机器人行驶路径分层规划系统设计

Design of Hierarchical Path Planning System for Driving of Robot

对轮式机器人自主行驶路径规划系统进行分析,采用分层路径规划方法,搭建了由全局规划、局部规划、路径跟踪组成的三层路径规划架构。全局规划采用栅格法对已知地图进行分割,采用搜索效率高的A*算法搜索出最短无碰撞路径;局部规划是在行驶过程中,当局部范围内有静态或动态障碍物出现时,进行局部路径的重新规划,结合机器人动力学、运动学等约束条件,将RRT算法用在局部路径规划中;路径跟踪位于控制层,控制机器人行驶速度和方向,使其无差跟踪规划好的期望速度和方向。通过仿真实验和实车测试,验证了该系统的有效性和工程实用性。

The autonomous path planning system of wheel robot was analyzed.Based on the method of hierarchical path planning,the three-layer path planning system architecture was built,which was composed of global planning,local planning and path tracking.The grid method was used to divide the known map in global planning,and the efficient A*algorithm was used to search the shortest collision free path.Local planning was used to re-plan the local path when there were new static or dynamic obstacles in the local area.The RRT algorithm was applied to local path planning with the constraints of robot dynamics and kinematics.Path tracking was located in the control layer,which controlled the speed and direction of the robot so that it can track the planned desired speed and direction without error.The simulation experiment and real vehicle test show that the system has a good effectiveness and engineering practicability.