An Improved RRT* Algorithm for Mobile Robot Path Planning
Junxiao Wen, Yi Zhang, Jie Tian, Qian Chen · 2025
The RRT* algorithm plays an important role in the path planning of mobile robots due to its path optimization ability. However, RRT* still has problems such as low sampling efficiency, insufficient safety and unsmooth path in narrow and complex environments, Inspired by the Fast-RRT principle, an improved GEO-RRT* algorithm based on RRT* is proposed. Firstly, the sampling efficiency is improved by the dynamic hybrid sampling strategy. Secondly, the path safety is enhanced by combining the expansion based on obstacle information. Finally, a path smoothing approach that integrates intermediate point interpolation with cubic B-spline curves is proposed, enhancing both the continuity and practicality of the path. Simulations were conducted across three distinct scenarios, and the results demonstrated that the enhanced algorithm outperformed traditional RRT* and Fast-RRT methods, notably in terms of search efficiency and path quality.