基于RRT算法的障碍物移动模型研究

Research on Obstacle Moving Model Based on RRT Algorithm

现有的移动模型大多基于理想状态来模拟实物的运动,而不考虑其中障碍物的存在,这往往导致模型与现实场景存在一定的差距。因此,设计了一种适用于障碍物环境下的移动模型。通过在仿真场景内设置若干圆形障碍区,在传统的随机路点移动模型(RWPM)的基础上,引入快速扩展随机树(RRT)算法来探测障碍物,找到一条通往终点的路径。首先介绍了RWPM模型的运行原理,并分析了将它运用在障碍环境下的缺陷,接着对RRT算法进行详细阐述,然后通过Mat⁃lab仿真工具,将改进后的移动模型与传统的RWPM模型应用在两种不同的障碍物场景中,分析其节点概率分布情况。结果显示,改进后的移动模型不仅能更好地适应现实中多障碍物存在的情况,还具有更优的概率分布。

Most of the existing movement models are based on the ideal state to simulate the movement of real objects without considering the existence of obstacles,which often results in a certain gap between the model and the real scene.Therefore,a mov⁃ing model suitable for obstacles is designed.By setting several circular obstacle areas in the simulation scene,based on the tradition⁃al random waypoint movement model(RWPM),a rapid extended random tree(RRT)algorithm is introduced to detect obstacles and find a path to the end point.First,the operation principle of the RWPM model is introduced,and the defects of using it in an ob⁃stacle environment are analyzed.Then,the RRT algorithm is explained in detail.And the improved mobile model and the traditional RWPM model are applied in Matlab simulation tools.In different obstacle scenarios,the node probability distribution is analyzed.The results show that the improved mobile model not only better adapts to the existence of multiple obstacles in reality,but also has a better probability distribution.