视觉引导的搬运机器手位置测量研究

为提高工作效率，满足搬运机器手自动化抓取汽车保险杠的需要，研究了基于结构光视觉原理的保险杠抓取点位置测量方法。该方法通过对相机拍摄图像进行图像处理，识别出投射在保险杠平坦区域的激光光斑中心点；利用透视变换原理对获取的激光光斑图像进行矫正，最后结合像素标定结果计算得到汽车保险杠抓取点沿图像X轴方向偏移量和高度信息。现场实验结果表明：该测量方法稳定性能好，测量误差小于±3mm，可以实现对汽车保险杠的非接触测量，实时引导搬运机械手准确抓取汽车保险杠，给自动化行业提供一种简单准确的定位识别系统和方法。

Research on Position Measurement of Vision - guided Handling

In order to improve working efficiency and meet the needs of automated handling Manipulator grasping the bumper, a position measurement method of the bumper gripping point based on structural light visual principle is studied. In this method, image of the camera is captured and the center of the laser spot projected in the flat area of the bumper is identified. The laser spot image is corrected by the perspective transformation principle, Finally, offset and height information of the car bumper gripping point along the X axis direction of the image are calculated by combining the pixel calibration results. The experimental results show that the stability of the measurement method is good and the measurement error is less than ± 3mm. It can realize non-contact measurement of the car bumper, and the real-time guided handling robot can accurately grasp the car bumper, and provide the automation industry with a simple and accurate positioning Identify systems and methods.