双机器人基准特征识别与位置补偿技术  

Dual Robot Datum Feature Detection and Position Compensation Technology

作　　者：柳峻达 李波[1] 田威[1] 廖文和[1] 张霖[1] 喻强 LIU Junda;LI Bo;TIAN Wei;LIAO Wenhe;ZHANG Lin;YU Qiang(College of Mechanical and Electrical Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]南京航空航天大学机电学院,江苏南京210016

出　　处：《机械制造与自动化》2024年第5期224-228,共5页Machine Building & Automation

基　　金：国家自然科学基金资助项目(51875287)。

摘　　要：针对自动化装配中由于产品制造、装夹和系统标定等多元累计误差造成的机器人加工位置不确定问题,提出基于线扫描仪双级模板匹配的基准特征识别和双机器人定位补偿算法。设计末端基准检测模块,实现基准孔点云获取;通过点云预处理、模板构建和双级模板匹配实现基准孔特征识别,确定圆心坐标;针对机翼中不同加工序列,利用直线式局部基准和四点式全局基准补偿方法实现双机器人协同位置补偿。针对系统基准检测精度和位置补偿效果进行验证试验。试验结果显示:系统基准检测精度达到0.048 mm,基准补偿后加工点位置精度提升了82.79%,满足机翼装配的精度指标。To address the issue of uncertain robot processing position due to cumulative errors in manufacturing,clamping and calibration,a method based on template matching for datum feature recognition and dual-robot position compensation is proposed.A benchmark detection module is designed to obtain the reference hole point cloud.The coordinates of the circle's center are determined using point cloud preprocessing,template construction,and two-stage matching.The compensation method of linear local datum and four-point global datum is applied to enable collaborative position compensation of two robots in different wing machining sequences.Verification experiments are conducted to evaluate benchmark detection accuracy and position compensation effect.The results show a benchmark detection accuracy of 0.048 mm and an 82.79%improvement in position accuracy after compensation.

关 键 词：模板匹配 位置补偿 双机器人 协同装配 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]