应用改进RRT_Connect算法的移动机器人路径规划
Path planning of mobile robot using improved RRT_Connect algorithm
史小旭 1张烈平 1唐柳 2董路熙 1彭建盛3
作者信息
- 1. 广西高校先进制造与自动化技术重点实验室,桂林 541006
- 2. 贺州学院人工智能学院,贺州 542899
- 3. 广西高校人工智能与信息处理重点实验室,河池 546300
- 折叠
摘要
针对复杂未知环境下的移动机器人路径规划问题,提出一种优化已搜索节点和已规划部分路径的改进RRT_Connect算法.算法引入了终点和已搜索节点目标偏向策略,该策略在随机采样函数中引入终点和已搜索节点偏向概率基准值,使随机采样点按随机概率设定为终点或已搜索节点;通过筛选有效新节点和一定范围内邻节点的父节点,优化路径规划成本,使规划路径趋于平滑.仿真结果表明,提出的改进RRT_Connect算法所规划的路径在平均转弯次数、平均规划路径长度和平均规划成功率等指标较改进前更优.
Abstract
Path planning is an indispensable technology in the field of robot research.Aiming at the path planning problem of mobile robot in complex unknown environment,an improved RRT_Connect algorithm is proposed to optimize the searched nodes and planned partial paths.Firstly,the algorithm introduces bias strategy of endpoint and searched node,which introduces bias probability reference value of endpoint and searched node in random sampling function,so that random sampling point is set as endpoint or searched node according to random probability.Then,by screening the effective new nodes and the parents of neighboring nodes in a certain range,the path planning cost is optimized to make the planned path tend to be smooth.The simulation results show that the path planned by the proposed improved RRT_Connect algorithm is better than before in terms of the average number of turns,average planned path length and average planning success rate.
关键词
RRT_Connect算法/移动机器人/路径规划/目标偏向策略Key words
RRT_Connect algorithm/mobile mobot/path planning/goal biased strategy引用本文复制引用
基金项目
国家自然科学基金(61741303)
广西空间信息与测绘重点实验室项目(19-185-10-08)
出版年
2024
NETL
NSTL
万方数据