摘要
由于工业机器人焊枪和焊接焊缝之间的位置偏差会导致焊接焊缝跟踪结果存在一定的误差,因此提出了一种基于激光扫描的工业机器人焊接焊缝跟踪方法。首先,采用激光扫描传感器采集工业机器人焊接焊缝图像,并对图像进行滤波、增强、黑白化预处理;其次,检测图像边缘后,采取激光扫描方法确定焊接焊缝边缘图像中激光条纹骨架的大概位置,经最小二乘法直线拟合出准确的激光条纹中心线,提取焊接焊缝特征点;再次,结合工业机器人手眼标定矩阵,将激光扫描传感器坐标系中的焊接焊缝特征点坐标转化为工业机器人坐标,从而形成焊接焊缝轨迹;最后,采取3次非均匀有理B样条方法得到工业机器人焊枪和焊接焊缝之间的位置偏差,修正机器人焊枪位置,从而实现工业机器人焊接焊缝的精准跟踪。实验结果表明,该方法不仅能够准确检测出焊接焊缝边缘,而且能够跟踪不同宽度、不同形状的焊接焊缝工件。
Due to the position deviation between the welding gun of the industrial robot and the welding seam,there will be some errors in the welding seam tracking results.Therefore,an industrial robot welding seam tracking method based on laser scanning is proposed.Firstly,the image of the industrial robot welding seam is collected by laser scanning sensor,and the image is filtered,enhanced and pre-processed in black and white.Secondly,after detecting the edge of the image,the approximate position of the laser fringe skeleton in the weld edge image is determined by laser scanning method.After the accurate laser fringe center line is fitted by the least square method,the feature points of the weld are extracted.Thirdly,combined with the hand-eye calibration matrix of the industrial robot,the coordinates of the welding seam feature points in the laser scanning sensor coordinate system are transformed into the industrial robot coordinate system to form the welding seam trajectory.Finally,after three times of non-uniform rational B-spline interpolation,the position deviation between the welding gun of the industrial robot and the welding seam is obtained.According to the position deviation,the position of the welding gun of the robot is corrected to complete the tracking of the industrial robot welding seam.The experimental results show that the method can not only accurately detect the edge of the weld,but also track the workpieces with different widths and shapes.
作者
胡石
HU Shi(Department of Electromechanical and Automotive,Chizhou Vocational and Technical College,Chizhou Anhui 247000,China;School of Mechanical and Power Engineering,Nanjing Technology University,Nanjing 211816,China)
出处
《重庆科技学院学报（自然科学版）》
CAS
2023年第5期69-75,共7页
Journal of Chongqing University of Science and Technology：Natural Sciences Edition
基金
安徽省高校优秀骨干教师国内访问研修项目“基于深度学习的图像文本区域的检测与应用研究”(GXGNFX2020161)
安徽省2022年高校科学研究重点项目“基于轨迹在线提取的3D折线焊缝机器人摆动GMAW实时跟踪技术研究”(2022AH052866)。
