Professional Documents
Culture Documents
Liu2019 Article AdaptiveNeuralNetworkTrackingC PDF
Liu2019 Article AdaptiveNeuralNetworkTrackingC PDF
https://doi.org/10.1007/s11071-019-05170-8
ORIGINAL PAPER
Received: 27 April 2018 / Accepted: 27 July 2019 / Published online: 8 October 2019
© The Author(s) 2019
Abstract This paper studies neural network-based non-collocated subset is approximated through neural
tracking control of underactuated systems with networks within a local region. Finally, simulation stud-
unknown parameters and with matched and mis- ies on an underactuated manipulator and an underac-
matched disturbances. Novel adaptive control schemes tuated vibro-driven system are conducted to verify the
are proposed with the utilization of multi-layer neu- effectiveness of the proposed control schemes.
ral networks, adaptive control and variable structure
strategies to cope with the uncertainties containing Keywords Adaptive tracking control ·
approximation errors, unknown base parameters and RBF neural networks ·
time-varying matched and mismatched external distur- Underactuated systems · Auxiliary control variables ·
bances. Novel auxiliary control variables are designed Matched and mismatched disturbances
to establish the controllability of the non-collocated
subset of the underactuated systems. The approxima-
tion errors and the matched and mismatched external 1 Introduction
disturbances are efficiently counteracted by appropri-
ate design of robust compensators. Stability and con- Underactuated mechanical systems (UMSs) are rapidly
vergence of the time-varying reference trajectory are growing research fields that combine control and
shown in the sense of Lyapunov. The parameter updat- robotics societies [1–8]. They have extensive appli-
ing laws for the designed control schemes are derived cations such as UAVs, underground vehicles, space-
using the projection approach to reduce the tracking craft, humanoid robots and vibro-driven robots [9–12].
error as small as desired. Unknown dynamics of the UMSs have more degrees-of-freedom (DOF) n than
independent control inputs m; thus, (n-m) DOF are not
P. Liu
Cardiff School of Technologies, Cardiff Metropolitan directly controllable, which characterize the nature of
University, Cardiff CF5 2YB, UK underactuation. This nature is made possible for the
UMSs to undertake complicated tasks with a reduced
H. Yu (B) number of actuators that in turn implies the reduc-
School of Engineering and the Built Environment,
Edinburgh Napier University, 10 Colinton Road, tion in weight and energy consumption. Challenge that
Edinburgh EH10 5DT, UK faced by control of underactuated systems is the exis-
e-mail: h.yu@napier.ac.uk tence of underactuation and some undesirable prop-
erties such as being in a non-minimum phase and/or
S. Cang
School of Economics and Management, Yanshan University, possessing an undetermined relative degree, which
Qinhuangdao 066004, China makes conventional approaches not directly applica-
123
1448 P. Liu et al.
ble, particularly for the issue of trajectory tracking applied at the actuated subsystem and the resulting
control. Design of the control schemes for UMSs accelerations of the passive subsystem, with the intent
is intractable because of the internal dynamics and of achieving control globally. As such, in this paper,
non-holonomic property, and they are not feedback nonlinear control approach is investigated by employ-
linearizable [13]. Besides, the uncertainties in sys- ing multi-layer NNs. NNs have versatile features such
tem model, as well as the matched and mismatched as learning capability mapping and parallel process-
disturbances, make control of UMSs more challeng- ing. An attractive feature of NNs is that their synaptic
ing. weights are online updated without any offline learn-
The complexity of control problem related to UMSs ing phases. NNs have the property of robustness; thus,
can be reduced when the objective is merely to stabi- they have been widely applied in various robotic sys-
lize a subset of the UMSs’ DOF. In the literature, a tems to address the stabilization problem [26–28]. The
great number of existing control system designs for issue of tracking control of UMSs based on NNs has
UMSs explore the concept of linearization through attracted extensive attentions. Optimal motion control
partial feedback [14–18]. Linear systems can be uti- using NNs and stochastic adaptive concepts was stud-
lized to capture the underactuated dynamics within ied towards the Pendubot in [29] and a WIP system in
a local range; however, global stabilization of the [30]. For UMSs with full-state constraints containing
underactuated dynamics is still unavailable under this a Moore–Penrose inverse term, an adaptive NN con-
approach. Other prevailing techniques such as inverse trol system was proposed in [31]. In [31], the authors
dynamics [19,20], sliding mode/variable structure [14, developed two decentralized output feedback control
21,22], energy/passivity-based approaches [17,23,24] systems based on adaptive NN to tackle with immeasur-
have been extensively exploited. Furthermore, practical able states and unknown time delays in UMSs. Towards
requirements are raised from the current applications, a wheeled mobile robot that is non-holonomic with
in which the adaptability of UMSs is extremely cru- unknown parameters and uncertain dynamics, an adap-
cial when facing environments with uncertainties. For tive tracking control scheme was presented in [32] to
instance, microrobotic systems work across vulnerable tune the kinematic controller gain online and minimize
media in restricted space for minimally invasive sens- the tracking error in velocity. A bio-inspired tracking
ing and risk intervention in pipeline inspection, endo- control scheme based on NNs was developed in [33]
scopic assistance, underwater exploration, etc. How- for an underactuated surface vessel with unknown sys-
ever, an exact dynamic model is intractable to obtain tem dynamics. A cart–pendulum system with unknown
due to the presence of frictions, unknown disturbances, dynamics was studied in [34], and a trajectory track-
time-varying parameters, etc. As a result, adaptive con- ing control scheme of the pendulum subsystem based
trol schemes for generic UMSs have attracted great on adaptive NN was designed instead of considering
attentions. Considering uncertainties and ocean distur- the position of the cart. In [35], an output feedback
bances, a control system using leader–follower forma- control system based on NNs was proposed for track-
tion was studied in [25] for underactuated autonomous ing control of a spherical inverted pendulum. A com-
surface vehicles. Dynamic surface control technique bined PID and neural network compensation approach
and neural networks (NNs) were used to construct was proposed in [36] to control a wheel-driven mobile
the control scheme. A hierarchical sliding mode con- pendulum system, and the results were experimen-
trol system with adaptive and fuzzy inclusions was tally analysed. From the literature, it is noted that rela-
studied for uncertain UMSs in [14], where differ- tively few studies have addressed the issue of tracking
ent layers of sliding surface are constructed to cope control for UMSs, particularly when the disturbance
with the uncertainties and disturbances and fuzzy exists in the non-collocated subsystem, which is mis-
models are designed to approximate the nonlineari- matched with the control actions. Also, it is noted that
ties. very few reported studies towards this topic have pre-
It is evident that description of dynamic couplings sented rigorous analysis of trajectories of the closed-
between the actuated and passive subsystems of UMSs loop system for UMSs. Therefore, trajectory tracking
is typically highly nonlinear. Therefore, it is plau- control for UMSs with uncertainties and disturbances
sible to consider the employment of approximation is still an open problem and requires in-depth investi-
approaches to map the coupling between the torques gations.
