移动焊接机器人的工件识别及焊缝起始位置定位

Recognition of Work-piece and Location of Initial Seam Position for Mobile Welding Robot

考虑到非结构空间焊接的特殊性,通过视觉技术,基于圆投影和Zernike矩,提出了一种两步模板匹配法检测待焊工件是否出现在视觉区域;结合霍夫直线检测方法,设计了一套提取特征直线算法,并确定焊缝起始点在图像中的亚像素位置.由移动机器人在不同角度拍摄工件图像,圆投影和Zernike矩计算结果验证了其表征模板特性的有效性,具有旋转抗性,而且对于光照条件及摄像机视野的鲁棒性好,焊缝起始位置定位结果满足视觉计算需要.

Considering the environmental characters of welding in unstructured space, via vision technolo-gy, a two-steps template matching method for work-piece recognition based on circle projection and Zerni-ke moments was presented. An algorithm based on Hough line detection was designed to extract special lines which were used to calculate initial seam position with sub-pixel precision. Experiment was carried out by mobile welding robot, and series of templates were acquired by snapping work-piece images with different rotation angles. Circle projection results and Zernike moments were calculated, and they could ef- fectively represent features of templates for typical work-piece, meanwhile errors due to rotation could be eliminated. These algorithms were proved to perform satisfied robustness of illumination and camera view. Sub-pixel location of initial seam position can satisfy the need for visual calculation.