| Author Name | Affiliation | Postcode | | Rentao Xiong | School of Electromechanical Engineering, Guangdong University of Technology, Guangzhou 510006, China | 510006 | | Zengliang Lai | School of Electromechanical Engineering, Guangdong University of Technology, Guangzhou 510006, China | 510006 | | Yisheng Guan* | School of Electromechanical Engineering, Guangdong University of Technology, Guangzhou 510006, China | 510006 | | Yufeng Yang | Foshan Biowin Robotics and Automation Technology Co., Ltd., Foshan 528000, Guangdong, China | 528237 | | Chuanwu Cai | Foshan Biowin Robotics and Automation Technology Co., Ltd., Foshan 528000, Guangdong, China | 528237 |
|
| Abstract: |
| Burr is a common phenomenon in the manufacturing of metal parts, which directly affects the assembly accuracy and service performance of mechanical parts. It is necessary to replace traditional deburring methods with robotic deburring technology, and the improvement of robotic deburring efficiency has always been a hot research topic in recent years. In this paper, for two-dimensional planar workpiece, a vision-based method is proposed, where burr contour is recognized and coordinates sequence is real-time generated in X and Y directions, and meanwhile robotic deburring efficiency is improved based on the quantitative information of burr size. First, by utilizing the local deformable template matching algorithm, standard workpiece contour was matched with the workpiece contour to be processed, and the corresponding pixels distance between the two contours was calculated. Second, the distance thresholds was set to divide the burr contours into different levels, coordinates of the burr contours were extracted and mapped to the standard workpiece contour, and then the closed-loop robotic deburring path sequence was generated. Finally, on the basis of the quantitative information of burr size, the deburring speed was adjusted in real-time during the deburring process. Experiments show that the deburring time of the proposed method was shortened by 15.45%, compared with conventional off-line programming deburring method. For industrial mass production, the deburring efficiency could be greatly improved. |
| Key words: robotic deburring assembly accuracy contour recognition template matching deburring efficiency |
| DOI：10.11916/j.issn.1005-9113.19014 |
| Clc Number:TP 249 |
| Fund: |
|
| Descriptions in Chinese: |
| 毛刺是金属零件加工过程中出现的常见现象，其直接影响机械零部件的装配精度和使用性能。利用机器人去毛刺技术代替传统人工去毛刺显得尤为必要，而怎样提高机器人去毛刺的效率是近年来研究的热点。本文针对二维平面工件，提出一种基于视觉的毛刺轮廓识别和在X与Y方向上将毛刺轮廓坐标顺序实时生成的方法，然后基于毛刺大小量化信息提高机器人去毛刺的效率。首先，基于局部可变形模板匹配算法将标准工件轮廓与待加工工件轮廓进行配准，并计算两轮廓对应像素点之间的距离；其次，基于距离阈值将毛刺轮廓划分为不同的量级，并提取毛刺轮廓坐标将其映射到标准工件轮廓上，然后生成机器人去毛刺闭环加工路径顺序；最后，基于毛刺大小量化信息实时调整机器人去毛刺的速度。实验表明，所提出的方法相比传统离线编程去毛刺，去毛刺的时间缩短了13.53%。对于工业大批量生产，可以大大提高机器人去毛刺的效率。 |
