Fixed-time adaptive fault-tolerant tracking control for uncertain strict-feedback nonlinear systems via command filtered backstepping

In this article, the fault-tolerant tracking control is addressed for uncertain strict-feedback nonlinear systems with actuator faults. Neural networks are utilized to identify unknown dynamics in strict-feedback nonlinear systems, and the adaptive technique is employed to estimate the parameter of actuator effectiveness. More importantly, a command filtered backstepping control method is improved by introducing a fixed-time command filter and modifying virtual control laws with compensation mechanisms. By incorporating the adaptive neural networks into the command filtered backstepping design framework, a novel adaptive fault-tolerant control law is constructed. Under the presented control law, the negative influence of the actuator fault and unknown dynamics is effectively compensated simultaneously. Besides, the "explosion of complexity" and "singularity" problems of backstepping is avoided. Moreover, the practical fixed-time stability is guaranteed for the resulted closed-loop system.