Adaptive neural-based fault-tolerant formation control for multi-UAV systems with error constraints

In this paper, the problem of fault-tolerant cooperative formation control for multi-UAV systems with error constraints is studied. A novel neural-based adaptive backstepping control scheme is developed. Firstly, a nonlinear model of fixed-wing UAVs incorporating actuator faults and complex disturbances is established. Then, a deferred error conversion is proposed to eliminate the requirement for initial error values to be within constrained regions. Moreover, a constraint transformation is designed to ensure the tracking consensus errors remain within prescribed constraint boundaries strictly. Thirdly, an adaptive neural compensator is designed to compensate for actuator faults and composite disturbances including both external disturbances and internal unknown uncertainties. Furthermore, a formation control approach based on the backstepping strategy is presented for both velocity and altitude subsystems to guarantee the uniform ultimate boundedness of all closed-loop signals. Finally, the simulation results show the effectiveness of the proposed scheme.