Fuzzy15 00221
Fuzzy15 00221
Fuzzy15 00221
Article
Nonlinear Adaptive Fuzzy Control Design for Wheeled Mobile
Robots with Using the Skew Symmetrical Property
Yung-Hsiang Chen 1 and Yung-Yue Chen 2, *
Abstract: This research presents a nonlinear adaptive fuzzy control method as an analytical design
and a simple control structure for the trajectory tracking problem in wheeled mobile robots with skew
symmetrical property. For this trajectory tracking problem in wheeled mobile robots, it is not easy to
find an analytical adaptive fuzzy control solution due to the complicated error dynamics between
the controlled wheeled mobile robots and desired trajectories. For deriving the analytical adaptive
fuzzy control law of this trajectory tracking problem, a filter link is firstly adopted to find the solvable
error dynamics, then the research is based on the skew symmetrical property of the transformed
error dynamics. This proposed nonlinear adaptive fuzzy control solution has the advantages of low
computational resource consumption and elimination of modeling uncertainties. From the results
for tracking two simulation scenarios (an S type trajectory and a square trajectory), the proposed
nonlinear adaptive fuzzy control method demonstrates a satisfactory trajectory tracking performance
for the trajectory tracking problem in wheeled mobile robots with huge modeling uncertainties and
outperforms the existing H2 control method.
Keywords: nonlinear adaptive fuzzy control; wheeled mobile robots; skew symmetrical property;
energy consumption
Citation: Chen, Y.-H.; Chen, Y.-Y.
Nonlinear Adaptive Fuzzy Control
Design for Wheeled Mobile Robots
with Using the Skew Symmetrical
Property. Symmetry 2023, 15, 221. 1. Introduction
https://doi.org/10.3390/ The broad applications of wheeled mobile robots (WMRs) with a skew symmetrical
sym15010221 property have been attracting much attention because WMRs have been becoming more
Academic Editors: Chun-Yen Chang,
and more important for assisting daily human life and industry operations in recent years.
Teen-Hang Meen, Charles Tijus and These WMRs need to have a precise motion ability and effective control methods to achieve
Po-Lei Lee assigned missions [1–5]. Much of the literature on WMRs for trajectory tracking control
have been focused on fuzzy control [6–9], neural networks control [10–12], sliding mode
Received: 22 December 2022 control [13–16], backstepping control [17–19], and feedback linearization control [20–22].
Revised: 8 January 2023
However, in real applications, successfully implementing the above proposed control meth-
Accepted: 10 January 2023
ods requires huge design efforts due to controller complexities. Thus, an effective optimal
Published: 12 January 2023
control with a simple and easy-to-implement control structure is important for controlled
WMR, and there are several related research works on this [23–26]. For the above issue,
a simple optimal control structure should be developed, and this issue can be fixed with
Copyright: © 2023 by the authors.
a closed-form solution or analytical solution for this optimal trajectory tracking problem.
Licensee MDPI, Basel, Switzerland. However, it is a difficult task for the trajectory tracking problem of the WMR to obtain a
This article is an open access article closed-form solution or analytical solution due to complex error dynamics. Fortunately,
distributed under the terms and using a suitable mathematical arrangement, the closed-form solution or an analytical so-
conditions of the Creative Commons lution can be derived. In practice, the WMR must work under modelling uncertainties
Attribution (CC BY) license (https:// due to variations of payload and energy loss. Therefore, an innovative nonlinear adaptive
creativecommons.org/licenses/by/ fuzzy optimal control method that is a closed-form design will compensate for the issue
4.0/). of system parameter variations for the optimal trajectory tracking problem of the WMR
in this investigation. Adaptive fuzzy control designs with parameters of on-line learning
ability have been widely used in many applications. For example, a neural network based
adaptive control design was proposed for dealing with nonlinear nonaffine systems with
modeling uncertainties [27]; a trajectory-tracking and path-following design was developed
for underactuated autonomous vehicles with parametric modeling uncertainty [28]; an
iterative learning control (ILC) of constrained multi-input multi-output (MINO) nonlinear
systems under the state alignment condition with varying trail lengths was developed to
meet the alignment condition by adjusting the reference trajectory [29]; and a data-driven
control of networked nonlinear systems with event-triggered output was proposed by
integrating the estimated disturbances, the true and the estimated tracking error and the
resultant tracking error systems, which were uniformly ultimately bounded [30]. These
published results indicate the powerful modeling elimination capability of adaptive control
laws with on-line parameter learning ability. For these reasons, for tackling varying system
parameters of the WMR, a parameterized formulation is used on-line to precisely estimate
time-varying system parameters, including mass, inertia, etc., in this investigation. This
proposed adaptive fuzzy control law facilities engineers and researchers to realize control
design of the WMR practically, because it is a closed-form solution and has the simplest
control structure. Due to these promising characteristics, this centralized control law will
possess a low calculation consumption and thus save energy. Furthermore, this proposed
method can mitigate the effect of varying modelling uncertainties based on the on-learning
ability of the derived adaptive fuzzy control law for system parameters. This research is
arranged as follows. Section 1 presents the introduction and literature review. In Section 2,
the wheeled mobile robot dynamics are described. Section 3 presents the adaptive fuzzy
with H2 control design. Section 4 illustrates the simulation results obtained for tracking the
S type and square trajectory by adopting the proposed adaptive fuzzy control method and
the H2 control method. Finally, the conclusions are given in Section 5.
