Prescribed performance adaptive neural output control for a class of switched nonstrict‐feedback nonlinear time‐delay systems: State‐dependent switching law approach

Abstract

Summary This paper investigates the tracking problem for a class of uncertain switched nonlinear delayed systems with nonstrict‐feedback form. To address this problem, by introducing a new common Lyapunov function (CLF), an adaptive neural network dynamic surface control is proposed. The state‐dependent switching rule is designed to orchestrate which subsystem is active at each time instance. In order to compensate unknown delay terms, an appropriate Lyapunov‐Krasovskii functional is considered in the constructing of the CLF. In addition, a novel switched neural network–based observer is constructed to estimate system states through the output signal. To maintain the tracking error performance within a predefined bound, a prescribed performance bound approach is employed. It is proved that by the proposed output‐feedback control, all the signals of the closed‐loop system are bounded under the switching law. Moreover, the transient and steady‐state tracking performance is guaranteed by the prescribed performance bound. Finally, the effectiveness of the proposed method is illustrated by two numerical and practical examples.

Event‐triggered adaptive finite‐time prescribed performance tracking control for uncertain nonlinear systems

“…It can be seen that the proposed AFTPF adjusts its parameters according to the amplitude of the tracking error for t < T d and does not change when t ≥ T d . That is, (4) not only converges faster than the performance functions in References 15‐26 but also reaches its steady‐state at finite time.…”

“…">1.An AFTPF, for the first time, is proposed, whose parameters are determined by initial states at t = 0 and adjusted online according to tracking errors subsequently. So it is not only more applicable but also converges faster than the PFs 15‐17,19‐26 . Besides, the proposed function has the property of finite‐time convergence, which is an improvement on the PFs 27,28 …”

“…14 During the development of control theory and applications, another useful method named prescribed performance control (PPC) was proposed 15 and has been used to design many nonlinear controllers. [16][17][18][19][20] Two adaptive controllers were developed for switched nonlinear systems 16,17 by using PPC. Using a linear observer 18 and dynamic surface, 19 two different adaptive control schemes were developed for switched nonlinear systems via PPC.…”

“…So it is not only more applicable but also converges faster than the PFs. [15][16][17][19][20][21][22][23][24][25][26] Besides, the proposed function has the property of finite-time convergence, which is an improvement on the PFs. 27,28 2.…”

Observer-based adaptive consensus control for nonlinear multi-agent systems with time-delay

“…Consider the nonlinear MASs (1) under Assumptions 1-3 and bounded initial conditions. Using the distributed consensus adaptive control laws (27), (37) and 44, NN weights updating laws (28), (29), (38) and (45), assuming that the leader has directed paths to all the followers, one has that the consensus tracking errors are CSUUB and all the signals in the closed-loop system remain bounded.…”