All Issue

2025 Vol.38, Issue 2 Preview Page

Research Paper

Automated Crack Mapping Utilizing a Mobile Robot Equipped with a 2-DOF Manipulator

균열 지도 작성을 위한 2자유도 매니퓰레이터 기반 모바일 로봇

Sunkyung Kim1
,
Jonghyeok Kim2
,
Haemin Jeon3*
김 선경1
,
김 종혁2
,
전 해민3*
1Graduate Student, Department of AI Convergence, Gwangju Institute of Science and Technology, Gwangju, 61005, Korea
2Graduate Student, Department of Mechanical and Robotics Engineering, Gwangju Institute of Science and Technology, Gwangju, 61005, Korea
3Associate Professor, Department of Civil and Environmental Engineering, Hanbat National University, Daejeon, 34158, Korea
1광주과학기술원 AI융합학과 석사과정
2광주과학기술원 기계로봇공학과 석사과정
3한밭대학교 건설환경공학과 부교수
*Corresponding Author
30 April 2025. pp. 085-090
PDF XML Citation
Abstract

This paper presents a crack mapping technique using a ground mobile robot equipped with a 2-degree-of-freedom (DOF) manipulator. Stereo vision sensors are mounted on the front and side of the robot. Point cloud data from the front-mounted sensor is used to localize the robot and generate a map, while image data from the side-mounted sensor is used to detect wall cracks. A coordinate transformation method unifies the coordinate systems of the two sensors, enabling the real-time integration of crack detection data into the generated map to accurately record and manage damage information. The 2-DOF manipulator allows the mobile robot to detect cracks outside its movement range by capturing images beyond its field of view. Deep learning techniques are applied to detect cracks in the captured images. If only a portion of a crack is initially detected, the manipulator adjusts its position to estimate and capture the remaining crack segments, enabling the stitching of a complete crack image. Indoor experiments were conducted to validate the system’s performance, confirming the real-time mapping of crack information onto the generated map.

Keywords
crack mapping
2-degree-of-freedom (DOF) manipulator
mobile robot
stereo vision sensors
deep learning

본 논문에서는 2자유도 매니퓰레이터(manipulator)가 탑재된 지상형 이동 로봇을 활용한 균열 지도 작성 기법을 소개한다. 로봇의 앞·측면에 각각 스테레오 비전 센서를 설치하였으며, 앞면에 설치된 센서의 포인트 클라우드 정보를 이용하여 로봇의 위치를 인식하고 지도를 작성하며, 측면에 설치된 센서의 영상 정보를 바탕으로 벽면 내 균열을 검출한다. 이때, 두 센서의 좌표계를 좌표 변환식을 통해 통일하여 정합 및 검출된 균열 정보를 생성된 지도에 실시간으로 표기하고, 손상의 정보가 기록 및 관리될 수 있도록 하였다. 2자유도 움직임이 가능한 매니퓰레이터 장치를 이동로봇에 탑재하고 사각지역의 제한 없이 로봇의 진행 방향을 벗어난 균열을 촬영할 수 있도록 하였다. 촬영된 영상 내 딥러닝 기법을 적용하여 균열을 검출하고, 해당 균열이 촬영된 영상 내 일부만 존재한다고 판단하는 경우 매니퓰레이터를 동작하여 남은 균열의 위치를 추정 및 추가 촬영, 스티칭할 수 있도록 하였다. 본 시스템의 성능 확인을 위하여 실내 환경에서 실험을 진행하였으며, IoU기반 검출율 0.6 이상의 정확도로 실시간 균열 정보를 구축된 지도 위에 작성하는 것을 확인하였다.

키워드
균열 지도 작성
2자유도 매니퓰레이터
이동 로봇
스테레오 비전 센서
딥러닝
References
1

Baek, S., Oh, J., Woo, H., Kim, I., Jang, S. (2023) Localization of Cracks in Concrete Structures Lacking Reference Objects and Feature Points Using an Unmanned Aerial Vehicle, Appl. Sci., 13(17), p.9918.

10.3390/app13179918
2

Bang, H., Min, J., Jeon, H. (2021) Deep Learning-based Concrete Surface Damage Monitoring Method using Structured Lights and Depth Camera, Sens., 21(8), p.2759.

10.3390/s2108275933919733PMC8070667
3

Hsieh, Y.A., Tsai, Y.J. (2020) Machine Learning for Crack Detection: Review and Model Performance Comparison, J. Compu. Civil Eng., 34(5), p.04020038.

10.1061/(ASCE)CP.1943-5487.0000918
4

Ji, B. (2021) Crack Detection on Bridge Deck Using Generative Adversarial Networks and Deep Learning, J. Korean Soc. Constr. Recycl. Resour., 9(3), pp.303~310.

5

Kim, B., Kim, G., Jin,S., Cho, S. (2019) A Comparative Study on Performance of Deep Learning Models for Vision-based Concrete Crack Detection according to Model Types, J. Korean Soc. Saf., 34(6), pp.50~57.

6

Kim, E., Cho, S., Sim, S. (2015) A Recent Research Summary on Smart Sensors for Structural Health Monitoring, J. Korea Inst. Struct. Maint. & Insp., 19(3), pp.010~021.

10.11112/jksmi.2015.19.3.010
7

Kim, J., Jung, Y., Rhim, H. (2017) Study on Structure Visual Inspection Technology using Drones and Image Analysis Techniques, J. Korea Inst. Build. Constr., 17(6), pp.545~557.

8

Li, Y., Ma, R., Liu, H., Cheng, G. (2023) Real-time High-Resolution Neural Network with Semantic Guidance for Crack Segmentation, Autom. Constr., 156, p.105112.

10.1016/j.autcon.2023.105112
9

McNeal, G.S. (2016) Drones and the Future of Aerial Surveillance, George Washington Law Rev., 84, p.64.

10

Meng, S., Gao, Z., Zhou, Y., He, B., Djerrad, A. (2023) Real‐Time Automatic Crack Detection Method based on Drone, Comput.‐Aided Civil& Infrastr. Eng., 38(7), pp.849~872.

10.1111/mice.12918
11

Nguyen, H., Kam, T., Cheng, P. (2014) An Automatic Approach for Accurate Edge Detection of Concrete Crack Utilizing 2D Geometric Features of Crack, J. Signal Process. Syst., 77(3), pp. 221~240.

10.1007/s11265-013-0813-8
12

Sohn, H., Lim, Y., Yun, K., Kim, G. (2004) Monitoring Crack Changes in Concrete Structures, Comput.-Aided Civil & Infrastr. Eng., 20, pp.52~61.

10.1111/j.1467-8667.2005.00376.x
13

Yu, S., Jang, J., Han, C. (2007) Auto Inspection System using a Mobile Robot 364 for Detecting Concrete Cracks in a Tunnel, Autom. Constr., 16(3), pp.255~261.

10.1016/j.autcon.2006.05.003
14

Yuan, C., Xiong, B., Li, X., Sang, X., Kong, Q. (2022) A Novel Intelligent 366 Inspection Robot with Deep Stereo Vision for Three-Dimensional Concrete Damage Detection and 367 Quantification, Struct. Health Monit., 21(3), pp.788~802.

10.1177/14759217211010238
Information
  • Publisher :Computational Structural Engineering Institute of Korea
  • Publisher(Ko) :한국전산구조공학회
  • Journal Title :Journal of the Computational Structural Engineering Institute of Korea
  • Journal Title(Ko) :한국전산구조공학회 논문집
  • Volume : 38
  • No :2
  • Pages :085-090
  • Received Date : 2024-10-30
  • Revised Date : 2024-12-03
  • Accepted Date : 2024-12-03
  • DOI :https://doi.org/10.7734/COSEIK.2025.38.2.85
Journal Informaiton Journal of the Computational Structural Engineering Institute of Korea Journal of the Computational Structural Engineering Institute of Korea
Online Submission
  • NRF
  • KOFST
  • crossref crossmark
  • crossref cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close