Professional Documents
Culture Documents
DOI: 10.5772/56059
© 2013 Ding et al.; licensee InTech. This is an open access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
Figure 1. Mechanism design of AmphiRobot‐II. (a) Concept
design. (b) Fish‐like swimming mode. (c) Dolphin‐like
swimming mode.
In synthesizing the motion characteristics of both ground
mobile robots and underwater robots, this paper mainly
presents a novel amphibious robot called “AmphiRobot‐
II” with a pair of bio‐inspired wheel‐propeller‐fins. It is Figure 2. Configuration of the CPG model for the AmphiRobot‐II.
an improved version of the AmphiRobot‐I developed in
In this paper, CPG‐based locomotion control in
[12]. As shown in Figure 1, mechanically the robot is
conjunction with Lagrangian dynamics is employed to
composed of a head with wheel‐propeller‐fins, three fish‐
tackle the aquatic propulsion of AmphiRobot‐II. In
like propelling units and a flapping caudal fin. When
particular, the integration of the Lagrange dynamics and
moving on land, the robot implements wheel‐based
the CPG‐based control guides the search for CPG
movements, whereby the wheel‐propeller‐fins function as
parameters, steady locomotion gaits and preliminary
wheels; when swimming in water, the robot performs
validation, greatly facilitating the creation of versatile
fish‐like (lateral oscillations) or dolphin‐like (dorsoventral
gaits. As a result, the amphibious robot equipped with
oscillations) propulsion, where the wheel‐propeller‐fins
wheel‐propeller‐fins successfully possesses more novel
function as pectoral fins. Specifically, the robot takes fish‐
amphibious gaits, enhancing its motion ability and
like swimming as the primary underwater propulsor by
possible operations.
utilizing the travelling waves of body undulation
transmitted from head to tail. That is, several fish‐like
The rest of the paper is organized as follows. In Section 2,
oscillating hinge joints (J1–J4) comprise the drive
the adopted CPG‐based locomotion control is briefly
mechanism (yaw control) while coordinated with the
described. The formulated Lagrangian dynamics are
pectoral fins (pitch control). A specially‐designed
detailed in Section 3. The simulated and experimental
swivelling mechanism is further designed to rotate the
results are provided in Section 4. Finally, Section 5
posterior body by ±90° (roll control) to switch gaits
concludes the paper with an outline of future work.
between fish‐ and dolphin‐like modes.
2. CPG‐based multimodal locomotion control
To generate a fish body wave, Lighthill proposed a
carangiform propulsive wave equation in the 1960s [13]. To estimate the oscillation property of a fish body,
In 1996, Barrett et al. developed an experimental model of Lighthill’s elongated body theory has been widely used
the RoboTuna’s swimming kinematics in terms of a set of to generate a propulsive travelling wave. Many other
seven key parameters that can determine swimming aspects of fish‐swimming behaviour have been studied to
performance [14]. These sine‐based approaches use shed light on the propulsion mechanism, which
simple time‐indexed sine‐based functions for generating demonstrates an implied undulatory wave travelling
www.intechopen.com Rui Ding, Junzhi Yu, Qinghai Yang and Min Tan: Dynamic Modelling 3
of a CPG-Controlled Amphibious Biomimetic Swimming Robot
Figure 3. Dynamic model of the amphibious swimming robot. (a) System configuration of a five‐link serial mechanism and reference
frames description. (b) Pressure on the ith link. (c) Approach stream pressure on the head. (d) Friction drag. (e) Pressure on the left
pectoral fin.
www.intechopen.com Rui Ding, Junzhi Yu, Qinghai Yang and Min Tan: Dynamic Modelling 5
of a CPG-Controlled Amphibious Biomimetic Swimming Robot
d L L produces the perpendicular pressure and parallel friction
FX dt X X drag in relation to the surface. The thickness of the
d L L pectoral fin is small and the friction drag can be ignored
FY
Y (7) simply by considering the perpendicular component that
dt Y
d L L contributes to propulsion (see Figure 3e). As with the
M dt
moving links, the pressure on the pectoral fin is given by:
FL L v L |v L | , L C1SL / 2 , where C1 is the shape
where FX and FY are the compositions of the coefficient adopting the flat plate‐type, SL is the effective
hydrodynamic forces on the X axis and Y axis, area of the left pectoral fin and v L is the normal velocity
respectively. MΘ is the moment about the point (X, Y). perpendicular to the surface of the left pectoral fin which
Notice that X, Y and MΘ are the functions of time can be calculated from (5).