The dynamic system of the WMR is illustrated in Figure 1 and this WMR usually
moves heading toward the direction of the axis of the driving wheels. Then, the kinematics
of the WMR with constraints can be presented as Equation (2).
.
xR
cos θ −d sin θ
. . v
a = y.R = sin θ d cos θ l (2)
ω
θ 0 1
.. . .
P( a) a + Q( a, a) a + O( a) = W ( a)τ (3)
.
where P( a) ∈ R3×3 is the inertia matrix, Q( a, a) ∈ R3×3 is the centripetal and coriolis
matrix, O( a) ∈ R3×3 is the gravitational vector, W ( a) ∈ R3×2 is the input transformation
Symmetry 2023, 15, x FOR PEER REVIEW . .. 3 of 21
matrix, and τ ∈ R2×1 is the torque of input control vector. The a and a are velocity and
acceleration vectors of the WMR, respectively.
Figure1.1.Wheeled
Figure Wheeled mobile
mobile robot
robot dynamic
dynamicsystem.
system.
According
In Equationto(3),
thethere
WMR kinematics
are modelwhich
two properties in Equation (2), it canasbeP1
are described transferred to the
and P2 [31,32].
dynamics
P1. Theof inertia
the controlled
matrix PWMR
( a) is by applyingpositive
symmetric the Euler–Lagrange
definite. method. Then, Equa-
. .
tion P2.
(2) can be rewritten
Skew symmetry asproperty:
Equation (3):
The matrix P( a) − 2Q( a, a) is skew symmetric and
. .
x T ( P( a) − 2Q( a, a)) x = 0Pfor ) a x+∈QR(na. , a ) a + O ( a ) = W ( a )τ
( a all (3)
The O( a) can be ignored because this WMR moves on the horizontal plane. Then,
where P(3)
Equation (a) ∈ 3×be
can
3
is the inertia
rewritten as matrix,
Equation a,a) ∈3×3 is the centripetal and coriolis matrix,
Q((4).
O(a) ∈3×3 is the gravitational vector,.. W (a) ∈ . 3×.2 is the input transformation matrix, and
P( a) a + Q( a, a) a = W ( a)τ (4)
τ ∈ 2×1 is the torque of input control vector. The a and a are velocity and accelera-
tion The
vectors of the descriptions
WMR, respectively. .
detailed of P( a) ∈ R3×3 , Q( a, a) ∈ R3×3 and W ( a) ∈ R3×2 are
givenIn Equation (3), there are two properties which are described as P1 and P2 [31,32].
below:
P1. The inertia matrix P(a) is symmetric
m positive
0 definite.
ml sin θ
P2. Skew symmetry ) = 0The matrixm P (a) - −
P( aproperty: 2Qml a) θis skew symmetric and(5)
(a,cos
x (P (a) - 2Q(a, a))x = 0 for all x ∈ml
T n
sin . θ −ml cos θ IP
The O ( a ) can be ignored because this WMR moves on the horizontal plane. Then,
.
0 0 ml θ cos θ
Equation (3) can be rewritten as . .
QEquation
( a, a) = (4). (6)
0 0 ml θ sin θ
P ( a ) a + Q (0a , 0a ) a = 0W ( a )τ (4)
where m is the WMR mass, θ is the angle of heading attitude, l is the distance between
position R and a, IP is the inertia, τr and τl are the torques of right and left wheel, respec-
tively. The system parameters (m, l, IP ) of WMR are modeling uncertainty due to varying
payloads in this research.
where T
ar = [ x d yd θd ] (10)
The following dynamics of the trajectory tracking error can be obtained by Equations
(4) and (9).
. .. . .
− P−1 ( a) Q( a, a) 03×3 − ar − P−1 ( a) Q( a, a) ar
−1
. P ( a )W ( a ) τ
e= e+ + (11)
I3×3 03 × 3 03 × 3 03 × 3
In order to analyze this trajectory tracking problem of a WMR with Equation (11), a
PD type filter link f (t) and a state space transformation matrix T are defined as Equations
(12) and (13) to simplify this dynamic of the trajectory tracking error.
.
f (t) = ϑ â + v â (12)
Differentiating the PD type filter link f (t) and Equation (12) can be described with
following equation:
. .
f (t) = − P−1 ( a) Q( a, a) f (t) + ϑP−1 ( a)[− F (e, t) + W ( a)τ ] (14)
where ϑ.
.. . . ϑ
F (e, t) = P( a)( ar −
â) + Q( a, a, )( ar − â) (15)
v v
From Equations (12) to (14), a more solvable error dynamics can be presented as
". #
. −1 f. ( t )
= M(e, t)e(t) + B(e, t) − F (e, t) + τ 0
e=T (16)
â(t)
in which .
− P−1 ( a) Q( a, a)
03×3
M (e, t) = T −1 1 T (17)
v I3×3 − vϑ I3×3
B(e, t) = T −1 DP−1 ( a) (18)
I
D = 3×3 (19)
03×3
τ 0 = W ( a)τ (20)
Symmetry 2023, 15, 221 5 of 18
The nonlinear function F (e, t) contains perturbed factors due to the variations of the
WMR system parameters. For precisely estimating F (e, t), a fuzzy approximation which
includes a set of adjustable parameters Θ f is integrated as an important control part of τ 0
as follows:
τ 0 = τe + τ f (e, Θ f ) (21)
where τ f (e, Θ f ) is a fuzzy approximator and Θ f is the adaptive parameter vector. This
τ f (e, Θ f ) is used for approximating the perturbed term F (e, t). The control term τe is used
to eliminate the influence of the environmental disturbances.
