0018-9286 (c) 2016 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TAC.2017.2700384, IEEE
Transactions on Automatic Control
1
Event-Triggered Global Robust Output Regulation
for a Class of Nonlinear Systems
Wei Liu, Member, IEEE and Jie Huang, Fellow, IEEE
Abstract—This note studies the global robust output regulation
problem for a class of nonlinear systems by an output-based
event-triggered control law. First, we convert the problem into
the event-triggered global robust stabilization problem of a well
defined augmented system based on the internal model approach.
Then, we design an output-based event-triggered control law
together with an output-based event-triggered mechanism to solve
the stabilization problem, which in turn leads to the solution to
the original problem by an output-based event-triggered control
law. Finally, we illustrate our approach by an example.
Index Terms—Event-triggered control, nonlinear systems, out-
put regulation, robust control.
I. INTRODUCTION
The robust output regulation problem has been one of
the basic and important control problems since 1970s. The
problem aims to design a feedback control law for an un-
certain system so that the output of the closed-loop system
asymptotically tracks some reference inputs in the presence
of some external disturbances, where both the reference inputs
and the external disturbances are generated by an exosystem.
The problem has been studied for both linear uncertain systems
in, say, [4], [7], [8], and nonlinear uncertain systems in, say,
[3], [14], [15], [16]. In this note, we will further study the
global robust output regulation problem for nonlinear systems
in normal form with unity relative degree by the output-based
event-triggered control law.
In practice, a control law is often implemented in a dig-
ital platform. One approach is the traditional sampled-data
implementation [2], [9], where the control actuation updates
periodically even after the system has achieved the control
goal with sufficient accuracy. In contrast, the event-triggered
control approach generates the samplings and control actuation
when the system state or output deviates more than a certain
threshold from an acceptable value [11]. Thus the event-
triggered control approach is more efficient in utilizing the lim-
ited control task execution capabilities and energy resources.
An important issue with the event-triggered control is to
guarantee the existence of a positive dwelling time to prevent
the so-called Zeno behavior, i.e., the execution times become
arbitrarily close and result in an accumulation point [25]. The
This work has been supported by the Research Grants Council of the Hong
Kong Special Administration Region under grant No. 14200515.
Wei Liu and Jie Huang are with the Department of Mechanical and Automa-
tion Engineering, The Chinese University of Hong Kong, Shatin, New Territo-
ries, Hong Kong. E-mail: wliu@mae.cuhk.edu.hk, jhuang@mae.cuhk.edu.hk
Corresponding author: Jie Huang.
The main result of this paper without any proof will be presented at the
20th World Congress of the International Federation of Automatic Control,
July 9-14, 2017, Toulouse, France.
event-triggered control problems have been first studied for
linear systems [6], [11], [19], [26]. Specifically, reference [11]
gave an introduction to the event-triggered control and studied
the stabilization problem for a class of linear systems by
a state-feedback event-triggered control law. Reference [26]
further studied the stabilization problem for a class of LTI
systems by an output-based event-triggered control law. In [6],
an output-based event-triggered control law was proposed to
guarantee the closed-loop stability and the L
∞
-performance
for a class of linear systems. Reference [19] further studied the
robust practical output regulation problem for a class of linear
uncertain minimum-phase systems by an output-based event-
triggered control law. Recently, the event-triggered control
problems have also been studied for various nonlinear sys-
tems. For example, under the assumption of the input-to-state
stability with respect to the measurement error, reference [25]
proposed a state-based event-triggered control law to solve
the stabilization problem for a class of nonlinear systems.
In [18], by using the cyclic small gain theorem, the robust
stabilization problem for a class of nonlinear systems subject
to external disturbances was solved by a state-based event-
triggered control law. Reference [27] studied the asymptotic
tracking problem for a class of nonlinear systems by a state-
based event-triggered control law and it was shown that the
tracking error was uniformly ultimately bounded. Reference
[5] analyzed the stability and L
p
-performance for a class of
nonlinear systems based on an output feedback event-triggered
control law. In [1], an output-based event triggered control law
was proposed to solve the stabilization problem for nonlinear
systems. Some other efforts can be found in [10], [23], [28],
[31].
So far, the event-triggered control problems for nonlinear
systems are mainly limited to stabilization problem [1], [10],
[18], [23], [25], or tracking problem [27], [28]. The event-
triggered global robust output regulation problem for nonlinear
systems has never been studied before. Compared with those
existing results, our problem is more challenging in the fol-
lowing three ways. First, we need to address both the tracking
problem and disturbance rejection problem simultaneously.
Second, we need to handle uncertain parameters that belong
to some arbitrarily large prescribed compact set. Third, our
control law is a dynamic output feedback control law. To
implement such a control law in a digital platform, we need
to seek a specific form of control law and sample not only
the output of the plant but also the state of the dynamic
compensator. Thus, the stability analysis of the closed-loop
system is more challenging than the static state or static output
feedback case. To overcome these challenges, we combine the