Sliding mode control for network control systems with packets
dropout
Luyang Zhang
1
, Ling Zhao
2,3
,LiLi
1
1. Institute of Electrical Engineering, Yanshan University, Qinhuangdao 066004, China.
2. Hebei Province Key Laboratory of Heavy Machinery Fluid Power Transmission and Control, Yanshan
University, Qinhuangdao 066004, China.
E-mail: zhaoling@ysu.edu.cn
3. Key Laboratory of Advanced Forging & Stamping Technology and Science (Yanshan University) Ministry of
Education of China, College of Mechanical Engineering, Yanshan University, Qinhuangdao 066004, China.
Abstract: In this paper, we investigate the problems of stabilization of networked control systems with packets dropout via
delta operator approach. A sliding mode controller is constructed to analyze the stochastic stability of the closed-loop networked
control systems. The process of packets dropout is modeled as a Bernoulli process. Sufficient LMI conditions are developed to
guarantee the existence of the discrete linear sliding surfaces. By solving the conditions, a linear sliding surface is characterized
and a sliding mode controller is designed. Simulation results are given to demonstrate the effectiveness of the proposed approach.
Key Words: Sliding mode control, network control systems(NCSs), delta operator system, packets dropout.
1 Introduction
During the past two decades, networked control system-
s (NCSs) have been widely used, and they are defined as
a feedback control system that the control loops are closed
through a realtime network [1, 2]. It is well known that net-
work is an unreliable communication channel and it has been
well recognized that using network in NCSs leads to several
practical issues, such as network-induced delays and pack-
ets dropout [3, 4]. Those issues make the analysis of NCSs
more complex and importantly, and they are the major caus-
es for NCSs performance deterioration and potential system
instability. However, these issues have not been considered
by conventional control theory as so far.
In recent years, delta operator based implementations
have gained interest due to their excellent finite word length
performance under fast sampling, please refer to [5–7]. The
delta operator model has the advantage of better numerical
properties at high sampling rates in finite word length com-
puter especially. Delta operator was constructed to instead of
traditional shift operator for sampling continuous systems at
high sampling rate in [8, 9]. Analysis and synthesis of delta
operator systems has been discussed in the book [10]. The
problem of robust state feedback control using delta operator
approach for a class of uncertain systems with time-varying
delays was presented in [11]. A strictly positive real con-
trol with a low frequency range has been presented in [12].
Stability analysis for high frequency networked control sys-
tems has been considered by delta operator approach in [13].
Stability analysis of Markovian jump delta operator systems
over networks has been given in [14].
It is worth noting that sliding mode control (SMC) is well
known for its robustness in handling external disturbances
and system modeling errors. Therefore, the SMC method
is employed in this paper to compute the future control se-
quences. The estimated state errors are transformed into sys-
tem disturbances, and the stability of the closed-loop NCSs
is guaranteed by constructing the robust SMC controller. By
using SMC in robust systems has attractive features such as
fast response and good transient performance, see for exam-
ple [15–17]. In addition, the problem of SMC of uncertain
state continuous systems with input delay has been investi-
gated in [18]. A discrete version of SMC is important when
the implementation of the control is realized digitally using a
relatively slow sampling period. In [19], the problem of dis-
crete variable structure control (VSC) was first considered.
In this paper, we consider a discrete sliding mode con-
troller to analyze the stochastic stability of the closed-loop
networked control systems. The packets dropout process is
modeled as a Bernoulli process. Sufficient LMI conditions
are developed to guarantee the existence of the discrete s-
liding mode controller. By solving the conditions, we can
characterize a linear sliding surface. The SMC method is
employed in this paper to compute the future control se-
quences. Furthermore, a sliding mode controller is also de-
signed. Simulation and experimental results are given to
demonstrate the effectiveness of the proposed approach.
2 Problem formulation and preliminaries
The plant, which is completely controllable and complete-
ly observable, is described as follows:
˙x(t)=A
s
x(t)+B
s
(u(t)+w(t, x(t))) (1)
This work was supported by the National Basic Research Program of
China (973 Program) under Grant 2014CB046405. The work of Ling Zhao
was supported partially by the Hebei Provincial Natural Science Fund un-
der Grand E2014203122. The work of Li Li was supported by the Nation-
al Natural Science Foundation of China under Grant 61403330, the Post-
doctoral Science Foundation of China under Grant 2014M551052, and the
Natural Science Foundation of Hebei Education Department under Grant
QN2014068.
Proceedings of the 34th Chinese Control Conference
Jul
28-30, 2015, Han
zhou, China
6592