【作者】 李华；

【导师】 钟佩思；

【作者基本信息】 山东科技大学 ， 机械制造及其自动化， 2017， 硕士

【摘要】 我国制造业未来十几年要向自动化、智能化的方向发展。焊接技术在现代制造业中占据非常重要的地位,因此焊接过程的自动化和智能化是焊接技术未来发展的方向。目前,焊接机器人大多通过示教再现的方式进行焊接作业,当外部环境发生变化时就需要对机器人进行重新示教,效率低下,自动化程度也达不到要求。把视觉技术应用到焊接中,利用视觉传感器识别焊缝路径,再对焊缝路径进行轨迹规划,可以在一定程度上提高焊接机器人的自动化程度。本文采用CMOS摄像机与6自由度机器人形成“Eye-to-hand”的手眼关系,搭建图像采集实验平台获取平面曲线焊缝图像。对视觉识别系统进行建模,介绍了摄像机的标定模型和标定方法以及手眼标定原理,利用平面棋盘格作为靶标对摄像机进行了标定实验和手眼标定实验,得到了摄像机的内外参数和手眼关系矩阵。对采集到的平面曲线焊缝进行图像预处理和特征信息提取,提出了一种基于局部阈值分割的图像二值化方法,并利用求平均值的方法得到了焊缝路径中心线的坐标,最后经过拟合得到了焊缝路径中心线在图像坐标系下的方程,再经过坐标转换得到了焊缝路径中心线在世界坐标系下的方程。介绍了一种焊缝特征坐标系建立的方法,并建立了平面曲线焊缝的特征坐标系;介绍了6自由度机器人的运动学模型,分析了其正运动学方程和逆运动学方程;提出了一种改进的等间距直线逼近法并利用该方法生成了平面曲线焊缝路径轨迹,并进行了运动学仿真。开发了基于视觉的焊缝路径识别与轨迹规划的系统软件,进行了焊接实验,并对实验结果进行了误差分析。更多还原

【Abstract】 China’s manufacturing industry will be in the direction of automation and intelligence in the future ten years. Welding technology takes an important part in modern manufacturing industry. So automatic and intelligent welding process is the direction of welding technology in the future. At present, most of the welding robots are conducted by the way of teaching and reappearance. When external environment changed, robots must be conducted again. So the production efficiency is low and the level of welding automation can not meet the requirement. The visual technology is applied to welding. Weld path is identified by the visual sensor. Then the weld path for trajectory planning is implemented. To some extent, all these can improve the automation of welding robot."Eye-to-hand" relationship was formed with the CMOS camera and the six degrees of freedom robot. Experimental platform was set up for the acquisition of images of plane curve welding line. The model of visual identification system was built. The model and calibration method of camera and the hand-eye calibration principle were introduced. The camera calibration and the hand-eye calibration were experimented with a target of planar checkerboard, obtaining the inside and outside parameters of the camera and the hand-eye relation matrix. The image preprocessing and feature information extraction were conducted with the collected plane curve weld. A method of image binary-coding was proposed based on the local threshold segmentation. And the coordinates of the welding center line were obtained by using the method of average. Finally, after fitting, the equation of the weld path center line in the image coordinate system was obtained. And after coordinate transformation,its equation in the world coordinate system was got. A method of establishing characteristic coordinate system of weld was given and that of the plane curve weld was also built. The kinematic model of six degrees of freedom robot was presented and its forward kinematic equations and inverse kinematic equations were analyzed.The equal interval line-approaching method was proposed and the plane curve weld path trajectory was generated by this method and its kinematical simulation was carried on. The system software of welding path recognition and trajectory planning based on vision was developed, about which the weld test was conducted. And errors about the test results were analyzed.更多还原