Fuzzy Adaptive Fault-Tolerant Control of Unknown Nonlinear Systems With Time-Varying Structure

This paper explores the fault-tolerant control problem for a family of nonlinear systems in Brunovsky form. The presence of process and actuator faults leads to an unknown and time-varying system structure, such that neural networks or fuzzy logic systems cannot be utilized for approximation directly. To solve this problem, a new-type control strategy is proposed. A zero-overshoot error constraint technique is first introduced to keep the sign of a certain variable invariant, while ensuring the premise on a compact set. Then, a fuzzy logic system is adopted to approximate the bound of the unknown and time-varying nonlinear dynamics. Further a group of adaptive mechanisms are employed to compensate for the approximation error as well as actuator faults. It is proved that by applying the presented method, both the tracking performance and the boundedness of closed-loop signals can be guaranteed. Finally, simulation results are given to illustrate the established theoretical findings intuitively.