Adaptive neural optimal path following control for rotor-assisted vehicles via an improved event-triggered mechanism

Abstract

This paper proposes an adaptive neural optimal control scheme for the path following of rotor-assisted vehicles (RAVs) in the presence of external disturbances, integrated with an improved event-triggered mechanism. Without requiring the precise knowledge of the system dynamics, a critic-only adaptive dynamic programming (ADP) framework combined with the robust neural damping technique is developed, which significantly reduces the overall algorithm complexity and enhances the robustness of the system. To mitigate the issues of frequent signal updates and excessive actuator wear inherent in conventional control algorithms, a dynamic integral-event-triggered (DIET) mechanism is proposed, which automatically adjusts the triggering thresholds based on the output states. Meanwhile, owing to its integral-type structure, invalid triggering behaviors induced by the transient disturbances can be effectively eliminated. Furthermore, the semi-global uniform ultimate boundedness (SGUUB) of all signals in the closed-loop system is rigorously demonstrated via the Lyapunov theory. Finally, the effectiveness and feasibility of the proposed algorithm are validated through two numerical simulations.