Substituting Equation (21) into Equation (16), we can obtain Equation (22).
.
h i
e = M (e, t)e(t) + B(e, t)τe + B(e, t) − F (e, t) + τ f (e, Θ f ) (22)
The proposed fuzzy approximation τ f (e, Θ f ) with adjustable parameters can be de-
scribed in Equation (23):
τ f (e, Θ f ) = ξ (e)Θ f (23)
where T
Θ f = Θ1 Θ2 Θ3
(24)
T
ξ 1 (e) 01× M 01× M
ξ (e) = 01× M ξ 2T (e) 01× M (25)
01× M 01× M ξ 3T (e)
T
Θi = Θi1 · · · ΘiM
i = 1, 2, 3 (26)
T
ξ i e) = ξ i1 (e) · · · ξ iM (e) i = 1, 2, 3 (27)
Furthermore, the optimal parameter vector Θ f (t) must be defined before deriving the
adaptive law. The optimal parameter vector is defined as
where k.k denotes the Euclidean norm, ΩΘ and Ωe are the set of Θ f (t) and e(t), respectively.
Based on this optimal parameter vector Θ∗f (t), the perturbed term F (e, t) can be formulated
as follows:
F (e, t) = ξ (e)Θ∗f + ε(t) (30)
where
e f (t) = Θ f (t) − Θ∗ (t)
Θ (32)
f
where L f , L, Z and X are the weighting matrix which are positive definite matrices for all
t f ∈ [0, ∞).
If one unique solution, C (e, t) satisfies the nonlinear time-varying differential Equation
(35). According to the mathematical analysis, the trajectory tracking problem of a WMR
will be guaranteed to find a closed-form solution.
.
C (e, t) + C (e, t) M (e, t) + M T (e, t)C (e, t) + L − C (e, t) B(e, t) X −1 B T (e, t)C (e, t) = 0 (35)
3.2. Analytical Solution C (e, t) of Adaptive Fuzzy with H2 Trajectory Tracking Problem
The trajectory tracking problem is a closed-form solution if the analytical solution
C (e, t) is to be solved mathematically. It is tough to solve Equation (35) and find an
analytical solution for C (e, t) because the differential Equation (35) is a very complex time-
varying equation. Fortunately, we can obtain the solution of C (e, t) by treating C (e, t) as
the following mathematical form:
M (e, t) 03×3
C (e, t) = R T R (40)
03 × 3 O
where
0 O
Λ = 3×3 (42)
O 03 × 3
Equation (43) can be obtained by using Equations (18) and (40):
According to Equations (41) and (43), the time-varying Equation (35) can be described
as the following Equation (44):
Λ + L − T T DX −1 D T T = 0 (44)
Symmetry 2023, 15, 221 7 of 18
Suppose
X = δ2 I3×3 (45)
where δ > 0.
The symmetric matrix L in Equation (44) is a diagonal matrix and can be further
factorized as the following form:
l11 T l11
l12
L= T (46)
l21 l22 T l22
Applying T and D defined in Equations (13) and (19), Equation (44) can be described
as Equation (47):
" #
l11 T l11 − δ12 T11 T T11 O + l12 − δ12 T11 T T12
=0 (47)
O + l12 T − δ12 T12 T T11 l22 T l22 − δ12 T12 T T12
According to Equation (13), above submatrices T11 and T12 can be rewritten as
and
ϑ=v=δ (52)
From the results of Equation (47), the nonlinear adaptive fuzzy control law τ 0 (e, t) can
be presented as Equations (53) and (55), respectively:
1
τe∗ (e, t) = − Π T e(t) (53)
δ
where T
Π = l11
l22 (54)
e f (t) = Z −1 ξ (t)Πe(t)
Θ (55)
Then, the trajectory tracking problem of adaptive fuzzy control can be solved by the
following adaptive fuzzy control law.
1
Symmetry 2023, 15, 221 τ ′(e, t) = ξ (t)Θ*f (t) + τe* (e, t) (58) 8 of 18
ϖ
3.3. Closed-Loop3.3.
Control Diagram
Closed-Loop of ThisDiagram
Control ProposedofMethod
This Proposed Method
Assuming all system states
Assuming allof the controlled
system states of WMR can be fully
the controlled WMR measured, the measured,
can be fully control the control
block diagramblock
of thediagram
trajectory tracking design of a WMR is illustrated in Figure
of the trajectory tracking design of a WMR is illustrated 2. There
in Figure 2. There
are five function
areelements, including
five function a desired
elements, trajectory
including generator,
a desired the proposed
trajectory generator,adaptive
the proposed adaptive
control law, optimal
control law, optimal control term, system parameter adapterdynamics
control term, system parameter adapter law and of the of the WMR.
law and dynamics
WMR.
4. Verification4.Results
Verification Results
There are twoThere
testingare two testing
scenarios, i.e., scenarios,
an S type i.e.,
and an S type
square and square
trajectory, trajectory,
to verify the to verify the
tracking performance in this section. The desired S type trajectory
tracking performance in this section. The desired S type trajectory has a radius of 5 m. For has a radius of 5 m. For
the desired square trajectory, there are four straight lines
the desired square trajectory, there are four straight lines and four corners. The mathe- and four corners. The mathe-
matical
matical equations equations
of the desired Softype
the desired
and squareS type and square
trajectory trajectory in
are presented areSection
presented
4.1. in Section 4.1.
