, the external disturbance changed abruptly in magnitude and

direction. However, these changes did not significantly affect the controller's
performance, as shown in Figure 63 and Figure 64.
Figure 63. Experimental results for depth station-keeping with a time-varying external disturbance
77
Figure 64. Experimental results for depth station-keeping with a time-varying external disturbance
Eighteen experiments were performed, including the ones mentioned earlier, and
compared with the control test. Results for the thrusters’ coefficient and depth RMSE are
summarized in Table 11.
Table 11. Experimental results for the vertical thruster coefficients in the station-keeping phase.
Thrusters’
Experiment 𝒖𝟓 𝒖𝟔 𝑹𝑴𝑺 Depth RMSE (m) Disturbance Thruster Coefficient
Saturation
Control (C-T) 0.1275 0.011 0.00 No
1 0.1609 0.011 0.15 No
2 0.1720 0.013 0.25 No
3 0.1674 0.012 0.35 No
4 0.3546 0.024 0.50 No
5 0.5877 0.043 0.75 Yes
6 0.5810 0.040 1.00 Yes
7 0.1591 0.010 −0.15 No
8 0.1627 0.013 −0.25 No
9 0.2397 0.027 −0.35 No
10 0.2249 0.020 −0.50 No
11 0.2518 0.017 −0.75 No
12 0.3106 0.021 −1.00 No
13 0.1453 0.009 0.15 to 0.25 No
14 0.1275 0.009 0.15 to 0.25 No
15 0.1698 0.012 −0.15 to −0.25 No
16 0.2081 0.018 −0.15 to −0.25 No
17 0.2683 0.020 −0.35 to + 0.35 No
18 0.2312 0.017 −0.35 to + 0.35 No
As can be seen, the coefficients of the vertical thrusters increase when the
magnitude of the external disturbance increases. The demand for the BlueROV2 vertical
thrusters has increased up to 461% compared to the control test, but the controller keeps
the vehicle at a small distance from its reference. The RMSE for depth station-keeping
78
slightly increases. The maximum RMSE is 2.7 cm when the vertical thrusters do not reach
saturation and 4.3 cm when they do. The results for the control signal 𝜏𝑧 are summarized
in Table 12.
Table 12. Experimental results for the 𝜏𝑧 control signal in the station-keeping phase.
Disturbance Thruster
Experiment 𝝉𝒛 RMS % vs C-T Thrusters’ Saturation
Coefficient
Control (C-T) 5.88 - 0.00 No
1 8.95 152% 0.15 No
2 10.97 187% 0.25 No
3 11.63 198% 0.35 No
4 28.36 482% 0.50 No
5 49.57 843% 0.75 Yes
6 50.18 853% 1.00 Yes
7 7.98 136% −0.15 No
8 9.79 166% −0.25 No
9 16.03 273% −0.35 No
10 14.95 254% −0.50 No
11 20.14 342% −0.75 No
12 25.12 427% −1.00 No
13 10.39 177% 0.15 to 0.25 No
14 8.81 150% 0.15 to 0.25 No
15 10.43 177% −0.15 to −0.25 No
16 13.64 232% −0.15 to −0.25 No
17 21.46 365% −0.35 to +0.35 No
18 18.50 314% −0.35 to +0.35 No
A comparison with the control signal in the control test is made in the third column
(% vs C-T). There, an increase of up to 853% in the energy demanded by the controller
can be observed. This higher energy demand demonstrates that the controller effectively
reacts to compensate for the disturbances introduced.
6.3. Chapter conclusions
It is critical to have a controller that can hold an AUV in position while performing
manipulation tasks. An MFSOSMC with finite-time convergence in a predefined time,
which has already been evaluated and validated for trajectory tracking in Section 5, was
evaluated for the station-keeping of the BlueROV2 AUV in the presence of unknown
external disturbances. The external disturbance, in a range of about −40 to +50 N, was
directly introduced in the 𝑧 − 𝑎𝑥𝑖𝑠 of the BlueROV2. This external disturbance remained
unknown to the controller, and the weight of the additional thruster also acted as an
unknown disturbance. The controller maintained the same parameters and gains to prove
there was no need for tuning to achieve robust performance. Other controllers found in
the literature are either model-based, require an observer for the external disturbances,
need to adjust their gains online, or do not perform as well as the controller proposed in
this work in the position RMSE. Simulation results showed that the tracking error for the
vehicle’s position was maintained at zero throughout the station-keeping task. The
79
vehicle's position did not change when an external disturbance was introduced, as with
other station-keeping controllers reported in the literature. Eighteen experiments were
conducted in a semi-Olympic swimming pool to evaluate the proposed controller. The
results showed the robustness of the controller in the position RMSE, which was less than
2.7 cm in all cases where there was no saturation in the vertical thrusters, even in cases
where the external disturbance abruptly changed its magnitude and direction. Related
works where experiments were also performed reported up to 20 cm RMSE. When the
external disturbance was large enough to cause the saturation of the vertical thrusters,
the RMSE was kept below 4.5 cm.
80
Chapter 7: Coordinated manipulation task
Previous chapters have detailed the control algorithm design and validated that it
can achieve finite-time convergence of the tracking errors to a practical zero error in a
time-base arbitrarily defined by the user. That controller feature is used in this Chapter to
coordinate two BlueROV2 vehicles in a collaborative manipulation task of approaching,
grabbing, transporting, and releasing an object. Vehicle coordination and finite-time
tracking are crucial for the designed task since no V2V communication is considered. The
following sections describe the coordinated manipulation task, the assumptions and
considerations made, and the simulation setup. The performance of the proposed
MFSOSMC in vehicle coordination during the collaboration task is compared with two
state-of-the-art finite-time controllers, which are also introduced next. The results in this
Chapter were published in the full research paper [196]:
J. González-García, N. A. Narcizo-Nuci, A. Gómez-Espinosa, L. G. García-Valdovinos,

and T. Salgado-Jiménez, “Finite-Time Controller for Coordinated Navigation of Unmanned
Underwater Vehicles in a Collaborative Manipulation Task,” Sensors, vol. 23, no. 1, p. 239,
Dec. 2022, doi: 10.3390/s23010239.
7.1. Coordinated manipulation task design
A collaborative manipulation task was designed to validate the proposed

