Adaptive command filter control for switched time-delay systems with dead-zone based on an exact disturbance observer

The focus of this article lies in the adaptive command filter tracking control problems for a class of switched time-delay systems with an asymmetric dead-zone and external disturbance under arbitrary switching. First, the external disturbance is estimated via an exact disturbance observer. Then, an echo state network (ESN) was adopted to approximate unknown function in system without tuning the weights between a reservoir and an input layer. Second, the time-varying input of the dead-zone model is considered as uncertain physical quantity of the system based on the dead-zone slope boundary information. Third, an adaptive command filtered controller is employed to conquer the matter of "explosion of complexity" encountered in the traditional backstepping method. The filtering errors present in a dynamic surface control (DSC) approach are effectively eliminated by means of error compensation signals. A prescribed performance control (PPC) is resorted to confine the system output to meet prescribed steady-state and transient tracking performances. The time delay terms are compensated by a Lyapunov-Krasovskii functional. The semi-globally ultimately uniformly boundedness of all states in the closed-loop system is ensured by the Liapunov's stability criterion. Finally, the validity of the developed control scheme is further verified through a simulation example.