初始化参数对RRT算法性能影响研究

Effect of Initialization Parameters on RRT Algorithm Performance

由于采用随机采样方式进行路径规划时RRT算法无需对任务空间内的障碍物进行数学建模,因而该算法在路径规划领域得到了广泛应用,但该算法的初始化参数步长和偏置概率对路径规划结果有直接影响。为了揭示初始化参数对RRT算法性能的影响规律,为参数选取提供依据,通过计算机模拟不同复杂程度任务空间下的路径规划问题,并利用RRT算法进行路径规划实验,从算法耗时、路径长度、规划失败概率3个方面统计分析初始化参数对RRT算法性能的影响。实验结果表明,初始化参数步长和偏置概率对RRT算法性能有直接影响,且不同复杂程度任务空间需要设置不同的初始化参数。

Due to the random sample method for path planning,RRT algorithm does not need mathematical modeling of the internal obstacles in the task space,so it has been widely used in the field of path planning.However,the initial parameter step size and the sampling probability of the goal point have a direct impact on the path planning results.In order to reveal the influence of initialization parameters on the performance of RRT algorithm,and provide the basis for parameter selection,path planning problems in different complexity task space has been simulated through computer,and RRT algorithm was used to carry out path planning,and the influence of initial parameters on the planning time,path length,and failure times in the planning results,the influence of five aspects of RRT algorithm performance was analyzed.The experimental results have shown that the initialization parameters,which contained the step size and the bias probability,have a direct impact on the performance of RRT,and the initialization parameters should be set according to the complexity of the workspace.