MFSOSMC performance in such a task. Due to a no V2V communication restriction, the
vehicles must coordinate their navigation to follow predefined trajectories accurately. In
this task, two BlueROV2 vehicles approach an object whose position on the bottom of a
water tank is known beforehand. Then, the BlueROV2 vehicles grab the object, lift it, and
transport it to a predetermined destination to release it. Both vehicles must converge to
their trajectories faster than the specified time for grabbing the object. The vehicles must
maintain a constant distance from each other when carrying the object and face each
other. The following are all considerations for the collaborative manipulation task:
• A water tank of 6 m × 6 m × 3 m is considered the working space. It is
considered clear of obstacles or entities other than the two BlueROV2 and
the object to manipulate;
• The initial position for the BlueROV2 vehicles is irrelevant for control and
does not affect the convergence time;
• An ideal abstract model of the object to be manipulated is considered. Its
physical and dynamic properties are neglected;
• The position of the object is known beforehand;
• Grasping the object is considered solved since it is out of the scope of this
research. The grasping occurs when the vehicles reach their respective
grasping positions;
• The vehicles grasp the object from opposite sides, i.e., they should align
their orientations to face each other and maintain a distance of 50 cm;
• There is no communication between the vehicles;
The design of the collaborative task considers nine intervals as follows:
• Interval 1: 𝑡 ≤ 6 s
81
Approach: The vehicles start from an arbitrary initial position and approach
the object. The initial position of the object is provided by 𝑂𝑖(1.5, 3.5, 3). The
vehicles reach a position 1 m above 𝑂𝑖 with a distance of 1 m in the 𝑦 −
𝑎𝑥𝑖𝑠 measured from their Body-fixed-frame origin. The 𝑥 − 𝑎𝑥𝑖𝑠 coordinate
corresponds to the corresponding grasp point. The final orientations of the
𝜋 𝜋
vehicles are provided by 𝜓1 = − 2 rad and 𝜓2 = 2 rad;
• Interval 2: 6 s < 𝑡 ≤ 9 s
Descending: The vehicles descend to 𝑧 = 3 m;
• Interval 3: 9 s < 𝑡 ≤ 12 s
Grasping: The vehicles move slowly along the 𝑦 − 𝑎𝑥𝑖𝑠 and reach their
respective grasping points. The grasping of the object takes place;
• Interval 4: 12 s < 𝑡 ≤ 15 s
Lifting: The vehicles rise to 𝑧 = 2 m to lift the object from the bottom of the
water tank;
• Interval 5: 15 s < 𝑡 ≤ 21 s
Transporting: The vehicles coordinately move the object to a position 1 m
above the releasing point, defined as 𝑂𝑓 = (4.5,1.5,3);
• Interval 6: 21 s < 𝑡 ≤ 24 s
Descending: The vehicles lower the object at its final position 𝑂𝑓 ;
• Interval 7: 24 s < 𝑡 ≤ 27 s
Release: The vehicles release the object and move 0.5 m on the 𝑦 − 𝑎𝑥𝑖𝑠 to
gain a slight separation from it;
• Interval 8: 27 s < 𝑡 ≤ 30 s
Ascending: The vehicles move to 𝑧 = 2 m to gain separation from the object;
• Interval 9: 30 s < 𝑡 ≤ 40 s;
Return: The vehicles move to their home positions at opposite corners of
the water tank.
All intervals are parameterized using the TBG provided in Equation 47 to ensure
smooth and coordinated movement of the vehicles. The velocities are parameterized by
𝑚
Equation 48, which guarantees that the initial and final velocities in each interval are 0 𝑠 .
The complete set of trajectories is shown in Figure 65.
82
Figure 65. Cooperative manipulation task. Trajectory design for both vehicles.
7.2. Comparing with other controllers
Two relevant state-of-the-art controllers [155], [156] that address AUV's finite-time
tracking are used for completeness. Those controllers and the proposed MFSOSMC were
evaluated in the simulations for comparative analysis. A brief introduction to these
controllers is provided in the following subsections.
7.2.1. Finite-time second-order SMC
Liu et al. [156] presented a Finite-Time Second-Order SMC (FTSOSMC) for
trajectory tracking of underwater vehicles subject to system uncertainties and unknown
disturbances. The model expressed in the Body-fixed frame is given by
𝜂̇ = 𝐽(𝜂)𝑣, (94)
′𝜐̇ ′ (𝜐)𝜐 ′ (𝜐)𝜐 ′ (𝜂)
𝑀 = +𝐶 +𝐷 +𝑔 = 𝜏 + 𝜏𝑒𝑥𝑡 . (95)
Then, the hydrodynamical model can be rewritten as
𝜂̈ = 𝐹(𝜂, 𝜂̇ ) + 𝐺(𝜂)𝜏 + 𝜏𝑑 , (96)
where
𝐹(𝜂, 𝜂̇ ) = 𝑀−1 (−𝐶𝜂̇ − 𝐷𝜂̇ − 𝑔), (97)
−1
𝐺(𝜂) = 𝑀 , (98)
𝑀 = 𝑀′𝐽−1 (𝜂), (99)
′ ′𝐽 𝑇 𝐽̇ −1
𝐶(𝜐, 𝜂̇ ) = [𝐶 − 𝑀 ]𝐽 (𝜂), (100)
′𝐽 −1
𝐷(𝜐, 𝜂̇ ) = 𝐷 (𝜂), (101)
𝑔(𝜂) = 𝑔′(𝜂). (102)
The tracking error is given by
𝑒(𝑡) = 𝜂(𝑡) − 𝜂𝑑 (𝑡), (103)
and the sliding surface is defined as
𝑠 = 𝑒̇ + 𝜆𝑒, (104)
where 𝜆 is a positive scalar.
Finally, the control law is given by
𝜏 = 𝜏𝑒 + 𝑀𝜏𝑠 , (105)
where
𝜏𝑒 = −𝐺 −1 (𝐹 + 𝜆𝑒̇ − 𝜂̈ 𝑑 ), (106)
83
𝑡
𝑏 (𝑠)
𝜏𝑠 = −𝑘1 𝑠𝑖𝑔𝑛 − ∫ (𝑘2 𝑠𝑖𝑔𝑛2𝑏−1 (𝑠(𝜏)) + 𝑘3 𝑠(𝜏) + 𝑘4 𝑠𝑖𝑔𝑛𝑏 𝑠(𝜏)𝑑𝑡. (107)
0
7.2.2. Non-singular Terminal Sliding Mode Control
Liu et al. [155] presented a Non-Singular Terminal SMC (NSTSMC). This controller
includes a non-singular sliding manifold defined by
𝑠 = 𝑒2 + 𝛽𝐹𝑓 (𝑒1 ), (108)
with
𝑒1 = 𝜂 + 𝜂𝑑 , (109)
𝑒2 = 𝜂̇ + 𝜂̇ 𝑑 , (110)
where 𝜂𝑑 and 𝜂̇ 𝑑 are the desired positions and velocities, 𝛽 is a positive constant and
𝐹𝑓 (𝑒1 ) is given by
(𝐼+𝑒 −𝑒1 )2
√|𝐴(𝑒1 )| 𝑠𝑖𝑔𝑛(𝑒1 ) 𝑖𝑓 |𝑒1 | ≥ 𝜖1
𝑒 −𝑒1
(𝑒 )
𝐹𝑓 1 = { 𝜋𝑒 𝜖 𝑒
, (111)
𝐴1 sin ( 2𝜖 1) + 2𝜋1 𝐴2 sin (2𝜋 𝜖1 ) 𝑖𝑓 |𝑒1 | ≤ 𝜖1
1 1
where 𝜖1is a vector of constants, 𝐼 is a vector of ones with compatible dimensions, |𝑒1 | is
the absolute value of each element of 𝑒1 , and
𝐼 − 𝑒 −𝑒1
𝐴(𝑒1 ) = , (112)
𝐼 + 𝑒 −𝑒1
(𝐼+𝑒 −𝜖1 )2
𝐴1 = √|𝐴(𝑒1 )| −𝜖1 , (113)
𝑒
1 𝐼−𝑒 −2𝜖1
𝐴2 = + √|𝐴(𝑒1 )| . (114)
√|𝐴(𝑒1 )| 𝑒 −𝜖1
Considering an AUV subject to model uncertainty and external disturbances, the
action of control law with the sliding manifold and the adaptive rate is
𝑢 = 𝑢𝑒𝑞 + 𝑢𝑎𝑑 , (115)
𝜕𝐹 (𝑒 )
𝑢𝑒𝑞 = 𝐵 + 𝐽𝑇 ((𝐶̂𝑅𝐵𝜂 + 𝐶̂𝜂 )𝜂̇ + 𝐷
̂𝜂 𝜂̇ + 𝑔̂𝜂 ) + 𝐵 + 𝐽𝑇 𝑀
̂𝜂 (𝜂̇ 𝑑 − 𝛽 𝑓 1
), (116)
𝜕𝑒 1
̂𝑇
𝑢𝑎𝑑 ̂𝜂 (𝛼1 𝑠 + 𝛼2 |𝑠|𝛾 𝑠𝑖𝑔𝑛(𝑠) + 𝐿𝑚Θ 𝑠).
= −𝐵 𝐽 𝑀 + 𝑇
(117)
2𝜖 0
with
‖𝑠‖2
𝐿̂̇𝑚 = 𝜆−1 2 ̂
𝑚 ( 2𝜖 2 Θ − 𝑘𝑚 𝐿𝑚 ), (118)
0
𝑘̇𝑚 = −𝜆−1 𝑝 𝑘𝑚 , (119)
where the superscript + denotes a pseudo-inverse operation; 𝛼1 and 𝛼2 are diagonal
matrices; 𝜖0 is a positive constant; 𝛾 is a constant between 0 and 1 and
Θ = [1; ‖𝜂̇ ‖; ‖𝜂̇ ‖2 ]2 , (120)
𝜕𝐹𝑓 (𝑒1 )
Finally, the partial derivative is given by
𝜕𝑒1
1 𝐼 − 𝑒 −2𝜖1
+ √|𝐴(𝑒1 )| 𝑠𝑖𝑔𝑛(𝑒1 ) 𝑖𝑓 |𝑒1 | ≥ 𝜖1
𝜕𝐹𝑓 (𝑒1 ) √|𝐴(𝑒 )| 𝑒 −𝜖1
= 1 . (121)
𝜕𝑒1 𝜋 𝜋𝑒1 𝜖1 𝑒1
𝐴1 cos ( )+ 𝐴 cos (2𝜋 ) 𝑖𝑓 |𝑒1 | ≤ 𝜖1
{ 2𝜖1 2𝜖1 2𝜋 2 𝜖1
84
7.3. Numerical simulations
Simulations were performed to test and evaluate the performance of the proposed
MFSOSMC against the controllers presented in the previous sections. The effects of
ocean currents were not considered at first. Then, the high ocean currents described in
Equation 74 were introduced as external disturbances during the mission. The
parameters used for the simulations are summarized in Table 13.
Table 13. Parameters for the controllers used in the collaborative manipulation simulations
Controller Parameters
MFSOSMC 𝑘𝑑 = 𝑑𝑖𝑎𝑔[800,800,800,800,0,800],
𝑘𝑖 = 𝑑𝑖𝑎𝑔[1,1,1,1,0,1],
𝜅 = 5,
𝛼0 = 1.01,
𝛼𝑐 = 25,
𝛿 = 0.001,
𝑡𝑏 = 6𝑠.
NSTSMC 𝛽 = 0.5,
𝜀0 = 2,
2
𝛾= ,
3
𝜀1 = 0.01[1,1,1,1,1,1]𝑇 ,
𝜆𝑚 = 𝑑𝑖𝑎𝑔[0.5,0.5,0.5],
𝜆𝑝 = 𝑑𝑖𝑎𝑔[0.5,0.5,0.5],
𝛼1 = 𝑑𝑖𝑎𝑔[5,5,5,30,30,30],
𝛼2 = 𝑑𝑖𝑎𝑔[5,5,5,30,30,30],
𝐿̂𝑚 (0) = [2.5,0.5, 0.5]𝑇 ,
̂𝑚 (0) = 𝑑𝑖𝑎𝑔[0.1,0.1,0.1].
𝐾
FTSOSMC 𝜆 = 0.8,
𝑏 = 0.7,
𝑘1 = 30,
𝑘2 = 20,
𝑘3 = 15,
𝑘4 = 20.
7.3.1. Simulations without external disturbances

Initially, a simulation was performed without considering the effects of ocean
currents. The 𝑥, 𝑦, and 𝑧 trajectories of BlueROV2 #1 (UUV1) and BlueROV2 #2 (UUV2)
are shown in Figure 66. In both cases, convergence to the desired trajectory occurs
smoothly at the user-defined time-base 𝑡𝑏 , which was arbitrarily set to 6 s. After reaching
the time-base, the vehicles maintained their respective desired trajectories by keeping a
practical zero value in the tracking errors.
85
Figure 66. MFSOSMC x,y, and z trajectory tracking (no external disturbances). BlueROV2 #1 (left).
BlueROV2 #2 (right).
The 𝜙, 𝜃, and 𝜓 trajectories of UUV1 and UUV2 are shown in Figure 67. A smooth
transition to the desired trajectory is observed for 𝜙 and 𝜓, in the user-defined time-base.
The pitch angle 𝜃 shows a slight deviation from the desired value, which is assumed to
be caused by the vehicle motion. As stated before, this DoF is not controllable.
Figure 67. MFSOSMC ϕ,θ, and ψ trajectory tracking (no external disturbances). Left: BlueROV2 #1.
Right: BlueROV2 #2.
The desired and controlled velocities are shown in Figure 68 and Figure 69 for
UUV1 and UUV2, respectively. The velocities of the vehicles are brought to the desired
trajectories at the time-base 𝑡𝑏 = 6 s. Then, the tracking error of these velocities is kept
at a practical zero value. Since the angular velocity 𝑞̇ cannot be controlled, it is assumed
that the observed small deviations are due to the vehicle’s motion.
86
Figure 68. BlueROV2 #1 velocity tracking with the MFSOSMC (no external disturbances).
Figure 69.BlueROV2 #2 velocity tracking with the MFSOSMC (no external disturbances).
The 3D trajectory followed by the vehicles controlled by the MFSOSMC during the
designed collaborative manipulation task is shown in Figure 70. The orientation of the
vehicles can be observed as their Body-fixed frame is included in the figure. Note that the
orientation of the vehicles is initially aligned and later changes to meet the task
requirement of facing each other to manipulate the object.
87
Figure 70. UUV1 and UUV2 trajectory tracking the MFSOSMC in the collaborative manipulation of an
object (no external disturbances). (a) Isometric view. (b) Top view.
From the results presented earlier, the MFSOSMC guides the vehicles to the
desired trajectories in the expected time-base, guaranteeing the vehicles’ synchronization
to perform the collaborative manipulation task in a coordinated manner.
The NSTSMC and FTSOSMC presented in Section 7.2 were programmed in the
simulator to complete the same collaborative manipulation task. The tracking errors for
the 𝑥, 𝑦, and 𝑧 positions are shown in Figure 71 for all three controllers. In the tracking
error for the 𝑥 position case, both vehicles converge to a practical zero value
simultaneously when driven by the MFSOSMC. Convergence happens at different times
for the other controllers as it depends on the initial error condition, which is different for
both vehicles. In the case of FTSOSMC, this is more evident as the difference is about
two seconds. In the tracking error for the 𝑦 position case, the proposed MFSOSMC is the
only method that achieves coordinated convergence as expected, even when the two
vehicles started at the same 𝑦 coordinate and thus with the same 𝑦 initial error condition.
In the tracking error for the 𝑧 position case, the two vehicles appear to converge
simultaneously. However, the MFSOSMC smoothly controls the vehicles, while the
88
convergence of the other controller is not so smooth and even exhibits an overshoot. This
overshoot is not desirable and could harm the vehicle or the object as it could cause the
vehicle to collide with the bottom of the water tank or hit the object.
Figure 71. Tracking error for all the controllers (no external disturbances). (a) 𝑥 − 𝑎𝑥𝑖𝑠. (b) 𝑦 − 𝑎𝑥𝑖𝑠. (c)
𝑧 − 𝑎𝑥𝑖𝑠.
The tracking errors for the 𝜙 and 𝜓 orientations are shown in Figure 72. In the 𝜙
orientation case, the initial error condition is zero, which means that the vehicles are
already on their reference. The MFSOSMC does not move the vehicles away from the
desired 𝜙, so the tracking error remains at a zero value. However, NSTSMC and
FTSOSMC move the vehicles away from their reference up to 42° and 21°, respectively,
which is not desired. In the 𝜓 orientation case, something similar to the 𝑥 and 𝑦 tracking
errors is observed. The MFSOSMC smoothly tracks the vehicles to the reference with the
same time-base, but the NSTSMC and the FTSOSMC converge at different times for
UUV1 and UUV2, and the transition is not as smooth because it has ripples. These ripples
represent a higher vehicle effort and, therefore, higher energy consumption.
89
Figure 72. Tracking error for all the controllers (no external disturbances). Left: 𝜙 orientation. Right:
𝜓 orientation.
As described earlier, the distance and relative orientation between vehicles must
remain constant at 0.5 m and −180°, respectively, as they approach and manipulate the
object. If this is accomplished and tracking errors are kept to zero, the collaborative
manipulation task can be successful. These performance indicators are shown in Figure
73. All controllers drive the vehicles to meet these values before the manipulation begins.
However, the NSTSMC and FTSOSMC have some difficulty controlling the relative
orientation of the vehicles in the transitory.
Figure 73. Performance indicators for all the controllers (no external disturbances). Left: Euclidean
distance in the 𝑥 − 𝑎𝑥𝑖𝑠. Right: relative heading orientation.
7.3.2. Simulations with external disturbances

