Robust Adaptive Fault-Tolerant Control for a Class of Unknown Nonlinear Systems

This paper studies the fault-tolerant tracking control problem for a class of strict-feedback nonlinear systems subjected to actuator faults and external disturbances. The prior knowledge for actuator fault, nonlinearity, and external disturbance is totally unknown, besides the control directions. Based on a backstepping approach, an adaptive fault-tolerant control scheme is developed, without utilizing neural networks. In the control design, a group of new feedback mechanisms are proposed to compensate for the unknown system dynamics and actuator faults. Furthermore, to relax a requirement of the initial system states, a modification technique is designed to adjust the reference signal and virtual control laws for a short time. It is shown that the global closed-loop stability is guaranteed and the tracking performance is achieved. The above result is illustrated via simulations on a one-link manipulator and a ship autopilot.