作者：陆艺¹, 郭小娟¹, 郭斌², 沈添秀¹

作者单位：1. 中国计量大学计量测试工程学院, 浙江 杭州 310018;
2. 杭州沃镭智能科技股份有限公司, 浙江 杭州 310018

关键词：手眼系统;视觉定位;手眼系统标定;定位精度;不确定度

摘要：

针对自动化产品装配中的视觉定位问题，搭建一套基于手眼（eye-in-hand）系统的机器视觉装配实验平台。在Halcon软件平台上运用张正友标定法对相机内参数和外参数进行标定，并利用手眼关系方程CX=XD求解出机器人的手眼关系以完成手眼系统标定的过程，最终将手眼系统标定结果应用于视觉装配平面的目标点定位。其中以机器人定位到的目标点坐标为理想位置坐标，对视觉定位与机器人定位得到的目标点位置进行比较分析，验证该手眼系统视觉定位的准确性并对视觉系统进行不确定度评定，实验重复性良好。实验结果表明，用此方法得到的手眼系统视觉定位精度能达到0.6 mm，满足机器视觉生产线上产品装配的需求。

Research on visual positioning based on hand-eye system

LU Yi¹, GUO Xiaojuan¹, GUO Bin², SHEN Tianxiu¹

1. College of Metrology & Measurement Engineering, China Jiliang University, Hangzhou 310018, China;
2. Hangzhou Wolei Intelligent Technology Co., Ltd., Hangzhou 310018, China

Abstract: To solve the problem of visual positioning in automated product assembly, a set of machine vision assembly experiment platform based on eye-in-hand system was built. Using the Zhang Zhengyou calibration method on the Halcon software platform to calibrate the camera parameters and external parameters, and using the hand-eye relationship equation CX=XD to solve the hand-eye relationship of the robot and to complete the calibration process of the hand-eye system, and eventually hand-eye calibration results were applied to the visual target point positioning in the plane of the assembly. The coordinates of the target points where the robot was located were the ideal position coordinates, and the position of the target points obtained by the visual positioning and robot positioning were compared and analyzed to verify the accuracy of the visual positioning of the hand-eye system, and the visual system was evaluated with uncertainty, and the experiment was repeatable. The experimental results show that the visual positioning accuracy of the hand-eye system obtained by this method can reach 0.6 mm and meet the requirements of product assembly on the machine vision line.

Keywords: hand-eye system;visual positioning;hand-eye system calibration;positioning accuracy;uncertainty

2018, 44(12): 117-121  收稿日期: 2018-06-05;收到修改稿日期: 2018-08-12

基金项目: 国家自然科学基金项目（51405463）；浙江省基础公益研究计划项目（LGG18E050009）

作者简介: 陆艺(1979-),男,江苏扬州市人,副教授,硕士,研究方向为精密检测技术、汽车零部件自动化测试

参考文献

[1] TSAI R Y. An efficient and accurate camera calibration technique for 3D machine vision[C]//Proceeding of IEEE Conference of Computer Vision and Pattern Recognition, Miami, America, 1986:364-374.
[2] DE M S. A self-calibration technique for active vision systems[J]. IEEE Transactions on Robotics & Automation, 1996, 12(1):114-120
[3] WEI G Q, MA S D. Two plane camera calibration:a unified model[C]//Computer Vision and Pattern Recognition, 1991. Proceedings CVPR'91. IEEE Computer Society Conference on. IEEE, 1991:133-138.
[4] ZHANG Z. A flexible new technique for camera calibration[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2000, 22(11):1330-1334
[5] 程玉立, 熊蓉. 面向工业应用的机器人手眼标定与物体定位[D]. 杭州:浙江大学, 2016.
[6] SHAH S V, SAHA S K, DUTT J K. Denavit-hartenberg parameterization of euler angles[J]. Journal of Computational & Nonlinear Dynamics, 2012, 7(2):021006-021006-10
[7] JOHN J. 机器人学导论[M]. 3版. 北京:机械工业出版社, 2006:62-135.
[8] 周丽凤, 陈幼平. 面向三维检测的相机标定方法研究[D]. 武汉:华中科技大学, 2015.
[9] 黄敏高. 基于单目视觉的工业机器人定位系统设计[J]. 现代电子技术, 2017, 40(18):114-116
[10] 解则晓, 牟楠, 迟书凯, 等. 结构光内参数和机器人手眼关系的同时标定[J]. 激光与红外, 2017, 47(9):1142-1148
[11] 卢泉奇, 苗同升, 汪地, 等. 基于HALCON的机械手视觉抓取应用研究[J]. 光学仪器, 2014, 36(6):492-498
[12] 谢俊, 吴荥荥, 朱广韬, 等. 基于机器视觉的二维尺寸检测[J]. 工具技术, 2017, 51(1):98-100
[13] 测量不确定度评定与表示:JJF1059-2012[S]. 北京:中国质检出版社, 2012.