首页 | 本学科首页   官方微博 | 高级检索  
     

移动机器人基于激光点云定位建图的汽车宽度与方位估计
引用本文:薛连杰,张彪,张霄远,齐臣坤. 移动机器人基于激光点云定位建图的汽车宽度与方位估计[J]. 机械与电子, 2018, 0(6): 76-80
作者姓名:薛连杰  张彪  张霄远  齐臣坤
作者单位:(上海交通大学机械与动力工程学院 机械系统与振动国家重点实验室, 上海 200240)
摘    要:在移动机器人抓取搬运作业中,识别待操作目标物体的宽度和方位是成功的关键。针对室外环境下停放的汽车搬运作业需求,提出一种识别待搬运汽车宽度和方位的方法。该方法基于三维激光传感器,先构建整个环境的三维点云地图,然后从地图中分离出目标汽车点云,最后将汽车点云向地面投影,根据投影的点云包围盒进行汽车的宽度和方位估计。采用室外实验验证了该方法的识别效果,其精度满足应用要求。

关 键 词:移动机器人  汽车搬运  点云分割  宽度识别  方位识别

Width and Orientation Identification of Cars Based on Lidar Localization and Mapping
XUE Lianjie,ZHANG Biao,ZHANG Xiaoyuan,QI Chenkun. Width and Orientation Identification of Cars Based on Lidar Localization and Mapping[J]. Machinery & Electronics, 2018, 0(6): 76-80
Authors:XUE Lianjie  ZHANG Biao  ZHANG Xiaoyuan  QI Chenkun
Affiliation:(State Key Laboratory of Mechanical System and Vibration, School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China)
Abstract:The key for a mobile robot to successfully grasp and transport target object is accurate identification of the width and orientation of the target object.?A method of identifying the width and orientation of an object is proposed for a mobile robot to transport cars outdoors.?This method is based on a three-dimensional laser sensor and SLAM technique.?Firstly, a three-dimensional point cloud map of the car and the environment was constructed. Then, the point cloud of the car was separated from the map.?Finally, the point cloud of the car was projected to the ground, and the width and orientation of the car was determined according to the bounding box of the projected point cloud.?The experiment shows that the proposed method is effective and its accuracy meets the application requirements.
Keywords:mobile robot  car transportation  point cloud segmentation  width recognition  orientation identification
本文献已被 CNKI 等数据库收录!
点击此处可从《机械与电子》浏览原始摘要信息
点击此处可从《机械与电子》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号