Data-driven fault-tolerant formation control for nonlinear quadrotors under multiple simultaneous actuator faults

This paper addresses the problem of data-driven fault-tolerant formation control for quadrotors with nonlinearities, unknown system parameters, and multiple actuator faults in the vehicle dynamics. A distributed fault-tolerant formation control law is developed including a distributed observer to generate the position reference for each vehicle, a fault-tolerant position control law to track the position reference, and a fault-tolerant attitude control law to regulate the attitude. Reinforcement learning approaches are implemented to update the optimal control weights in the fault-tolerant formation control law design. Stability of the proposed fault-tolerant formation control law is proven and simulation results of quadrotors under multiple actuator faults are provided to demonstrate the effectiveness of the proposed method.