한국산업정보학회논문지 (Journal of Korea Society of Industrial Information Systems)

한국산업정보학회 (Korea Society of Industrial Information Systems)
권호 표지
DOI QR코드

DOI QR Code

산업제조현장 스마트 안전 시스템용 레이다 및 IMU 센서를 이용한 앙상블 부스팅 모델 기반 작업자 탐지 기술

Worker Detection Based on Ensemble Boosting Model Using a Low-cost Radar and IMU for Smart Safety System in Manufacturing

  • 송승언 (대구경북과학기술원 첨단레이다연구소) ;
  • 김상동 (대구경북과학기술원 첨단레이다연구소) ;
  • 김봉석 (대구경북과학기술원 첨단레이다연구소) ;
  • 류정탁 (대구대학교 전자전기공학부) ;
  • 이종훈 (대구경북과학기술원 첨단레이다연구소)
  • 투고 : 2024.10.01
  • 심사 : 2024.10.14
  • 발행 : 2024.10.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문은 산업 제조 현장에서 작업자의 안전을 위협하는 사각지대를 해결하기 위해서 저가형 CW(Continuous Wave) 레이다와 IMU(Inertial Measurement Unit)센서를 결합한 스마트안전시스템을 제안하였다. 24GHz 레이다와 6축 IMU 센서를 사용하여 작업자의 움직임을 감지하고, 기계 학습 모델을 통해 작업자 상황을 인식할 수 있었다. 레이다와 IMU 특징점과 앙상블 부스팅 트리 기반 기계학습모델을 사용한 결과, 92.8% 이상의 작업자 탐지율을 확보하였다.

This paper proposes a smart safety system that combines low-cost CW(Continuous Wave) radar and IMU sensors to enhance blind spots that pose safety risks to workers in industrial manufacturing environments. The system employs a 24 GHz radar and a 6-axis IMU sensor to detect worker movements and utilizes a machine learning model to recognize worker situations in vibrating manufacturing sites. The ensemble boosting tree-based model achieved over 92.8% worker detection accuracy, demonstrating its effectiveness in improving safety in industrial settings.

키워드

과제정보

이 논문은 2024년 행정안전부에서 지원하는 재난안전산업 기술사업화 지원사업(RS-2024-00416793)과 2024년 과학기술정보통신부 재원으로 DGIST 기관사업 (2024-IT-01)의 지원에 의해 연구되었음.

참고문헌

  1. Ajakwe, S. O., Ihekoronye, V. U., Kim, D.-S. and Lee, J. M. (2022). DRONET: Multi-tasking Framework for Real-time Industrial Facility Aerial Surveillance and Safety, Drones, 6(2), 46. https://doi.org/10.3390/drones6020046
  2. Ergun, S., Han, D. G. and Kim, J. S. (2021). A Unified Perception Benchmark for Capacitive Proximity Sensing towards Safe Human-robot Collaboration (HRC), 2021 IEEE International Conference on Robotics and Automation (ICRA), May 30-Jun. 05, Xi'an, China, pp. 3634-3640. https://doi.org/10.1109/ICRA48506.2021.9561224
  3. Yeo, J., Lee, J. and Ryu, J. T. (2024). Study on Effects of Separation Distance between Flat Cover and Radar for 24 GHz Band Radar, Journal of Korea Society of Industrial Information Systems, 29(2), 27-33. http://dx.doi.org/10.9723/jksiis.2024.29.2.027
  4. Kim, B., Kim, S. and Lee, J. (2018). A Novel DFT-Based DOA Estimation by a Virtual Array Extension Using Simple Multiplications for FMCW Radar, Sensors, 18(5), 1560-1576. https://doi.org/10.3390/s18051560
  5. Kim, S., Kim, B. and Lee, J. (2017). Low-complexity Spectral Partitioning Based Music Algorithm for Automotive Radar, Elektronika ir Elektrotechnika, 23(4), 33-38. https://doi.org/10.5755/j01.eie.23.4.18719
  6. Linder, T., Vaskevicius, N., Schirmer, R. and Arras, K. O. (2021). Cross-modal Analysis of Human Detection for Robotics: An Industrial Case Study, 2021 IEEE/ RSJ International Conference on Intelligent Robots and Systems (IROS), Sep. 27-Oct. 01, Prague, Czech Republic, pp. 971-978. https://doi.org/10.1109/IROS51168.2021.9636158
  7. Liu, G., Li, X., Xu, C., Ma, L. and Li, H. (2023). FMCW Radar-based Human Sitting Posture Detection, IEEE Access, 11, 102746-102756. https://doi.org/10.1109/ACCESS.2023.3312328
  8. Rey-Merchan, M. del C., Gomez-de-Gabriel, J. M., Fernandez-Madrigal, J. A. and Lopez-Arquillos, A. (2020). Improving the Prevention of Fall from Height on Construction Sites through the Combination of Technologies, International Journal of Occupational Safety and Ergonomics, 28(1), 590-599. https://doi.org/10.1080/10803548.2020.1815393
  9. Song, S., Kim, B., Kim, S. and Lee, J. (2021). Detection of Stationary and Moving Obstacles within a Critical Safety Zone Using 24GHz Multiple Radars and IMU Sensors in a Safety-assisted Module for Heavy Construction Vehicles, 2021 KSAE Annual Conference & Exhibition, Nov. 17, Yeosu, Korea.
  10. Shen, Z., Nunez-Yanez, J. and Dahnoun, N. (2024). Advanced Millimeter-wave Radar System for Real-time Multiple-human Tracking and Fall Detection, Sensors, 24(11), 3660. https://doi.org/10.3390/s24113660
  11. Shi, T., Guo, P., Wang, R., Ma, Z., Zhang, W., Li, W., Fu, H. and Hu, H. (2024). A Survey on Multi-sensor Fusion Perimeter Intrusion Detection in High-speed Railways, Sensors, 24(17), 5463. https://doi.org/10.3390/s24175463
  12. Skog, M. (2024). Human Detection in Low-visibility Industrial Settings Using Automotive 4D Radar and Deep Learning, Dissertation, Graduate School of Orebro University, Orebro, Sweden.
  13. Sommer, P., Rigner, A. and Zlatanski, M. (2020). Radar-based Situational Awareness for Industrial Safety Applications, 2020 IEEE Sensors, Oct. 25-28, Rotterdam, Netherlands, pp. 1-4. https://doi.org/10.1109/SENSORS47125.2020.9278603
  14. Song, S., Kim, S., Kim, B., Kweon, H. and Lee, J. (2022). Machine Learning in a Human Detection based on a Multi-radar and IMU Sensor for a Blind Spot Detection of a Smart Construction Vehicle, ISI ITA 2022, S5-16.