123
Adaptive neural network tracking control 1449
Through the utilization of the unique physical prop- iary control variables that contain NN approximator
erties of the UMSs, the overall underactuated system and robust compensator.
breaks down into two subsystems, i.e. a fully actu- 2. The parametric uncertainties and the matched and
ated subsystem and an unactuated (passive) subsys- mismatched disturbances are considered in the
tem. Radial basis neural network (RBFNN) has sim- design of the adaptive control schemes, which fea-
ple structure and fast convergence rate, and it can ture a generic model for the studies on underac-
overcome the local minimum problem; therefore, it tuated systems. It is noted that the mismatched
is utilized as a nonlinear function approximator of disturbances have been neglected in most of the
uncertain dynamics of the unactuated (passive) sub- existing approaches for the tracking control of
system of the UMSs. The NN control has the abil- UMSs.
ity of universal approximation, and it has been thor- 3. Employing the adaptive control approach, com-
oughly studied on discrete-time system [37–40] and bined with variable structure and NNs, exact values
continuous-time systems [41–44]. There are very lim- of the system base parameters are not required to
ited studies using NN to approximate the system be known a priori.
dynamics of UMSs. In this paper, we develop NN- 4. Designing robust compensators to counteract the
based adaptive tracking control schemes to cope with matched and mismatched disturbances, and func-
the internal uncertain dynamics and external distur- tion approximation error of NNs and nonlinear
bances, and auxiliary control variables are explic- frictions can reduce the tracking error as small as
itly designed to close the unactuated feedback loops. desired in finite time through selecting appropriate
RBFNN is adopted to approximate the mismatched parameters for the controller.
system uncertainties, and the adaptive control algo- The rest of this paper is organized as follows.
rithm is constructed to estimate the NNs approximation Notations, assumptions, system dynamic model for
error and the bounded mismatched disturbance. The UMSs and preliminaries are presented in Sect. 2.
combination of NN approximation, variable structure Section 3 gives the main theoretical results con-
control and adaptive approach makes the constructed cerning the adaptive NN tracking control systems
new controller more robust, and as such, errors result- design for a UMSs. Validations of the effective-
ing from trajectory tracking, parameter uncertainties, ness of the proposed approaches are presented in
mismatched external disturbances and NN approxi- Sect. 4 through simulation studies on an underactu-
mation are counteracted. Theoretical background of ated manipulator and a vibro-driven mobile system.
these methods is presented with rigorous analysis and Finally, concluding remarks and perspectives are given
developed in detail for some examples. The schemes in Sect. 5.
promote the utilization of linear filters in the control
input such that the system robustness is improved.
Stability of the system dynamics and convergence 2 Preliminaries and problem description
of the time-varying reference trajectories are demon-
strated using Lyapunov analysis. In addition, adapta- 2.1 Notations
tion laws for the NNs weights of the proposed con-
trol systems are derived from the above procedure. The Let . denote any suitable vector Euclidean norm.
main contributions of this paper are summarized as fol- Specifically, . p represents the p-norm of a given
lows: vector. The Frobenius norm of the given matrix H =
[h i j ] ∈ Rn×m
is defined as H 2F = T r (H T H ) =
T r (H H ) = i, j h i2j with T r (.) denoting the trace
T
1. Stabilization for fully actuated systems is well operator. The Frobenius norm is associated with the
established in terms of the time-varying trajectories 2-norm in a manner that H x2 ≤ H F x2 with
through adaptive control. However, its application H ∈ Rn×m and x ∈ Rm . The trace operator has the
and extension to the UMSs are not straightforward. property of A T B = T r (AB T ) with ∀A, B ∈ Rn .
This paper proposes the adaptive control schemes to λmin (.) and λmax (.) are, respectively, the minimum and
encompass the conventional approaches and stabi- maximum eigenvalue of the given matrix. In represents
lize the UMSs’ state space through design of auxil- the identity matrix of dimension n × n.