Simulations were performed applying the high ocean currents described in
Equation 74 and were considered from the beginning and throughout the collaborative
manipulation task. These disturbances remained unknown to the controllers. Tracking
errors for the 𝑥, 𝑦, and 𝑧 positions are shown in Figure 74. The robustness of the
MFSOSMC to strong external disturbances can be observed as it maintains its
performance and drives the vehicles to their references smoothly and without any visible
90
problems in the provided time-base. The difficulties of the NSTSMC and the FTSOSMC
observed in the previous section were amplified in this case.
Figure 74. Tracking error for all the controllers (with external disturbances). (a) 𝑥 − 𝑎𝑥𝑖𝑠. (b) 𝑦 − 𝑎𝑥𝑖𝑠. (c)
𝑧 − 𝑎𝑥𝑖𝑠.
The external disturbances also exacerbated the problems of the NSTSMC and the
FTSOSMC in keeping the vehicles on the 𝜙 = 0° reference, as can be seen in Figure 75
(left), because it not only deviates from the reference but also has multiple oscillations.
The NSTSMC even lost the reference during the last interval of the task. The MFSOSMC
maintains the 𝜙 = 0° reference all the time. The tracking error for the 𝜓 orientation is
shown in Figure 75 (right). Multiple ripples are observed for the NSTSMC and the
FTSOSMC, while the MFSOSMC smoothly converges to zero. The NSTSMC lost both
vehicle references in the last intervals of the task, which is a worse performance than the
results without external disturbances.
91
Figure 75. Tracking error for all the controllers (with external disturbances). Left: 𝜙 orientation. Right): 𝜓
orientation.
Regarding the Euclidean distance indicator, all controllers maintained a constant

distance of 0.5 m in the 𝑦 − 𝑎𝑥𝑖𝑠 during the grasp, transport, and release intervals, as
shown in Figure 76 (left). The relative heading orientation is shown in Figure 76 (right),
the NSTSMC and the FTSOSMC struggles are noticeable, but the −180° difference is
maintained during the manipulation of the object.
Figure 76. Performance indicators for all the controllers (with external disturbances). Left: Euclidean
distance in the 𝑥 − 𝑎𝑥𝑖𝑠. Right: relative heading orientation.
Another performance indicator is the control signal 𝜏, shown in Figure 77 for the
UUV1 control of position and orientation. It can be observed that the control signals of the
proposed MFSOSMC start at a small value and increase slightly depending on the
manipulation task requirements. The changes in these control signals are smooth, and
no chattering is observed at any time. In the case of the FTSOSMC, the initial values of
the control signals are considerably higher, which imposes a high demand on the
thrusters and higher energy consumption. The NSTSMC provides the worst performance.
In this case, the energy demand is higher, and the chattering effect is observed at some
intervals.
92
Figure 77.Control signals for UUV1 by all the controllers (with external disturbances).
7.4. Chapter conclusions
In this Chapter, a coordinated navigation scheme for two UUVs without

communication between them was developed to perform a collaborative manipulation
task. A Model-free second-order sliding mode control (MFSOSMC) with finite-time
convergence is used to coordinate the navigation of the vehicles so that they converge to
their desired trajectories simultaneously. This controller has the advantage that the user
can arbitrarily choose the convergence time by simply specifying a time-base parameter
𝑡𝑏 . This convergence time does not depend on other control parameters, vehicle
hydrodynamics, or initial error conditions, as with other state-of-the-art controllers. The
coordinated vehicles then perform a collaborative manipulation task of approaching,
grasping, lifting, transporting, and releasing an object in a water tank. Numerical
simulations validated the performance of the proposed controller, which was compared
93
with two state-of-the-art finite-time controllers, a Finite-Time Second-Order Sliding Mode
Control (FTSOSMC) and a Non-Singular Terminal Sliding Mode Control (NSTSMC). The
results showed superior performance of the MFSOSMC, as the tracking error of all
controlled DoFs converged coordinately and smoothly in the predefined time-base and
did not deviate from its references, even in the presence of high ocean currents. The
FTSOSMC and the NSTSMC had difficulties driving the vehicles to their respective
trajectories. Both vehicles converged at different times and exhibited ripples, overshoots,
and oscillations. This is highly undesirable because it represents a collision risk that could
damage the vehicles or the object and causes a higher energy demand on the thrusters.
The proposed MFSOSMC has the best performance in terms of energy consumption
observed in the calculated control signals. Higher energy demand was observed in the
FTSOSMC and especially in the NSTSMC cases, where chattering also occurred, which
is not present in the proposed controller. A constant distance of 0.5 m in the 𝑦 − 𝑎𝑥𝑖𝑠 and
a relative heading difference of -180° between the vehicles during the manipulation
intervals were used as further indicators of the success of the collaborative manipulation
task performed by the three controllers, with the MFSOSMC showing the best
performance, as described previously.
94
General conclusions
A model-free high-order sliding mode controller was introduced in this project as
an advanced control algorithm for autonomous underwater vehicles. Unlike similar
controllers, this guarantees finite-time convergence of the tracking errors to a practical
zero value in a time-base that the user can arbitrarily set. Numerical simulations and
experiments in a semi-Olympic swimming pool were made to evaluate the proposed
controller's performance and validate it for the trajectory tracking and station-keeping
problems in a single BlueROV2 vehicle. Results demonstrated the finite-time
convergence in all the cases, in the different time-bases that were arbitrarily set. The
tracking errors remained at practical zero in all simulations and were limited to a few
centimeters in the experiments. The sensor's accuracy and measurement rate can
explain those small errors. However, the proposed controller outperformed classic
controllers and its asymptotic version regarding RMS tracking error values and energy
consumption. It also demonstrated robustness when external disturbances were
introduced.
Then, the proposed controller was used to coordinate two BlueROV2 vehicles to
perform a collaborative manipulation task autonomously where communications were
null. The cooperative manipulation task was simulated, and the performance of the
proposed controller was compared with other state-of-the-art finite-time controllers. The
proposed controller drove the vehicles to the desired predefined trajectories in the given
time-base, no matter the initial position of the vehicles. On the other hand, the state-of-
the-art controllers converged at different times depending on the controller parameters,
the vehicle's hydrodynamics, and initial error condition, among other factors which
complicate the coordination of the vehicles in the task.
Future work
In the short term, future work will include a more detailed simulation setup. The
coordinated navigation of the vehicles using the proposed MFSOSMC and the
collaborative manipulation task can be tested and validated by including more detailed
features such as sensors and noise models and considering the object’s physical
characteristics. In the long term, the research will address some challenges for validation
in an actual experimental setup. Those challenges include:
• Implementing an object identification algorithm;
• Implementing an autonomous path planning algorithm;
• Implementing an autonomous object grasping method;
• Including the physical and geometrical parameters of the object;
• Implementing a method for the localization of the vehicle inside a swimming
pool;
• Implementing obstacle avoidance algorithms;
• Implementing an auxiliary control algorithm to deal with vehicle-to-vehicle
disturbances.
Validation of coordinated autonomous navigation of multiple (2+) vehicles using
the proposed MFSOSMC is also considered for future works.
95
Abbreviations and acronyms
AHRS Attitude and Heading Reference System

AUV Autonomous Underwater Vehicle
BC Backstepping Control
BITAN Beijing university of aeronautics and astronautics Inertial Terrain-
Aided Navigation
BK Bandler and Kohout
CG Center of Gravity
CRNN Convolution Recurrent Neural Network
DGPS Differential GPS
DR Dead-Reckoning
DSO Direct Sparse Odometry
DT Distance Traveled
DVL Doppler Velocity Loggers
EKF Extended Kalman Filter
ELC Extended Loosely Coupled
ESC Electronic Speed Controllers
FBL Feed-Back Linearization
FLS Forward-Looking SONAR
FPS Frames Per Second
FTPS Fitting of Two Point Sets
FTSOSMC Finite-Time Second-Order Sliding Mode Control
GBNN Glasius Bio-inspired Neural Network
GN Geophysical Navigation
GPS Global Positioning System
HSV Hue Saturation Value
I-AUV Intervention AUV
ICCP Iterated Closest Contour Point
IMU Inertial Measurement Units
INS Inertial Navigation Systems
KF Kalman Filter
LBL Long-BaseLine
LC Loosely Coupled
LCI Language-Centered Intelligence
MEMS Micro Electro Mechanical Systems
MFSOSMC Model-Free Second-Order SMC
NN Neural Network
NRT Near-Real-Time
NSTSMC Non-Singular Terminal Sliding Mode Control
PF Particle Filters
96
PID Proportional Integral Derivative
PL-SLAM Point and Line SLAM
PMF Point Mass Filter
PS Pressure Sensor
PSO Particle Swarm Optimization
PTAM Parallel Tracking and Mapping
PWM Pulse Wide Modulation
RMS Root Mean Square
RMSE Root-Mean-Square Error
RNN Recurrent Neural Network
ROS Robot Operating System
ROV Remotly Operated Vehicle
RPI Raspberry Pi
SBL Short BaseLine
SINS Strapdown Inertial Navigation System
SLAM Simultaneous Location and Mapping
SMC Sliding Mode Control
SOG Sum of Gaussian
SONAR Sound Navigation and Ranging
SVO Semi-direct Visual Odometry
TAN Terrain-Aided Navigation
TBG Time-Base Generator
TBN Terrain-Based Navigation
TC Tightly Coupled
TRN Terrain-Referenced Navigation
UKF Unscented Kalman Filter
USBL Ultra Short BaseLine
USV Unmanned Surface Vehicles
UUV Unmanned Underwater Vehicles
UWSIM UnderWater SIMulator
V2I Vehicle-to-Infrastructure
V2V Vehicle-to-Vehicle
VIO Visual Inertial Odometry
VO Visual Odometry
97
Bibliography
[1] A. Palomer, P. Ridao, and D. Ribas, “Inspection of an underwater structure using