evolution ‐ i.e., X(t), Y(t) and MΘ(t) ‐ which are
fundamentally decided by the joint actuation laws. Similarly, FL is resolved into components along the
direction of the head and perpendicular to the surface of
3.3 Hydrodynamic force approximation the pectoral fin. The perpendicular component mainly
The interaction of a robot swimming underwater with the contributes to an up‐and‐down locomotion, which will
fluid flow is a crucial issue that remains unresolved. The not be considered here. The force along the head can be
hydrodynamic aspects of swimming have been described expressed as below:
in various models [25,26]. In many cases, some
simplifications are necessary. To calculate the resultant FL sin L (FL sin L cos1 )i (FL sin L sin 1 ) j
forces, a large Reynolds number is applied and all of the
where i and j are the unit vectors along the X axis and Y
forces acting on any propulsive element are due to the
motion of that element in the fluid. The movement of any axis, respectively. Corresponding results for the right
object through a stationary fluid causes an increase in pectoral fin can be obtained by analogy, as denoted by FR .
pressure in front of the object and a decrease behind it.
3.3.2 Approach stream pressure
The different pressures on the two sides also give a net
drag force to the object, counteracting the movement. During swimming, the robot suffers from a resulting
Forces acting on the body from the surrounding water force which acts upon the head in opposition to the
depend upon the speed of the body relative to the water. motion ‐ i.e., the approach stream pressure (see Figure
The movement of robot’s propulsive elements through 3c). Owing to a larger cross‐section of the head with small
the fluid causes pressure differentials to develop on either oscillation amplitudes of the rear body, the drag forces
side of the elements, resulting in drag forces in opposition acting on the other links can be ignored, except for the
to the motion. Next, each link is acted upon by three head. The drag force on the head is given by:
forces: pressure on the links, the approach stream F1 1 v1 |v1 |, where v1 is the component of the
pressure and friction drag. velocity parallel to the surface of the head relative to the
fluid, which is determined by (3). 1 C2S / 2 is the
3.3.1 Pressure on the links
drag coefficient with C2 of the bullet‐type [25] ‐ i.e., C2 =
While oscillating, the hydrodynamic force acting 0.295. S is the cross‐section area of the head.
perpendicular to the surface of the ith link is the thrust for
3.3.3 Friction drag
advancement (see Figure 3b), which is given by:
Fi i v i |vi | , where v i is the component of the Friction drag arises as a result of the viscosity of the fluid
velocity perpendicular to the surface of ith link relative to flow between the fish body and the boundary layer of the
the fluid, which can be obtained from (3). Likewise, water (see Figure 3d). The movement of the robot’s
i C1S i / 2 is the drag coefficient with C1 of a flat propelling units through the fluid causes friction drag
plate‐type (i.e., C1 = 1.28 [27]. Si is the effective area of the
parallel to the propelling units, resulting in drag forces
ith link).
adverse to the motion. It is empirically evaluated as
20−50% of the approach stream pressure (i.e.,
Many fish balance vertical forces by continually
Ff (0.2 0.5)F1 ). In view of the matte body of
swimming, with their pectoral fins flapping and while
AmphiRobot‐II, a factor of 50% is adopted here.