This proposedThis proposed
method method
will also will also with
be compared be compared
H2 control with H2 control
approach [24],approach
which was [24], which was
developed baseddeveloped basedWMR
on the same on themodel
same WMR as thismodel as this investigation.
investigation. In addition, In addition,
Matlab ver-Matlab version
of 2022a
sion of 2022a will will to
be used beverify
used to verify
the the trajectory
trajectory trackingtracking performance
performance of this of this proposed method
proposed
and the compared
method and the compared method. method.
In the desired S type trajectory scenario, the original states are x0 = 0(meter),
◦
y0 = 0(meter), and ωd = 3 /s. This desired trajectory (DT) is generated by Equation (59)
and is shown in Figure 3.
x = x0 + rd cos(2θd )
(59)
y = y0 + 2rd sin(θd )
Rt
where rd is the radius of the desired S type trajectory, and θd = 0 ωd dt is the desired
rotation angle with the desired constant angular velocity ωd .
x = x0 + rd cos(2θd )
(59)
y = y0 + 2rd sin(θd )
t
Symmetry 2023, 15, 221 where rd is the radius of the desired S type trajectory, and θ d = ωd dt is the desired
9 of 18
0
rotation angle with the desired constant angular velocity ω d .
Figure3.3.The
Figure Thedesired
desiredSStype
typetrajectory
trajectorywith
withaaradius
radiusof
of55m
mfrom
fromxx00== 00 m,
m, yy00 == 00 m.
m.
In
Inanother
another desired
desiredsquare trajectory
square scenario,
trajectory this desired
scenario , thistrajectory
desired (DT) is constructed
trajectory (DT) is with
con-
four straight lines and four corners, which are d1, d2, . . . , d8 from
structed with four straight lines and four corners, which are d1, d2, …, d8 from Equation (60) to (67). The
Equation
initial condition
(60) to (67). of this desired square trajectory is configured
The initial condition of this desired square trajectory to x 0 = − 7 ( m ) , y 0 =
is configured − 8 ( ),
mto
Symmetry 2023, 15, x FOR PEER REVIEW ◦ 11 of 21
and = 3 /s. These sub-trajectories are described in the following statements
x0 = −r 7(m) , y0 = −8(m) , and ω r = 3 ° / s . These sub-trajectories are described in the fol-
ω and are
shown in Figure 4.
lowing statements and are shown in Figure 4.
Figure4.4.The
Figure Thedesired
desiredsquare
squaretrajectory
trajectorywith
withaaradius
radiusof
of88m
mfrom
fromxx00== −
−77 m,
m, yy00==−8−m.
8 m.
The first line d1 starts from ( x0 , y0 ) and moves forward to the right-hand side with
a moving velocity vr and can be described by the following equation.
xr1 = x0 + vr t1
d1 : yr1 = y0 (60)
Symmetry 2023, 15, 221 10 of 18
The first line d1 starts from ( x0 , y0 ) and moves forward to the right-hand side with a
moving velocity vr and can be described by the following equation.
xr1 = x0 + vr t1
d1 : y = y0 (60)
r1
θr1 = 0
The first corner d2 starts from ( xr1 , yr1 ) with a radius rr and rotation angle θr2 which
is presented in Equation (61).
◦
xr2 = xr1 + rr cos(θr2 + 270◦ )
d2 : yr2 = yr1 + rr sin(θr2 + 270 ) (61)
θ = R t2 ωr dt
r2 t
1
From Equation (62), the second line d3 starts from ( xr2 , yr2 ) with a moving velocity vr
◦
and a θr3 = 90 .
xr3 = xr2
d3 : y = yr2 + vr t3 (62)
r3 ◦
θr3 = 90
As in corner d2, the second corner d4 can be generated with the following equation.
The initial position of d4 is ( xr3 , yr3 ) and θr4 is the rotation angle of d4.
xr4 = xr3 + rr cos(θr4 )
d4 : yr4 = yr3 + rr sin(θr4 ) (63)
θr4 = t4 ωr dt + 90◦
R
3 t
◦
The third line d5 starts from ( xr4 , yr4 ) with a velocity vr and a θr5 = 90 is expressed as
xr5 = x4 − vr t5
d5 : y = yr4 (64)
r5 ◦
θr5 = 180
In Equation (65), the third corner d6 of this desired square trajectory starts from
( xr5 , yr5 ) with a radius rr and rotation angle θr6 .
◦
xr6 = xr5 + rr cos(θr6 + 90◦ )
d6 : yr6 = yr5 + rr sin(θr6 + 90 ) (65)
θ = R t6 ωr dt + 180◦
r6 t5
The fourth line d7 of this desired square trajectory starts from ( xr6 , yr6 ) with a velocity
◦
vr and a constant θr5 = 270 is described as following equation.
xr7 = xr6
d7 : y = yr6 − vr t7 (66)
r7 ◦
θr7 = 270
Finally, the fourth corner d8 of this desired square trajectory starts from ( xr7 , yr7 ) with
a radius rr and rotation angle θr8 can be presented as
◦
xr8 = xr7 + rr cos(θr8 + 180◦ )
d8 : yr8 = yr7 + rr sin(θr8 + 180 ) (67)
θ = R t8 ωr dt + 270◦
r8 t7
Initial conditions of the controlled WMR with respect to these two control laws and
two tracking trajectories are given in Table 2.
