基于粗-精立体匹配的激光焊接机器人三维视觉定位研究

Research on three-dimensional visual positioning of laser welding robot based on coarse-refined three-dimensional matching

激光焊接机器人在开展焊接任务时,难以找到最佳特征匹配点,导致目标三维视觉定位精度下降,直接影响焊接效果。为了提升激光焊接机器人目标定位精度,提出基于粗-精立体匹配的激光焊接机器人三维视觉定位方法。通过机器人搭载的摄像机获取目标图像,并对目标图像实施轮廓增强处理。根据轮廓增强结果提取目标特征点,结合粗-精立体匹配寻找最佳特征匹配点,建立激光焊接机器人目标的三维视觉定位模型,实现目标的三维视觉定位。实验结果表明,使用该方法进行激光焊接机器人目标定位时的时间短、效果好。

When laser welding robot carries out the welding task,it is difficult to find the best feature matching point,which leads to the decline of the three-dimensional visual positioning accuracy of the target and directly affects the welding effect.In order to improve the target positioning accuracy of laser welding robot,a three-dimensional vi-sion positioning method of laser welding robot based coarse-refined three-dimensional matching is proposed.The tar-get image is acquired by the camera mounted on the robot,and the contour enhancement processing is performed on the target image.According to the contour enhancement results,the target feature points are extracted,and the best feature matching points are found by combining coarse-refined three-dimensional matching.The three-dimensional visual positioning model of the laser welding robot target is established to realize the three-dimensional visual positio-ning of the target.The experimental results show the target positioning of laser welding robot using this method is short and the effect is good.