123
1450 P. Liu et al.
2.2 Dynamic model and properties Property 3 The above matrixes D (q, α) and
C (q, q̇, α) have the following skew-symmetric
The dynamics of n-DOF UMSs can be expressed in interconnection
the generalized coordinates via the Euler–Lagrangian’s
approach, given by x T Ḋ (q, α) − 2C (q, q̇, α) x = 0, ∀x ∈ Rn−m (4)
123
Adaptive neural network tracking control 1451
χ2(z)
z2 Σ W N ), ε (z) ∈ R is a bounded approximation error
over the compact set, ε N is an upper bound (posi-
tive constant) of the approximation error
which sat-
φk-1(•) isfies ε N = sup χ̂ (z, W ∗ ) − χ (z) and φ (z) =
χn(z) [φ1 (z) , φ2 (z) , ... , φk (z)]T is the NN basis function
zn Σ
which is conventionally chosen as Gaussian functions
φk(•) as
Input Layer Neuron Layer Output Layer z − Ci 2
φi (z) = exp − , i = 1, 2, ... , k (10)
Fig. 1 Structure of the RBFNN 2bi2
123
1452 P. Liu et al.
where W ∗ is typically defined as the optimal value of W bounds of the desired reference trajectories. Noting that
such that the approximation error ε (z) could be mini- the design of ϑ1 and ϑ2 has to satisfy the zero dynamics
mized for all z ∈ z as based on the non-holonomic dynamics, we have
W ∗ := arg min W ∈Rk sup z∈z χ (z) − W ∗T φ (z) Dnc (q, α) ϑ̈1 + Dnn (q, α) ϑ̈2 + Cnc (q, q̇, α) ϑ̇1
(14) + Cnn (q, q̇, α) ϑ̇2 + G n (q, α)
+ Fvn (α) ϑ̇2 + Fcn (q, q̇, α) + τdn = 0 (18)
where qc ∈ Rm and qn ∈ Rn−m denote the actuated and where c , δc ∈ Rm and n , δn ∈ Rn−m . δ denotes the
unactuated coordinate vector, respectively. The sub- filtered error signal and describes the measure of track-
scripts “c” and “n,” respectively, indicate collocated ing accuracy, is referred to as vector of the reference
and non-collocated subsets. trajectory, = diag[c Im×m , n I(n−m)×(n−m) ] with
Without loss of generality, the underactuated system c and n be positive constants selected by design-
(1) can be rewritten into a partitioned form as ers. Ii×i denotes i × i identity matrix. It is noted that
the error dynamics of the underactuated systems can be
⎧
⎪
⎪ Dcc (q, α) q̈c + Dcn (q, α) q̈n + Ccc (q, q̇, α) q̇c obtained by firstly introducing the tracking error from
⎪
⎪ + Ccn (q, q̇, α) q̇n + G c (q, α) + Fvc (α) q̇c
⎪
⎪
collocated and non-collocated loops and then filtering
⎨
+ Fcc (q, q̇, α) + τdc = τ out the error signals. In this regard, we can encompass
⎪
⎪ D nc (q, α) q̈c + Dnn (q, α) q̈n + C nc (q, q̇, α) q̇c the conventional adaptive control approaches and sta-
⎪
⎪
⎪
⎪ + Cnn (q, q̇, α) q̇n + G n (q, α) bilize the state space of underactuated systems. The
⎩
+ Fvn (α) q̇n + Fcn (q, q̇, α) + τdn = 0 choice of c > 0 and n >0 guarantees that (20) is an
(16) exponentially stable system for q. Therefore, the tra-
jectory q converges to an adjacent of qd exponentially
where τdc and τdn denote the bounded unknown dis- fast as long as the control system drives δ to an adjacent
turbances and unmodelled dynamics to the collocated of zero.
and non-collocated subsets, respectively. Applying the defined variables in the system dynam-
Let the reference trajectories for the collocated and ics (17), we have
non-collocated subsets be descried by the vector-valued
functions qcd ∞ ≤ ϑ1 and qnd ∞ ≤ ϑ2 , respec- ⎧
⎪ Dcc (q, α) δ̇c + ˙ c + Dcn (q, α) δ̇n + ˙ n
tively, and assume that these functions are bounded in ⎪
⎪
⎪
⎪
norm and uniformly continuous on R+ , and homoge- ⎪
⎪ + Ccc (q, q̇, α) (δc + c ) + Ccn (q, q̇, α) (δn
⎪
⎪
nously on the same set, its first- and second-order ⎪
⎪
⎪ + n ) + G c (q, α) + Fvc (α) q̇c + Fcc (q, q̇, α)
⎪
derivatives are bounded, well defined and uniformly ⎨
+ τdc = τ
continuous. Introduce the trajectory tracking error ⎪
⎪
⎪
⎪ Dnc (q, α) δ̇c + ˙ c + Dnn (q, α) δ̇n + ˙ n
as ⎪
⎪
⎪
⎪
⎪
⎪ + Cnc (q, q̇, α) (δc +c )+Cnn (q, q̇, α) (δn +n )
q̃c = qc − qcd , q̃n = qn − qnd (17) ⎪
⎪
⎪
⎩
which is to be stabilized to zero without the knowledge + G n (q, α)+ Fvn (α) q̇n + Fcn (q, q̇, α)+τdn = 0
of the system parameters α. ϑ1 and ϑ2 are positive upper (21)
123
Adaptive neural network tracking control 1453
123
1454 P. Liu et al.
123
Adaptive neural network tracking control 1455
δn 2 δn 2 δ̇ ∈ L n∞ , q̈ = [q̈c q̈n ]T ∈ L n∞ and q̃ = [q̃c q̃n ]T ∈ L 2n
∞.
