Event-triggered-based Adaptive Output Feedback Control with
Prescribed Performance for Strict-feedback Nonlinear Systems
Lijie Wang
1
and C. L. Philip Chen
2
Abstract— This paper mainly investigates the problem of
event-triggered output feedback control for a class of strict
feedback systems that satisfies a prescribed performance. With
the help of introducing the prescribed performance functions,
the constraint conditions on the tracking error can be ob-
tained, which can converge to an arbitrarily small residual set,
meanwhile the maximum overshoot is less than a pre-specified
value. For the case of unmeasurable state, a state observer is
used as approximation model to estimate them. Moreover, in
order to reduce the computation burden, the event-triggering
mechanism is successfully combined with backstepping . To
guarantee the prescribed performance for tracking error and
the boundedness property for all signals in the closed loop,
an effective event-triggered adaptive output feedback control
method is developed based on the relative threshold strategy,
which needs to update only one parameter online. In simulation,
the comparative results are provided to illustrate the validity
of the control scheme.
I. INTRODUCTION
The recent years, interest in the study of nonlinear systems
with unstructured uncertainties has been ever increasing.
To deal with the unstructured uncertainties, the approxima-
tion structures, such as fuzzy logic systems [1] and neural
networks [2], [3], have been combined with backstepping
control [4], [5]. Moreover, extensive results of adaptive
fuzzy/neural network control have been reported for nonlin-
ear systems as described in [6]–[11] and references therein.
In the works of [8] and [9], the authors respectively proposed
an efficient adaptive control scheme for constrained nonlinear
systems being subject to free time-delay and time-delay. In
contrast with the state feedback control schemes [8] and
[9], the authors in [10], [11] addressed the problem of
unmeasurable states for nonstrict feedback systems by means
of backstepping control.
In order to handle the problem of limited bandwidth for
network system and the computational burden for nonlinear
system, tool like the event-triggered control [12]–[15] gains
more attention due to its significant advantages. It can reduce
the amount of the transmissions from sensors to controllers
or from controllers to actuators, besides ensuring the system
stability. Therefore, many researchers have been interest in
such control strategy and developed some effective event-
triggered control methods [16]–[22]. To name a few, based
1
L. Wang is with the Department of Computer and Information Science,
Faculty of Science and Technology, University of Macau, Macau 99999,
China lijiewang1@gmail.com
2
C. L. Philip Chen is with School of Computer Science and Engineering,
South China University of Technology, Guangzhou 510641, China, with
Navigation College, Dalian Maritime University, Dalian 116026, China, and
also with the Faculty of Science and Technology, University of Macau,
Macau 99999, China philip.chen@ieee.org
on an assumption that the considered system is input-to-
state stable, the authors in [16], [17] investigated the event-
triggered control issue. Further, remove such assumption
for considered nonlinear systems, the authors in [18], [19]
developed two different event-triggered control schemes. In
subsequent work, the authors in [20]–[22] developed some
efficient control schemes for nonlinear system with the state
being not available for measurement.
It’s worth noting that the above mentioned results are
no considering the transient and steady-state performance
of the system. However, in practical engineering, some
certain quality performance indicators, such as overshoot,
convergence rate and steady-state error, are always required
to be satisfied. Therefore, the investigation on the problem of
performance constraint is of significance not only in theory
but also in practice. To achieve the performance constraint
for nonlinear systems, the authors in [23] firstly developed a
prescribed performance control (PPC). Subsequently, many
solutions [24]–[29] for solving various PPC problems are
proposed. For instance, In [25], the authors proposed a
solution to the tracking problem for flexible joint robot
based on PPC. In [26], [27] and [28], [29], the authors
respectively addressed the prescribed performance issue for
nonlinear systems based on state feedback control and output
feedback control. Unfortunately, there is no event-triggered
output feedback results existed for strict feedback nonlinear
systems satisfying a prescribed performance.
Based on the above analysis, the tracking problem of
strict feedback system is investigated and a novel event-
triggered output feedback controller is proposed. We focus
on reducing the frequency of signal updates from controller
to actuator and the prescribed performance of tracking error
while briefly discussing the issue of disturbance. Compared
with the existing results, the main contributions of our
paper can be summarized as follows: 1) Compared with
the existing results [30], [31], the tracking error is not only
guaranteed to be bounded, but also within the bounded
residual errors. 2) Compared with the nonlinear results [32],
[33], the proposed approach can save the on-line learning
time. For the part of approximating the unknown functions
by FLSs, only one parameter needs to be adjusted on-line,
which is independent of the system order and the number of
fuzzy rule bases. 3) Compared with [26], [34] which con-
sider strict feedback systems, the proposed event-triggered
output controller can reduce the number of control input
transmissions from controller to actuator, thereby reducing
the computational burden. Moreover, the system states we
consider are unmeasured, which estimates by a state ob-
2019 IEEE International Conference on Systems, Man and Cybernetics (SMC)
Bari, Italy. October 6-9, 2019
978-1-7281-4569-3/19/$31.00 ©2019 IEEE 2927