SCARA robot path planning based on the adaptive goal-biased heuristic-guided probabilistic RRT algorithm

To address the challenges of high-dimensional computational complexity, path safety, and planning stability in SCARA robot trajectory planning, this paper proposes an Adaptive Goal-biased Heuristic-guided Probabilistic RRT (AGHP-RRT) algorithm based on spatial grid indexing and a gravity-guided mechanism. Building upon the traditional RRT framework, this method introduces a Gravity-guided Heuristic Point (GHP) mechanism. By incorporating the directional guidance between the goal and the current sampling point, it enables more goal-oriented expansion and reduces randomness in path generation. An adaptive goal-bias probability model is also designed, allowing the sampling strategy to flexibly adapt to varying environmental complexities and significantly improve search efficiency. During the search process, the proposed spatial grid indexing acceleration structure partitions the high-dimensional space into uniformly spaced grids, which accelerates the retrieval of neighboring nodes and substantially reduces search redundancy. Experimental results show that the proposed algorithm significantly reduces path length, improves path smoothness, and demonstrates superior stability and repeatability, offering a more efficient, safe, and robust path planning solution for SCARA robotic arms.