摘要
针对快速拓展随机树算法(rapidly-exploring random trees,RRT)存在采样随机、重复搜索、偏离目标点和节点冗余等问题,提出一种强化快速拓展随机树算法(intensity-guide rapidly-exploring random trees,IG-RRT)。采用覆盖剔除机制强化算法搜索能力,将已搜索区域进行覆盖,覆盖后不再进行搜索和产生新节点,避免重复搜索,提高搜索能力和搜索效率。后续加入目标引导概率,根据地图难度对目标引导概率进行调整,强化算法目标趋向性,对末端节点采用贪婪思想,强化算法收敛性。通过简化路径,去除冗余点,利用三次B样条曲线平滑拐点,提高路径质量。仿真试验表明,IG-RRT算法性能优于传统RRT算法及其相关衍生算法。IG-RRT算法可以增强对复杂约束空间的搜索能力,加快算法的收敛速度,提高路径规划的成功率。
Aiming at the problems of random sampling,repeated search,deviation from target points and node redundancy in rapidly-exploring random trees(RRT) algorithm,an intensity-guide rapidly-exploring random trees(IG-RRT) algorithm was proposed.The coverage elimination mechanism was used to intensify the search ability of the algorithm and cover the searched area.No search and new nodes were generated after coverage to avoid repeated search and improve search ability and search efficiency.The target guidance probability that was adjusted according to the map difficulty was added to intensify the target tendency of the algorithm.The greedy idea was added at the end nodes to intensify the convergence of the algorithm.Simplifying the path,removing redundant points,and cubic B-spline curve were used to smooth the inflection points and improve the path quality.Simulation result showed that the performance of the IG-RRT algorithm was better than the traditional RRT algorithm and its related derivative algorithms.The IG-RRT algorithm could enhance the search ability of complex constraint space,speed up the convergence speed of the algorithm and improve the success rate of path planning.
作者
王雨
刘延俊
贾华
薛钢
WANG Yu;LIU Yanjun;JIA Hua;XUE Gang(School of Mechanical Engineering,Shandong University,Jinan 250061,Shandong,China;Key Laboratory of High-Efficiency and Clean Mechanical Manufacture of Ministry of Education,Shandong University,Jinan 250061,Shandong,China;National Demonstration Center for Experimental Mechanical Engineering Education,Shandong University,Jinan 250061,Shandong,China;Institute of Marine Science and Technology,Shandong University,Qingdao 266237,Shandong,China)
出处
《山东大学学报（工学版）》
CAS
CSCD
北大核心
2022年第6期123-130,138,共9页
Journal of Shandong University（Engineering Science）
基金
国家自然科学基金项目(52001186)
山东省自然科学基金项目(ZR2020QE292)。
关键词
路径规划
RRT算法
机械臂
趋势强化
路径平滑
path planning
RRT algorithm
mechanical arm
intensify tend
path smoothing
