摘要
为解决移动机器人在动态环境下的路径规划问题,将Informed-RRT*和人工势场法相融合,提出全局与局部规划算法相融合的路径规划方法。首先,针对Informed-RRT*算法采样效率低,以及得到路径不满足机器人运动学约束的问题,采用目标偏置法与自适应步长法,减少冗余搜索与不必要树的生长;同时,引入走廊优化与时间重分配法,优化路径节点,使路径更加平滑。其次,针对人工势场法易陷入局部极小值和目标点附近不可达的问题,采用平滑窗格策略,增设全局路径子目标点,使机器人能够逃离局部极小值,完成规划任务。仿真结果表明,静态环境中自适应步长Informed-RRT*算法相比于Informed-RRT*算法求解时间缩短了71.98%;动态环境中,混合算法相比于人工势场法,搜索时间缩短了15.4%,路径长度缩短了11.1%。
In order to solve the path planning problem of mobile robot in the dynamic environment,a hybird strategy based on the global and local algorithms is proposed by combining the Informed RRT*and artificial potential field method.Firstly,the target offset method and adaptive step size method are designed in the process of informed RRT*random point selection to reduce the redundant search and unnecessary tree growth.Corridor optimization and time redistribution method are introduced to optimize the nodes of path for making the path more smooth.Secondly,the artificial potential field method based on subtarget points is employed for the local path planning,and the problem of easy to fall into local minimum and unreachable near the target point for the artificial potential field method is solved.The sub target point is set to make sure that the mobile robot can escape the local minimum and complete the mission.The simulation results show that the time to find a path in a static environment was 71.98%shorter by using adaptive step-length Informed-RRT*algorithm compared to the Informed-RRT*algorithm.Compared to the artificial potential field method,the search time and the path length are decreased 15.4%and 11.1%,respectively,by using the bybird algorithm.
作者
郑维
王昊
王洪斌
ZHENG Wei;WANG Hao;WANG Hong-bin(Institute of Electric Engineering,Yanshan University,Qinhuangdao,Hebei 066004,China)
出处
《计量学报》
CSCD
北大核心
2023年第1期26-34,共9页
Acta Metrologica Sinica
基金
国家自然科学基金(62203379)
河北省自然科学基金(F2021203083,F2021203104)
河北省教育厅高等学校科技计划自然科学基金(QN2021138)。
作者简介
第一作者:郑维(1989-),女,河北保定人,燕山大学副教授,主要研究方向为机器人控制、计算机实时控制、自动化技术与应用、智能信息处理等。Email:weizheng@ysu.edu.cn;通讯作者:王洪斌(1965-),男,黑龙江齐齐哈尔人,燕山大学教授,博士生导师,主要从事机器视觉,机器人控制技术等方面的研究。Email:hbwang@ysu.edu.cn。
