The Journal of The Korea Institute of Intelligent Transport Systems (한국ITS학회 논문지)

The Korea Institute of Intelligent Transport Systems (한국ITS학회)
권호 표지
DOI QR코드

DOI QR Code

Development of Predictive Pedestrian Collision Warning Service Considering Pedestrian Characteristics

보행자 특성을 고려한 예측형 보행자 충돌 경고 서비스 개발

  • Ka, Dongho (Dept. of Civil & Environment Eng, KAIST) ;
  • Lee, Donghoun (Center for Connected & Automated Driving Research, KOTI) ;
  • Yeo, Hwasoo (Dept. of Civil & Environment Eng, KAIST)
  • 가동호 (한국과학기술원 건설 및 환경공학과) ;
  • 이동훈 (한국교통연구원 자율협력주행연구센터) ;
  • 여화수 (한국과학기술원 건설 및 환경공학과)
  • Received : 2019.05.22
  • Accepted : 2019.06.25
  • Published : 2019.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

The number of pedestrian traffic accident fatalities is three times the number of car accidents in South Korea. Serious accidents are caused especially at intersections when the vehicle turns to their right. Various pedestrian collision warning services have been developed, but they are insufficient to prevent dangerous pedestrians. In this study, P2CWS is developed to warn approaching vehicles based on the pedestrians' characteristics. In order to evaluate the performance of the service, actual pedestrian data were collected at the intersection of Daejeon, and comparative analysis was carried out according to pedestrian characteristics. As a result, the performance analysis showed a higher accordance when the characteristics of the pedestrian is considered. Accordingly, we can conclude that identifying pedestrian characteristics in predicting the pedestrian crossing is important.

국내 보행자 교통사고 건당 사망자수는 차대차 사고의 3배에 달한다. 해당 사고의 약 40%가 횡단 중 발생하며 특히 교차로에서는 차량의 우회전시 보행자-차량간 상충 가능성이 높기에 심각한 사고를 초래할 수 있다. 이에 다양한 보행자 충돌 경고 서비스가 개발되었지만 교차로에서 돌발적인 행동을 하는 보행자와 차량의 충돌을 막기에는 역부족이었다. 이에 본 연구에서는 횡단 이전의 보행자들을 관찰하고 추출된 보행자의 특성을 토대로 보행자의 횡단여부를 예측하여 접근 차량에 경고하는 예측형 보행자 충돌 경고 서비스(P2CWS, Predictive Pedestrian Collision Warning Service)를 개발하였다. 서비스 성능 평가를 위해 대전광역시 유성구 교차로에서 실제 보행자 데이터를 수집하였고 보행자 특성(나이, 성별, 회두여부)의 유무에 따른 비교 분석을 수행하였다. 분석 결과 보행자 특성을 반영한 서비스가 반영하지 않은 서비스보다 성능이 뛰어났으며 이로써 보행자의 횡단 여부를 예측하는데 보행자의 특성을 파악하는 것의 중요성을 확인하게 되었다.

Keywords

References

  1. Bradley A.(1997), "The Use of the Area under the ROC Curve in the Evaluation of Machine Learning Algorithms," Pattern Recognition, vol. 30, no. 7, pp.1145-1159. https://doi.org/10.1016/S0031-3203(96)00142-2
  2. Dalal N. and Triggs W.(2004), "Histograms of Oriented Gradients for Human Detection," 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition CVPR05, vol. 1, no. 3, pp.886-893.
  3. Flohr F., Dumitru-Guzu M., Kooij J. F. P. and Gavrila D. M.(2014), "A Probabilistic Framework for Joint Pedestrian Head and Body Orientation Estimation," Proceedings of the IEEE Intelligent Vehicles Symposium, vol. 16, no. 4, pp.617-622.
  4. Gandomi A. H., Yang X. S. and Alavi A. H.(2013), "Cuckoo Search Algorithm: A Metaheuristic Approach to Solve Structural Optimization Problems," Engineering with Computers, vol. 29, no. 1, pp.17-35. https://doi.org/10.1007/s00366-011-0241-y
  5. Intelligent Transport Society of Korea(2017), Next Generation ITS Service Standard Part 1. Features and Performance Requirements, p.57.
  6. Kongsorot Y. and Horata P.(2014), "Multi-Label Classification with Extreme Learning Machine," In Proceedings of the 6th International Conference on Knowledge and Smart Technology (KST), pp.81-86.
  7. Kotte J., Schmeichel C., Zlocki A., Gathmann H. and Eckstein L.(2017), "Concept of an Enhanced V2X Pedestrian Collision Avoidance System with a Cost Function-Based Pedestrian Model," Traffic Injury Prevention, vol. 18, pp.S37-S43. https://doi.org/10.1080/15389588.2017.1310380
  8. Lee D. and Yeo H.(2015), "A Study on the Rear-End Collision Warning System by Considering Different Perception Reaction Time using Multi-Layer Perceptron Neural Network," IEEE Intelligent Vehicles Symposium, pp.24-30.
  9. Redmon J., Divvala S., Girshick R. and Farhadi A.(2015), "You Only Look Once: Unified, Real-Time Object Detection," IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV.
  10. Rehder E., Kloeden H. and Stiller C.(2014), "Head Detection and Orientation Estimation for Pedestrian Safety," 2014 IEEE International Conference on Intelligent Transportation Systems, Qingdao, China, pp.2292-2297.
  11. Ren S., He K., Girshick R. and Sun J.(2017), "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp.1137-1149. https://doi.org/10.1109/TPAMI.2016.2577031
  12. Traffic Accident Analysis System in Korea(2019.05.01), Statistics of Traffic Accident, http://taas.koroad.or.kr
  13. Wojke N., Bewley A. and Paulus D.(2018), "Simple Online and Realtime Tracking with a Deep Association Metric," 2018 International Conference on Image Processing, pp.3645-3649.
  14. Zebala J., Ciepka P. and Reza A.(2012)., "Pedestrian Acceleration and Speeds," Problems of Forensic Sciences, vol. 91, pp.227-234.
  15. Zhu Q. Y., Qin A. K., Suganthan P. N. and G.(2005), "Bin Huang. Evolutionary Extreme Learning Machine," Pattern Recognition, vol. 38, no. 10, pp.1759-1763. https://doi.org/10.1016/j.patcog.2005.03.028