基于轨迹匹配的焊缝离线纠偏研究  

Research on Off-Line Correction of Weld Seam Based on Trajectory Matching

作　　者：杨宏韬[1] 熊凤銮 于微波[1] 苏航 李秀兰[1] YANG Hong-tao;XIONG Feng-luan;YU Wei-bo;SU Hang;LI Xiu-lan(School of Electrical and Electronic Engineering,Changchun University of Technology,Changchun 130012,China)

机构地区：[1]长春工业大学电气与电子工程学院,长春130012

出　　处：《组合机床与自动化加工技术》2022年第9期88-91,95,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：国家自然科学基金(62106023);吉林省科技发展计划项目(20200401118GX);吉林省教育厅科学技术研究项目(JJKH20210744KJ)。

摘　　要：针对焊接机器人示教偏差较大导致结构光传感器离线提取焊缝轨迹缺失的问题,提出一种基于轨迹匹配的焊缝离线纠偏方法。该方法主要包括初始匹配和精细匹配两个阶段,在初始配准阶段,通过曲率相似度进行对应点对的匹配,采用主成分分析(PCA)算法获取初始转换矩阵,进而将示教轨迹进行更新;在精细匹配阶段,引入曲率相似度结合欧式距离的思想改进迭代最近点(ICP)算法,同时采用K-D树加速对应点的查找和随机采样(RACSAC)进行剔除错误匹配点,将初始匹配阶段更新的示教轨迹匹配到焊缝轨迹上。结果表明,所提出的方法解决了示教偏差问题,纠偏后的示教轨迹与焊缝轨迹的平均偏差小于0.5 mm,满足离线纠偏精度的需求。Aiming at the problem of missing weld track extracted offline by structured light sensor due to large teaching deviation of welding robot,an off-line correction method of weld track matching was proposed in this paper.The method mainly includes initial matching and fine matching.In the initial registration stage,the corresponding point pairs are matched by curvature similarity,and the initial transformation matrix is obtained by principal component analysis(PCA)algorithm,and then the teaching trajectory is updated.At the stage of fine matching,the curvature similarity combined with the idea of European distance was introduced to improve the iterative nearest point(ICP)algorithm,and the K-D tree was used to accelerate the search of corresponding points and random sampling(RACSAC)was used to eliminate the wrong matching points,and the updated teaching track in the initial matching stage was matched to the weld track.The results show that the proposed method solves the problem of teaching deviation,and the average deviation between teaching trajectory and weld trajectory after correction is less than 0.5 mm,which meets the requirements of off-line correction accuracy.

关 键 词：焊接机器人 轨迹匹配 迭代最近点(ICP) 离线纠偏 

分 类 号：TH165[机械工程—机械制造及自动化] TG659[金属学及工艺—金属切削加工及机床]