Event-triggered robust neural control for unmanned sail-assisted vehicles subject to actuator failures

Abstract This note focus on the waypoints-based path-following control for the unmanned sail-assisted vehicles (USAV), aiming to release the constraints of the actuator failures and gain uncertainties. The proposed scheme is formulated as two components, i.e., the composite guidance part and the control part. By utilizing the sign self-selection algorithm, the composite Logic Virtual Ship (LVS) guidance law is developed in the scheme to program the real-time heading angle for the USAV. The main superiorities of this design are to ensure the USAV navigating efficiently and choose the corresponding sailing mode: upwind mode, downwind mode or crosswind mode. Furthermore, to improve the effectiveness of the closed-loop control system, an event-triggered robust neural control algorithm is targetly designed for the rudder actuator and the sail actuator by fusing the robust neural damping technique and the input event-triggered mechanism. In this algorithm, the unknown terms of the system are tackled requiring no information of the system model and the external disturbances. The transmission burden from the controller to the actuator is reduced. And the unknown actuator failures and the gain uncertainties are compensated through four adaptive updated parameters. Based on the Lyapunov analysis, sufficient effort has been made to guarantee that all the signals of the closed-loop control system are the semi-global uniform ultimate bounded (SGUUB). Finally, the simulated results demonstrate the validity of the proposed control strategy.