Adaptive Finite Time Control Of Nonlinear Switched System Based On Neural Network

This paper considers the stability of nonlinear switched systems under adaptive finite-time theory.First of all,the approximation theory of neural networks as an effective approximation tool,we use it to approximate nonlinear systems with unknown structures.Secondly,in order to realize the control of complex system,we construct appropriate Lyapunov function,and realize the control target by combining the backstepping technology.The main contributions of this paper include the following two aspects:1.Neural networks-based adaptive finite-time control of switched nonlinear systems under time-varying actuator failuresThis part is distributed in the third chapter of this paper.It mainly studies the stability of a class of nonlinear switched systems with time-varying actuator faults.In this part,we introduce RBF neural network to approximate the unknown nonlinear term.Compared with the actuator failure designed before,the actuator failure designed by us is time-varying,that is,the system is not based on healthy driving.Then,the adaptive controller of the nonlinear switched system is designed by the backstepping technique to ensure the semi-global real finite-time stability of the system.2.Adaptive finite-time tracking control for nonlinear switched systems with time-varying state constraintsThis part is distributed in the fourth chapter of this paper.It mainly studies the problem of nonlinear switched systems with full time-varying state constraints.Firstly,the asymmetric time-varying barrier Lyapunov function is introduced to ensure that the time-varying state of the system does not violate the constraint region.Then using the finite-time stability theory and adaptive backstepping technology,the adaptive tracking controller is designed to ensure the stability of the nonlinear system.