基于单目视觉的双机器人协同标定技术

Collaborative calibration of dual mobile robots based on monocular vision

  • 摘要: 由于机器人的基坐标系隐藏在底座内部,导致双机器人基坐标系间位姿关系的测量变得十分困难,而该关系作为机器人离线编程、协同控制的基石,对于多机器人系统的正常工作至关重要。为此提出了基于单目视觉的协同标定方法,利用单目视觉和棋盘格标定板求解双机器人基坐标系间位姿相对关系并使用激光跟踪仪标定法作为对照对双机器人基坐标系关系进行了标定,并通过双机器人位姿协同实验对标定的结果进行验证,最终得到基于单目视觉的协同标定方法所对应的双机器人位置协同误差最大值为1.278 mm,最小值为0.601 mm;姿态协同误差最大值则为0.481°。

     

    Abstract: Because the base coordinate system of the robot is hidden inside the base, it is very difficult to measure the pose relationship between the base coordinate system of the two robots. As the cornerstone of offline robot programming and cooperative control, this relationship is very important for the normal work of the multi-robot system. A collaborative calibration method based on monocular vision was proposed to solve the relative relation of pose between the base coordinate systems of the two robots by using monocular vision and checkerboard calibration plate. The laser tracker calibration method was used as a control to calibrate the relationship between the coordinate systems of the two robots, and the calibration results were verified by the cooperative experiment of the two robots' pose. Finally, the maximum and minimum position cooperative error of the two robots corresponding to the collaborative calibration method based on monocular vision was 1.278 mm and 0.601 mm. The maximum attitude synergy error is 0.481°.

     

/

返回文章
返回