A robotic arm RRT* path planning algorithm based on adaptive target point sampling and path smoothing optimization

Path planning is essential for robotic arms to perform tasks in complex environments, where efficiency, feasibility, and path quality are critical. This paper proposes an improved RRT* (Rapidly-exploring Random Tree Star) algorithm to address the slow convergence and poor path quality of traditional RRT*. An adaptive target sampling strategy dynamically adjusts the sampling probability, balancing global exploration and local convergence. Path optimization combines neighborhood search, optimal parent selection, and rewiring to reduce costs and improve quality. Quintic B-spline interpolation ensures smooth, continuous paths. The algorithm integrates inverse kinematics and joint space validity checks to meet kinematic constraints and practical execution. Simulations with a six-DOF robotic arm show the improved RRT* outperforms the traditional algorithm in path length, smoothness, and efficiency, especially in complex environments. This approach offers an effective solution for robotic arm planning in industrial automation, medical robotics, and similar fields.