+ (κ − εT ) κ̇ − − εT
δn + μ δn + μ Therefore, q̃c and q̃n are uniformly continuous and
≤ − δcT K 1 δc − δc δn |η| − K 2 δc δn q̃ = [q̃c q̃n ]T ∈ L 2n
∞ , and it is evident that q̃ → 0
as t → ∞.
− δcT ξ − K 3 δn 2 − δcT τdc
δn 2 Remark 3 The NNs are adopted to approximate the
+ δn εT − εT mismatched system uncertainties, and the adaptive con-
δn + μ
trol algorithm is constructed to estimate the NN approx-
= − δcT K 1 δc − δc δn |η| −K 2 δc δn imation error and the bounded mismatched distur-
δn μεT bance. The combination of variable structure control,
− δcT ξ − K 3 δn 2 − δcT τdc +
δn + μ NN approximation and adaptive approach makes the
≤ −δcT K 1 δc − δc δn |η| − K 2 δc δn constructed new controller more robust, and such errors
resulting from trajectory tracking, parameter uncertain-
− δcT ξ − K 3 δn 2 − δcT τdc + μεT
ties, mismatched external disturbances and NN approx-
≤ K 1 δc 2 − K 3 δn 2 − δcT ξ − δcT τdc + μεT imation are compensated.
= − δT K δ − δcT ξ − δcT τdc + μεT (31)
For the case τdc = 0 and τdc < βm , i.e. the sys-
tem is subject to both matched and mismatched distur-
K1 0 bances, one can only conclude that δ is bounded from
where K = .
0 K3 (26) and (31), but α̃c and W̃ may become unbounded
When no disturbance exerts on the collocated sub- as (31) merely contains a negative definite component
system (τdc = 0), i.e. the system is only subject to mis- of δ2 and no negative terms of α̃c and W̃ are appar-
matched disturbances, we design the collocated robust ently included. As a result, the system may tend to be
compensator as ξ = 0 and integrate both sides of (31) unstable. To improve the robustness of Theorem 1, the
from t = 0 to t = T as following adaptation algorithm is therefore proposed.
Adaptation Algorithm 2. Consider the following
T T adaptation law
V (T ) − V (0) ≤ − δ K δ dt + εT
T
μ dt
0 0
(32) α̂˙ c = − α̃c −Yc δc (34)
∞
Considering that V (T ) ≥ 0 and 0 μdt < ∞, we Corollary 2 Consider the error equation (22) with the
have sliding surface designed in (20) under the adaptive
NNs-based robust control law in (25), the following
T T
1 1 corollary holds: If adaptation algorithm 2 is adopted,
lim T →∞ sup δ2 dt ≤ V (0) + εT μ dt
T 0 K 0 the system error signals q̃, q̃˙ and α̃ converge to zero
1 asymptotically. If τdc = 0 and τdc < βm , then the
lim T →∞ (33)
T system becomes globally uniformly ultimately stable
From the definition of the Lyapunov function V in (26) and the boundedness depends on τdc .
and V̇ derived from (31–33), the global uniform bound- Proof Adopting Adaptation Algorithm 2 in function
edness of the filtered tracking error δc for collocated (27), we have
subsystem and δn for non-collocated subsystem, the
parameter estimation error W̃ is guaranteed. From the
V̇ = δcT δnT
definition and assumption 1 of filtered tracking error
δ, it is evident that δ is bounded. The boundedness of −K 1 δc − Yc α̃c − sgn (δc ) δn |η| − K 2 sgn (δc ) δn − ξ
control input is obvious from (25). It can be concluded W φ+ε
T
that since δ = [δc δn ]T ∈ L n2 ∩ L n∞ , δc and δn are − δ T τd + α̂˙ cT −1 α̃c + tr W̃ T ϒ −1 W̃˙
continuous and δc → 0, δn → 0 as t → ∞, and −1
η ∈ L ∞ . From (25c), it can be shown that α̃c ∈ L ∞ .
p + (κ − εT )κ̇ + η 2n+1 η̇
This in turn implies, based on property 1 and (25c), that = −δcT K 1 δc − δc δn |η| − δcT ξ + δnT (ε − τdn )