Symmetry 2023, 15, 221 11 of 18
h 2 i
µ Fi2 = exp − e j − 2a j (69)
j
h 2 i
µ Fi3 = exp − e j − a j (70)
j
µ Fi4 = exp −e2j (71)
j
h 2 i
µ Fi5 = exp − e j + a j (72)
j
h 2 i
µ Fi6 = exp − e j + 2a j (73)
j
h 2 i
µ Fi7 = exp − e j + 3a j (74)
j
for i = 1, 2, 3 and j = 1, 2, 3, 4, 5, 6 where a1, a2, a3, a4, a5, a6 are the center for each
membership functions which will be defined for adjusting parameters. However, each
output of the fuzzy logic system is given 7 fuzzy rules corresponding to 6 state variables.
All fuzzy rules are defined in the following statements.
R(i1) : IF e1 is F1i1 , e2 is F2i1 · · · and e6 is F6i1 then τ f i is Gi1
R(i2) : IF e1 is F1i2 , e2 is F2i2 · · · and e6 is F6i2 then τ f i is Gi2
R(i3) : IF e1 is F1i3 , e2 is F2i3 · · · and e6 is F6i3 then τ f i is Gi3
R(i4) : IF e1 is F1i4 , e2 is F2i4 · · · and e6 is F6i4 then τ f i is Gi4
R(i5) : IF e1 is F1i5 , e2 is F2i5 · · · and e6 is F6i5 then τ f i is Gi5
R(i6) : IF e1 is F1i6 , e2 is F2i6 · · · and e6 is F6i6 then τ f i is Gi6
R(i7) : IF e1 is F1i7 , e2 is F2i7 · · · and e6 is F6i7 then τ f i is Gi7
for i = 1, 2, 3.
7 6
A = ∑k=1 ∏ j=1 µ Fik (e j ) (75)
j
ξ 1T (e)
01×7 01 × 7
ξ ( e ) = 01 × 7 ξ 2T (e) 01 × 7 (76)
01 × 7 01×7 ξ 3T (e)
T
Θ f = Θ1 Θ2 Θ3
(77)
where T
Θi = Θi1 Θi2 Θi3 Θi4 Θi5 Θi6 Θi7
(78)
6
∏ j=1 µ Fi1 (e j ) ∏6j=1 µ Fi2 (e j ) ∏6j=1 µ Fi7 (e j )
ξ i (e) = j j j (79)
A A ··· A
Symmetry 2023, 15, 221 12 of 18
for i = 1, 2, 3. In the following simulation, we will apply the same fuzzy rules and Gaussian
function for the fuzzy logic system.
Θ i = [Θ i1 Θ i 2 Θ i 3 Θ i 4 Θ i 5 Θ i 6 Θ i 7 ]
T
(78)
Figure 5. The result of the S type trajectory by the adaptive fuzzy control method from x = 0 m, y =
Figure
3.8 m. 5. The result of the S type trajectory by the adaptive fuzzy control method from x = 0 m,
Symmetry 2023, 15, x FOR PEER REVIEW 15 of 21
y = 3.8 m.
Figure 6. The result of the S type trajectory by the H2 control method from x = 0 m, y = 3.8 m.
Figure 6. The result of the S type trajectory by the H2 control method from x = 0 m, y = 3.8 m.
The tracking errors in the X-axis, Y-axis and the heading angle of the WMR are illus-
trated in Figures 7–9. The results depict the histories of tracking errors e X , e Y , and e θ
for the WMR using the adaptive fuzzy control method and H2 control method. From the
comparisons of these tracking results, it is easy to see that the trajectory tracking perfor-
mance of the H2 control method is worse than the proposed adaptive fuzzy control method
for the controlled WMR because of the existence of the effect of 20% modeling uncertainty
caused by Δ m . From these simulation results, both of these two control methods possess
rapid transient response. However, larger tracking errors are found in the whole tracking
Symmetry 2023, 15, 221 13 of 18
Figure 6. The result of the S type trajectory by the H2 control method from x = 0 m, y = 3.8 m.
The tracking errors in the X-axis, Y-axis and the heading angle of the WMR are
illustrated in Figures
The tracking 7–9.
errors in The results
the X-axis, depict
Y-axis and the histories
the heading angle ofofthetracking illus- e X ,eY, and
WMR areerrors
eθtrated
for the WMR7–9.
in Figures using
Thethe adaptive
results depict thefuzzy control
histories method
of tracking andeH
errors X 2 econtrol
, Y , and emethod.
θ From
the
for comparisons
the WMR using of thethese tracking
adaptive results,
fuzzy control it is and
method easyH2 to see that
control method.theFrom
trajectory
the tracking
comparisons ofofthese
performance the Htracking results,
2 control it is easy
method to see that
is worse thanthethetrajectory
proposed tracking perfor-
adaptive fuzzy control
mance offor
method the H 2 control method is worse than the proposed adaptive fuzzy control method
the controlled WMR because of the existence of the effect of 20% modeling
for the controlled WMR because of the existence of the effect of 20% modeling uncertainty
uncertainty caused by ∆m. From these simulation results, both of these two control methods
caused by Δ m . From these simulation results, both of these two control methods possess
possess rapid response.
rapid transient transientHowever,
response. However,
larger tracking larger tracking
errors are found inerrors are tracking
the whole found in the whole
tracking
period forperiod
the H2 for
control H2 control
the method. On themethod.
whole, On the whole,
the proposed the proposed
adaptive adaptive fuzzy
fuzzy control
control
methodmethod
obviouslyobviously
outperforms outperforms
the H2 controlthe H2 control
method method in the WMR.
in the WMR.
