Adaptive Neural Control For A Class Of Nonstrictfeedback Nonlinear Systems

Adaptive nonlinear control as an important part of nonlinear control theory has been widely concerned by researchers in the past decades. Backstepping-based adaptive neural control method is especially effective in dealing with uncertain systems, and many research results have been obtained. However, adaptive backstepping design method has a drawback that nonlinear system must satisfy the structure of strict feedback. Most of the existing results are confined to this restriction. For the field of nonstrict-feedback nonlinear systems, it still needs further study. This paper mainly studies the control problems of a class of nonstrict-feedback nonlinear systems.The tracking control problem of a class of nonstrict-feedback nonlinear systems is first investigated in this paper. Due to unknown and unavailable system state, a novel input-driven filter is firstly introduced. Based on the Lyapunov function and backstepping method, the controller with adaptive laws is designed, obtaining the result of SGUUB(Semi-Globally Uniformly Ultimately Bounded). The result of this research breaks through the limitation of non-strict feedback structure of system and relaxes the conditions of unknown nonlinear functions, which greatly reduces conservatism.Then, the output feedback tracking control problem for a class of multi input multi output nonlinear systems with time delay is studied. The studied systems consider the factor of time delay and the problem of interconnected states in each subsystem. By using neural network approximation, adaptive control and backstepping technique, the controller is finally designed. Based on Lyapunov functional stability analysis, the closed-loop system signals are ultimately bounded. This study further solve the problem of time delay and the control problem of complexity state interconnection.