基于3D视觉的散乱堆放工件机器人抓取技术研究

Research on vision-based robot's grabbing attitude for scattered stacking workpieces

  • 摘要: 自动化生产线上工件的上下料,尤其是工件处于散乱堆放的情形下,其自动上料一直是一个尚未很好解决的难题。利用机器视觉二维、三维测量及信息融合的方法,采用双目立体视觉的技术,对散乱堆放状态下的工件进行二维、三维测量,获取堆放状态下工件的三维点云数据;再通过对整个三维点云数据的分割,获得属于注意力集中条件下单个工件可见部分的三维数据;最后通过与已建立的工件CAD模型离散得到的单个工件的三维数据,进行三维点云的匹配,确定目标工件在散乱堆放中的三维位姿,并自动生成控制六关节机器人执行抓取的运动轨迹与动作程序,从而完成工件的分拣工作。

     

    Abstract: The loading and unloading of the workpiece on the automatic production line is one of the problems that need to be solved urgently in the industrial production, especially when the workpiece is in a disorderly stacking situation. The two-dimensional and three-dimensional measurement and information fusion methods of machine vision are used, and the binocular stereoscopic vision technology is adopted to carry out two-dimensional and three-dimensional measurement of the workpiece in the scattered stacking state, so as to obtain the three-dimensional point cloud data of the workpiece in the stacking state. Then through the segmentation of the entire three-dimensional point cloud data, the three-dimensional data belonging to the visible part of a single workpiece under the condition of concentration can be obtained. Finally, through the three-dimensional data of a single workpiece obtained discretely by the CAD model of the established workpiece, the three-dimensional point cloud is matched to determine the three-dimensional pose of the target workpiece in the disorderly stacking, and the motion track and action program of the six-joint robot is automatically generated to control the grasping, so as to complete the sorting of the workpiece.

     

/

返回文章
返回