point-cloud SLAM with an AUV and a laser scanner,” J. F. Robot., vol. 36, no. 8,
pp. 1333–1344, 2019, doi: 10.1002/rob.21907.
[2] A. G. Rumson, “The application of fully unmanned robotic systems for inspection of
subsea pipelines,” Ocean Eng., vol. 235, no. January, p. 109214, 2021, doi:
10.1016/j.oceaneng.2021.109214.
[3] J. Hwang, N. Bose, H. D. Nguyen, and G. Williams, “Acoustic search and detection
of oil plumes using an autonomous underwater vehicle,” J. Mar. Sci. Eng., vol. 8,
no. 8, 2020, doi: 10.3390/JMSE8080618.
[4] J. D. Hernández et al., “Autonomous underwater navigation and optical mapping in
unknown natural environments,” Sensors (Switzerland), vol. 16, no. 8, 2016, doi:
10.3390/s16081174.
[5] P. J. B. Sánchez, M. Papaelias, and F. P. G. Márquez, “Autonomous underwater
vehicles: Instrumentation and measurements,” IEEE Instrum. Meas. Mag., vol. 23,
no. 2, pp. 105–114, 2020, doi: 10.1109/MIM.2020.9062680.
[6] E. Simetti, “Autonomous Underwater Intervention,” Curr. Robot. Reports, vol. 1, no.
3, pp. 117–122, 2020, doi: 10.1007/s43154-020-00012-7.
[7] W. Gorma et al., “Development of modular bio-inspired autonomous underwater
vehicle for close subsea asset inspection,” Appl. Sci., vol. 11, no. 12, 2021, doi:
10.3390/app11125401.
[8] H. A. Dawson and M. Allison, “Requirements for Autonomous Underwater Vehicles
(AUVs) for scientific data collection in the Laurentian Great Lakes: A questionnaire
survey,” J. Great Lakes Res., vol. 47, no. 1, pp. 259–265, 2021, doi:
10.1016/j.jglr.2020.11.004.
[9] H. Zhang et al., “Subsea pipeline leak inspection by autonomous underwater
vehicle,” Appl. Ocean Res., vol. 107, no. July 2020, p. 102321, 2021, doi:
10.1016/j.apor.2020.102321.
[10] J. González-García, A. Gómez-Espinosa, E. Cuan-Urquizo, L. G. García-
Valdovinos, T. Salgado-Jiménez, and J. A. Escobedo Cabello, “Autonomous
underwater vehicles: Localization, navigation, and communication for collaborative
missions,” Appl. Sci., vol. 10, no. 4, 2020, doi: 10.3390/app10041256.
[11] J. Evans, P. Redmond, C. Plakas, K. Hamilton, and D. Lane, “Autonomous docking
for intervention-AUVs using sonar and video-based real-time 3D pose estimation,”
Ocean. 2003 Celebr. Past... Teaming Towar. Futur., vol. 4, pp. 2201–2210, 2003,
doi: 10.1109/OCEANS.2003.178243.
[12] P. Cieslak and P. Ridao, “Adaptive Admittance Control in Task-Priority Framework
for Contact Force Control in Autonomous Underwater Floating Manipulation,” IEEE
Int. Conf. Intell. Robot. Syst., pp. 6646–6651, 2018, doi:
10.1109/IROS.2018.8593542.
[13] E. Simetti et al., “Autonomous Underwater Intervention: Experimental Results of
the MARIS Project,” IEEE J. Ocean. Eng., vol. 43, no. 3, pp. 620–639, 2018, doi:
10.1109/JOE.2017.2733878.
[14] D. Ribas et al., “I-AUV Mechatronics Integration for the TRIDENT FP7 Project,”
IEEE/ASME Trans. Mechatronics, vol. 20, no. 5, pp. 2583–2592, 2015, doi:
98
10.1109/TMECH.2015.2395413.
[15] E. Kim, S. Fan, N. Bose, and H. Nguyen, “Path Following for an Autonomous
Underwater Vehicle (AUV) by Using a High-Gain Observer based on an AUV
Dynamic Model,” IFAC-PapersOnLine, vol. 52, no. 21, pp. 218–223, 2019, doi:
10.1016/j.ifacol.2019.12.310.
[16] D. Li and L. Du, “Auv trajectory tracking models and control strategies: A review,”
J. Mar. Sci. Eng., vol. 9, no. 9, 2021, doi: 10.3390/jmse9091020.
[17] H. Qin, C. Li, Y. Sun, X. Li, Y. Du, and Z. Deng, “Finite-time trajectory tracking
control of unmanned surface vessel with error constraints and input saturations,” J.
Franklin Inst., vol. 357, no. 16, pp. 11472–11495, 2020, doi:
10.1016/j.jfranklin.2019.07.019.
[18] N. Ding, Y. Tang, Z. Jiang, Y. Bai, and S. Liang, “Station-Keeping Control of
Autonomous and Remotely-Operated Vehicles for Free Floating Manipulation,” J.
Mar. Sci. Eng., vol. 9, no. 11, pp. 1–25, 2021, doi: 10.3390/JMSE9111305.
[19] J. Bosch, N. Gracias, P. Ridao, K. Istenič, and D. Ribas, “Close-range tracking of
underwater vehicles using light beacons,” Sensors (Switzerland), vol. 16, no. 4, pp.
1–26, 2016, doi: 10.3390/s16040429.
[20] Z. Ren, L. Chen, H. Zhang, and M. Wu, “Research on geomagnetic-matching
localization algorithm for unmanned underwater vehicles,” Proc. 2008 IEEE Int.
Conf. Inf. Autom. ICIA 2008, pp. 1025–1029, 2008, doi:
10.1109/ICINFA.2008.4608149.
[21] J. J. Leonard and A. Bahr, “Autonomous underwater vehicle navigation,” Springer
Handb. Ocean Eng., no. Dvl, pp. 341–357, 2016, doi: 10.1007/978-3-319-16649-
0_14.
[22] H. Rice, S. Kelmenson, and L. Mendelsohn, “Geophysical navigation technologies
and applications,” Rec. - IEEE PLANS, Position Locat. Navig. Symp., pp. 618–624,
2004, doi: 10.1109/plans.2004.1309051.
[23] M. F. Fallon, M. Kaess, H. Johannsson, and J. J. Leonard, “Efficient AUV navigation
fusing acoustic ranging and side-scan sonar,” Proc. - IEEE Int. Conf. Robot.
Autom., pp. 2398–2405, 2011, doi: 10.1109/ICRA.2011.5980302.
[24] T. Nicosevici, R. Garcia, M. Carreras, and M. Villanueva, “A review of sensor fusion
techniques for underwater vehicle navigation,” pp. 1600–1605, 2005, doi:
10.1109/oceans.2004.1406361.
[25] L. Paull, S. Saeedi, M. Seto, and H. Li, “AUV navigation and localization: A review,”
IEEE J. Ocean. Eng., vol. 39, no. 1, pp. 131–149, 2014, doi:
10.1109/JOE.2013.2278891.
[26] X. Che, I. Wells, G. Dickers, P. Kear, and X. Gong, “Re-evaluation of RF
electromagnetic communication in underwater sensor networks,” IEEE Commun.
Mag., vol. 48, no. 12, pp. 143–151, 2010, doi: 10.1109/MCOM.2010.5673085.
[27] J. Melo and A. Matos, “Survey on advances on terrain based navigation for
autonomous underwater vehicles,” Ocean Eng., vol. 139, no. September 2015, pp.
250–264, 2017, doi: 10.1016/j.oceaneng.2017.04.047.
[28] C. Jekeli, “Gravity on Precise, Short-Term, 3-D Free- Inertial Navigation,” J. Inst.
Navig., vol. 44, no. 3, pp. 347–357, 1997.
[29] M. Perlmutter and L. Robin, “High-performance, low cost inertial MEMS: A market
in motion!,” Rec. - IEEE PLANS, Position Locat. Navig. Symp., pp. 225–229, 2012,
99
doi: 10.1109/PLANS.2012.6236884.
[30] C. Jekeli, “Precision free-inertial navigation with gravity compensation by an
onboard gradiometer,” J. Guid. Control. Dyn., vol. 30, no. 4, pp. 1214–1215, 2007,
doi: 10.2514/1.29808.
[31] H. Rice, L. Mendelsohn, R. Aarons, and D. Mazzola, “Next generation marine
precision navigation system,” Rec. - IEEE PLANS, Position Locat. Navig. Symp.,
pp. 200–206, 2000, doi: 10.1109/plans.2000.838303.
[32] “ISM3D - Underwater AHRS Sensor - Impact Subsea.”
http://www.impactsubsea.co.uk/ism3d-2/ (accessed Dec. 26, 2019).
[33] “Underwater 9-axis IMU/AHRS sensor - Seascape Subsea BV.”
https://www.seascapesubsea.com/product/underwater-9-axis-imu-ahrs-sensor/
(accessed Dec. 26, 2019).
[34] “Inertial Labs Attitude and Heading Reference Systems (AHRS) - Inertial Labs.”
https://inertiallabs.com/products/ahrs/ (accessed Dec. 26, 2019).
[35] “Elipse 2 series - Miniature Inertial Navigation Sensors.” https://www.sbg-
systems.com/products/ellipse-2-series/#ellipse2-a_miniature-ahrs (accessed Dec.
26, 2019).
[36] “DSPRH - Depth and AHRS sensor | TMI-Orion.com.” https://www.tmi-
orion.com/en/robotics/underwater-robotics/dsprh-depth-and-ahrs-sensor.htm
[37] “VN-100 - VectorNav Technologies.” https://www.vectornav.com/products/vn-100
(accessed Jan. 21, 2020).
[38] “XSENS - MTi 600-series.” https://www.xsens.com/products/mti-600-series
(accessed Jan. 21, 2020).
[39] S. J. Hurtós Natalia, Ribas David, Cufi Xavier, Petillot Yvan, “Fourier-based
Registration for Robust Forward-looking Sonar Mosaicing in Low-visibility
Underwater Environments,” J. F. Robot., vol. 32, no. 1, pp. 123–151, 2014, doi:
10.1002/rob.
[40] C. Galarza, I. Masmitja, J. Prat, and S. Gomariz, “Design of obstacle detection and
avoidance system for Guanay II AUV,” 24th Mediterr. Conf. Control Autom. MED,
vol. 5, pp. 410–414, 2016, doi: 10.1109/MED.2016.7535959.
[41] B. Braginsky and H. Guterman, “Obstacle avoidance approaches for autonomous
underwater vehicle: Simulation and experimental results,” IEEE J. Ocean. Eng., vol.
41, no. 4, pp. 882–892, 2016, doi: 10.1109/JOE.2015.2506204.
[42] C. Lin, H. Wang, J. Yuan, D. Yu, and C. Li, “An improved recurrent neural network
for unmanned underwater vehicle online obstacle avoidance,” Ocean Eng., vol.
189, no. September, p. 106327, 2019, doi: 10.1016/j.oceaneng.2019.106327.
[43] Z. Li, S. E. Dosso, and D. Sun, “Motion-compensated acoustic localization for
underwater vehicles,” IEEE J. Ocean. Eng., vol. 41, no. 4, pp. 840–851, 2016, doi:
10.1109/JOE.2015.2503518.
[44] Y. Zhai, Z. Gong, L. Wang, R. Zhang, and H. Luo, “Study of underwater positioning
based on short baseline sonar system,” Int. Conf. Artif. Intell. Comput. Intell. AICI,
vol. 2, pp. 343–346, 2009, doi: 10.1109/AICI.2009.83.
[45] “LBL Positioning Systems | EvoLogics.” https://evologics.de/lbl#products
[46] “GeoTag.”
100
http://www.sercel.com/products/Pages/GeoTag.aspx?gclid=EAIaIQobChMI5sqPk
Y3S5gIV2v_jBx0apwTAEAAYAiAAEgILm_D_BwE (accessed Dec. 26, 2019).
[47] “MicroPAP Compact acoustic positioning system - Kongsberg Maritime.”
https://www.kongsberg.com/maritime/products/Acoustics-Positioning-and-
Communication/acoustic-positioning-systems/pap-micropap-compact-acoustic-
positioning-system/ (accessed Dec. 26, 2019).
[48] “Subsonus | Advanced Navigation.”
https://www.advancednavigation.com/product/subsonus?gclid=EAIaIQobChMI5sq
PkY3S5gIV2v_jBx0apwTAEAAYASAAEgJNpfD_BwE (accessed Dec. 26, 2019).
[49] “Underwater GPS - Water Linked AS.” https://waterlinked.com/underwater-gps/
[50] P. Batista, C. Silvestre, and P. Oliveira, “Tightly coupled long baseline/ultra-short
baseline integrated navigation system,” Int. J. Syst. Sci., vol. 47, no. 8, pp. 1837–
1855, 2016, doi: 10.1080/00207721.2014.955070.
[51] A. Vasilijević, D. Nad, F. Mandi, N. Miškovi, and Z. Vukić, “Coordinated navigation
of surface and underwater marine robotic vehicles for ocean sampling and
environmental monitoring,” IEEE/ASME Trans. Mechatronics, vol. 22, no. 3, pp.
1174–1184, 2017, doi: 10.1109/TMECH.2017.2684423.
[52] E. I. Sarda and M. R. Dhanak, “Launch and Recovery of an Autonomous
Underwater Vehicle from a Station-Keeping Unmanned Surface Vehicle,” IEEE J.
Ocean. Eng., vol. 44, no. 2, pp. 290–299, 2019, doi: 10.1109/JOE.2018.2867988.
[53] I. Masmitja, S. Gomariz, J. Del Rio, B. Kieft, and T. O’Reilly, “Range-only
underwater target localization: Path characterization,” Ocean. 2016 MTS/IEEE
Monterey, OCE, no. group 1, pp. 1–7, 2016, doi: 10.1109/OCEANS.2016.7761246.
[54] M. Bayat, N. Crasta, A. P. Aguiar, and A. M. Pascoal, “Range-Based Underwater
Vehicle Localization in the Presence of Unknown Ocean Currents: Theory and
Experiments,” IEEE Trans. Control Syst. Technol., vol. 24, no. 1, pp. 122–139,
2016, doi: 10.1109/TCST.2015.2420636.
[55] G. Vallicrosa and P. Ridao, “Sum of Gaussian single beacon range-only localization
for AUV homing,” Annu. Rev. Control, vol. 42, pp. 177–187, 2016, doi:
10.1016/j.arcontrol.2016.09.007.
[56] T. Zhang, Z. Wang, Y. Li, and J. Tong, “A passive acoustic positioning algorithm
based on virtual long baseline matrix window,” J. Navig., vol. 72, no. 1, pp. 193–
206, 2019, doi: 10.1017/S0373463318000590.
[57] F. Zhang, X. Chen, M. Sun, M. Yan, and D. Yang, “Simulation study of underwater
passive navigation system based on gravity gradient,” Int. Geosci. Remote Sens.
Symp., vol. 5, no. C, pp. 3111–3113, 2004, doi: 10.1109/igarss.2004.1370357.
[58] E. Wei et al., “A Robust Solution of Integrated SITAN with TERCOM Algorithm:
Weight-Reducing Iteration Technique for Underwater Vehicles’ Gravity-Aided
Inertial Navigation System,” Navig. J. Inst. Navig., vol. 64, no. 1, pp. 111–122, 2017,
doi: 10.1002/navi.176.
[59] D. Vaman, “TRN history, trends and the unused potential,” AIAA/IEEE Digit. Avion.
Syst. Conf. - Proc., pp. 1–16, 2012, doi: 10.1109/DASC.2012.6382278.
[60] Y. Pei, Z. Chen, and J. C. Hung, “BITAN-II: an improved terrain aided navigation
algorithm,” IECON Proc. (Industrial Electron. Conf., vol. 3, pp. 1675–1680, 1996,
doi: 10.1109/iecon.1996.570665.
101
[61] Y. Han, B. Wang, Z. Deng, and M. Fu, “A combined matching algorithm for
underwater gravity-Aided navigation,” IEEE/ASME Trans. Mechatronics, vol. 23,
no. 1, pp. 233–241, 2018, doi: 10.1109/TMECH.2017.2774296.
[62] C. Guo, Li Anliang, H. Cai, and Yang Huabo, “Algorithm for geomagnetic navigation
and its validity evaluation,” Proc. - 2011 IEEE Int. Conf. Comput. Sci. Autom. Eng.
CSAE 2011, vol. 1, pp. 573–577, 2011, doi: 10.1109/CSAE.2011.5953286.
[63] J. Zhao, S. Wang, and A. Wang, “Study on underwater navigation system based
on geomagnetic match technique,” ICEMI 2009 - Proc. 9th Int. Conf. Electron.
Meas. Instruments, pp. 3255–3259, 2009, doi: 10.1109/ICEMI.2009.5274338.
[64] S. Wang, H. Zhang, Y. Kun, and C. Tian, “Study on the underwater geomagnetic
navigation based on the integration of TERCOM and K-means clustering
algorithm,” Ocean. IEEE Sydney, Ocean. 2010, pp. 1–4, 2010, doi:
10.1109/OCEANSSYD.2010.5603583.
[65] L. Jaulin, “Isobath following using an altimeter as a unique exteroceptive sensor,”
CEUR Workshop Proc., vol. 2331, pp. 105–110, 2018.
[66] R. Cowie, M and Wilkinson, N and Powlesland, “Latest Development of the
TERPROM ® Digital Terrain System ( DTS ),” IEEE/ION Position, Locat. Navig.
Symp., pp. 1219–1229, 2008.
[67] D. K. Meduna, “Terrain Relative Navigation for Sensor-Limited Systems with
Applications to Underwater Vehicles,” Standford University, 2011.
[68] L. Zhao, N. Gao, B. Huang, Q. Wang, and J. Zhou, “A novel terrain-aided navigation
algorithm combined with the TERCOM algorithm and particle filter,” IEEE Sens. J.,
vol. 15, no. 2, pp. 1124–1131, 2015, doi: 10.1109/JSEN.2014.2360916.
[69] G. Salavasidis et al., “Terrain-aided navigation for long-endurance and deep-rated
autonomous underwater vehicles,” J. F. Robot., vol. 36, no. 2, pp. 447–474, 2019,
doi: 10.1002/rob.21832.
[70] D. K. Meduna, S. M. Rock, and R. S. McEwen, “Closed-loop terrain relative
navigation for AUVs with non-inertial grade navigation sensors,” IEEE/OES Auton.
Underw. Veh. AUV 2010, 2010, doi: 10.1109/AUV.2010.5779659.
[71] G. Nootz, E. Jarosz, F. R. Dalgleish, and W. Hou, “Quantification of optical
turbulence in the ocean and its effects on beam propagation,” Appl. Opt., vol. 55,
no. 31, p. 8813, 2016, doi: 10.1364/ao.55.008813.
[72] F. Eren, S. Pe’Eri, M. W. Thein, Y. Rzhanov, B. Celikkol, and M. R. Swift, “Position,
orientation and velocity detection of Unmanned Underwater Vehicles (UUVs) using
an optical detector array,” Sensors (Switzerland), vol. 17, no. 8, pp. 1–18, 2017,
doi: 10.3390/s17081741.
[73] L. Zhong, D. Li, M. Lin, R. Lin, and C. Yang, “A fast binocular localisation method
for AUV docking,” Sensors (Switzerland), vol. 19, no. 7, 2019, doi:
10.3390/s19071735.
[74] S. Liu, H. Xu, Y. Lin, and L. Gao, “Visual navigation for recovering an AUV by
another AUV in shallow water,” Sensors (Switzerland), vol. 19, no. 8, pp. 1–19,
2019, doi: 10.3390/s19081889.
[75] J. A. Monroy-Anieva, C. Rouviere, E. Campos-Mercado, T. Salgado-Jimenez, and
L. G. Garcia-Valdovinos, “Modeling and control of a micro AUV: Objects follower
approach,” Sensors (Switzerland), vol. 18, no. 8, pp. 1–18, 2018, doi:
10.3390/s18082574.
102
[76] M. Prats et al., “Reconfigurable AUV for intervention missions: A case study on
underwater object recovery,” Intell. Serv. Robot., vol. 5, no. 1, pp. 19–31, 2012, doi:
10.1007/s11370-011-0101-z.
[77] H. Durrant-Whyte and T. Bailey, “Simultaneous localization and mapping: Part I,”
IEEE Robot. Autom. Mag., vol. 13, no. 2, pp. 99–108, 2006, doi:
10.1109/MRA.2006.1638022.
[78] Y. Lu and D. Song, “Visual Navigation Using Heterogeneous Landmarks and
Unsupervised Geometric Constraints,” IEEE Trans. Robot., vol. 31, no. 3, pp. 736–
749, 2015, doi: 10.1109/TRO.2015.2424032.
[79] H. Carrillo, P. Dames, V. Kumar, and J. A. Castellanos, “Autonomous robotic
exploration using occupancy grid maps and graph SLAM based on Shannon and
Rényi Entropy,” Proc. - IEEE Int. Conf. Robot. Autom., vol. 2015-June, no. June,
pp. 487–494, 2015, doi: 10.1109/ICRA.2015.7139224.
[80] M. Pizzoli, C. Forster, and D. Scaramuzza, “REMODE: Probabilistic, monocular
dense reconstruction in real time,” Proc. - IEEE Int. Conf. Robot. Autom., pp. 2609–
2616, 2014, doi: 10.1109/ICRA.2014.6907233.
[81] C. Brand, M. J. Schuster, H. Hirschmüller, and M. Suppa, “Stereo-vision based
obstacle mapping for indoor/outdoor SLAM,” IEEE Int. Conf. Intell. Robot. Syst., no.
Iros, pp. 1846–1853, 2014, doi: 10.1109/IROS.2014.6942805.
[82] A. Palomer, P. Ridao, and D. Ribas, “Multibeam 3D underwater SLAM with
probabilistic registration,” Sensors (Switzerland), vol. 16, no. 4, 2016, doi:
10.3390/s16040560.
[83] C. Roman and H. Singh, “Consistency based error evaluation for deep sea
bathymetric mapping with robotic vehicles,” Proc. - IEEE Int. Conf. Robot. Autom.,
no. May, pp. 3568–3574, 2006, doi: 10.1109/ROBOT.2006.1642247.
[84] R. Gomez-Ojeda, F. A. Moreno, D. Zuñiga-Noël, D. Scaramuzza, and J. Gonzalez-
Jimenez, “PL-SLAM: A Stereo SLAM System Through the Combination of Points
and Line Segments,” IEEE Trans. Robot., vol. 35, no. 3, pp. 734–746, 2019, doi:
10.1109/TRO.2019.2899783.
[85] R. Wang, X. Wang, M. Zhu, and Y. Lin, “Application of a Real-Time Visualization
Method of AUVs in Underwater Visual Localization,” Appl. Sci., vol. 9, no. 7, p.
1428, 2019, doi: 10.3390/app9071428.
[86] M. Ferrera, J. Moras, P. Trouvé-Peloux, V. Creuze, and D. Dégez, “The Aqualoc
Dataset: Towards Real-Time Underwater Localization from a Visual-Inertial-
Pressure Acquisition System,” IROS Work. - New Horizons Underw. Interv.
Mission. from Curr. Technol. to Futur. Appl., pp. 20–22, 2018, [Online]. Available:
http://arxiv.org/abs/1809.07076.
[87] B. Joshi et al., “Experimental Comparison of Open Source Visual-Inertial-Based
State Estimation Algorithms in the Underwater Domain,” IEEE/RSJ Int. Conf. Intell.
Robot. Syst. (IROS), Macau, China, pp. 7221–7227, 2019, [Online]. Available:
http://arxiv.org/abs/1904.02215.
[88] B. Allotta et al., “A comparison between EKF-based and UKF-based navigation
algorithms for AUVs localization,” MTS/IEEE Ocean. 2015 - Genova Discov.
Sustain. Ocean Energy a New World, pp. 1–5, 2015, doi: 10.1109/OCEANS-
Genova.2015.7271681.
[89] W. Li, L. Zhang, F. Sun, L. Yang, M. Chen, and Y. Li, “Alignment calibration of IMU
103
and Doppler sensors for precision INS/DVL integrated navigation,” Optik (Stuttg).,
vol. 126, no. 23, pp. 3872–3876, 2015, doi: 10.1016/j.ijleo.2015.07.187.
[90] A. Rossi, M. Pasquali, and M. Pastore, “Performance analysis of an inertial
navigation algorithm with DVL auto-calibration for underwater vehicle,” 2014 DGON
Inert. Sensors Syst. ISS 2014 - Proc., pp. 1–19, 2014, doi:
10.1109/InertialSensors.2014.7049481.
[91] W. Gao, J. Li, G. Zhou, and Q. Li, “Adaptive Kalman filtering with recursive noise
estimator for integrated SINS/DVL systems,” J. Navig., vol. 68, no. 1, pp. 142–161,
2015, doi: 10.1017/S0373463314000484.
[92] P. Liu, B. Wang, Z. Deng, and M. Fu, “INS/DVL/PS Tightly Coupled Underwater
Navigation Method with Limited DVL Measurements,” IEEE Sens. J., vol. 18, no. 7,
pp. 2994–3002, 2018, doi: 10.1109/JSEN.2018.2800165.
[93] A. Tal, I. Klein, and R. Katz, “Inertial navigation system/doppler velocity log
(INS/DVL) fusion with partial dvl measurements,” Sensors (Switzerland), vol. 17,
no. 2, pp. 1–20, 2017, doi: 10.3390/s17020415.
[94] T. Zhang, H. Shi, L. Chen, Y. Li, and J. Tong, “AUV positioning method based on
tightly coupled SINS/LBL for underwater acoustic multipath propagation,” Sensors
(Switzerland), vol. 16, no. 3, 2016, doi: 10.3390/s16030357.
[95] T. Zhang, L. Chen, and Y. Li, “AUV underwater positioning algorithm based on
interactive assistance of SINS and LBL,” Sensors (Switzerland), vol. 16, no. 1,
2016, doi: 10.3390/s16010042.
[96] A. Manzanilla, S. Reyes, M. Garcia, D. Mercado, and R. Lozano, “Autonomous
navigation for unmanned underwater vehicles: Real-time experiments using
computer vision,” IEEE Robot. Autom. Lett., vol. 4, no. 2, pp. 1351–1356, 2019, doi:
10.1109/LRA.2019.2895272.
[97] D. Li, D. Ji, J. Liu, and Y. Lin, “A Multi-Model EKF Integrated Navigation Algorithm
for Deep Water AUV,” Int. J. Adv. Robot. Syst., vol. 13, no. 1, 2016, doi:
10.5772/62076.
[98] Y. Chen, D. Zheng, P. A. Miller, and J. A. Farrell, “Underwater Inertial Navigation
with Long Baseline Transceivers: A Near-Real-Time Approach,” IEEE Trans.
Control Syst. Technol., vol. 24, no. 1, pp. 240–251, 2016, doi:
10.1109/TCST.2015.2429613.
[99] M. Ferrera, V. Creuze, J. Moras, and P. Trouvé-Peloux, “AQUALOC: An
underwater dataset for visual–inertial–pressure localization,” Int. J. Rob. Res., vol.
38, no. 14, pp. 1549–1559, 2019, doi: 10.1177/0278364919883346.
[100] “Autonomous Fiel Robotic Laboratory - Datasets.”
https://afrl.cse.sc.edu/afrl/resources/datasets/ (accessed Dec. 30, 2019).
[101] N. Farr, A. Bowen, J. Ware, C. Pontbriand, and M. Tivey, “An integrated,
underwater optical/acoustic communications system,” Ocean. IEEE Sydney,
Ocean., no. December, 2010, doi: 10.1109/OCEANSSYD.2010.5603510.
[102] M. Stojanovic and P.-P. J. Beaujean, “Acoustic Communication,” in Springer
Handbook of Ocean Engineering, M. R. Dhanak and N. I. Xiros, Eds. Cham:
Springer International Publishing, 2016, pp. 359–386.
[103] “920 Series ATM-925 - Acoustic Modems - Benthos.”
http://www.teledynemarine.com/920-series-atm-925?ProductLineID=8 (accessed
Dec. 26, 2019).
104
[104] “Micromodem :: Acoustic Communications Group.”
https://acomms.whoi.edu/micro-modem (accessed Dec. 26, 2019).
