视觉引导工业机器人的标定与工件识别方法研究  

Research on Calibration and Workpiece Recognition Methods for Visually Guided Industrial Robot

作　　者：许鸿萍 XU Hongping(Minjiang Teacher College,Fuzhou 350108)

机构地区：[1]闽江师范高等专科学校,福州350108

出　　处：《现代制造技术与装备》2024年第11期210-212,共3页Modern Manufacturing Technology and Equipment

摘　　要：针对自动化生产线上不同工件摆放重叠或位姿杂乱导致产线自动化控制困难的问题,研究视觉引导工业机器人的标定与目标工件识别方法。使用固定相机的手眼标定方法,建立工业机器人与相机之间的仿射变换矩阵。利用VisionPro软件对平台拍摄的目标工件图像进行特征提取和形状匹配识别。研究结果表明,该工件识别方法的识别速度达到16.8 ms,可同时识别多个不同位姿的工件,有效引导机械臂抓取工件,能够满足自动化生产线的需求。Aiming at the difficulty of automatic control of production line caused by overlapping of different workpieces or disorganized positions in automatic production line,the calibration and target workpiece recognition methods of visual guided industrial robots were studied.An affine transformation matrix between an industrial robot and a camera is established by using the hand-eye calibration method of a fixed camera.The feature extraction and shape matching recognition of target workpiece images captured by VisionPro software were carried out.The research results show that the recognition speed of the workpiece recognition method can reach 16.8ms,and it can recognize multiple parts with different poses at the same time,effectively guide the robot arm to grasp the workpiece,and can meet the needs of the automatic production line.

关 键 词：机器视觉 工业机器人 手眼标定 工件识别 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP242.2[自动化与计算机技术—计算机科学与技术]