Dynamic RRT* with bridge guidance for robot path planning in dense and narrow passages environment
Jinze Li, Xin Zhou, Jianliang Mao, Chuan‐Lin Zhang · Journal of Control and Decision · 2025
The rapidly-exploring random trees (RRT) algorithm and its variants have been extensively studied because of their high efficiency and asymptotic optimality in path planning. Yet, they still struggle in complex environments with a dense arrangement of short and closely spaced narrow passages. In terms of the issue, this paper proposes a dynamic RRT* with bridge guidance (DBRRT*) to accelerate feasible path searching by adding a dynamic sampling method and an improved bridge test. To validate the effectiveness of DBRRT*, a series of experiments on a 6-DOF manipulator are conducted in reference to some typical path planning algorithms. Experimental results demonstrate that DBRRT* generally enhances the resolution and optimisation speed, and reduces path planning time in 2D and 3D environments, indicating its capability to address the challenges of path planning in environments with numerous short, dense and narrow passages.