基于引导扩展的快速随机搜索树算法

Rapidly-Exploring Random Trees Algorithm with Guided Extension

针对移动机器人在路径规划过程中,快速扩展随机树(RRT)算法随机性强、搜索无偏向性以及在较窄出口环境下搜索效率明显下降等问题,提出一种基于障碍物有效顶点引导扩展的改进算法。该算法通过对碰撞障碍物分析,选取障碍物有效顶点引导随机树扩展,从而提高随机树的搜索效率。最后,在机器人操作系统(ROS)上进行仿真实验,使用ROS可视化工具(RVIZ)显示规划结果,结果显示基于障碍物有效顶点引导扩展的算法性能更优。

In order to solve the problems such as strong randomness,non-bias of search and obvious decrease of search efficiency of Rapidly-explor⁃ing Random Tree(RRT)algorithm in the path planning process of mobile robot,an improved algorithm based on obstacle effective vertex guidance expansion is proposed.By analyzing the collision obstacles,the proposed algorithm selects the effective vertices of the obstacles to guide the random tree expansion,so as to improve the search efficiency of the random tree.Finally,the simulation experiment is carried out on the Robots Operating System(ROS),and the 3D visualization tool for ROS(RVIZ)is used to display the path planning results.The experimental results show that the proposed algorithm based on obstacle effective vertex guidance extension has achieved better perfor⁃mance.