Adaptive fault-tolerant control for aircraft against control surface failure diagnosed by set-membership estimation

This paper focuses on the fault diagnosis method and the fault-tolerant control method of the aircraft with control surface damage. The fault diagnosis method is based on the set-membership estimation. To diagnose the fault of control surface under the condition of data containing noise and uncertainty, a multi-index optimized fault residual observer based on sensitivity and robustness is designed for the observation of aircraft fault information. Based on the observed fault residual information, the set-membership estimation theory is used to determine whether the aircraft control surface fault occurs. The fault-tolerant control method is based on dynamic inverse control via aerodynamic parameter identification. Furthermore, to reduce the nonlinear effects of unmodeled dynamics and aerodynamic identification errors after a fault, an adaptive disturbance rejection observer is designed for control compensation. The proposed methods are evaluated by numerical simulation. Results demonstrate that the fault diagnosis method can effectively diagnose under the influence of uncertain noise and the attitude control has the robustness of different faults.