当前位置:
X-MOL 学术
›
Autom. Constr.
›
论文详情
Our official English website, www.x-mol.net, welcomes your
feedback! (Note: you will need to create a separate account there.)
Automated system of scaffold point cloud data acquisition using a robot dog
Automation in Construction ( IF 9.6 ) Pub Date : 2024-12-24 , DOI: 10.1016/j.autcon.2024.105944
Duho Chung, Juhyeon Kim, Sunwoong Paik, Seunghun Im, Hyoungkwan Kim
Automation in Construction ( IF 9.6 ) Pub Date : 2024-12-24 , DOI: 10.1016/j.autcon.2024.105944
Duho Chung, Juhyeon Kim, Sunwoong Paik, Seunghun Im, Hyoungkwan Kim
This paper introduces Automated system of Scaffold Point cloud data Acquisition using a Robot dog (ASPAR), a method for automating scaffold point cloud data acquisition using a quadruped robot. The method consists of three stages: (1) Initial Exploration, where the robot autonomously explores the site and detects scaffolds in real-time; (2) Scan Plan Generation, which uses 3D SLAM data and scaffold detection results to determine optimal scan positions and generate paths between them; and (3) Scan Plan Execution, where the robot follows these paths and performs scans at the designated positions. ASPAR demonstrated its effectiveness in scanning scaffold structures on construction sites without prior information. Experimental results showed that, compared to manual scans by skilled workers, it secured an average of 0.7 additional scan positions, achieving a coverage rate of 106.1 %. In a large-scale outdoor construction site experiment, it recorded a coverage rate of 96.8 %, validating its real-world applicability.
中文翻译:
使用机器狗的脚手架点云数据采集自动化系统
本文介绍了使用机器狗的脚手架点云数据采集自动化系统 (ASPAR),这是一种使用四足机器人自动采集脚手架点云数据的方法。该方法包括三个阶段:(1) 初步探索,机器人自主探索现场并实时检测脚手架;(2) 扫描计划生成,它使用 3D SLAM 数据和脚手架检测结果来确定最佳扫描位置并生成它们之间的路径;(3) 扫描计划执行,机器人遵循这些路径并在指定位置执行扫描。ASPAR 证明了它在没有事先信息的情况下扫描建筑工地脚手架结构的有效性。实验结果表明,与熟练工人的人工扫描相比,它平均获得了 0.7 个额外的扫描位置,覆盖率达到 106.1%。在大规模户外施工现场实验中,它记录了 96.8% 的覆盖率,验证了其实际适用性。
更新日期:2024-12-24
中文翻译:

使用机器狗的脚手架点云数据采集自动化系统
本文介绍了使用机器狗的脚手架点云数据采集自动化系统 (ASPAR),这是一种使用四足机器人自动采集脚手架点云数据的方法。该方法包括三个阶段:(1) 初步探索,机器人自主探索现场并实时检测脚手架;(2) 扫描计划生成,它使用 3D SLAM 数据和脚手架检测结果来确定最佳扫描位置并生成它们之间的路径;(3) 扫描计划执行,机器人遵循这些路径并在指定位置执行扫描。ASPAR 证明了它在没有事先信息的情况下扫描建筑工地脚手架结构的有效性。实验结果表明,与熟练工人的人工扫描相比,它平均获得了 0.7 个额外的扫描位置,覆盖率达到 106.1%。在大规模户外施工现场实验中,它记录了 96.8% 的覆盖率,验证了其实际适用性。