通用机器人视觉检测系统的全局校准技术

Global calibration of visual inspection system based on universal robots

介绍了基于通用机器人的车身在线柔性视觉检测系统的工作原理和在复杂现场环境下使用激光跟踪仪和测量臂进行全局校准的实用方法。基于测量姿态的全局校准方法,利用机械约束将传感器坐标系和靶标坐标系统一起来,实现测量姿态时传感器坐标系与系统基准坐标系的坐标转换。基于机器人运动学模型的全局校准方法,利用机器人末端关节位姿可通过正向运动学获知的特点,在现场构建机器人基坐标系,并通过坐标变换获得与机器人运动精度无关的手眼关系,实现任意姿态时的全局校准。实际测量结果表明,应用两种校准方法后系统的重复性精度都十分理想,测量装置重复测量标准差和系统重复测量标准差分别为0.07mm和0.13mm,能够满足在线监控白车身加工尺寸变化的精度要求,系统进行坐标测量时的误差分别为±0.2mm和±0.8mm。

An on-line and flexible visual inspection system based on universal robots was introduced for the car body-in white （BIW） and two practical global calibration methods by using a laser tracker and a measuring arm were described in complexity field environments. The method based on a measuring posture realized the coordinate transformation between the sensor coordinates and the system reference coordinate by combining the sensor coordinates with the target coordinates through the mechanical constraint. And the method based on the robot kinematics realized the global calibration at arbitrary postures by establishing a robot coordinate frame in fields and by obtaining hand-eye calibration rela- tion through the coordiate transformation. Experimental results show that the standard deviations of repeatability for the measurement equipment and the system are 0.07 mm and 0.13 mm respectively, which indicates that both the methods can meet the requirement of monitoring the manufacture dimension changes of BIW. Futhermore, the coordinate measuring errors of the system are ±0.2 mm and ±0.8 mm from the two methods, respectively.