A Dual-Strategy Optimized JP - RRT Algorithm Based on Jump Point Sampling : Exploring Efficiency and Stability Enhancement in Robotic Path Planning

This paper addresses low-efficiency and high-randomness node sampling in RRT - Connect for path planning by proposing a quadratic optimization JP - RRT algorithm using jump point sampling. The improved algorithm optimizes the path through jump point connection and nested loop, reducing the generation of redundant nodes and shortening the path length. Two optimization strategies, namely multiple iterative jump point sampling (MI) and adaptive step size jump point sampling (ASS), are adopted to further improve performance. In the simulation tests based on ROS combined with MoveIt, the JP - RRT algorithm significantly improved the path planning efficiency compared with RRT - Connect. In complex environments, JP - RRT (MI) and JP - RRT (ASS) reduce the planning path length by 16.83 % and 27.53 % respectively; in simple environments, they reduce it by 18.46 % and 20.88 % respectively. The verification through the Gluon robot arm in the ROS simulation environment proves that this improved algorithm has higher path planning efficiency and practicability, providing a more efficient solution for the path planning of robot arms.