摘要
针对Informed-RRT*算法在路径规划中的采样盲目性、转折角过大和扩展新节点效率低等问题,提出了一种改进方法。该方法限制采样方向和转角,引入二次扩展策略,并有条件地选取新节点。首先,生成新节点时引入启发式信息,并根据Metropolis准则以一定概率接受新节点;其次,为满足运动学约束和有目的性搜索,将采样点限制在起点和目标点连线方向且以起点为轴左右小于最大转角的范围内,在重选父节点和重布线时也要符合最大转角约束;最后,采用二次扩展策略,第一次扩展在限定范围内采样,第二次扩展在新节点和目标点之间的椭圆区域内采样。通过多组实验表明,改进算法提高了搜索效率,降低了路径长度,使运行更稳定。
An improvement method is proposed for the Informed-RRT algorithm to address the problems of sampling blindness,excessively large turning angles and low efficiency of expanding new nodes in path planning.The method restricts the sampling direction and turning angle,introduces a quadratic expansion strategy,and selects new nodes conditionally.Firstly,heuristic information is introduced when generating new nodes,and new nodes are accepted with a certain probability according to the Metropolis criterion.Second,to satisfy the kinematic constraints and purposeful search,the sampling points are restricted to the direction of the line connecting the starting point and the target point and less than the maximum turning angle around the axis of the starting point,and the maximum turning angle constraints are also complied with when re-selecting parent nodes and rewiring.Finally,the quadratic expansion strategy is adopted,where the first expansion samples within a constrained range,and the second expansion samples within the elliptical region between the new node and the target point.It is shown through several sets of experiments that the improved algorithm improves the search efficiency,reduces the path length and makes the operation more stable.
作者
但远宏
季勇
程东
DAN Yuanhong;JI Yong;CHENG Dong(College of Computer Science and Engineering,Chongqing University of Technology,Chongqing 400054,China)
出处
《组合机床与自动化加工技术》
北大核心
2025年第5期14-19,共6页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国防科技创新特区项目资助(20-163-ZT-01)
重庆市科委重点攻关计划项目(2021CCB03)
南京理工大学重点实验室基金资助项目(2022-JCJQ-LB-061-07)。
