A multi-RRT based hierarchical path planning method

Abstract

The Rapidly-Exploring Random Tree (RRT) algorithm is an efficient approach to solve the path planning problem with nonholonomic constraint of wheeled mobile robot. However, the efficiency of the RRT algorithm will be reduced dramatically when a large amount of obstacles scatter in the environment. On the other hand, the RRT path planner is not optimized, which becomes the bottleneck of the traditional RRT algorithm. In this paper, a two-leveled hierarchical path planning method is presented to overcome the difficulties mentioned above. On the top level of the method, a multi-RRT framework is created by artificial-guided points to exploit and generate traversable area through RRT local exploration and merge function; on the bottom level, a heuristic search algorithm is adopted to effectively search for a feasible trajectory in the traversable area. The experimental results verify the effectiveness of the proposed method in solving the problem of wheeled robot path planning in complex environments.