Virtual Sliding Spline Control for Obstacle Avoidance in Leader-Follower Unmanned Surface Vehicle Formations

A novel virtual sliding spline framework is proposed for leader-follower Unmanned Surface Vehicle (USV) formations with collision avoidance in unknown environments. The method generates USV-mounted virtual splines that self-adjust using obstacle detection data. Adaptive heading and lateral adjust-ments via spline centroids produce collision-free paths, with LiDAR measurements enabling real-time handling of irregular obstacles in maritime environments. A backstepping technology baesd control law ensures precise path tracking in unstructured settings after trajectory optimization.