中国机械工程

• 机械基础工程 • 上一篇    下一篇

基于镜面反射的手眼标定方法

张旭1;贾君慧1;张刚2   

  1. 1.上海大学机电工程与自动化学院,上海,200072
    2.华中科技大学无锡研究院机器人技术中心,无锡,214147
  • 出版日期:2018-05-25 发布日期:2018-05-25
  • 基金资助:
    国家自然科学基金资助项目(51575332,61673252);
    科技部重点研究项目(2016YFC0302401)
    National Natural Science Foundation of China(No. 51575332,61673252)

Eye-on-Hand Calibration Method Based on Specular Reflection

ZHANG Xu1;JIA Junhui1;ZHANG Gang2   

  1. 1.School of Mechanical Engineering and Automation, Shanghai University, Shanghai, 200072
    2.Robot Technology Center, HUST-Wuxi Research Institute, Wuxi, Jiangsu, 214147
  • Online:2018-05-25 Published:2018-05-25
  • Supported by:
    National Natural Science Foundation of China(No. 51575332,61673252)

摘要: 针对Eye-on-Hand机器人手眼系统的标定问题,提出一种基于镜面反射的手眼标定方法。通过镜面反射获取图像,根据镜像原理,通过线性求解算法确定机器人末端执行器和相机坐标系的位姿关系。与以往的标定方法不同,该方法只需要3次改变平面镜的姿态,就能同时获得相机的内参数和手眼视觉系统的位姿关系,不会引入传统标定方法中利用机器人运动参数而产生的机器人运动误差。最后将其与Halcon手眼标定算法进行对比,验证了所提方法的正确性和可行性。

关键词: 手眼系统, 镜面反射, 镜像原理, 标定

Abstract: For the Eye-on-Hand calibration problems of eye-on-hand robots, a Eye-on-Hand calibration method was proposed based on specular reflection.Images were captured through specular reflection, and a linear solution algorithm was used to determine the pose relationship between ends of the robot and camera coordinate system according to the image theory.In contrast to the conventional calibration method, this method only needed to change the attitudes of plane mirrors 3 times,and might simultaneously get the camera??s internal parameters and pose relationship of the Eye-on-Hand vision systems.It will eliminate robot motion errors caused by the robot's motion parameters in traditional calibration methods.Finally,compared with the Halcon Eye-on-Hand calibration algorithm, the correctness and feasibility of the proposed method were verified.

Key words: Eye-on-Hand, specular reflection, image theory, calibration

中图分类号: