Path planning of RRT* algorithm with subregional dynamic probabilistic sampling based on artificial potential field in radiation environments
Yanjun Wang, Jinjia Cao, Xiaochang Zheng, Yulong Zhang, Yadong Zhang, Wei Chen · Nuclear Engineering and Technology · 2025
An improved path planning method, ASD-RRT*, is proposed to address the path planning problem in complex static radiation environments. First, the artificial potential field (APF) method is introduced to optimize the RRT* algorithm, guiding the RRT* tree nodes to shift away from the radiation sources. Next, a subregional dynamic probabilistic sampling strategy is employed, improving the goal-directedness of the path search process while considering the effects of radiation dose. Finally, the Douglas-Peucker algorithm is used to smooth the path. This study builds radiation fields using Geant4, simulates multiple radiation scenarios, and performs path planning using A*, PRM, APF-PRM, RRT*, and ASD-RRT* algorithms. The research results indicate that the ASD-RRT* algorithm performs better in navigating narrow regions and excels in complex radiation environments. It can find a safer path with the lowest cumulative dose within a reasonable time, providing a reference solution for path planning in radiation field environments.