Figure 7. The X axis errors of the adaptive fuzzy and H2 control method for S type trajectory from
Figure 7. The X axis errors of the adaptive fuzzy and H2 control method for S type trajectory from x
x==0 0m,m,
y =y3.8
= 3.8
m. m.
Figure 8.
8. The
TheYYaxis errors
axis of the
errors ofadaptive fuzzy and
the adaptive H2 control
fuzzy and Hmethod for S type trajectory from x
Figure 2 control method for S type trajectory from
= 0 m, y = 3.8 m.
x = 0 m, y = 3.8 m.
Symmetry 2023, 15, 221 14 of 18
Figure 8. The Y axis errors of the adaptive fuzzy and H2 control method for S type trajectory from x
= 0 m, y = 3.8 m.
Figures 10
Figures 10 and
and 11
11 reveal
reveal the
the square
square trajectory
trajectory tracking
tracking results
results ofof the
the WMR
WMR driven
driven byby
the adaptive fuzzy control method (AF) and H method (H ). The performance
the adaptive fuzzy control method (AF) and H2 method (H2 ). The performance of the square
2 2 of the
square trajectory
trajectory trackingtracking can be obtained
can be obtained for this for
WMRthisinWMR in four corners
four corners of the desired
of the desired square
trajectory driven by these two control methods. The WMR controlled by theby
square trajectory driven by these two control methods. The WMR controlled the pro-
proposed
posed adaptive
adaptive fuzzy method
fuzzy control control method (AF) demonstrated
(AF) demonstrated outstanding
outstanding performance
performance for
for tracking
tracking
the thesquare
desired desiredtrajectory
square trajectory
compared compared
to the H2tomethod
the H2 method (H2) the
(H2 ) under under theof
effect effect
a 20%of
a 20% modeling
modeling of m. of m .
uncertainty
uncertainty
Figure 10.
Figure 10. The
The result of the square trajectory by the adaptive
adaptive fuzzy
fuzzy control
control method
method from
from xx==−9
−9m,m,y
y= =9 9m.m.
Symmetry 2023, 15, 221 15 of 18
Figure 10. The result of the square trajectory by the adaptive fuzzy control method from x = −9 m, y
= 9 m.
The tracking
The tracking error
error results
results in the X-axis, X-axis,
in theY-axis andY-axis andangle
the heading the heading
of the WMR angle
can of the WMR
can be found
be found in Figures
in Figures 12–14.
12–14. The Theillustrate
results resultstheillustrate the performance
performance of tracking errors of tracking
eX , errors eX ,
eYe,Y and
, andeθefor
θ
the
for desired
the desired square
square trajectory
trajectory tracking
tracking of the of
WMR the WMR
by by
adopting adopting
the adap- the adaptive
fuzzy control
tive fuzzy method
control method and
and H H2 2control
control method.
method. According
According to the compared
to the compared results, it results, it is
obvious
is obviousthat
that the proposed
the proposed adaptive
adaptive fuzzyfuzzy
controlcontrol method
method has a betterhas a better
trajectory trajectory tracking
tracking
performance than
performance thanthe H2Hcontrol
the 2 controlmethod
methodfor the controlled
for the WMR due
controlled WMR to the
due Δ m
to the ∆m effect of 20%
effect
of 20% modeling uncertainty. From these tracking error results, we can see that smaller
modeling uncertainty. From these tracking error results, we can see that smaller tracking
tracking errors always can be obtained throughout the entire tracking period for the pro-
errors always can be obtained throughout the entire tracking period for the proposed
posed adaptive fuzzy control method. On the whole, the H2 control method is obviously
adaptive
inferior tofuzzy controladaptive
the proposed method. On control
fuzzy the whole,
methodtheinH 2 control
the trajectorymethod
trackingisofobviously
the inferior
toWMR.
the proposed adaptive fuzzy control method in the trajectory tracking of the WMR.
Figure 12. The X axis errors of the adaptive fuzzy and H2 control method for the square trajectory
Figure 12. The X axis errors of the adaptive fuzzy and H2 control method for the square trajectory
from x = −9 m, y = 9 m.
from x = −9 m, y = 9 m.
Symmetry 2023,15,
Symmetry2023, 15,221
x FOR PEER REVIEW 16 of
19 of 18
21
Figure13.
Figure 13.The
TheYYaxis
axiserrors
errorsofofthe
theadaptive
adaptivefuzzy
fuzzyand
andHH2 2control
controlmethod
methodfor
forsquare
squaretrajectory
trajectoryfrom
from
x = −9 m, y = 9 m.
x = −9 m, y = 9 m.
Figure14.
Figure 14.The
Theangle
angleerrors
errorsofofthe
theadaptive
adaptivefuzzy
fuzzyand
andHH2 2control
controlmethod
methodfor
forthe
thesquare
squaretrajectory
trajectory
fromxx==−−9
from m, yy == 99 m.
9 m, m.
