Abstract: To solve the problems in A* algorithm path planning,such as redundant turning points,low security and inconsistency with non-holonomic kinematics model of mobile robots,a path planning algorithm based on motion constraints was proposed.The path points planned by A* algorithm were moved to ensure a proper distance from the obstacles.Then,the nodes between the accessible path points were eliminated to reduce redundant steering,thus the necessary turning points were obtained.Finally,the robot motion constraints were considered for quadratic path planning,and the turning point pose was taken as the local target to guide the hybrid A* algorithm for path planning.The results showed that the safety distance of the improved algorithm was increased by 65% by comparing with the A* algorithm,the path with pose continuity and had no redundant steering,which could ensure the smooth,efficient and safe movement of the robot.

Key words: mobile robot, path planning, A*algorithm, motion constraints, hybrid A*algorithm

摘要： 针对A*算法路径规划中存在的转折冗余、安全性低、不符合移动机器人非完整约束模型等问题,提出一种基于运动约束的路径规划算法。首先,通过挪移处理使A*算法规划的路径点适度远离障碍物。然后,剔除无障碍路径点间节点,获取路径必经转折点。最后考虑机器人运动约束,以转折点位姿为局部目标,引导混合A*二次规划路径。实验结果表明,改进算法比传统A*算法提高了65%的安全距离,无冗余转向,路径具有位姿连续性,可以保证机器人的平稳、高效、安全的移动。

关键词: 移动机器人, 路径规划, A*算法, 运动约束, 混合A*算法