位置不确定下移动机器人自主抓取方法

Autonomous Grasping Method of Mobile Robot with Uncertain Position

往返取物递送移动机器人在入位定点抓取物体时,自身重复定位的不确定性将影响抓取准确性。为此提出一种克服位置不确定性的移动机器人物体抓取新方法。通过机器人体外定点位置深度视觉传感器与机器人手臂系统的关系标定,基于机器人本体激光传感器数据采用迭代最近点算法补偿机器人位置偏差,基于顶抓策略简化描述抓取位姿。通过待抓物识别与定位、确定抓取姿态以及机械臂运动规划等过程,实现了对平整支撑面上形状规则物体的自主抓取。在移动机器人往返取物作业场景下,实验验证了该方法可以显著提升物体抓取的成功率。

When a mobile robot repeatedly executes object pick and delivery tasks,its repeated-localization uncertainty affects the object grasping accuracy.A novel object grasping method was proposed for mobile robot that contain positional uncertainty.Based on the relationship between the depth vision sensor of the robot's external fixed point position and the robot arm system,based on the data of the robot body laser sensor,the iterative nearest point algorithm was used to compensate the robot position deviation,a top-grasp strategy was employed to reduce the complexity of representing the grasp pose.By conducting a series of procedures that contains object segmentation,extracting grasp pose features and motion planning,automatic grasping of regular shaped objects were achieved.In the mobile robot round-trip fetching operation scenario,the results show that the method can significantly improve the success rate of object grabbing.