计算机集成制造系统 ›› 2021, Vol. 27 ›› Issue (1): 118-127.DOI: 10.13196/j.cims.2021.01.010

• 当期目次 • 上一篇    下一篇

基于视觉定位与轨迹规划的机器人打磨工作站

万国扬1,王国峰1,李福东2,朱文俊3   

  1. 1.大连海事大学船舶电气工程学院
    2.扬州大学信息学院
    3.南京工业大学信息学院
  • 出版日期:2021-01-31 发布日期:2021-01-31
  • 基金资助:
    国家自然科学基金资助项目(51609033);辽宁省自然科学基金资助项目(20180520005);辽宁省重点研发指导计划资助项目(2019JH8/10100100);大连市软科学研究计划资助项目(2019J11CY014);中央高校基本科研业务费专项基金资金资助项目(3132019005,3132019311)。

Robotic grinding station based on visual positioning and trajectory planning

  • Online:2021-01-31 Published:2021-01-31
  • Supported by:
    Project supported by the National Natural Science Foundation,China(No.51609033),the Natural Science Foundation of Liaoning Province,China(No.20180520005),the Key Development Guidance Program of Liaoning Province,China(No.2019JH8/10100100),the Soft Science Research Program of Dalian City,China(No.2019J11CY014),and the Fundamental Research Funds for the Central Universities,China(No.3132019005,3132019311).

摘要: 针对铸件打磨过程中的毛坯件定位困难、打磨轨迹示教工作繁琐等问题,结合工业机器人与机器视觉技术,设计并实现了从上件到打磨的机器人打磨工作站。首先,在有环境光干扰的情况下,使用模板匹配实现对毛坯工件的机器人定位引导与抓取上料工作。然后,使用边缘提取与轮廓拟合技术,在屏蔽微小毛刺干扰的情况下,完成了工件轮廓的位置测量与工件表面主要毛刺位置的判断。在此基础上,提出网格法实现了对工件表面主要毛刺的检测,并结合机器人的坐标转化方法,完成了机器人离线打磨轨迹的规划。系统完成了打磨过程中上料、打磨、下料全过程的自动化作业。实验数据表明,机器人打磨工作站具备上件抓取精度高,打磨工件质量一致性好等优点,并可替代3~4名人工实现打磨作业。

关键词: 工业机器人, 模板匹配, 轨迹规划, 机器视觉

Abstract: Targeting the difficulty of positioning the rough workpiece in the process of casting grinding and the problem of cumbersome work during the grinding track teaching,a robot grinding system which included the whole process from loading part to grinding part with the combination of industrial robots and machine vision technology was designed and realized.The method of template matching was applied to achieve the robotic positioning and grabbing guidance of the rough workpiece under the disturbance of ambient light.Subsequently,edge extraction and contour fitting technology were used to measure the position of the contour of workpiece and to estimate the main burr on the surface after eliminating the influence of burr.On this basis,the grid method was put forward to detecting the main burrs on the surface of the workpiece.Moreover,the offline grinding trajectory of the robot was planned with the guidance of the coordinate transformation method.The system greatly improved the automaticity throughout the entire process of loading,grinding and unloading.As measured,the proposed system could work stably and efficiently.

Key words: industrial robot, template matching, trajectory planning, machine vision

中图分类号: