基于改进平滑A*算法的移动机器人路径规划  

Path Planning of Mobile Robot Based on Improved Smoothing A* Algorithms

作　　者：付丽霞[1] 任玉洁 张勇[1] 毛剑琳[1] FU Li-xia;REN Yu-jie;ZHANG Zhang-yong;MAO Jian-lin(Faculty of Information Engineering&Automation,Kunming University of Science and Technology,Kunming Yunnan 650500,China)

机构地区：[1]昆明理工大学信息工程与自动化学院,云南昆明650500

出　　处：《计算机仿真》2020年第8期271-276,共6页Computer Simulation

基　　金：国家自然科学基金(61163051);云南省教育厅科学研究基金(2015Y071)。

摘　　要：针对传统A*算法用于路径规划时存在路径长、转折角度大、实时性差等问题,本文提出了基于改进平滑A*算法的移动机器人路径规划方法。首先对机器人视觉传感器获取的图像进行处理,得到机器人前方视线范围内障碍物的边缘信息。然后将这些边缘信息中的像素坐标转换为世界坐标,并得到障碍物的实际坐标,从而实现障碍物的识别和定位。机器人行驶过程中,通过融合人工势场的平滑A*算法对检测到的障碍物进行动态避障。实验结果表明,所提出的方法有效避免了传统人工势场法因陷入局部极小值点而导致的路径规划失败问题,并优于D*算法,能够使机器人在移动过程中识别动态障碍物位置,并规划出一条到达目标点的安全无碰撞较优路径。In path planning,traditional A*algorithm often leads to many problems,such as long path,large angle,and low real-time performance.Therefore,the paper put forward a method of path planning of mobile robot based on improved smoothing A*algorithm.At first,the image captured by vision sensor was processed to find the edge information of the obstacle in the sight of robot.Then,the pixel coordinates of edge information were transformed into world coordinates,and the actual coordinates of obstacles were obtained,so that the obstacles can be identified and located.When the robot was moving,a smoothing A*algorithm based on artificial potential field was adopted for dynamic obstacle avoidance.Experimental results prove that the proposed algorithm effectively avoids the path planning failure caused by the local minimum points in traditional artificial potential method,so it is better than D*algorithm.In addition,this algorithm can make the robot identify the position of dynamic obstacle during the move and plan a safe path without collision to reach the target point.

关 键 词：机器视觉 人工势场法 路径规划 动态避障 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]