An improved Predefined-Time Neuroadaptive Control for Maritime Quadrotor UAV Autonomous Shipboard Landing*

The collaborative system including unmanned aerial vehicle (UAV) and unmanned surface vehicle (USV) has been perceived as a significant role in marine exploration and monitoring. This paper proposes an adaptive predefined-time control scheme for quadrotor autonomous shipboard landing (ASL) on a moving USV. By introducing neural networks (NNs) into a command-filtered backstepping framework, the developed algorithm achieves both translational and rotational tracking control while estimating unknown system dynamics. Theoretical analysis demonstrates that the closed-loop system is practical predefined-time stable (PPTS), with all tracking errors converging to near-zero values within a preset time. Numerical simulations are carried out to confirm the performance of the proposed control scheme.