一种移动机器人激光模型全局路径规划方法

A Global Path Planning Method for Mobile Robot Laser Model

为了使移动机器人在全局路径规划中更好地适应地图环境,提出了一种改进A*算法和射线模型的激光模型全局路径规划方法。该方法借鉴了平面激光雷达实时扫描原理,由路径节点向目标点散发多条虚拟激光射线来感知障碍物边界,从而快速越过障碍物并到达目标点,同时结合Floyd优化算法,在较短时间内得出一条安全可靠稳定的全局路径。在实验凹形环境中,激光模型搜索时间快于A*算法和射线模型99%,搜索过程减少96%;在实验不可行区域中,激光模型快于前两种算法99%以上,搜索过程减少97%;同时与双蚁群交叉凹形环境算法对比,表现也相对较优。实验结果表明,所提方法对解决凹形陷阱及不可行区域耗时问题比较有效。

Objective:In order to improve the adaptability of mobile robots to complex map environments,we propose a global path planning method called laser model,which improves the A*algorithm and ray model.Methods:This method draws inspiration from the real-time scanning principle of planar light detec-tion and ranging,which emits multiple virtual laser rays from the path node to the target point to perceive the boundary of obstacles,thereby quickly crossing obstacles and reaching the target point.At the same time,combined with the Floyd optimization algorithm,a safe,reliable,and stable global path is obtained in a relatively short time.Results:In the concave environment of the experiment,the search time of the la-ser model is 99%faster than that of the A*algorithm and ray model,and the search process is reduced by 96%.In the experimental infeasible region,the laser model is 99%faster than the A*algorithm and ray model,and the search process is reduced by 97%.Compared with the double ant colony crossover algo-rithm,the laser model also performs better in concave environments.Conclusions:The experimental re-sults show that the proposed method is effective in solving the time consumption problems of concave traps and infeasible areas.