maintaining a steady swimming speed. Thanks to the
pectoral fins mounted on both sides of the head, the robot 3.3.4 Composition of the hydrodynamic forces
is more agile for turning or maintaining balance in the
water. The propulsive forces generated by the pectoral By resolving all of the above forces in the WRCS, the
fins in conjunction with the rear body promote the robot components of the forces in the direction of the X‐axis
in moving forwards. While oscillating, the pectoral fin and the Y‐axis can be rewritten as:
5 By assigning the same drive between the left and right
FX Fix FLx FRx F1x Ffx
sides of the model ‐ e.g., dL=dR=3 ‐ the corresponding
i 1
5
oscillatory frequencies and amplitudes for moving joints
FY Fiy FLy
FRy
F1y Ffy can be obtained from the CPG model, which can then be
i 1 (10) imported into the Lagrangian model. Figure 4a shows the
5 5
M [Fix (y if Y)] Fiy (xif X) FLx
(y Lf Y) displacement of the point (X,Y), where the robot moved
i 1 i 1 in the negative‐direction of the X‐axis. The orientation is
not strictly in the negative direction, for the reason that
FRx
(y Rf Y) FLy
(x fL X) FRy
(xfR X)
the forces acting upon the robot are not absolutely
symmetrical at the start time, as illustrated in Figures 4c
To further solve the above equations, some initial values
and 4d. The oscillation curve of Θ in Figure 4b partly
should be given. Suppose that the robot is at the origin at
demonstrates the relative accuracy of the hydrodynamic
the initial moment and that the robot body maintains
model. The moment acting on the point (X, Y) is plotted
itself motionless in a line ‐ i.e., X(0)=0, Y(0)=0, Θ(0)=0,
in Figure 4e. It is a vital parameter for the servomotor
X(0)
0 , Y(0) 0 , (0) 0 ‐ and that the
selection and the maximum moment of the servomotor
hydrodynamic model is then decided by (10). should satisfy the simulation result. With the simulated
Furthermore, the body parameters such as masses, results, the average velocity can be estimated by:
lengths and lateral areas for all the links and the pectoral
fins, are needed. The parameters measured and estimated
φ(rad)
from the physical robot are listed in Table 2. In particular, φR
the joint angles i , L and R ‐ i.e., oscillation motion 10
suitable CPG control parameters and conceived gaits can φ2
4
be searched and eventually better locomotion φ3
performance will be acquired. 2
φ4
0
Notice that, in a similar fashion to the hydrodynamic t (s)
5 10 15 20
modelling of fish‐like lateral motions, dolphin‐like -2
dorsoventral motions can be dynamically simulated with 9cm Y (m)
the same kinematic analysis and hydrodynamic force X (m)
-6 -5 -4 -3 -2 -1 0
approximation [28]. Hence, a comparison between fish‐
(a)
and dolphin‐like swimming in the same robotic platform
becomes available.
www.intechopen.com Rui Ding, Junzhi Yu, Qinghai Yang and Min Tan: Dynamic Modelling 7
of a CPG-Controlled Amphibious Biomimetic Swimming Robot
Θ (rad)
Simulated data
0.2 Experimental data
Start point Y (m)
0.1
X (m)
-1.0 -0.5 0 0.5 1.0
0 t (s)
5 10 15 20
-0.1
-0.5
-0.2
-1.0
(b)
Fx (N)
1.0 -1.5
0.5
0 t (s)
5 10 15 20 -2.0
-0.5
-1.0
(b)
-1.5
Figure 5. Simulated turning gait. (a) Actuated joint angle signals
(c) from the CPG model. (b) Comparison of turning trajectories.
Fy (N)
3 (X(t s T) X(t s ))2 (Y(t s T) Y(t s ))2 / T (11)
2
1 where ts is an arbitrary time after the simulation is steady
0 t (s)
and T is the oscillating period indirectly obtained from
5 10 15 20
-1
the oscillating frequency. The velocity curve obtained in
Figure 4f indicates the motion stability after several
-2
periods, and the steady average velocity is 0.335 m/s.
-3
(d) Meanwhile, turning can be induced when asymmetrical
drives are applied between the left and right sides of the
MΘ (N·m)
1.0
CPG model. The robot will turn towards the side
receiving the higher drive. The posterior body keeps
0.5 oscillating at a common frequency in terms of
synchronization, which is in vain for modulating the
0 t (s) direction. The turning gait will mostly be credited to
5 10 15 20
asymmetrical oscillation amplitudes of all the joints in the
-0.5
posterior body, with the side receiving higher input drive
-1.0 oscillating at higher amplitudes, as shown in Figure 5a.
