Observer-based adaptive synchronization of uncertain
time-delay chaotic systems
Xiao-Shi Xiao, Changzhong Pan
School of Information and Electrical Engineering, Hunan University of Science and Technology, Xiangtan, Hunan, 411201, China
E-mail: yezhiheng10@sina.com
Abstract: For a class of chaotic systems with uncertain nonlinear time-delay functions and exogenous disturbance, the problem
of observer-based adaptive synchronization is investigated. Based on Lyapunov-Krasovskii fucntional method, a delay-free
observer is presented by using sliding mode control and adaptive control techniques. Two numerical simulations are given to
show the effectiveness of the proposed chaos synchronization scheme.
Key Words: Observer, Chaos synchronization, Time delay, Adaptive control, Uncertainty
1 Introduction
Synchronization of chaotic systems has been receiving ex-
tensive research interest since two decades ago due to its po-
tential applications in secure communication, chemical re-
actions, chaos generator design, biological systems, etc. So
far, various techniques have been developed by using linear
or nonlinear feedback control methods [1–8]. However, the
majority of them were based on exact knowledge of system’s
structure and parameters. In practice, obtaining a system’s
accurate model is usually difficult and sometimes even im-
possible. With this fact in mind, some adaptive or robust
synchronization methods have been developed for chaotic
systems with parameter uncertainties [9–12]. Nevertheless,
when chaotic systems suffer from strong nonlinear uncer-
tainties, especially strong nonlinear uncertainties with time
delays, the existing results are not feasible.
On the other hand, chaos phenomena are frequently found
in some time-delay systems [13–16]. The introducing of
time delay renders the problem of chaos synchronization
much more complicated than delay-free chaotic systems, e-
specially when the time delay is unknown. In [17], an H
∞
synchronization scheme was presented for a class of certain
chaotic systems with multiple time delays by using linear
matrix inequality technique and Lyapunov-Krasovskii (L-K)
functional method. In [18], the problem of inverse chaos
synchronization between two unidirectionally linearly and
nonlinearly coupled chaotic systems with multiple time de-
lays was considered. By using linear and nonlinear feed-
back, chaos synchronization was realized for a class of time-
delay chaotic systems in [19]. Unfortunately, these results
are feasible only when the considered chaotic systems are
certain and time delays are known.
Motivated by the aforementioned fact, the problem of
chaos synchronization of uncertain time-delay chaotic sys-
tems is investigated in this study. The considered chaotic
systems not only have nonlinear uncertainties with unknown
time delay, but also have exogenous disturbance. By us-
ing L-K functional method, sliding mode control (SMC) and
adaptive control technique, a synchronization scheme is pro-
posed. The main contributions of this study can be summa-
rized as: (1) The nonlinearities in the considered systems
This work is supported by National Natural Science Foundation (NNS-
F) of China under Grant 61403134,61403135.
are assumed to be uncertain and bounded by general nonlin-
ear functions, this fact means that the designed observer has
more extensive field of application than the existing results
where the nonlinearities were assumed to be certain and lo-
cally or globally Lipschitz and the observers were designed
by duplicating the considered systems [20–22]; (2) The pro-
posed observer is delay-independent, which means that it is
feasible when the time delay is unknown. This characteristic
is very important for chaos synchronization in secure com-
munication since the network-induced delay is often time-
varying and uncertain; (3) The proposed observer can be ap-
plied to non-square systems (i.e., the dimensions of input and
output are not equal). Finally, two numerical examples are
provided to show the effectiveness of the presented scheme.
Notations: The notations used throughout this paper are
quite standard, the n-dimensional Euclidean space will be
denoted by R
n
, R
n×m
refers to the set of all n × m real ma-
trices. Matrices, if their dimensions are not explicitly stated,
are assumed to be compatible for algebraic operations, Su-
perscript T denotes the transpose of a matrix. Finally, we
use ·to denote the Euclidean norm. Functions’ variables
sometimes are omitted when no confusion can arise.
2 Problem formulation
Consider the following uncertain time-delay chaotic sys-
tems (drive systems)
˙x(t)=Ax(t)+Bf(x(t),x(t − τ)) + Bw(t)
y(t)=Cx(t)
(1)
where x(t) ∈ R
n
and y(t) ∈ R
p
are system’s state vec-
tor and output vector, respectively, f(x(t),x(t − τ)) ∈ R
m
(m ≤ p ≤ n) is the nonlinear uncertainty with unknown
time delay τ, w(t) is the exogenous disturbance bounded by
δ (i.e., w(t)≤δ), A ∈ R
n×n
,B ∈ R
n×m
and C ∈ R
p×n
are known constant matrices. In this study, the following
assumptions are imposed.
Assumption 1 The couple (A, C) is completely observable.
Accordingly, there exists a constant matrix L ∈ R
n×p
such
that the matrix
¯
A = A − LC is Hurwitz.
Assumption 2 The uncertain nonlinear time-delay function
vector f(x(t),x(t − τ)) satisfies the following inequality
f(x(t),x(t − τ))≤α
T
a(y(t)) + β
T
b(y(t − τ)) (2)
Proceedings of the 36th Chinese Control Conference
Jul
26-28, 2017, Dalian, China
3321