Event-triggered adaptive fixed-time NN control for constrained nonstrict-feedback nonlinear systems with prescribed performance

Abstract

In this paper, an observer-based adaptive fixed-time control strategy is put forward for uncertain nonstrict-feedback nonlinear systems subject to prescribed performance, full-state constraints and event-triggered mechanism. The difficulty of control design is to use the state observer to estimate unmeasurable states for nonstrict-feedback form in the fixed-time convergence setting. Neural networks are implemented to model the unknown nonlinearities of system. Via introducing fixed-time theory and asymmetric barrier Lyapunov function (ABLF), the fixed-time control problem of the full-state contrained nonlinear system with prescribed performance is solved. Meanwhile, the problem of “explosion of complexity” caused by backstepping technique is averted by utilizing the dynamic surface control technique. Furthermore, an event-triggered controller is devised, which can significantly save communication resources. Moreover, it is concluded that all signals involved are bounded, full-state constraints are not transgressed, tracking error remains within a prescribed domain and Zeno phenomenon is completely circumvented. Finally, the effectiveness of the proposed algorithm is verified by some simulation results.