123
1456 P. Liu et al.
δn 2 τ = τc + τn (40a)
− K 2 δc δn − δcT τdc − κ
δn + μ τc = Yc α̂c − K 1 δc − ξ, τn = −sgn (δc ) δn |η|
− K 3 δn 2 + (κ − εT )κ̇ + tr W̃ T ϒ −1 (W̃˙ + ϒδnT φ) − K 2 sgn (δc ) δn (40b)
− α̃cT −1 α̃c
with the Adaptation Algorithm 2 designed in (34), and
≤ −δcT K 1 δc − δc δn |η| − δcT ξ − δcT τdc
the collocated robust compensator ξ designed using
− K 2 δc δn − K 3 δn 2 + μεT − α̃cT −1 α̃c (35) hyperbolic tangent function as
−1 α̃ 2
V̇ ≤ −K 1 δc − K 3 δn −
2 2 c
− δcT ξ − δcT τdc + μεT nηr βm δc
T
ξ = βm tanh (40c)
= − δ K δ − δcT ξ − δcT τdc + μεT (36) ρ
Case 2. For the case when τdc = 0 and τdc < βm , the 1 T 1
V = δ Dδ + α̃cT −1 α̃c
collocated robust compensator ξ is designed to satisfy 2 2
1 2n + 1 2n
the following conditions + tr W̃ T ϒ −1 W̃ + η 2n+1 (41)
2 2n
δcT ξ ≥ 0 The derivative of Lyapunov candidate function is
(39)
βm δc − δ cT ξ ≤ ρ yielded as
123
Adaptive neural network tracking control 1457
V̇ = δcT δnT
−K 1 δc − Yc α̃c − sgn (δc ) δn |η| − K 2 sgn (δc ) δn − ξ λmin (K 3 ) δn + β W̃ W̃ − Wmax − εT
W φ+ε
T F F
−δ T τd + α̂˙ cT −1 α̃c + tr W̃ T ϒ −1 W̃˙ + η 2n+1 η̇
−1
Wmax 2 W2
= β W̃ − − β max
F 2 4
= − δc δn |η| − δcT K 1 δc − K 2 δc δn
+λmin (K 3 ) δn − εT (45)
−δcT ξ − δcT τdc − K 3 δn 2
To guarantee V̇1 ≤ 0, the following inequality needs
+tr W̃ T ϒ −1 W̃˙ + ϒδnT φ
to be satisfied
+δnT (ε − τdn ) − α̃cT −1 α̃c βWmax2 + 4ε
T
δn > or W̃
≤ − δc δn |η| − K 2 δc δn − δcT K 1 δc − δcT ξ 4λmin (K 3 ) F
+ − K 3 δn 2 Wmax Wmax 2 εT
> + + (46)
+tr W̃ T ϒ −1 (−ϒφδnT + βϒ δn Ŵ + ϒδnT φ) 2 4 β
δn εT − δcT τdc − α̃cT −1 α̃c Therefore, V̇1 is negative outside a compact set.
Based on the standard Lyapunov
theorem extension,
V̇ ≤ −δcT K 1 δc − −1 α̃c 2 − K 3 δn 2 − δcT ξ
the UUB of both δn and W̃ is demonstrated.
−δcT τdc + β δn tr W̃ T W − W̃ + δnT εT (42) F
Through (43b), the time derivative of V2 can be given
by
Let us decompose (42) into the following two parts V̇2 ≤ − −1 α̃c 2 − δcT K 1 δc + βm δc − δcT ξ (47)
123
1458 P. Liu et al.
4 Simulations
123
Adaptive neural network tracking control 1459
3
1.2
2 0.8
0.6
1.5
0.4
1
0.2
0.5
0
0 -0.2
0 20 40 60 80 100 120 0 20 40 60 80 100 120
Time(s) Time (s)
350
q2 can be achieved through convenient computation,
300
and it should satisfy the following constraint equation
Control Torque (Nm)
250
D21 q̈1 + D22 q̈2 + C21 q̇1 + C22 q̇2 200
+ G 2 + Fv1 q̇ + Fc1 + τd1 = 0 (51)
150
123
1460 P. Liu et al.
200
χ
180 χ^
160
140
120
χ and χ^
100
80
60
40
40 T
represents the gravitational torques, F = [cθ̇ f ] is
20
a1 sin(t)
0 the friction forces, τd = [τdc τdn ]T =
a2 sin(t)
-20 denotes the external matched and mismatched distur-
-40 bances, B = [1 0]T is the input force matrix and τ ∈ R1
-60 denotes the control input applied to the system.
-80
In the simulation, the rationality of the parameter
values selection in this section is specified as follows:
-100
0 20 40 60 80 100 120 the system parameter values are configured from the
Time (s) studies in literature as reported in [48,49] as M =
Fig. 7 NN approximation error
0.5 kg, m = 0.138 kg, l = 0.3 m, g = 9.81 m/s 2 ,
μ = 0.01 N /ms. Initial conditions of the system are
set as θ (0) = θ 0 = π /3, θ̇ (0) = 0, x (0) = 0 and
and the actuated dynamics of the first link are cou- ẋ (0) = 0. The simulation is conducted in 6.6s which
pled; thus, the unmodelled dynamics of the second link is one full motion cycle. The parameter values for
may contribute additional time-varying inertia and non- the matched and mismatched external disturbances are
linearity to the manipulator dynamics. In this subsec- chosen as a1 = a2 = 0.2. The bandwidth of the first-
tion, the context of an underactuated mobile robotic order filter is set as = [1 2 ]T = [15 20]T . In the
model is considered as shown in Fig. 8. This under- simulation, the controller parameters are chosen to be
actuated vibro-driven robotic system was proposed in K 1 = 10I , K 2 = 20I and K 3 = 50I . The adaptation
[3] for which the actuated and unactuated dynamics are gains are chosen as = 10I and = 6I . Parame-
strongly coupled. ter values for the collocated robust compensator are set
In the presence of matched and mismatched external as βm = 20, ρ = 0.5. In addition, the weight tuning
disturbances, the underactuated dynamics of the vibro- parameter of the proposed control system is selected as
driven system are given as ϒ = 0.005 and β = 0.1. The rationality of these selec-
tions is configured using iterative simulations.
