Path Planning of a Mobile Robot Based on the Improved RRT Algorithm

Abstract

RRT (rapidly exploring random tree) is a sampling-based planning algorithm that has been widely used due to its simple structure and fast speed. However, the RRT algorithm has several issues such as low planning efficiency, high randomness, and poor path quality. To address these issues, this paper proposes a novel method, the adjustable probability and sampling area and the Dijkstra optimization-based RRT algorithm (APSD-RRT), which consists of the following two modules: an APS-RRT planner and an optimizer. The APS-RRT planner can reach a feasible path quickly using the proposed adjustable probability and sampling area strategies, while the optimizer applies the Dijkstra algorithm to prune and improve the initial path generated by the APS-RRT planner and smooths-out sharp nodes based on the interpolation method. A series of experiments are conducted to demonstrate that our method can perform much better in terms of the balance between the computing cost and performance.