5.5.Conclusions
Conclusions
AAnonlinear
nonlinearadaptive
adaptivefuzzy
fuzzycontrol
controldesign
designwith
withthe
theskew
skewsymmetrical
symmetricalproperty
propertyisis
successfully developed for the trajectory tracking problem in wheeled mobile robots
successfully developed for the trajectory tracking problem in wheeled mobile robotsin thisin
Symmetry 2023, 15, 221 17 of 18
research. The main contribution of this study is the closed-form solution with adaptive
fuzzy control method for the trajectory tracking problem in wheeled mobile robots. Based
on the related literature, the closed-form solution for the trajectory tracking problem in
wheeled mobile robots has still not been perfected due to the lack of optimal control meth-
ods for complex dynamics under modeling uncertainties. For achieving this investigative
purpose, this study has several key contributions, i.e., the simplest and most easy-to-
implement control structure for the trajectory tracking problem in wheeled mobile robots
with a closed-form solution. According to the above results, this proposed method provides
an energy consumption saving with complex control structures in practice. From the two
simulation scenarios, the proposed adaptive fuzzy control method has better tracking
performance of the desired S type and square trajectory than the H2 method, no matter the
convergence of tracking errors in X-axis, Y-axis and the heading angle, and in handling of
modelling uncertainties.
Author Contributions: Conceptualization, Y.-H.C. and Y.-Y.C.; methodology, Y.-H.C. and Y.-Y.C.;
software, Y.-H.C. and Y.-Y.C.; validation, Y.-H.C. and Y.-Y.C.; formal analysis, Y.-H.C. and Y.-Y.C.;
investigation, Y.-H.C. and Y.-Y.C.; writing—original draft preparation, Y.-H.C. and Y.-Y.C.; writing—
review and editing, Y.-H.C. and Y.-Y.C.; supervision, Y.-H.C. and Y.-Y.C.; visualization, Y.-H.C. and
Y.-Y.C.; funding acquisition, Y.-H.C. All authors have read and agreed to the published version of
the manuscript.
Funding: This research was funded by the MOST (Ministry of Science and Technology of Taiwan),
project number is MOST 111-2221-E-020-023.
Data Availability Statement: All data revealed in this paper are made by our team.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Wang, C.; Liu, X.; Yang, X.; Hu, F.; Jiang, A.; Yang, C. Trajectory Tracking of an Omni-Directional Wheeled Mobile Robot Using a
Model Predictive Control Strategy. Appl. Sci. 2018, 8, 231. [CrossRef]
2. Ullah, Z.; Xu, Z.; Lei, Z.; Zhang, L. A Robust Localization, Slip Estimation, and Compensation System for WMR in the Indoor
Environments. Symmetry 2018, 11, 149. [CrossRef]
3. Li, Y.; Dai, S.; Zhao, L.; Yan, X.; Shi, Y. Topological Design Methods for Mecanum Wheel Configurations of an Omnidirectional
Mobile Robot. Symmetry 2019, 11, 1268. [CrossRef]
4. Dosoftei, C.; Popovici, A.; Sacaleanu, P.; Gherghel, P.; Budaciu, C. Hardware in the Loop Topology for an Omnidirectional Mobile
Robot Using Matlab in a Robot Operating System Environment. Symmetry 2021, 13, 969.
5. Eyad, A.; Mustafa, K. Modeling and Trajectory Planning Optimization for the Symmetrical Multi- Wheeled Omnidirectional
Mobile Robot. Symmetry 2021, 13, 1033.
6. Chen, Y.H.; Chao, C.H.; Cheng, C.F.; Huang, C.J. A Fuzzy Control Design for the Trajectory Tracking of Autonomous Mobile
Robot. Adv. Rob. Mech. Eng. 2020, 2, 206–211.
7. Nie, J.; Wang, Y.; Miao, Z.; Jiang, Y.; Zhong, H.; Lin, J. Adaptive Fuzzy Control of Mobile Robots with Full-State Constraints and
Unknown Longitudinal Slipping. Non. Dyn. 2021, 106, 3315–3330. [CrossRef]
8. Du, O.; Sha, L.; Shi, W.; Sun, L. Adaptive Fuzzy Path Tracking Control for Mobile Robots with Unknown Control Direction.
Discret. Dyn. Nat. Soc. 2021, 2021, 1–7. [CrossRef]
9. Brahim, M.; Hicham, A.; Mohammed, D. Fuzzy Adaptive Sliding Mode Controller for Electrically Driven Wheeled Mobile Robot
for Trajectory Tracking Task. J. Con. Dec. 2022, 9, 71–79.
10. Amani, A.; Abderrazak, C. Indirect Neural Adaptive Control for Wheeled Mobile Robot. Int. J. Inn. Tech. Exp. Eng. 2019, 9,
2138–2145.
11. Chiraz, B.; Hassene, S. Design of a PID Optimized Neural Networks and PD Fuzzy Logic Controllers for a Two-Wheeled Mobile
Robot. Asian J. Con. 2021, 23, 23–41.
12. Mateusz, S.; Marcin, S. Neural Tracking Control of a Four-Wheeled Mobile Robot with Mecanum Wheels. Appl. Sci. 2022, 12, 5322.
13. Zhai, J.Y.; Song, Z.B. Adaptive Sliding Mode Trajectory Tracking Control for Wheeled Mobile Robots. Int. J. Con. 2019, 92,
2255–2262. [CrossRef]
14. Krishanu, N.; Asifa, Y.; Anirban, N.; Manas, K.B. Event-Triggered Sliding-Mode Control of Two Wheeled Mobile Robot: An
Experimental Validation. IEEE J. Emerg. Sel. Top. Ind. Elec. 2021, 2, 218–226.
