Adaptive Control For Uncertain Nonlinear Switched Systems With Dynamic Uncertainties

In the past decades,the nonlinear systems with dynamic uncertainties have attracted much attention from many researchers in control field.Although much progress has been made in the study of nonlinear non-switched systems with dynamic uncertainties,there are still many problems to be solved for nonlinear switched systems with dynamic uncertainties.Thus,this thesis studies the controller design issues of several kinds of nonlinear switching systems with dynamic uncertainties by combining the backstepping technology and adaptive neural network(NN)control technology,and also considers the convergence and stability problems of the corresponding closed-loop systems.The main work of this thesis is as follows:(1)An adaptive neural-network-based tracking control strategy is proposed for a class of nonlinear switched non-strict feedback systems with completely unknown nonlinearties and unmodeled dynamics.The design difficulties arise mainly from the fact that the intercoupling between the non-strict feedback form and the unmodeled dynamics leads the switched system under consideration has a very complex structure.In order to get the desired state feedback controller,a mild assumption associated with a dynamic signal is utilized to deal with the unmodeled dynamics,and a separating variable method is presented to handle the system nonlinearties of all state variables in the framework of adaptive neural backstepping technique,respectively.The obtained result shows that all signals of the closed-loop switched system are semi-global bounded with the output tracking error can be guaranteed to enter a small region around the origin.In the end,twosimulation examples are given to demonstrate the feasibility and practicability of the presented design strategy.(2)The problem of the adaptive neural tracking control for a class of non-strict feedback nonlinear switched uncertain systems is considered.By using neural networks(NNs)to approximate the unknown smooth nonlinear functions,and employing the property of Gaussian functions,the design problem caused by the non-strict feedback form is solved.To avoid the situation where a common Lyapunov function is needed for all subsystems while the backstepping method is adopted,the multiple Lyapunov functions are introduced.Based on the developed scheme,it can be obtained that all signals remain bounded in the closed-loop switched system and the output tracking error can approximate a small region near the origin.Finally,two simulation examples are taken to prove the practicability and feasibility of the presented design scheme.In brief,although the adaptive control design of nonlinear switched systems with dynamic uncertainties is studied in this thesis,the study of nonlinear switched systems with dynamic uncertainties are still preliminary.There are many relevant problems can be explored,such as design of interconnected system with unmodeled dynamics,design of stochastic systems with unmodeled dynamics,and design of pure-feedback nonlinear systems with unmodeled dynamics,and so on.