[105] “LinkQuest.” http://www.link-quest.com/html/uwm1000.htm (accessed Dec. 26,
2019).
[106] “48/78 Devices | EvoLogics.” https://evologics.de/acoustic-modem/48-78
[107] “Mats 3G, Underwater Acoustics - Sercel.”
http://www.sercel.com/products/Pages/mats3g.aspx (accessed Dec. 26, 2019).
[108] “L3Harris | Acoustic Modem GPM300.” https://www.l3oceania.com/mission-
systems/undersea-communications/acoustic-modem.aspx (accessed Dec. 26,
2019).
[109] “Micron Modem | Tritech | Outstanding Performance in Underwater Technology.”
https://www.tritech.co.uk/product/micron-data-modem (accessed Dec. 26, 2019).
[110] “M64 Acoustic Modem for Wireless Underwater Communication.”
https://bluerobotics.com/store/comm-control-power/acoustic-modems/wl-11003-1/
[111] Z. Yan, L. Wang, T. Wang, Z. Yang, T. Chen, and J. Xu, “Polar cooperative
navigation algorithm for multi-unmanned underwater vehicles considering
communication delays,” Sensors (Switzerland), vol. 18, no. 4, 2018, doi:
10.3390/s18041044.
[112] S. Giodini and B. Binnerts, “Performance of acoustic communications for AUVs
operating in the North Sea,” Ocean. MTS/IEEE Monterey, OCE, pp. 1–6, 2016, doi:
10.1109/OCEANS.2016.7761011.
[113] T. Yang, S. Yu, and Y. Yan, “Formation control of multiple underwater vehicles
subject to communication faults and uncertainties,” Appl. Ocean Res., vol. 82, no.
September 2018, pp. 109–116, 2019, doi: 10.1016/j.apor.2018.10.024.
[114] A. Abad, N. DiLeo, K. Fregene, and Ieee, “Decentralized Model Predictive Control
for UUV Collaborative Missions,” Ocean. 2017 - Anchorage, 2017.
[115] N. J. Hallin, J. Horn, H. Taheri, M. O’Rourke, and D. Edwards, “Message
anticipation applied to collaborating unmanned underwater vehicles,” Ocean.
MTS/IEEE KONA, pp. 1–10, 2017, doi: 10.23919/oceans.2011.6107098.
[116] A. Rajala, M. O’Rourke, and D. B. Edwards, “AUVish: An application-based
language for cooperating AUVs,” Oceans, 2006, doi:
10.1109/OCEANS.2006.307128.
[117] J. Potter, J. Alves, D. Green, G. Zappa, I. Nissen, and K. McCoy, “The JANUS
underwater communications standard,” Underw. Commun. Networking, UComms,
pp. 0–3, 2014, doi: 10.1109/UComms.2014.7017134.
[118] R. Petroccia, J. Alves, and G. Zappa, “Fostering the use of JANUS in operationally-
relevant underwater applications,” 3rd Underw. Commun. Netw. Conf. Ucomms
2016, no. Figure 1, pp. 1–5, 2016, doi: 10.1109/UComms.2016.7583424.
[119] J. Alves et al., “Moving JANUS forward: A look into the future of underwater
communications interoperability,” Ocean. MTS/IEEE Monterey, OCE, pp. 1–6,
2016, doi: 10.1109/OCEANS.2016.7761094.
[120] R. Petroccia, J. Alves, and G. Zappa, “JANUS-based services for operationally
relevant underwater applications,” IEEE J. Ocean. Eng., vol. 42, no. 4, pp. 994–
1006, 2017, doi: 10.1109/JOE.2017.2722018.
105
[121] K. McCoy, V. Djapic, and M. Ouimet, “JANUS: Lingua Franca,” Ocean. MTS/IEEE
Monterey, OCE, pp. 1–4, 2016, doi: 10.1109/OCEANS.2016.7761459.
[122] T. F. Wiener and S. Karp, “The Role of Blue/Green Laser Systems in Strategic
Submarine Communications,” IEEE Trans. Commun., vol. 28, no. 9, pp. 1602–
1607, 1980, doi: 10.1109/TCOM.1980.1094858.
[123] J. J. Puschell, R. J. Giannaris, and L. Stotts, “The Autonomous Data Optical Relay
Experiment: First two way laser communication between an aircraft and
submarine,” 1992, doi: 10.1109/NTC.1992.267865.
[124] Z. Enqi and W. Hongyuan, “Research on spatial spreading effect of blue-green
laser propagation through seawater and atmosphere,” nternational Conf. E-bus. Inf.
Syst. Secur. EBISS, pp. 1–4, 2009, doi: 10.1109/EBISS.2009.5138057.
[125] R. S. Sangeetha, R. L. Awasthi, and T. Santhanakrishnan, “Design and analysis of
a laser communication link between an underwater body and an air platform,” Int.
Conf. Next Gener. Intell. Syst. ICNGIS, pp. 1–5, 2017, doi:
10.1109/ICNGIS.2016.7854012.
[126] G. Cossu et al., “Experimental demonstration of high speed underwater visible light
communications,” Proc. 2nd Int. Work. Opt. Wirel. Commun. IWOW, no. 1, pp. 11–
15, 2013, doi: 10.1109/IWOW.2013.6777767.
[127] L. Luqi, “Utilization and risk of undersea communications,” Ocean. MTS/IEEE
Monterey, OCE, pp. 1–7, 2016, doi: 10.1109/OCEANS.2016.7761504.
[128] T. Elmokadem, M. Zribi, and K. Youcef-Toumi, “Control for Dynamic Positioning
and Way-point Tracking of Underactuated Autonomous Underwater Vehicles Using
Sliding Mode Control,” J. Intell. Robot. Syst. Theory Appl., vol. 95, no. 3–4, pp.
1113–1132, 2019, doi: 10.1007/s10846-018-0830-8.
[129] Y. Li, Y. qing Jiang, L. feng Wang, J. Cao, and G. cheng Zhang, “Intelligent PID
guidance control for AUV path tracking,” J. Cent. South Univ., vol. 22, no. 9, pp.
3440–3449, 2015, doi: 10.1007/s11771-015-2884-0.
[130] W. Y. Gan, D. Q. Zhu, W. L. Xu, and B. Sun, “Survey of trajectory tracking control
of autonomous underwater vehicles,” J. Mar. Sci. Technol., vol. 25, no. 6, pp. 722–
731, 2017, doi: 10.6119/JMST-017-1226-13.
[131] J. Guerrero, J. Torres, V. Creuze, A. Chemori, and E. Campos, “Saturation based
nonlinear PID control for underwater vehicles: Design, stability analysis and
experiments,” Mechatronics, vol. 61, no. May 2018, pp. 96–105, 2019, doi:
10.1016/j.mechatronics.2019.06.006.
[132] L. Qiao, B. Yi, D. Wu, and W. Zhang, “Design of three exponentially convergent
robust controllers for the trajectory tracking of autonomous underwater vehicles,”
Ocean Eng., vol. 134, no. February, pp. 157–172, 2017, doi:
10.1016/j.oceaneng.2017.02.006.
[133] J. Rojas, G. Baatar, F. Cuellar, M. Eichhorn, and T. Glotzbach, “Modelling and
Essential Control of an Oceanographic Monitoring Remotely Operated Underwater
Vehicle,” IFAC-PapersOnLine, vol. 51, no. 29, pp. 213–219, 2018, doi:
10.1016/j.ifacol.2018.09.495.
[134] M. Dong, J. Li, and W. Chou, “Depth control of ROV in nuclear power plant based
on fuzzy PID and dynamics compensation,” Microsyst. Technol., vol. 26, no. 3, pp.
811–821, 2020, doi: 10.1007/s00542-019-04605-x.
[135] M. Yang, Z. Sheng, Y. Che, J. Hu, K. Hu, and Y. Du, “Design of Small Monitoring
106
ROV for Aquaculture,” pp. 1–9, 2019, doi: 10.1109/oceanse.2019.8867487.
[136] L. Zhang, L. Zhang, S. Liu, J. Zhou, and C. Papavassiliou, “Low-level control
technology of micro autonomous underwater vehicle based on intelligent
computing,” Cluster Comput., vol. 22, pp. 8569–8580, 2019, doi: 10.1007/s10586-
018-1909-5.
[137] F. Rezazadegan, K. Shojaei, F. Sheikholeslam, and A. Chatraei, “A novel approach
to 6-DOF adaptive trajectory tracking control of an AUV in the presence of
parameter uncertainties,” Ocean Eng., vol. 107, pp. 246–258, 2015, doi:
10.1016/j.oceaneng.2015.07.040.
[138] C. Yu, X. Xiang, Q. Zhang, and G. Xu, “Adaptive Fuzzy Trajectory Tracking Control
of an Under-Actuated Autonomous Underwater Vehicle Subject to Actuator
Saturation,” Int. J. Fuzzy Syst., vol. 20, no. 1, pp. 269–279, 2018, doi:
10.1007/s40815-017-0396-9.
[139] G. R. Cho, J. H. Li, D. Park, and J. H. Jung, “Robust trajectory tracking of
autonomous underwater vehicles using back-stepping control and time delay
estimation,” Ocean Eng., vol. 201, no. August 2019, p. 107131, 2020, doi:
10.1016/j.oceaneng.2020.107131.
[140] L. G. García-Valdovinos, T. Salgado-Jiménez, M. Bandala-Sánchez, L. Nava-
Balanzar, R. Hernández-Alvarado, and J. A. Cruz-Ledesma, “Modelling, Design
and Robust Control of a Remotely Operated Underwater Vehicle,” Int. J. Adv.
Robot. Syst., vol. 11, no. 1, 2014, doi: 10.5772/56810.
[141] Z. Yan, Z. Yang, J. Zhang, J. Zhou, A. Jiang, and X. Du, “Trajectory tracking control
of uuv based on backstepping sliding mode with fuzzy switching gain in diving
plane,” IEEE Access, vol. 7, pp. 166788–166795, 2019, doi:
10.1109/ACCESS.2019.2953530.
[142] Z. Yan, M. Wang, and J. Xu, “Integrated guidance and control strategy for homing
of unmanned underwater vehicles,” J. Franklin Inst., vol. 356, no. 7, pp. 3831–3848,
2019, doi: 10.1016/j.jfranklin.2018.11.042.
[143] J. Zhou, X. Zhao, T. Chen, Z. Yan, and Z. Yang, “Trajectory Tracking Control of an
Underactuated AUV Based on Backstepping Sliding Mode with State Prediction,”
IEEE Access, vol. 7, pp. 181983–181993, 2019, doi:
10.1109/ACCESS.2019.2958360.
[144] B. Huang and Q. Yang, “Double-loop sliding mode controller with a novel switching
term for the trajectory tracking of work-class ROVs,” Ocean Eng., vol. 178, no.
February, pp. 80–94, 2019, doi: 10.1016/j.oceaneng.2019.02.043.
[145] T. Lv, J. Zhou, Y. Wang, W. Gong, and M. Zhang, “Sliding mode based fault tolerant
control for autonomous underwater vehicle,” Ocean Eng., vol. 216, no. June 2019,
2020, doi: 10.1016/j.oceaneng.2020.107855.
[146] L. G. García-Valdovinos, F. Fonseca-Navarro, J. Aizpuru-Zinkunegi, T. Salgado-
Jiménez, A. Gómez-Espinosa, and J. A. Cruz-Ledesma, “Neuro-Sliding Control for
Underwater ROV’s Subject to Unknown Disturbances,” Sensors, vol. 19, no. 13, p.
2943, 2019, doi: 10.3390/s19132943.
[147] M. R. Ramezani-al and Z. Tavanaei Sereshki, “A novel adaptive sliding mode
controller design for tracking problem of an AUV in the horizontal plane,” Int. J. Dyn.
Control, vol. 7, no. 2, pp. 679–689, 2019, doi: 10.1007/s40435-018-0457-4.
[148] V. Parra-Vega, L. Garcia-Valdovinos, and O. A. Dominguéz-Ramiréz, “Sliding PID
107
Control for Tracking in Finite Time for Robot Arms,” 11th Int. Conf. Adv. Robot., no.
Proceedings of ICAR 2003, pp. 1526–1531, 2003.
[149] L. G. García-Valdovinos, V. Parra-Vega, and M. A. Arteaga, “Bilateral Cartesian
sliding PID force/position control for tracking in finite time of master-slave systems,”
Proc. Am. Control Conf., vol. 2006, pp. 369–375, 2006, doi:
10.1109/acc.2006.1655383.
[150] Z. Yan, H. Yu, W. Zhang, B. Li, and J. Zhou, “Globally finite-time stable tracking
control of underactuated UUVs,” Ocean Eng., vol. 107, pp. 132–146, 2015, doi:
10.1016/j.oceaneng.2015.07.039.
[151] H. Yu, C. Guo, and Z. Yan, “Globally finite-time stable three-dimensional trajectory-
tracking control of underactuated UUVs,” Ocean Eng., vol. 189, no. March, p.
106329, 2019, doi: 10.1016/j.oceaneng.2019.106329.
[152] Z. Chu, D. Zhu, and S. X. Yang, “Observer-based adaptive neural network
trajectory tracking control for remotely operated vehicle,” IEEE Trans. Neural
Networks Learn. Syst., vol. 28, no. 7, pp. 1633–1645, 2017, doi:
10.1109/TNNLS.2016.2544786.
[153] L. Qiao and W. Zhang, “Double-Loop Integral Terminal Sliding Mode Tracking
Control for UUVs with Adaptive Dynamic Compensation of Uncertainties and
Disturbances,” IEEE J. Ocean. Eng., vol. 44, no. 1, pp. 29–53, 2019, doi:
10.1109/JOE.2017.2777638.
[154] J. Guerrero, J. Torres, V. Creuze, and A. Chemori, “Trajectory tracking for
autonomous underwater vehicle: An adaptive approach,” Ocean Eng., vol. 172, no.
August 2018, pp. 511–522, 2019, doi: 10.1016/j.oceaneng.2018.12.027.
[155] X. Liu, M. Zhang, C. Yang, and B. Yin, “Finite-time tracking control for autonomous
