基于结构光视觉传感器的弧焊机器人视觉检测方法及试验结果分析  

Visual Inspection Method of Arc Welding Robot Based on Structured Light Vision Sensor and Experimental Results Analysis

作　　者：莫毅[1] 

机构地区：[1]广西工业职业技术学院,广西南宁530001

出　　处：《热加工工艺》2017年第1期238-242,共5页Hot Working Technology

基　　金：广西工业职业技术学院2014年度科学研究项目(2014056007);2015年度广西高校科学技术研究项目(KY2015YB444)

摘　　要：针对自动焊接机器人焊缝精确检测的问题,研究了基于结构光视觉传感器的视觉检测方法,对结构光视觉传感器实时采集图像进行理论分析,经过焊缝图像预处理、图像边缘检测和轮廓提取等处理过程,运用最小二乘法对焊缝边缘拟合,通过求平均值的方法提取结构光条纹的中心直线,找到数字图像中焊缝的位置,最后利用机器人手眼标定焊缝的位置,并进行焊缝跟踪。结合试验,验证了该视觉检测方法检测精度较好,能够满足实际的工作要求。Aiming at the problem of welding seam precise detection of automatic welding robot, the visual inspection method based on structured light vision sensor was studied. Theoretical analysis was made on the real time images collected by the structured light vision sensor. Aider the weld image preprocessing, image edge detection and contour extraction, the weld edge was fitted by using the least square method, the center line of the structured light stripe was extracted by calculating the average value, and the weld position in the digital image was found. Finally, the weld position was calibrated by the robot＇s hand and eye, and the seam tracking was carried out. The V-shape groove weld detection experiments show that the visual platform has favorable detection precision, and it can meet actual requirements.

关 键 词：焊缝检测 弧焊机器人 结构光视觉传感器 V型坡口 

分 类 号：TG409[金属学及工艺—焊接] TP212.6[自动化与计算机技术—检测技术与自动化装置]