基于改进RRT^(*)算法的移动机器人路径规划  

Path planning of mobile robots based on improved RRT^(*) algorithm

作　　者：郭利进[1] 李强[1] GUO Lijin;LI Qiang(School of Control Science and Engineering,Tiangong University,Tianjin 300387,China)

机构地区：[1]天津工业大学控制科学与工程学院,天津300387

出　　处：《智能系统学报》2024年第5期1209-1217,共9页CAAI Transactions on Intelligent Systems

基　　金：国家自然科学基金项目(61973234).

摘　　要：针对传统快速随机搜索树^(*)(rapidly-exploring random tree^(*),RRT^(*))算法收敛速率较慢,且不适用于动态场景等问题,提出一种基于目标点偏置和冗余节点删除的改进RRT*算法,用于解决移动机器人快速找到无碰撞最优路径的问题。此算法在RRT^(*)算法基础上,首先对采样点进行优化处理,保证路径最优的同时减少搜寻时间;其次引入路径节点最大值概念,删除扩展树冗余节点以提高算法效率;最后结合动态窗口(dynamic window approaches,DWA)算法提高路径的安全性和平滑性,实现对动态障碍物的避障。通过3种不同地图下的仿真验证,改进算法能有效提升路径质量,且大幅降低运行时间。To address the problems associated with the traditional rapidly-exploring random tree^(*)(RRT^(*))algorithm,including its slow convergence rate and unsuitability for dynamic environments,we propose an improved RRT^(*)algorithm.This new approach biases toward the target point and eliminates redundant nodes to efficiently find collisionfree optimal paths for mobile robots.The improved algorithm begins by optimizing the sampling points,which ensures the discovery of the optimal path while reducing search time.Then,it introduces the concept of path maximization to identify and remove redundant nodes from the extended tree,thereby enhancing algorithm efficiency.Additionally,the improved algorithm is integrated with the dynamic window approaches algorithm,which enhances the safety and smoothness of the path and enables effective obstacle avoidance in dynamic environments.Simulations conducted on three different maps demonstrate that the improved algorithm significantly improves path quality and greatly reduces running time.

关 键 词：移动机器人 动态环境 最优路径 RRT^(*)算法 动态窗口算法 冗余节点 安全性 平滑性 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]