The turning motion simulation is displayed in Figure 5b
(e) at dL=1.5 and dR=3.5, and the average turning radius is
n (m/s) around 1.15 m. Notice that the pectoral fins will oscillate
0.32
in their own intrinsic frequencies due to unattained
critical input drive during swimming. The differences of
0.30
the oscillatory frequencies and amplitudes of the pectoral
0.28
fins on both sides will also help turning and assist with
0.26 propulsion. The successful simulations of forward gaits
0.24 and turning gaits show the effectiveness of
0.22 t (s)
hydrodynamic modelling in describing yaw motions.
5 10 15 20
(f) In addition, by modulating the attack angle of the
pectoral fins, the robot is able to achieve an up‐and‐down
Figure 4. Simulated forward gait. (a) Forward swimming
trajectory. (b) Oscillatory angle signal of Θ. (c) Hydrodynamic
motion, termed ʹpitching gaitʹ. Figure 6 shows the
forces acting on the X‐axis. (d) Hydrodynamic forces acting on comparative results between simulations and
the Y‐axis. (e) Moment acting on (X,Y). (f) Average velocity at experiments on the pitching gait, where the attack angle
every period. increases from 0 to 60º, with a step of 10º. The simulated
Figure 6. Comparison of simulated and actual pitching gait via 0.2
modulating attack angle of the pectoral fins.
0.15
0.5
0.1
0.45 Simulated fish−like propulsion
Simulated dolphin−like propulsion
Actual fish−like propulsion 0.05
0.4 Actual dolphin−like propulsion
0
0.35 1 1.5 2 2.5 3 3.5 4
0.3 dL=dR
v (m/s)
0.25 Figure 8. Propulsion comparison of different forward gaits.
0.2
0.15
Furthermore, a series of terrestrial and aquatic
locomotion gaits were successfully implemented (see
0.1
Figure 9), which can significantly expand the operational
0.05
field and capability of the amphibious robot. Currently,
0
1 1.5 2 2.5 3 3.5 4 4.5 5 the gait transition is primarily induced by modulating the
d (dimensionless) activity of the CPG, the attack angle of the mechanical
Figure 7. Comparison of simulated and actual propulsive speeds pectoral fins or else the swivelling mechanism. As a
between fish‐like and dolphin‐like modes. quantitative case, the average turning radius shown in
Figure 9d is approximately 0.95 m, which is minor in
4.2 Experiments
contrast with Figure 5b. This mainly owes to the larger
The experiments were performed in a lake and the drive bending of the rear body and the approach stream
signals were generated through wireless communication pressures produced by the rear body, which are ignored
by a human operator. By applying the same drive in the dynamic model. In particular, the experimental
between the left and right sides of the CPG model, the trajectory of the turning robot exhibits an irregular
AmphiRobot could realize forward swimming. As the circular path. This phenomenon might be caused by the
same drive is increasing and is applied to both sides, the drive difference that undermines the balance of the robot
swimming distances and corresponding consuming time body tilting to the side of the small drive.
www.intechopen.com Rui Ding, Junzhi Yu, Qinghai Yang and Min Tan: Dynamic Modelling 9
of a CPG-Controlled Amphibious Biomimetic Swimming Robot
Figure 9. Multimodal amphibious locomotion in different environments. (a) Crawling in terrestrial environments, including stony
ground, grassland and a bumpy road. (b) Gait transition between fish‐ and dolphin‐like swimming. (c) Multiple swimming gaits,
including backward swimming, turning on the spot and braking. (d) Turning underwater.
