RRT-GPMP2: A Motion Planner for Mobile Robots in Complex Maze Environments

摘要

With the development of science and technology, mobile robots are playing a significant role in the new round of world revolution. Mobile robots could serve as assistants or substitutes for humans across a wide range of applications. To enhance mobile robot automation, advanced motion planners must be integrated to handle diverse environments. Navigating complex maze environments is a key challenge for mobile robots in various practical scenarios. Therefore, this article proposes a novel hierarchical motion planner named the rapidly exploring random tree-based Gaussian process motion planner 2, which aims to tackle the motion planning problem for mobile robots in complex maze environments. Specifically, the proposed motion planner successfully combines the advantages of the trajectory optimisation motion planning method and sampling-based motion planning method. To validate the performance and practicability of the proposed motion planner, we tested it in a series of self-constructed maze simulations and applied it on a surface marine robot in a high-fidelity maritime simulation environment in the Robot operating system.