基于手眼标定方程AX=XB的精度影响因素研究

发布时间：2018-07-18 15:21

【摘要】：机器人手位姿数据对手眼标定精度的影响不可忽略,将对基于手眼标定方程AX=XB的精度影响因素进行分析。通过手眼标定仿真和实测实验验证上述两个因素对手眼标定精度的影响与理论分析的一致性。通过仿真与实测实验,总结得出了减小摄像机与靶标间距离、减小机器人手的运动前后到基坐标空间距离的相差距离,可提高手眼标定精度,通过四元数法和矩阵直积法验证了此规律在解AX=XB标定方程时的通用性,并且在摄像机与靶标间距约为230 mm以及机器人手的运动前后到基坐标空间距离的相差距离为3.240 1 mm时,手眼标定平移向量相对误差最高精度可达0.040 3%。
[Abstract]:The influence of the robot hand posture data on the accuracy of the eye calibration can not be ignored, and the influence factors of the accuracy based on the hand eye calibration equation AX=XB are analyzed. The agreement between the influence of the two factors on the calibration accuracy of the opponent's eye calibration and the theoretical analysis is verified by the hand eye calibration simulation and the measured experiment. The results are summarized by the simulation and the measured experiments. The distance between the camera and the target is reduced and the distance between the robot hand's motion and the distance between the base coordinates is reduced. The accuracy of the hand eye calibration can be improved. The four element method and the matrix direct product method are used to verify the generality of this rule when the AX=XB calibration equation is solved, and the distance between the camera and the target is about 230 mm and the robot hand. The relative error of hand eye calibration vector can reach 0.040 3%. when the distance between the base and the coordinate space is 3.2401 mm before and after the movement.
【作者单位】：北京信息科技大学光电测试技术北京市重点实验室;北京邮电大学信息光子学与光通信研究院;
【基金】：国家重大科学仪器设备开发专项项目(2013YQ22089304) 教育部“长江学者与创新团队”发展计划资助项目(IRT1212) 北京市组织部青年骨干个人项目
【分类号】：TP242

【参考文献】