提出并设计了一款基于组合激光结构光的新型多功能单目视觉传感器，介绍了其研究开发思路和结构设计过程. 基于单目视觉和单幅图像处理，可实现焊接坡口截面尺寸、焊枪高度和空间位姿检测等功能. 传感器的标定充分利用了其内部固有参数，在焊枪空间位姿发生变化时，不需要进行重新标定，有效提高了其应用于复杂结构件焊接时的适应能力. 组合激光结构光的应用，有效解决了常规单目视觉存在的深度方向信息丢失问题. 针对V形坡口的截面尺寸和焊枪高度检测，推导了检测算法并进行了试验验证. 试验结果表明，最大检测误差不超过2.1%.
A novel multifunctional monocular visual sensor based on combined laser structured lights is proposed and designed; its research and development ideas and structural design process are detailed. The proposed visual sensor can realize multi-functions of detection such as the cross-sectional sizes of welding groove, the height, space position and posture of welding torch by using monocular vision and single image processing. The calibration of the sensor makes full use of its intrinsic parameters, and there is no need to re-calibrate the sensor when the position and posture of welding torch changes, which effectively improves the adaptability of the sensor in the welding process of complex structural workpiece and welding trajectory. With the use of combined laser structured lights, the problem of information loss in the depth direction existing in usual monocular visual sensor is solved effectively. Finally, the corresponding detection algorithms for the cross-sectional sizes of the V-type groove and the height of welding torch are derived, and the experiments are carried out to verify them. The experimental results show that the maximum relative detection error is not more than 2.1%.
[1] 王伟,邹奇仕,朱六妹,等.视觉传感焊缝跟踪技术的发展状况及实施方案探讨[J].电焊机, 2002, 32(5):1-8 Wang Wei, Zou Qishi, Zhu Liumei, et al. The development state and implementing method of seam tracking system with optical sensor[J]. Electric Welding Machine, 2002, 32(5):1-8
[2] 郭吉昌,朱志明,于英飞,等.焊接领域激光结构光视觉传感技术的研究及应用[J].中国激光, 2017, 44(12):1-10 Guo Jichang, Zhu Zhiming, Yu Yingfei, et al. Research and application of visual sensing technology based on laser structured light in welding industry[J]. Chinese Journal of Lasers, 2017, 44(12):1-10
[3] 吴珊珊,基于十字结构光的三维重建及精度分析[D].杭州:浙江理工大学, 2015.
[4] 乔东虓,郑军,潘际銮.双线结构光焊缝跟踪传感器及其特性[J].电焊机, 2010, 40(11):14-16 Qiao Dongxiao, Zheng Jun, Pan Jiluan. Dual structure laser vision sensor and its character[J]. Electric Welding Machine, 2010, 40(11):14-16
[5] 毛志伟,周少玲,赵滨,等.双线激光传感焊枪定位与焊缝走向识别[J].焊接学报, 2015, 36(2):35-38 Mao Zhiwei, Zhou Shaoling, Zhao Bin, et al. Welding torch position and seam orientation deviation based on two stripes laser vision sensing[J]. Transactions of the China Welding Institution, 2015, 36(2):35-38
[6] Shao W J, Huang Y, Zhang Y. A novel weld seam detection method for space weld seam of narrow butt joint in laser welding[J]. Optics and Laser Technology, 2018, 99:39-51.
[7] Sung K, Lee H, Choi Y S, et al. Development of a multiline laser vision sensor for joint tracking in welding[J]. Welding Journal, 2009, 88(4):79-85.
[8] 田忠运.基于双线结构光三维检测技术研究[D].南京:南京航空航天大学, 2012.
[9] 李学瑞.基于激光双目视觉的焊接机器人波纹板焊缝三维重建的研究[D].广州:华南理工大学, 2014.
[10] 陈熙引.基于双目视觉的机器人焊缝识别及轨迹规划研究[D].广州:华南理工大学, 2015.
[11] 邹媛媛,赵明扬,张雷,等.结构光视觉传感器误差分析与结构分析[J].仪器仪表学报, 2008, 29(12):2605-2610 Zou Yuanyuan, Zhao Mingyang, Zhang Lei, et al. Error analysis and structural analysis of structured-light visual sensor for seam tracking[J]. Chinese Journal of Scientific Instrument, 2008, 29(12):2605-2610
[12] 刘南生,郭昌荣,刘明友,等.爬行式弧焊机器人结构光视觉传感器内部布局实验研究[J].江西科学, 2005, 23(4):325-327 Liu Nansheng, Guo Changrong, Liu Mingyou., et al The internal layout experiment research of structural light visual sensor used in crawling arc welding robot[J]. Jiangxi Science, 2005, 23(4):325-327
[13] 邱茂林,马颂德.计算机视觉中摄像机定标综述[J].自动化学报, 2000, 26(1):43-54 Qiu Maolin, Ma Songde. Overview of camera calibration for computer vision[J]. Acta Autom Atica Sinica, 2000, 26(1):43-54