underwater vehicle based on an improved non-singular terminal sliding mode
manifold,” Int. J. Control, vol. 95, no. 3, pp. 840–849, 2020, doi:
10.1080/00207179.2020.1825818.
[156] S. Liu, Y. Liu, Z. Chen, and X. Liang, “Finite-time tracking control of underwater
vehicles with system uncertainties and unknown disturbances,” Chinese Control
Conf. CCC, pp. 6581–6585, 2017, doi: 10.23919/ChiCC.2017.8028400.
[157] G. Casalino, E. Simetti, N. Manerikar, A. Sperinde, S. Torelli, and F. Wanderlingh,
“Cooperative underwater manipulation systems: Control developments within the
MARIS project,” IFAC-PapersOnLine, vol. 28, no. 2, pp. 1–7, 2015, doi:
10.1016/j.ifacol.2015.06.001.
[158] N. Manerikar, G. Casalino, E. Simetti, S. Torelli, and A. Sperindé, “On autonomous
cooperative Underwater Floating Manipulation Systems,” Proc. - IEEE Int. Conf.
Robot. Autom., vol. 2015-June, no. June, pp. 523–528, 2015, doi:
10.1109/ICRA.2015.7139229.
[159] E. Simetti and G. Casalino, “Manipulation and Transportation with Cooperative
Underwater Vehicle Manipulator Systems,” IEEE J. Ocean. Eng., vol. 42, no. 4, pp.
782–799, 2017, doi: 10.1109/JOE.2016.2618182.
[160] J. Sakiyama and N. Motoi, “Position and attitude control method using disturbance
observer for station keeping in underwater vehicle,” Proc. IECON 2018 - 44th Annu.
Conf. IEEE Ind. Electron. Soc., vol. 1, pp. 5469–5474, 2018, doi:
10.1109/IECON.2018.8591108.
[161] M. T. Vu et al., “Station-Keeping Control of a Hovering Over-Actuated Autonomous
108
Underwater Vehicle under Ocean Current Effects and Model Uncertainties in
Horizontal Plane,” IEEE Access, vol. 9, pp. 6855–6867, 2021, doi:
10.1109/ACCESS.2020.3048706.
[162] J. Hwang, N. Bose, and S. Fan, “AUV Adaptive Sampling Methods: A Review,”
Appl. Sci., vol. 9, no. 15, p. 3145, 2019, doi: 10.3390/app9153145.
[163] Z. Yan, Z. Yang, L. Yue, L. Wang, H. Jia, and J. Zhou, “Discrete-time coordinated
control of leader-following multiple AUVs under switching topologies and
communication delays,” Ocean Eng., vol. 172, no. May 2018, pp. 361–372, 2019,
doi: 10.1016/j.oceaneng.2018.12.018.
[164] Z. Lichuan, F. Jingxiang, W. Tonghao, G. Jian, and Z. Ru, “A new algorithm for
collaborative navigation without time synchronization of multi-UUVS,” Ocean. -
Aberdeen, vol. 2017-Octob, pp. 1–6, 2017, doi: 10.1109/OCEANSE.2017.8084717.
[165] Z. Yan, D. Xu, T. Chen, W. Zhang, and Y. Liu, “Leader-follower formation control of
UUVs with model uncertainties, current disturbances, and unstable
communication,” Sensors (Switzerland), vol. 18, no. 2, 2018, doi:
10.3390/s18020662.
[166] L. Z. Cui, Jian, “Adaptive consensus tracking control for multiple autonomous
underwater vehicles with uncertain parameters,” ICIC Express Lett., vol. 13, no. 3,
pp. 191–200, 2019, doi: 10.1201/b10463.
[167] T. Y. Teck, M. Chitre, and F. S. Hover, “Collaborative bathymetry-based localization
of a team of autonomous underwater vehicles,” Proc. - IEEE Int. Conf. Robot.
Autom., pp. 2475–2481, 2014, doi: 10.1109/ICRA.2014.6907204.
[168] Y. T. Tan, R. Gao, and M. Chitre, “Cooperative path planning for range-only
localization using a single moving beacon,” IEEE J. Ocean. Eng., vol. 39, no. 2, pp.
371–385, 2014, doi: 10.1109/JOE.2013.2296361.
[169] D. De Palma, G. Indiveri, and G. Parlangeli, “Multi-vehicle relative localization
based on single range measurements,” IFAC-PapersOnLine, vol. 28, no. 5, pp. 17–
22, 2015, doi: 10.1016/j.ifacol.2015.06.457.
[170] J. Li, J. Zhang, G. Zhang, and B. Zhang, “An adaptive prediction target search
algorithm for multi-AUVs in an unknown 3D environment,” Sensors (Switzerland),
vol. 18, no. 11, 2018, doi: 10.3390/s18113853.
[171] S. lv and Y. Zhu, “A Multi-AUV Searching Algorithm Based on Neuron Network with
Obstacle,” Int. Symp. Auton. Syst., pp. 131–136, 2019, doi:
10.1109/isass.2019.8757793.
[172] D. Z. Bing Sun, “Complete Coverage Autonomous Underwater Vehicles Path
Planning Based on Glasius Bio-Inspired Neural Network Algorithm for Discrete and
Centralized Programming,” IEEE Trans. Cogn. Dev. Syst., vol. 11, 2019.
[173] Z. Yan, X. Liu, J. Zhou, and D. Wu, “Coordinated Target Tracking Strategy for
Multiple Unmanned Underwater Vehicles with Time Delays,” IEEE Access, vol. 6,
pp. 10348–10357, 2018, doi: 10.1109/ACCESS.2018.2793338.
[174] N. Palomeras et al., “I-AUV docking and panel intervention at sea,” Sensors
(Switzerland), vol. 16, no. 10, pp. 1–18, 2016, doi: 10.3390/s16101673.
[175] G. Casalino et al., “Underwater intervention robotics: An outline of the Italian
national project Maris,” Mar. Technol. Soc. J., vol. 50, no. 4, pp. 98–107, 2016, doi:
10.4031/MTSJ.50.4.7.
[176] E. Simetti, G. Casalino, N. Manerikar, A. Sperinde, S. Torelli, and F. Wanderlingh,
109
“Cooperation between autonomous underwater vehicle manipulations systems with
minimal information exchange,” MTS/IEEE Ocean. - Genova Discov. Sustain.
Ocean Energy a New World, 2015, doi: 10.1109/OCEANS-Genova.2015.7271700.
[177] R. Conti, E. Meli, A. Ridolfi, and B. Allotta, “An innovative decentralized strategy for
I-AUVs cooperative manipulation tasks,” Rob. Auton. Syst., vol. 72, pp. 261–276,
2015, doi: 10.1016/j.robot.2015.06.006.
[178] E. Cataldi, S. Chiaverini, and G. Antonelli, “Cooperative Object Transportation by
Two Underwater Vehicle-Manipulator Systems,” MED - 26th Mediterr. Conf. Control
Autom., pp. 161–166, 2018, doi: 10.1109/MED.2018.8442760.
[179] S. Heshmati-alamdari, G. C. Karras, and K. J. Kyriakopoulos, “A Distributed
Predictive Control Approach for Cooperative Manipulation of Multiple Underwater
Vehicle Manipulator Systems,” IEEE Int. Conf. Robot. Autom., pp. 4626–4632,
2019.
[180] M. Prats, J. Perez, J. J. Fernandez, and P. J. Sanz, “An open source tool for
simulation and supervision of underwater intervention missions,” IEEE Int. Conf.
Intell. Robot. Syst., pp. 2577–2582, 2012, doi: 10.1109/IROS.2012.6385788.
[181] S. Bhattacharya et al., “Blockchain for Internet of Underwater Things: State-of-the-
Art, Applications, Challenges, and Future Directions,” Sustainability, vol. 14, no. 23,
p. 15659, Nov. 2022, doi: 10.3390/su142315659.
[182] J. Khoury, J. Khoury, G. Zouein, and J. P. Arnaout, “A practical decentralized
access protocol for autonomous vehicles at isolated under-saturated intersections,”
J. Intell. Transp. Syst. Technol. Planning, Oper., vol. 23, no. 5, pp. 427–440, 2019,
doi: 10.1080/15472450.2018.1536859.
[183] BlueRobotics, “Affordable and capable Underwater Robot,” 2022.
https://bluerobotics.com/store/rov/bluerov2/ (accessed Apr. 03, 2022).
[184] T. I. Fossen, Handbook of Marine Craft Hydrodynamics and Motion Control.
Chichester, UK: John Wiley & Sons, Ltd, 2011.
[185] S. Lack, E. Rentzow, and T. Jeinsch, “Experimental Parameter Identification for an
open-frame ROV: Comparison of towing tank tests and open water self-propelled
tests,” IFAC-PapersOnLine, vol. 52, no. 21, pp. 271–276, 2019, doi:
10.1016/j.ifacol.2019.12.319.
[186] C. Wu, “6-DoF Modelling and Control of a Remotely Operated Vehicle,” Flinders
University, 2018.
[187] L. G. García-Valdovinos, T. Salgado-Jiménez, M. Bandala-Sánchez, L. Nava-
Balanzar, R. Hernández-Alvarado, and J. A. Cruz-Ledesma, “Modelling, Design
and Robust Control of a Remotely Operated Underwater Vehicle,” Int. J. Adv.
Robot. Syst., vol. 11, no. 1, 2014, doi: 10.5772/56810.
[188] V. Parra-Vega, “Second order sliding mode control for robot arms with time base
generators for finite-time tracking,” Dyn. Control, vol. 11, no. 2, pp. 175–186, 2001,
doi: 10.1023/A:1012535929651.
[189] V. Parra-Vega, S. Arimoto, Y. H. Liu, G. Hirzinger, and P. Akella, “Dynamic sliding
PID control for tracking of robot manipulators: Theory and experiments,” IEEE
Trans. Robot. Autom., vol. 19, no. 6, pp. 967–976, 2003, doi:
10.1109/TRA.2003.819600.
[190] A. Levant, “Higher-order sliding modes, differentiation and output-feedback
control,” Int. J. Control, vol. 76, no. 9–10, pp. 924–941, 2003, doi:
110
10.1080/0020717031000099029.
[191] J. González-García et al., “Model-free high order sliding mode control with finite-
time tracking for unmanned underwater vehicles,” Appl. Sci., vol. 11, no. 4, pp. 1–
22, 2021, doi: 10.3390/app11041836.
[192] J. González-García, A. Gómez-Espinosa, L. G. García-Valdovinos, T. Salgado-
Jiménez, E. Cuan-Urquizo, and J. A. E. Cabello, “Experimental Validation of a
Model-Free High-Order Sliding Mode Controller with Finite-Time Convergence for
Trajectory Tracking of Autonomous Underwater Vehicles,” Sensors, vol. 22, no. 2,
2022, doi: 10.3390/s22020488.
[193] S. Zhang, J. Yu, A. Zhang, and F. Zhang, “Spiraling motion of underwater gliders:
Modeling, analysis, and experimental results,” Ocean Eng., vol. 60, pp. 1–13, 2013,
doi: 10.1016/j.oceaneng.2012.12.023.
[194] K. D. Do and J. Pan, Control of Ships and Underwater Vehicles. London: Springer
Science & Business Media, 2009.
[195] J. González-García, A. Gómez-Espinosa, L. G. García-Valdovinos, T. Salgado-
Jiménez, E. Cuan-Urquizo, and J. A. E. Cabello, “Model-Free High-Order Sliding
Mode Controller for Station-Keeping of an Autonomous Underwater Vehicle in
Manipulation Task: Simulations and Experimental Validation,” Sensors, vol. 22, no.
12, p. 4347, 2022, doi: 10.3390/s22124347.
[196] J. González-García, N. A. Narcizo-Nuci, A. Gómez-Espinosa, L. G. García-
Valdovinos, and T. Salgado-Jiménez, “Finite-Time Controller for Coordinated
Navigation of Unmanned Underwater Vehicles in a Collaborative Manipulation
Task,” Sensors, vol. 23, no. 1, p. 239, Dec. 2022, doi: 10.3390/s23010239.
111
Published papers
Autonomous underwater vehicles: Localization, navigation, and
communication for collaborative missions
112
Model-free high order sliding mode control with finite-time tracking for
unmanned underwater vehicles
113
Experimental Validation of a Model-Free High-Order Sliding Mode Controller
with Finite-Time Convergence for Trajectory Tracking of Autonomous Underwater
Vehicles
114
Model-Free High-Order Sliding Mode Controller for Station-Keeping of an
Autonomous Underwater Vehicle in Manipulation Task: Simulations and
Experimental Validation
115
Finite-Time Controller for Coordinated Navigation of Unmanned Underwater
Vehicles in a Collaborative Manipulation Task
116
Resume
Josué González García was born in Tamaulipas, México, on July 4th, 1990. He
earned the Mechatronics Engineering degree from the Instituto Tecnológico de
Querétaro, in June 2013. He was accepted in the graduate programs in Manufacturing
Systems at the Instituto Tecnológico y de Estudios Superiores de Monterrey and earned
his master’s degree in May 2015. He worked as a lecturer in the same institution until he
was accepted to their Doctoral program in Engineering Science in August 2019.
This document was typed in using Microsoft Word by Josué González García
117

