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

Predicting a Queue Length Using a Deep Learning Model at Signalized Intersections

딥러닝 모형을 이용한 신호교차로 대기행렬길이 예측

  • 나다혁 (아주대학교 교통연구센터) ;
  • 이상수 (아주대학교 교통공학과) ;
  • 조근민 (아주대학교 교통연구센터) ;
  • 김호연 (아주대학교 교통공학과)
  • Received : 2021.11.10
  • Accepted : 2021.12.03
  • Published : 2021.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, a deep learning model for predicting the queue length was developed using the information collected from the image detector. Then, a multiple regression analysis model, a statistical technique, was derived and compared using two indices of mean absolute error(MAE) and root mean square error(RMSE). From the results of multiple regression analysis, time, day of the week, occupancy, and bus traffic were found to be statistically significant variables. Occupancy showed the most strong impact on the queue length among the variables. For the optimal deep learning model, 4 hidden layers and 6 lookback were determined, and MAE and RMSE were 6.34 and 8.99. As a result of evaluating the two models, the MAE of the multiple regression model and the deep learning model were 13.65 and 6.44, respectively, and the RMSE were 19.10 and 9.11, respectively. The deep learning model reduced the MAE by 52.8% and the RMSE by 52.3% compared to the multiple regression model.

본 연구는 영상검지기에서 수집되는 정보를 활용하여 딥러닝 기반으로 대기행렬길이를 예측하는 모형을 개발하였다. 그리고 통계적 기법인 다중회귀 모형을 추정하여 평균절대오차와 평균제곱근오차의 두 지표를 이용하여 비교·평가하였다. 다중회귀분석 결과, 시간, 요일, 점유율, 버스 교통량이 유효한 변수로 도출되었고, 이 중에서 독립변수들의 종속변수에 대한 영향력은 점유율이 가장 큰 것으로 나타났다. 딥러닝 최적 모형은 은닉층이 4겹, Look Back이 6으로 결정되었고, 평균절대오차와 평균제곱근오차가 6.34와 8.99로 나타났다. 그리고 두 모형을 평가한 결과, 다중회귀 모형과 딥러닝 모형의 평균절대오차는 각각 13.65와 6.44, 평균제곱근오차는 각각 19.10과 9.11로 계산되었다. 이는 딥러닝 모형이 다중회귀 모형과 비교하여 평균절대오차가 52.8%, 평균제곱근오차는 52.3% 감소된 결과이다.

Keywords

Acknowledgement

이 논문은 2021년도 정부(경찰청)의 재원으로 과학치안진흥센터의 지원을 받아 수행된 연구임. ( No.092021C28S01000, 자율주행 혼재 시 도로교통 통합관제시스템 및 운영기술 개발)

References

  1. Cho G., Lee S. and Nam D.(2020), "Forecasting of Rental Demand for Public Bicycles using a Deep Learning Model," Journal of the Korean Institute of Intelligent Transport Systems, vol. 19, no. 3, pp.28-37. https://doi.org/10.12815/kits.2020.19.3.28
  2. Do C.(2010), Principles of Transportation Engineering, Gyomoon Publishers.
  3. Fu R., Zhang Z. and Li L.(2016), "Using LSTM and GRU Neural Network Methods for Traffic Flow Prediction," 31st Youth Academic Annual Conference of Chinese Association of Automation, pp.324-328.
  4. Gao K., Huang S., Han F., Li S., Wu W. and Du R.(2020), "An Integrated Algorithm for Intersection Queue Length Estimation Based on IoT in a Mixed Traffic Scenario," Applied Sciences, vol. 10, no. 6, 2078. https://doi.org/10.3390/app10062078
  5. Han J., Kim S. and Choi B.(2000), "Training Sample of Artificial Neural Networks for Predicting Signalized Intersection Queue Length," Proceedings of the KOR-KST Conference, vol. 18, no. 4, p.75.
  6. Kang J. and Oh Y.(2005), "The Development of Traffic Queue Length Measuring Algorithm Using the Occupancy Rates," Proceedings of the KOR-KST Conference, vol. 48, p.309.
  7. Ki B., Han J. and Kim S.(2002), "A Study on Prediction of Queue Length Based on the Neural Network and Genetic Algorithm for a Signalized Intersection," Journal of the Korean Society of Civil Engineering D, vol. 22, no. 4, p.595.
  8. Kim S.(2018), Coding Chef's 3-minute Deep Learning, Kerass flavor, Hanbit Publishing Network.
  9. Kim T.(2017), Python Deep Learning Keras with Blocks, Digitalbooks.
  10. Lee C.(2000), Introducing ATIS to Seoul Metropolitan, Seoul Development Institute.
  11. Lee Y., Son B., Kim H.(2003), "Development of A Simple Method for Determining Queue-EndLocation", Proceedings of the KOR-KST Conference 44, pp.1-6.
  12. Lee Y. and Oh Y.(2002), "A Development of Queue Length Estimation in COSMOS," Proceedings of the KOR-KST Conference, vol. 42, p.1.
  13. Lee Y., Hwang J., Kim S. and Lee C.(2018), "Development of Vehicle Queue Length Estimation Model Using Deep Learning," Journal of the Korean Institute of Intelligent Transport Systems, vol. 17, no. 2, pp.39-57.
  14. Liu H. X., Wu X., Ma W. and Hu H.(2009), "Real-time Queue Length Estimation for Congested Signalized Intersections", Transportation Research Part C, 17(4), p.412. https://doi.org/10.1016/j.trc.2009.02.003
  15. Luo X., Ma D., Jin S., Gong Y. and Wang D.(2019), "Queue Length Estimation for Signalized Intersections Using License Plate Recognition Data," IEEE Intelligent Transportation System Magazine, p.209.
  16. Nikhil B.(2017), Fundamentals of Deep Learning (1st ed.), O'Reilly Media.
  17. Sim S., Lee C. and Choi K.(2002), "A Queue Length Prediction Algorithm using Kalman Filter," Journal of Korean Society of Transportation, vol. 20, no. 5, p.145.
  18. Yim Y. and Ygnace J. L.(1996), "Link Flow Evaluation Using Loop Detector Data: Traveler Response to Variable-Message Signs", TRR, Volume 1550 Issue 1, pp.58-64.
  19. Yin J., Sun J. and Tang K.(2018), "A Kalman Filter-Based Queue Length Estimation Method with Low-Penetration Mobile Sensor Data at Signalized Intersections," TRR, vol. 2672, no. 45, pp.253-264.