密集障碍环境下机器人路径智能搜索方法仿真

摘要

It needs to make intelligent path search of robot under the serried obstacle environment in order to improve the environmental adaptation and efficient dynamic collision avoidance ability of robot path planning better.However,current method cannot work out the shortest path under the serried obstacle environment during the intelligent path searching.It leads to big error of path searching.Thus,we propose an intelligent path search method of robot under the serried obstacle environment based on the leapfrogging.Firstly,the serried dynamic value of obstacle integrated with the leapfrogging method is set,the grid method is used to acquire the connection length between two adjacent grids,and the selectable grid set is obtained.Then,the path length is taken as the fitness function of intelligent path searching based on the particle swarm theory.Moreover,the path smoothness is introduced and the range limit of intelligent path searching central zone is defined as the middle position of maximum range away from the individual extremum and global extremum found by the individual particle and global particle.Finally,the intelligent robot path searching under the serried obstacle environment is completed.The simulation results show that the method has high precision of path searching.%为了更好的提升机器人路径规划的环境适应和高效动态避碰的能力,需要进行密集障碍环境下机器人路径智能搜索.但是采用当前方法进行机器人路径智能搜索时,无法计算出密集障碍环境下机器人的最短路径,存在路径搜索误差大的问题.为解决上述问题,提出一种基于蛙跳的密集障碍环境下机器人路径智能搜索方法.上述方法先融合于蛙跳方法设定障碍物的密集动态值,利用栅格法获取两个相邻栅格间的连接长度,给出可选栅格集,利用粒子群思想理论将路径长度作为对路径智能搜索的适应度函数,引入路径平滑度,将机器人路径智能搜索中心区域的范围界限定义为个体粒子和全局粒子所发现的距个体极值和全局极值最大距离的中间位置,以此为依据完成对密集障碍环境下机器人路径智能搜索.实验结果表明,所提方法路径搜索精度高,可以为开发智能机器人导航功能奠定基础.