Resilient flight control for unmanned aerial vehicle based on neural network and disturbance observer

In this paper, a resilient flight control scheme is proposed for the unmanned aerial vehicle (UAV) with system uncertainties, external disturbances, actuator faults and controller gain perturbations. Radial basis function neural network (RBFNN) is employed to tackle actuator faults and system uncertainties. The nonlinear disturbance observer (NDO) is used to estimate the approximation errors of RBFNN and the external disturbances. By using RBFNN and NDO, the resilient flight control scheme is developed for the UAV attitude system based on the command filtered backstepping control method. An optimization method using the linear matrix inequality is studied for the feedback gain matrix design under gain pertubations. Under the resilient flight control scheme, the uniformly ultimate bounded convergence of all closed-loop signals is guaranteed via Lyapunov analysis. Simulation results show the effectiveness of the proposed resilient flight control scheme.