The trajectory tracking performance of the actuated
D (q) q̈ +C (q, q̇) q̇ + K (q) q + G (q)+ F +τd = Bτ subsystem is presented in Fig. 9. It is observed from
(52) the figure that although the response of the proposed
123
Adaptive neural network tracking control 1461
9 9
Tracking performance
8 Reference trajectory
Tracking performance (rad/s)
8
7 9
6
5
5
4 4
6
3
3
2
5
2 1
0
1
-1 4
0 0.1 0.2 0.3 0.4 0.5 0.6
0 Time (s)
-1 3
0 1 2 3 4 5 6 0 1 2 3 4 5 6
0.25 20
0.2
15
0.15
Tracking error (rad/s)
10
Control torque (Nm)
0.1
0.05 5
0
0
-0.05
-5
-0.1
-0.15 -10
-0.2 -15
-0.25
0 1 2 3 4 5 6 -20
0 1 2 3 4 5 6
Time (s)
Time (s)
Fig. 10 Trajectory tracking error
Fig. 12 Control torque
control scheme is slightly slower, the controlled pen- control scheme is capable of guaranteeing accurate tra-
dulum trajectory tracks the reference trajectory accu- jectory tracking of the actuated subsystem, and mean-
rately. The reference trajectory for the actuated subsys- while, the passive subsystem can maintain a forward
tem is chosen as shown in Fig. 9. The proposed control locomotion at some desired velocity. Therefore, it is
system has a learning process that makes the estimated concluded that the designed control system is efficient
parameters adapt to appropriate values. in the presence of unknown nonlinear dynamic systems
The tracking error is shown in Fig. 10, from which and environmental disturbances.
the tracking error converges to an adjacent and bounded
compact set near zero in finite time. The trajectory of
the vibro-driven system is presented in Fig. 11 showing 5 Conclusions
that the cart travels at the speed about 7cm within 6.6s.
The control torque is shown in Fig. 12 that demonstrates In this paper, novel NNs-based adaptive tracking con-
the boundedness of the torque input. As demonstrated trol schemes for underactuated systems with matched
in the system performance, the developed NN adaptive and mismatched disturbances have been presented. The
123
1462 P. Liu et al.
parametric uncertainties and matched and mismatched 2. Seifried, R.: Integrated mechanical and control design
external disturbances have been considered in the con- of underactuated multibody systems. Nonlinear
Dyn. 67, 1539–1557 (2012). https://doi.org/10.1007/
troller design, which feature a generic model in the s11071-011-0087-2
research of underactuated systems. The mismatched 3. Liu, P., Yu, H., Cang, S.: Geometric analysis-based trajec-
disturbances have been omitted in most of the existing tory planning and control for underactuated capsule sys-
approaches for the tracking control of UMSs. Auxil- tems with viscoelastic property. Trans. Inst. Meas. Control.
(2017). https://doi.org/10.1177/0142331217708833
iary control variables have been designed to establish 4. Zhang, X., Fang, Y., Sun, N.: Minimum-time trajectory plan-
the controllability of non-collocated subset of underac- ning for underactuated overhead crane systems with state
tuated systems by using a universal approximation of and control constraints. IEEE Trans. Ind. Electron. 61, 6915–
RBFNN, and approximation errors and external distur- 6925 (2014). https://doi.org/10.1109/TIE.2014.2320231
5. Liu, P., Yu, H., Cang, S.: Modelling and dynamic analy-
bances can be efficiently counteracted through design sis of underactuated capsule systems with friction-induced
of robust compensators. Employing the adaptive con- hysteresis. In: 2016 IEEE/RSJ international conference on
trol approach, combined with variable structure and intelligent robots and systems (IROS), pp. 549–554. IEEE
NNs, the exact values of the parameters of the under- (2016)
6. Fang, Y., Ma, B., Wang, P., Zhang, X.: A motion planning-
actuated systems are not required to be known a pri-
based adaptive control method for an underactuated crane
ori. The stability of the overall system has been proved system. Control Syst. Technol. IEEE Trans. 20, 241–248
by Lyapunov analysis, and it is shown that the track- (2012)
ing error can be reduced as small as desired in finite 7. Nguyen, K.-D., Dankowicz, H.: Adaptive control of under-
actuated robots with unmodeled dynamics. Robot. Auton.
time by choosing appropriate controller parameters. Syst. 64, 84–99 (2015). https://doi.org/10.1016/j.robot.
The simulation studies on an underactuated manipu- 2014.10.009
lator and an underactuated vibro-driven system have 8. Liu, P., Yu, H., Cang, S.: On periodically pendulum-driven
shown the effectiveness of the proposed adaptive con- systems for underactuated locomotion: a viscoelastic jointed
model. In: 2015 21st International Conference on Automa-
trol systems. tion and Computing (ICAC). pp. 1–6 (2015)
9. Liu, P., Yu, H., Cang, S.: Modelling and analysis of dynamic
Acknowledgements This work was partially supported by frictional interactions of vibro-driven capsule systems with
the National Natural Science Foundation of China (61803396), viscoelastic property. Eur. J. Mech.-A Solids. 74, 16–25
European Commission Marie Skłodowska-Curie SMOOTH (2019)
(smart robots for fire-fighting) project (H2020-MSCA-RISE- 10. Liu, P., Huda, M.N., Tang, Z., Sun, L.: A self-propelled
2016-734875) and Royal Society International Exchanges robotic system with a visco-elastic joint: dynamics and
Scheme (Adaptive Learning Control of a Cardiovascular Robot motion analysis. Eng. Comput. (2019). https://doi.org/10.
