基于双目视觉的钻杆接头定位系统设计

Design of Pipe Joint Positioning System Based on Binocular Vision

文章针对某地质钻杆摩擦焊接工作站钻杆接头自动上下料需求,在不改变原有工件存料方式的基础上,设计基于双目视觉的钻杆接头定位系统。利用大恒MER-500系列摄像机采集图像信息,并将其上传到工控机进行图像预处理及边缘检测;然后采用最小二乘法拟合椭圆,确定工件圆心坐标位置,并对圆心点进行排序;最终将工件坐标信息传给ABB机器人,实现工件的视觉定位及抓取。结果表明,利用最小二乘法拟合椭圆能够满足工件定位精度的要求。该定位系统可为类似工件的机器视觉定位提供借鉴。

In this paper,aiming at the requirement of automatic loading and unloading of drill pipe joints in a geological drill pipe friction welding workstation,a drill pipe joint positioning system based on binocular vision is designed without changing the original material storage mode.Firstly,the image information is collected by Daheng MER-500 series cameras,and then uploaded to industrial computer for image preprocessing and edge detection.After that the ellipse is fitted by the least square method to determine the coordinate position of the workpiece's center,and the center points are sorted.Finally,the coordinate information of the workpiece is transmitted to the ABB robot to realize the visual positioning and grasping of the workpiece.The results show that ellipse fitting by least squares method can meet the requirements of workpiece positioning accuracy,and the system can provide reference for similar workpiece machine vision positioning.