基于平滑A*人工势场法的机器人动态路径规划
2018-02-01任玉洁付丽霞张勇毛剑琳
任玉洁+付丽霞+张勇+毛剑琳


摘要:针对动态环境下移动机器人的路径规划问题,提出了平滑A*人工势场法的路径规划方法。首先采用平滑A*算法在静态障碍物环境中进行全局路径规划;其次,在机器人遇到动态障碍物时采用A*人工势场法进行局部动态路径规划,并以此调整全局路径规划结果;最后对路径规划结果进行平滑处理。将平滑A*人工势场法应用于机器人动态路径规划,并与D*算法进行对比。实验结果表明,该算法能够在动态环境下规划出一条更为优化的路径,有效缩短了路径长度,提高了规划效率。
关键词:A*算法;人工势场法;路径规划;机器人
DOIDOI:10.11907/rjdk.172192
中图分类号:TP301
文献标识码:A文章编号文章编号:16727800(2018)001000803
Abstract:A path planning method based on smoothed A* artificial potential field is proposed aiming the path planning problem of mobile robot in dynamic environment, In order to improve the planning efficiency, Firstly a smoothed A* algorithm was used to perform the global path planning in the static obstacle environment;Secondly the A* artificial potential field method was applied to perform the local dynamic path planning and adjusted then the global path planning results when the robot encounters a dynamic obstacle;Finally the path planning results were smoothed.The smoothed A* artificial potential field method is applied to the robot dynamic path planning and compared with the D*algorithm. Experimental results have shown that the proposed algorithm can be worked out a more optimized path in dynamic environment, effectively shorten the path length, improve the planning efficiency.
Key Words:A* algorithm; artificial potential field method; path planning; robot
0引言
路径规划是移动机器人控制领域的一个重要研究方向,它是指依据某些指标在一个障碍物环境下搜寻一条从起始点到目标点的最优或近似最优无碰撞路径[12]。依据环境信息,大体上可将路径规划分为环境信息已知的全局路径规划和基于传感器感知的局部路径规划[3]。其中,全局路径规划能处理障碍物信息已知情况下的路径规划,但当出现动态障碍物时,规划效果不理想;局部路径规划能够通过传感器反馈的信息,规避环境中的动态障碍物。
A*算法是一种经典的启发式搜索算法,适用环境已知的全局路径规划,它具备最优性、完备性和高效性[4],利用A*算法可以得到较好的路径规划效果。但是由于A*算法的计算特点,规划出的路径点一般存在折线多、累计转折角度大等问题。……
