Abstract: This paper proposed a learning-observer-based resilient fault-tolerant control scheme for the control problem of quadrotor unmanned aerial vehicles with sensors and actuators faults. By processing the fault output signals of sensors, and based on coordinate changes, an extended system is constructed, and then a learning observer is designed using sensor fault outputs to estimate the system fault function and actual output online. Further, an adaptive neural network backstepping resilient fault-tolerant controller is designed to achieve the actual system output tracking the reference signal in the event of sensor faults and additive actuator faults in the unmanned aerial vehicle. Based on the Lyapunov stability theory, the stability of closed-loop system is proven. Finally, simulations and outdoor experiments are conducted using quadrotor unmanned aerial vehicle and the results validated the effectiveness of the proposed method.

Key words: Quadrotor unmanned aerial vehicles, Resilient fault-tolerant control, Learning observer, Sensor fault, Actuator fault