Docking method for vehicle transfer robots with tracking virtual targets under non-preset precise target pose conditions

The increasing demand for automation in roll-on/roll-off (Ro-Ro) logistics terminals underscores the critical importance of vehicle transfer robots for efficient finished vehicle handlind. The inherent challenges of achieving precise and robust docking between a robot and a target vehicle, particularly under dynamic conditions, necessitate the implementation of advanced control strategies. The proposed method divides the docking operation into the approach and rendezvous phases, a virtual target is constructed to transform the traditionally docking problem into a dynamic tracking task by leveraging 3D and 2D LiDAR sensors for continuous target pose estimation. To mitigate this, a condition-based target pose update mechanism is proposed, which selectively triggers updates only when significant changes occur, thereby reducing unnecessary replanning while maintaining tracking accuracy. Subsequently, we propose a coupled motion planning and tracking control strategy that precisely generates the virtual target’s trajectory and ensures robust tracking, thereby guiding the robot to safely dock with the target vehicle. Extensive evaluations in both simulated environments and real-world field tests demonstrate the superior performance of the proposed method over conventional algorithms. This research significantly advances the field of logistics robotics by providing a robust docking paradigm for the autonomous transportation of large-scale finished vehicles.