5. Conclusions [5] N. Kato, “Swimming and walking of an amphibious
robot with fin actuators”, Mar Technol Soc J, vol. 45,
This paper presents the modelling and control issues of pp. 181‐197, 2011.
an amphibious swimming robot capable of multimodal [6] Y. Sun and S. Ma, “ePaddle mechanism: towards the
motions. To generate robust gait control for steady development of a versatile amphibious locomotion
swimming, a CPG‐based control law was designed to mechanism”, in Proc IEEE/RSJ Int Conf Intell Robot
produce the joint angle signals of the propelling units. A Syst, San Francisco, USA, pp. 5035‐5040, 2011.
hydrodynamic model based on a Lagrangian formulation [7] T. Saleh, Y. H. Hann, Z. Zhen, et al., “Design of a
has been proposed for underwater locomotion. By gyroscopically stabilized single‐wheeled robot”, in
introducing the CPG‐based control into the Proc IEEE Conf Robot, Autom Mechatronics, Singapore,
hydrodynamic solution, multiple gaits could be pp. 904‐908, 2004.
numerically estimated. Comparative results have been [8] J. Ayers, “Underwater walking”, Arthropod Struct Dev,
given to show the effectiveness of the formed dynamic vol. 33, pp. 347‐360, 2004.
model combined with the CPG‐based control model. [9] A. S. Boxerbaum, P. Werk, R. D. Quinn, et al., “Design
of an autonomous amphibious robot for surf zone
Further research will focus on mechanical and gait operation: part I mechanical design for multi‐mode
optimization for improved amphibious performance. mobility”, in Proc IEEE/ASME Int Conf Adv Intell
Furthermore, sensory feedback information will be Mechatronics, USA, pp. 1460‐1464, 2005.
integrated into the CPGs to improve amphibious [10] H. Yamada, S. Chigisaki, M. Mori, et al.,
locomotor performance. “Development of amphibious snake‐like robot ACM‐
R5,” in Proc Int Sympo Robot, Tokyo, Japan, pp. 433‐
6. Acknowledgements
440, 2005.
This work was supported by the National Natural Science [11] S. Yu, S. Ma, B. Li, et al., “An amphibious snake‐like
Foundation of China (60775053, 61075102) and, in part, by robot with terrestrial and aquatic gaits”, in Proc IEEE
the Beijing Natural Science Foundation (4102063, 4122084). Int Conf Robot Autom, Shanghai, China, pp. 2960‐2961,
2011.
7. References [12] Q. Yang, J. Yu, M. Tan, et al., “Preliminary
development of a biomimetic amphibious robot
[1] G. Dudek, P. Giguere, C. Prahacs, et al., “AQUA: an capable of multi‐mode motion”, in Proc IEEE Int Conf
amphibious autonomous robot”, Computer, vol. 40, Robot Biomim, Sanya, China, pp. 769‐774, 2007.
pp. 46‐53, 2007. [13] M. J. Lighthill, “Note on the swimming of slender
[2] A. Crespi and A. J. Ijspeert, “Online optimization of fish”, J Fluid Mech, vol. 9, pp. 305‐317, 1960.
swimming and crawling in an amphibious snake [14] D. Barrett, M. Grosenbaugh, and M. Triantafyllou,
robot”, IEEE Trans Robot, vol. 24, pp. 75‐87, 2008. “The optimal control of a flexible hull robotic
[3] T. R. Consi, S. Bingham, J. Chepp, et al., “Amphibious undersea vehicle propelled by an oscillating foil,” in
robots as rapidly deployable near‐shore observatories”, Proc 1996 Symp Aut Underwater Vehicle Tech, Montery
in Proc OCEANS 2010, Seattle, USA, pp. 1‐6, 2010. CA, USA, pp. 1‐9, 1996.
[4] J. Yu, Y. Tang, X. Zhang, et al., “Design of a wheel‐ [15] A. J. Ijspeert, “Central pattern generators for
propeller‐leg integrated amphibious robot”, in Proc locomotion in animals and robots: a review”, Neural
Int Conf Contr, Autom, Robot Vision, Singapore, pp. Network, vol. 21, pp. 642‐653, 2008.
1815‐1819, 2010.
www.intechopen.com Rui Ding, Junzhi Yu, Qinghai Yang and Min Tan: Dynamic Modelling 11
of a CPG-Controlled Amphibious Biomimetic Swimming Robot