基于双目视觉的钻机管柱识别与定位方法  

Recognition and Positioning Method of Pipe String Based on Binocular Vision

作　　者：牛柯 彭斌[1] NIU Ke;PENG Bin(Faculty of Mechanical and Electrical Engineering,Lanzhou University of Technology,Lanzhou,Gansu 730050,China)

机构地区：[1]兰州理工大学机电工程学院,甘肃兰州730050

出　　处：《西安石油大学学报(自然科学版)》2025年第2期106-115,134,共11页Journal of Xi’an Shiyou University(Natural Science Edition)

基　　金：国家自然科学基金项目(51966009);中央引导地方科技发展资金项目“智能高效石油钻机管柱排放机器人关键设备开发及产业化”(23ZYQA312)。

摘　　要：针对钻机管柱处理过程中存在的自动化程度低、高风险和低效率问题,提出了基于双目视觉的管柱识别与定位方法。该技术利用双目相机系统,通过图像处理、精确标定、立体校正和立体匹配等步骤,生成管柱的深度图。在此基础上,对YOLOv5s目标识别模型进行了优化,以提升管柱图像的特征识别能力,显著提高了识别精度和速度。通过整合深度信息与目标识别结果,实现了管柱的精确定位。搭建模拟实验平台测试,验证了该方法的有效性和可靠性。为视觉引导机械手自动输送管柱和铁钻工自动上、卸扣提供了技术保障,为钻井平台自动化技术的发展提供了理论参考。A pipe string recognition and positioning method based on binocular vision is proposed to address the issues of high risk and low efficiency in the processing of pipe string.This technology utilizes a binocular camera system to generate a depth map of the pipe string through steps such as image processing,precise calibration,stereo correction,and stereo matching.On this basis,the YOLOv5s object recognition model was optimized for significantly improving recognition ability,accuracy and speed of the features of the pipe string images.The precise positioning of the pipe string is achieved by integrating depth information with target recognition results.The effectiveness and reliability of the method are verified by testing on a simulation experiment platform.This method provides technical support for the visual guided robotic arm to automatically transport pipe strings and for the automatic loading and unloading of iron drillers,and provides theoretical reference for the development of drilling platform automation technology.

关 键 词：双目视觉 YOLOv5s 管柱自动化处理 目标识别 管柱检测 管柱定位 

分 类 号：TE928[石油与天然气工程—石油机械设备]