15. Xie, Y.; Zhang, X.; Meng, W.; Zheng, S.; Jiang, L.; Meng, J.; Wang, S. Coupled Fractional-Order Sliding Mode Control and Obstacle
Avoidance of a Four-Wheeled Steerable Mobile Robot. ISA Tran. 2021, 108, 282–294. [CrossRef]
Symmetry 2023, 15, 221 18 of 18
16. Brahim, M.; Hicham, A.; Mohammed, D. Extended State Observer-Based Finite-Time Adaptive Sliding Mode Control for Wheeled
Mobile Robot. J. Con. Dec. 2022, 9, 465–476.
17. Ibari, B.; Benchikh, L. Backstepping Approach for Autonomous Mobile Robot Trajectory Tracking. J. Elect. Eng. Comp. Sci. 2016, 2,
478–485.
18. Nasim, E.; Alireza, A.; Hossein, K. Balancing and Trajectory Tracking of Two-Wheeled Mobile Robot Using Backstepping Sliding
Mode Control: Design and Experiments. J. Int. Rob. Sys. 2017, 87, 601–613.
19. Muhammad, J.R.; Attaullah, Y.M. Trajectory Tracking and Stabilization of Nonholonomic Wheeled Mobile Robot Using Recursive
Integral Backstepping Control. Electronics 2021, 10, 1992.
20. Taniguchi, T.; Sugeno, M. Trajectory Tracking Controls for Non-Holonomic Systems Using Dynamic Feedback Linearization
Based on Piecewise Multi-Linear Models. J. Appl. Math. 2017, 47, 1–13.
21. Korayem, M.H.; Yousefzadeh, M.; Manteghi, S. Dynamics and Input–Output Feedback Linearization Control of a Wheeled
Mobile Cable-Driven Parallel Robot. Mult. Sys. Dyn. 2017, 40, 55–73. [CrossRef]
22. Welid, B.; Rabah, M.; Mohammed, S.B. The Impact of the Dynamic Model in Feedback Linearization Trajectory Tracking of a
Mobile Robot. Period. Pol. Elec. Eng. Comp. Sci. 2021, 65, 329–343.
23. Chen, Y.H.; Li, T.H.S.; Chen, Y.Y. A Novel Nonlinear Control Law with Trajectory Tracking Capability for Mobile Robots:
Closed-Form Solution Design. Appl. Math. Inf. Sci. 2013, 7, 749–754. [CrossRef]
24. Chen, Y.H.; Chen, Y.Y.; Lou, S.J.; Huang, C.J. Energy Saving Control Approach for Trajectory Tracking of Autonomous Mobile
Robots. Intel. Auto. Soft Comp. 2021, 30, 1–17. [CrossRef]
25. Chen, Y.H.; Lou, S.J. Control Design of a Swarm of Intelligent Robots: A Closed-Form H2 Nonlinear Control Approach. Appl. Sci.
2020, 10, 1055. [CrossRef]
26. Chen, Y.H.; Chen, Y.Y. Trajectory Tracking Design for a Swarm of Autonomous Mobile Robots: A Nonlinear Adaptive Optimal
Approach. Mathematics 2022, 10, 3901. [CrossRef]
27. Kasra, E.; Farzaneh, A.; Heidar, A.T. Adaptive Control of Uncertain Nonaffine Nonlinear Systems with Input Saturation Using
Neural Networks. IEEE Tran. Neu. Net. Lear. Sys. 2015, 26, 2311–2322.
28. Aguiar, A.P.; Hespanha, J.P. Trajectory-Tracking and Path-Following of Underactuated Autonomous Vehicles with Parametric
Modeling Uncertainty. IEEE Tran. Auto. Cont. 2007, 52, 1362–1379. [CrossRef]
29. Shen, M.; Wu, X.; Park, J.H.; Yi, Y.; Sun, Y. Iterative Learning Control of Constrained Systems with Varying Trial Lengths Under
Alignment Condition. IEEE Tran. Neu. Net. Lear. Syst. 2021, 1–7. Available online: https://ieeexplore.ieee.org/document/9664474
(accessed on 21 December 2022). [CrossRef]
30. Shen, M.; Wang, X.; Park, J.H.; Yi, Y.; Che, W.W. Extended Disturbance-Observer-Based Data-Driven Control of Networked
Nonlinear Systems with Event-Triggered Output. IEEE Tran. Sys. Man Cyb. 2022, 1–12. Available online: https://ieeexplore.ieee.
org/document/9966320 (accessed on 21 December 2022). [CrossRef]
31. Das, T.; Kar, I.N.; Chaudhury, S. Simple Neuron-Based Adaptive Controller for a Nonholonomic Mobile Robot Including Actuator
Dynamics. Neurocomputing 2006, 69, 2140–2151. [CrossRef]
32. Boukens, M.; Boukabou, A. Design of an Intelligent Optimal Neural Network-Based Tracking Controller for Nonholonomic
Mobile Robot Systems. Neurocomputing 2017, 226, 46–57. [CrossRef]
33. Chen, B.S.; Chang, Y.C. Nonlinear Mixed H2/H ∞ Control for Robust Tracking Design of Robotic Systems. Int. J. Cont. 1997, 67,
837–857. [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.