using Expert Surgeon Techniques) project (IE151224). FP7 Peo- 1007/s00366-019-00722-3
ple: Marie-Curie Actions (Grant Number PIRSES-GA-2012- 11. Liu, P., Yu, H., Cang, S.: Trajectory synthesis and optimiza-
318902). tion of an underactuated microrobotic system with dynamic
constraints and couplings. Int. J. Control Autom. Syst. 16,
Compliance with ethical standards 2373–2383 (2018)
12. Liu, P., Yu, H., Cang, S.: Optimized adaptive tracking control
Conflicts of interest The authors declare that they have no con- for an underactuated vibro-driven capsule system. Nonlinear
flict of interest. Dyn. 94, 1803–1817 (2018)
13. Brockett, R.W.: others: Asymptotic stability and feedback
Open Access This article is distributed under the terms of the stabilization. Differ. Geom. Control Theory. 27, 181–191
Creative Commons Attribution 4.0 International License (http:// (1983)
creativecommons.org/licenses/by/4.0/), which permits unre- 14. Hwang, C.-L., Chiang, C.-C., Yeh, Y.-W.: Adaptive fuzzy
stricted use, distribution, and reproduction in any medium, pro- hierarchical sliding-mode control for the trajectory track-
vided you give appropriate credit to the original author(s) and ing of uncertain underactuated nonlinear dynamic systems.
the source, provide a link to the Creative Commons license, and IEEE Trans. Fuzzy Syst. 22, 286–299 (2014)
indicate if changes were made. 15. Liu, P., Neumann, G., Fu, Q., Pearson, S., Yu, H.: Energy-
efficient design and control of a vibro-driven robot. In: 2018
IEEE/RSJ International Conference on Intelligent Robots
and Systems (IROS), pp. 1464–1469. IEEE (2018)
References 16. Liu, P., Yu, H., Cang, S.: Modelling and control of an elasti-
cally joint-actuated cart-pole underactuated system. In: 2014
1. Azimi, M.M., Koofigar, H.R.: Adaptive fuzzy backstepping 20th International Conference on Automation and Comput-
controller design for uncertain underactuated robotic sys- ing (ICAC) , pp. 26–31. IEEE (2014)
tems. Nonlinear Dyn. 79, 1457–1468 (2015). https://doi.org/
10.1007/s11071-014-1753-y
123
Adaptive neural network tracking control 1463
17. Valentinis, F., Donaire, A., Perez, T.: Energy-based motion 32. Mohareri, O., Dhaouadi, R., Rad, A.B.: Indirect adaptive
control of a slender hull unmanned underwater vehicle. tracking control of a nonholonomic mobile robot via neural
Ocean Eng. 104, 604–616 (2015) networks. Neurocomputing. 88, 54–66 (2012). https://doi.
18. Liu, P.: Bio-inspired robotic control in underactuation: prin- org/10.1016/j.neucom.2011.06.035
ciples for energy efficacy, dynamic compliance interactions 33. A biologically inspired approach to tracking control
and adaptability (2018). https://ethos.bl.uk/OrderDetails. of underactuated surface vessels subject to unknown
do?uin=uk.bl.ethos.732064 dynamics—ScienceDirect. http://www.sciencedirect.com/
19. Mistry, M., Buchli, J., Schaal, S.: Inverse dynamics control science/article/pii/S0957417414005958
of floating base systems using orthogonal decomposition. 34. Hsu, C.-F.: Adaptive backstepping Elman-based neural con-
In: 2010 IEEE International Conference on Robotics and trol for unknown nonlinear systems. Neurocomputing. 136,
Automation (ICRA), pp. 3406–3412. IEEE (2010) 170–179 (2014)
20. Blajer, W., Dziewiecki, K., Kołodziejczyk, K., Mazur, Z.: 35. Ping, Z.: Tracking problems of a spherical inverted pendu-
Inverse dynamics of underactuated mechanical systems: a lum via neural network enhanced design. Neurocomputing.
simple case study and experimental verification. Commun. 106, 137–147 (2013)
Nonlinear Sci. Numer. Simul. 16, 2265–2272 (2011) 36. Jung, S., Kim, S.S.: Control experiment of a wheel-driven
21. Yue, M., An, C., Du, Y., Sun, J.: Indirect adaptive fuzzy mobile inverted pendulum using neural network. IEEE
control for a nonholonomic/underactuated wheeled inverted Trans. Control Syst. Technol. 16, 297–303 (2008). https://
pendulum vehicle based on a data-driven trajectory planner. doi.org/10.1109/TCST.2007.903396
Fuzzy Sets Syst. 290, 158–177 (2016). https://doi.org/10. 37. Liu, D., Wang, D., Zhao, D., Wei, Q., Jin, N.: Neural-
1016/j.fss.2015.08.013 network-based optimal control for a class of unknown
22. Xu, J.-X., Guo, Z.-Q., Lee, T.H.: Design and implementation discrete-time nonlinear systems using globalized dual
of integral sliding-mode control on an underactuated two- heuristic programming. IEEE Trans. Autom. Sci. Eng. 9,
wheeled mobile robot. IEEE Trans. Ind. Electron. 61, 3671– 628–634 (2012)
3681 (2014) 38. Liu, Y.-J., Chen, C.P., Wen, G.-X., Tong, S.: Adaptive neural
23. Xin, X., Tanaka, S., She, J., Yamasaki, T.: New analytical output feedback tracking control for a class of uncertain
results of energy-based swing-up control for the Pendubot. discrete-time nonlinear systems. IEEE Trans. Neural Netw.
