Research on the Application of Intelligent Detection Technology in Metro Vehicle Maintenance
智能检测技术在地铁车辆维修中的应用研究

摘要

针对地铁车辆传统人工检修效率偏低、主观性较强、隐蔽缺陷漏检率相对较高，以及现有巡检设备地形适配性差、 无法有限、难以实现立体化检测的行业难题，本文研制一款适配轨交复杂检修工况的四足机器人智能检测系统。该系 统以四足机器人为移动载体，集成全景图像采集与轻量化三维重建技术，构建涵盖图像预处理、全景拼接、三维建模 与量化缺陷识别的全流程检测体系。通过发挥依托四足机器人多步态越障、狭小空间灵活通行的优势，可在一定程度 上实现地铁车辆构件全方位精细化检测。 

Abstract

In response to industry problems such as the low efficiency and strong subjectivity of traditional manual inspection for metro vehicles, the relatively high missed-detection rate of hidden defects, poor terrain adaptability of existing inspection equipment, limited operational capability, and difficulty in realizing three-dimensional detection, this paper develops an intelligent inspection system based on a quadruped robot and adapted to complex maintenance conditions in rail transit. With the quadruped robot as the mobile carrier, the system integrates panoramic image acquisition and lightweight three-dimensional reconstruction technology, and constructs a full-process inspection system covering image preprocessing, panoramic stitching, three-dimensional modeling, and quantitative defect identification. By leveraging the advantages of the quadruped robot in multi-gait obstacle crossing and flexible movement in narrow spaces, the system can, to a certain extent, realize all-round and refined inspection of metro vehicle components.