The Journal of the Korea institute of electronic communication sciences (한국전자통신학회논문지)

Korea Institute of Electronic Communication Science (한국전자통신학회)
권호 표지
DOI QR코드

DOI QR Code

Development of Incident Detection Algorithm using GPS Data

GPS 정보를 활용한 돌발상황 검지 알고리즘 개발

  • Received : 2021.07.14
  • Accepted : 2021.08.17
  • Published : 2021.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

Regular or irregular situations such as traffic accidents, damage to road facilities, maintenance or repair work, and vehicle breakdowns occur frequently on highways. It is required to provide traffic services to drivers by promptly recognizing these regular or irregular situations, various techniques have been developed for rapidly collecting data and detecting abnormal traffic conditions to solve the problem. We propose a method that can be used for verification and demonstration of unexpected situation algorithms by establishing a system and developing algorithms for detecting unexpected situations on highways. For the detection of emergencies on expressways, a system was established by defining the expressway contingency and algorithm development, and a test bed was operated to suggest a method that can be used for verification and demonstration of contingency algorithms. In this study, a system was established by defining the unexpected situation and developing an algorithm to detect the unexpected situation on the highway, and a method that can be used verifying and demonstrating unexpected situations. It is expected to secure golden time for the injured by reducing the effectiveness of secondary accidents. Also predictable accidents can be reduced in case of unexpected situations and the detection time of unpredictable accidents.

고속화도로 및 자동차전용도로와 같은 고속도로에서는 중대형 교통사고, 도로시설물 파손 및 유지/보수작업, 차량 고장 및 정지 등 규칙/불규칙한 상황이 빈번히 발생한다. 이러한 규칙/불규칙적 상황을 즉각적으로 인식하여 운전자들에게 교통 서비스를 제공하는 것이 요구되었으며, 이를 해결하기 위해 신속히 데이터를 수집하고 비정상적인 교통상황을 검지하는 것에 대한 다양한 기법들이 개발되었다. 하지만 인프라에 대한 유지/보수와 검지율, 위치에 대한 정확성 등 개선점이 요구되었다. 본 연구에서는 고속도로내 돌발상황 검지를 위해 기존 연구에 대한 고찰과 자동차 위치정보(GPS, Global Positioning System) 기술, 교통공학 이론적 관점의 연구를 통해 고속도로 돌발상황 정의와 알고리즘 개발로 시스템을 구축하고 테스트베드를 운영하여 돌발상황 알고리즘 검증과 실증에 활용할 수 있는 방안을 제시하였으며, 돌발상황 발생 시 예측 가능한 사고를 줄일 수 있는 2차 사고에 대한 효과와 예측 불가능한 사고의 검지 시간을 줄여 부상자에 대한 골든타임 확보할 것으로 기대된다.

Keywords

References

  1. C. Kwon, "Development of Automatic Incident Detection Model for Highway Using Machine Learning." Master's Thesis, Ajou University, 2019.
  2. Y. Kim, "Development of a Modified McMaster Incident Detection Algorithm by State Moving Distance Technique." Master's Thesis, Ajou University, 2008.
  3. S. Kim, "A Study on the Classification of Traffic Flow Areas for Detecting Sudden Situations." of Korean Society of Transportation, vol. 24, no. 3, 2006, pp. 39-50.
  4. S. Han, "Introduction of Algorithms for Detecting Sudden Situations in Tunnels." of Korean Society of Transportation, vol. 6, no. 3, 2009, pp. 141-149.
  5. J. Choi, "Development of an automatic detection algorithm for continuous flow abrupt situations using a simple arithmetic." Master's Thesis, Kyonggi University, 2011.
  6. X. Jin, R. Cheu, and D. Srinivasan, "Development and adaptation of constructive probabilistic neural network in freeway incident detection." Transportation Research Part C: Emerging Technologies, vol. 10, no. 5, 2001, pp.1173-1187.
  7. P. Chakraborty, C. Hegde, A. sharma, "Data-driven parallelizable traffic incident detection using spatio-temporally denoised robust thresholds." Transportation Research Part C: Emerging Technologies, vol. 105, 2019, pp. 81-99. https://doi.org/10.1016/j.trc.2019.05.034
  8. F. Yuan and R. Cheu, "Incident detection using support vector machines." Transportation Research Part C: Emerging Technologies, vol. 11, no. 3, 2003, pp. 309-328. https://doi.org/10.1016/S0968-090X(03)00020-2
  9. Y. Asakura, T. Kusakabe, X. Nguyen, and T. Ushiki, "Incident detection methods using probe vehicles with on-board GPS equipment." Transportation Research Part C, vol. 81, 2017, pp.330-341 https://doi.org/10.1016/j.trc.2016.11.023
  10. Q. Liu, E. Chung, and L. Zhai, "Fusing moving average model and stationary wavelet decomposition for automatic incident detection: case study of Tokyo Expressway." Journal of Traffic and Transportation Engineering, vol. 1, no. 6, 2014, pp.404-414 https://doi.org/10.1016/S2095-7564(15)30290-7