摘要
针对总装生产线精密装配现状,项目设计开发了一种机器视觉定位引导系统,主要实现工业机器人在运动过程中完成在线采集不规则工件图像、视觉定位、角度补偿及精密装配作业。该系统首先采集实时工件图像,利用改进优化的九点标定算法对机器人进行视觉标定可以提高算法的精度;通过预处理操作获取质量较好的图像,按顺序创建卡尺工具,采用加权最小二乘法拟合工件亚像素边缘,再根据相关二维测量算法获取工件中心坐标。最终将得到的位姿数据传送给6轴机器人进行精密装配任务。实验结果表明,该视觉定位引导系统具有现场环境下的高精度定位检测功能,自动化装配定位精度在X方向最大定位误差为0.0281mm、Y方向最大定位误差为0.026mm,足以达到自动化装配生产线的技术要求,具有较好的应用前景和参考价值。
Aiming at the present situation of precision assembly of assembly line,this project designs and develops a machine vision guiding positioning system,which mainly realizes the online acquisition of irregular workpiece images,visual positioning,Angle compensation and precision assembly during the movement of industrial robots.This system firstly collects online workpiece images,and visual calibration of the robot using the improved and optimized nine-point calibration algorithm can improve the accuracy of the algorithm;Obtain a good quality image through preprocessing operation,then create caliper tools in order,fit the sub-pixel edge of the workpiece by weighted least squares method,and next obtain the workpiece center coordinate according to the relevant two-dimensional measurement algorithm.Finally,transmit the obtained pose data to six-axis robot for precise assembly tasks.The experiment shows that the visual positioning system has the function of high-precision location detection in the field environment,and The positioning accuracy of the automated assembly is 0.0281mm in the X direction and 0.026mm in the Y direction,which is enough to meet the technical requirements of the automated assembly line,and has a good application prospect and reference value.
作者
卢军
孙姝丽
宋岳秦
LU Jun;SUN Shu-li;SONG Yue-qin(College of Mechanical and Electrical Engineering,Shaanxi University of Science&Technology,Xi′an 710021,China)
出处
《组合机床与自动化加工技术》
北大核心
2020年第2期111-114,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
陕西省科技厅自然科学基金项目(2016GY-049)
关键词
精密装配
不规则形状
视觉定位
亚像素边缘
二维测量
precision assembly
irregular shape
visual positioning
sub-pixel edge
2D-Measurement