基于改进动态窗口法的移动机器人避障研究  

Obstacle Avoidance of Mobile Robot Based on Improved Dynamic Window Method

作　　者：常新新 胡为 姬书得[1] 岳玉梅[1] CHANG Xin-xin;HU Wei;JI Shu-de;YUE Yu-mei(Aerospace Academy,Shenyang Aerospace University,Shenyang 110136,China;School of Automation,Shenyang Aerospace University,Shenyang 110136,China)

机构地区：[1]沈阳航空航天大学航空宇航学院,沈阳110136 [2]沈阳航空航天大学自动化学院,沈阳110136

出　　处：《组合机床与自动化加工技术》2021年第7期33-36,39,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：辽宁省科技重大专项项目(2019JH1/10100028)。

摘　　要：动态窗口法(Dynamic Window Approach,DWA)是一种移动机器人速度采样的局部避障算法,基于DWA算法的移动机器人避障存在时间长以及在密集障碍物区无法选择最佳路径的缺点。基于此,提出一种改进的DWA算法,对移动机器人避障效果进行优化。首先根据单线激光雷达观测到的障碍物的方位信息以及设计的计算准则确定较优方位角范围,然后选取出较优方位角范围内的轨迹,最后对这些轨迹进行评价,确定最优轨迹对应的执行速度。MATLAB仿真结果表明:改进的DWA避障算法不仅能够使机器人以合理的运行轨迹选择较优角度区域通过密集障碍物区,而且在一定程度上减少了需要评价的轨迹数目,运行效率有所提高。Dynamic window approach,DWA,is a local obstacle avoidance algorithm based on the velocity sampling of mobile robot.Mobile robot avoidance based on DWA algorithm has the disadvantages of long existence time and the inability to choose the best path in dense obstacle areas.Based on this,an improved DWA algorithm was proposed to optimize the obstacle avoidance effect of mobile robot.Firstly,according to the azimuth information of obstacles observed by single line Lidar and the designed calculation criteria,the better azimuth range was determined,and then the trajectory within the better azimuth range was selected.Finally,these trajectories were evaluated to determine the execution speed corresponding to the optimal trajectory.MATLAB simulation results showed that the improved DWA algorithm can not only make the robot choose the optimal angle area through the obstacle dense area with reasonable trajectory,but also reduce the number of tracks to be evaluated to a certain extent,and improved the operation efficiency.

关 键 词：移动机器人 局部避障 DWA算法 较优方位角范围 

分 类 号：TH165[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]