Networked Reset Control Systems with Time Delays in Both Forward
and Feedback Channels
Guanglei Zhao
1∗
, Jie Mi
2
, Jing Leng
1
, Changchun Hua
1
Abstract— In this paper, in order to adapt reset control to
networked control systems (NCS), multi-input multi-output
(MIMO) discrete-time reset control model is firstly established
by decomposing the controller output and defining appropriate
flow set and jump sets. Then, we apply discrete-time reset con-
trollers to NCS with time-varying delays in both forward and
feedback channels, the modelling framework of networked reset
control systems is given based on switched linear parameter-
varying (LPV) systems, and stability conditions are derived in
terms of linear matrix inequalities. Finally, experimental and
simulation examples are presented to illustrate the superior
performance of the proposed approaches.
I. INTRODUCTION
Reset control was firstly proposed in [7] as the so-called
Clegg integrator, whose state is reset to zero whenever its
input crosses zeros. The main reason of using reset con-
trollers is that they are able to overcome some fundamental
limitations of linear systems [5],[10],[6]. Due to the advan-
tages, reset control has attracted lots of researchers’ attention
[14],[11],[4],[18] and can be used in many engineering
applications [13],[4]. However, most of the existing results
have only concerned with continuous-time reset control, that
are not applicable to NCS due to computer or embedded
processor based controllers are usually used in NCS.
Networked control systems (NCS) are systems in which the
controllers, sensors, actuators are not connected via point-
to-point connections, but through communication networks.
The distributed structure provides some advantages, such as
increased system flexibility, reduced complexity in wiring
connections, and ease of installation and maintenance. The
NCS is a popular research area [1],[17]. However, it is well
known that networking the control systems also introduces
some communication imperfections, such as packet dropouts
and variable transmission delays. In particular, the network-
induced delays are inevitable owing to the period of time
it takes for the signals to transmit, and the delays may
introduce some performance limitations that will degrade
performance of the NCS or even destabilize the NCS. The
network-induced delay related problems have been con-
sidered in some recent work [8],[15]. In general, it is a
challenging problem of using nonlinear/hybrid controllers to
This work was supported by National Natural Science Foundation of
China under Grant 61603329, China Postdoctoral Science Foundation under
Grant 2016M601283, 2017T100167, Natural Science Foundation of Hebei
Province under Grant F2017203145.
1. The authors are with the Institute of Electrical Engineering, Yanshan
University, Qinhuangdao, China.
glzhao517@126.com
2. The author is with the Department of Human Resource, Yanshan
University, Qinhuangdao, China.
overcome performance limitations introduced by network-
induced delays. In this work, the reset control, which was
originally proposed to overcome fundamental limitations of
linear systems [7], will be used to deal with network-induced
delays (similar work can be found in [3]).
The concept of discrete-time reset control has been intro-
duced in some recent works (see [3]), where the discrete-time
reset controller is in the form of
⎧
⎪
⎨
⎪
⎩
x
c
(k +1)=A
c
x
c
(k)+B
c
u
c
(k),u
c
(k − 1)u
c
(k) > 0
x
c
(k +1)=A
r
x
c
(k),u
c
(k − 1)u
c
(k) < 0
u(k)=C
c
x
c
(k)
(1)
where x
c
(k) is short for x
c
(kh) (h is the sampling period).
The above discrete-time reset controller implicitly assumes
that each reset action takes one time step to perform, that is
essentially different from the continuous-time reset control
which assumes instantaneous reset. The reset controller (1)
was applied for the first time to NCS in [2], and the results
were further analyzed in [3], where delays in forward channel
was considered in the modelling and analysis. But the reset
controller (1) is only suitable for single-input single-output
(SISO) systems, which restricts its applications in general
NCS. Moreover, the approaches [12],[16] regarding reset
control of MIMO systems are not easy to be applied to
NCS with communication delays. Thus, in order to adapt
reset control to NCS, a discrete-time MIMO reset controller
model will be proposed.
The main contributions of this work are twofold: first,
discrete-time MIMO reset controller model is proposed based
on sector reset conditions. Then, the proposed discrete-
time reset controller is applied to NCS and the modelling
framework of networked reset control systems (NRCS) are
established based on switched LPV systems. The rest of
this paper is organized as follows. Section II presents the
problem formulation. The discrete-time MIMO reset control
model is proposed in Section III and then applied to NCS
in Section IV. Experimental and simulation examples are
presented in Section V to show the effectiveness of the
proposed approaches and Section VI concludes this paper.
Notation: The set of real (integer) numbers is denoted as R
(N). x denotes Euclidean norm of x. ∩ denotes intersec-
tion operation. diag{A
1
, ··· ,A
n
} denotes a block-diagonal
matrix with the entries A
1
, ··· ,A
n
on the diagonal. The
convex hull of a set S is denoted as coS. λ
max
(P )(λ
min
(P ))
denotes the maximal (minimal) eigenvalue of a positive
definite matrix P .
2017 IEEE 56th Annual Conference on Decision and Control (CDC)
December 12-15, 2017, Melbourne, Australia
978-1-5090-2873-3/17/$31.00 ©2017 IEEE 5985