0.1tesis 2023 03 20

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

0.1tesis 2023 03 20

Uploaded by

Copyright:

Available Formats

Instituto Tecnológico y de Estudios Superiores de Monterrey

School of Engineering and Sciences

Object Manipulation by Collaborative Autonomous Underwater

Josué González García

Monterrey Nuevo León, June 1st, 2023

School of Engineering and Sciences

Monterrey Nuevo León, June 1st, 2023

@2023 by Josué González García

Thanks for all your unconditional confidence, support, patience, and

I would like to express my deepest gratitude to Tecnológico de Monterrey for the

Josué González García

1.2 Problem statement and context

1.3 Research question

This research aims to answer the question:

Is it possible to set up a team of underwater vehicles to autonomously approach,

1.4 Solution overview

• State-of-the-art research: An extensive literature review regarding

1.5 Main contributions

This work expects to contribute to underwater robotics research, particularly to

Previous contributions are also expected to be published as research papers in

1.6 Thesis structure

The rest of the document is structured as follows:

Chapter 2: Literature review

J. González-García, A. Gómez-Espinosa, E. Cuan-Urquizo, L. G. García-

Other challenges must be addressed besides the problem of self-localization and

2.2 Underwater vehicle localization

This section summarizes the underwater localization methods found in the

Figure 2. Dead-Reckoning drift effect.

Figure 3. Vertical deflection.

Heading Pitch and Roll Data

Figure 5. Forward-Looking SONAR (FLS) placed at vertical and horizontal orientations.

(a) (b) (c)

A schematic diagram of an SBL system is represented in Figure 7. Three

Figure 7. The schematic diagram for Short Baseline (SBL) system.

Operating Depth Range

Figure 8. Range-only/Single-beacon positioning of a fixed target from a moving vehicle.

Figure 9. Geomagnetic map features.

Figure 10. Terrain-Referenced Navigation.

There are different SLAM representation methods used to reconstruct the

Method Type Description Applications

Underwater SLAM can be categorized as acoustic-based and vision-based [39].

Attitude error = √(𝑒𝑟𝑟𝑜𝑟_𝑥)2 + (𝑒𝑟𝑟𝑜𝑟_𝑦)2 + (𝑒𝑟𝑟𝑜𝑟_𝑧)2 (9)

Figure 14. Sensor fusion alternatives for AUV positioning.

2.2.7 Localization and navigation overview

After the literature review, it can be considered that acoustic-based technologies

2.3 Underwater communication

Max Bit Rate Frequency Band

The working principles of underwater acoustic communication can be described as

2.4 Autonomous navigation of AUVs

Autonomous navigation can be performed using waypoint tracking, path following,

Figure 15. Waypoint navigation representation.

Figure 16. Path following navigation representation.

In the trajectory tracking problem, the vehicle must follow a time-parametrized

Figure 17. Trajectory tracking navigation representation.

In manipulation operations, autonomous navigation drives the vehicle to the object

2.6 Collaborative AUVs

Collaborative work refers to an interaction of two or more AUVs to perform a

In a leader-follower scheme, as illustrated in Figure 19, the leader AUV is equipped

Figure 19. Leader-follower navigation of AUVs.

Figure 20. Collaborative mission of AUVs. a) Survey missions. b) Intervention missions.

Collaborative survey missions

3.1 Kinematics and hydrodynamics