Int. J. Non-Linear Mech. 52, 110–118 (2013). https://doi. 22, 1162–1167 (2011)
org/10.1016/j.ijnonlinmec.2013.02.003 39. Xu, B., Sun, F., Yang, C., Gao, D., Ren, J.: Adaptive discrete-
24. Cornejo, C., Alvarez-Icaza, L.: Passivity based control of time controller design with neural network for hypersonic
under-actuated mechanical systems with nonlinear dynamic flight vehicle via back-stepping. Int. J. Control 84, 1543–
friction. J. Vib. Control. (2011). https://doi.org/10.1177/ 1552 (2011)
1077546311408469 40. Zhang, H., Qin, C., Luo, Y.: Neural-network-based con-
25. Peng, Z., Wang, D., Chen, Z., Hu, X., Lan, W.: Adaptive strained optimal control scheme for discrete-time switched
dynamic surface control for formations of autonomous sur- nonlinear system using dual heuristic programming. IEEE
face vehicles with uncertain dynamics. IEEE Trans. Control Trans. Autom. Sci. Eng. 11, 839–849 (2014)
Syst. Technol. 21, 513–520 (2013). https://doi.org/10.1109/ 41. Wang, T., Gao, H., Qiu, J.: A combined adaptive neural net-
TCST.2011.2181513 work and nonlinear model predictive control for multirate
26. Cong, S., Liang, Y.: PID-like neural network nonlinear adap- networked industrial process control. IEEE Trans. Neural
tive control for uncertain multivariable motion control sys- Netw. Learn. Syst. 27, 416–425 (2016)
tems. IEEE Trans. Ind. Electron. 56, 3872–3879 (2009) 42. Zhang, H., Wang, Z., Liu, D.: A comprehensive review of
27. Sazonov, E.S., Klinkhachorn, P., Klein, R.L.: Hybrid LQG- stability analysis of continuous-time recurrent neural net-
neural controller for inverted pendulum system. In: Proceed- works. IEEE Trans. Neural Netw. Learn. Syst. 25, 1229–
ings of the 35th Southeastern Symposium on System Theory, 1262 (2014)
2003, pp. 206–210. IEEE (2003) 43. Zhang, H., Cui, L., Luo, Y.: Near-optimal control for
28. Sprangers, O., Babuška, R., Nageshrao, S.P., Lopes, G.A.: nonzero-sum differential games of continuous-time nonlin-
Reinforcement learning for port-Hamiltonian systems. IEEE ear systems using single-network ADP. IEEE Trans. Cybern.
Trans. Cybern. 45, 1017–1027 (2015) 43, 206–216 (2013)
29. Li, J., Guo, X., Li, Z., Chen, W.: Stochastic adaptive opti- 44. Zou, A.-M., Kumar, K.D., Hou, Z.-G., Liu, X.: Finite-time
mal control of under-actuated robots using neural networks. attitude tracking control for spacecraft using terminal slid-
Neurocomputing. 142, 190–200 (2014). https://doi.org/10. ing mode and Chebyshev neural network. IEEE Trans. Syst.
1016/j.neucom.2014.04.049 Man Cybern. Part B Cybern. 41, 950–963 (2011)
30. Yang, C., Li, Z., Cui, R., Xu, B.: Neural network-based 45. Pucci, D., Romano, F., Nori, F.: Collocated adaptive control
motion control of an underactuated wheeled inverted pen- of underactuated mechanical systems. IEEE Trans. Robot.
dulum model. IEEE Trans. Neural Netw. Learn. Syst. 25, 31, 1527–1536 (2015)
2004–2016 (2014). https://doi.org/10.1109/TNNLS.2014. 46. Yang, C., Li, Z., Li, J.: Trajectory planning and optimized
2302475 adaptive control for a class of wheeled inverted pendulum
31. Tong, S.C., Li, Y.M., Zhang, H.G.: Adaptive neural net- vehicle models. Cybern. IEEE Trans. 43, 24–36 (2013)
work decentralized backstepping output-feedback control 47. Spong, M.W.: Underactuated mechanical systems. In: Sicil-
for nonlinear large-scale systems with time delays. IEEE iano, B., Valavanis, K.P. (eds.) Control Problems in Robotics
Trans. Neural Netw. 22, 1073–1086 (2011). https://doi.org/ and Automation. Lecture Notes in Control and Information
10.1109/TNN.2011.2146274
123
1464 P. Liu et al.
Sciences, vol. 230, pp. 135–150. Springer, Berlin, Heidel- 49. Yu, H., Liu, Y., Yang, T.: Closed-loop tracking control of a
berg (1998) pendulum-driven cart-pole underactuated system. Proc. Inst.
48. Li, H., Furuta, K., Chernousko, F.L.: Motion generation of Mech. Eng. Part J. Syst. Control Eng. 222, 109–125 (2008)
the capsubot using internal force and static friction. In: Pro-
ceedings of the 45th IEEE Conference on Decision and Con-
trol, pp. 6575–6580. IEEE (2006) Publisher’s Note Springer Nature remains neutral with regard
to jurisdictional claims in published maps and institutional affil-
iations.
123