skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Two-Stream Multi-Channel Convolutional Neural Network for Multi-Lane Traffic Speed Prediction Considering Traffic Volume Impact

Abstract

Traffic speed prediction is a critically important component of intelligent transportation systems. Recently, with the rapid development of deep learning and transportation data science, a growing body of new traffic speed prediction models have been designed that achieved high accuracy and large-scale prediction. However, existing studies have two major limitations. First, they predict aggregated traffic speed rather than lane-level traffic speed; second, most studies ignore the impact of other traffic flow parameters in speed prediction. To address these issues, the authors propose a two-stream multi-channel convolutional neural network (TM-CNN) model for multi-lane traffic speed prediction considering traffic volume impact. In this model, the authors first introduce a new data conversion method that converts raw traffic speed data and volume data into spatial–temporal multi-channel matrices. Then the authors carefully design a two-stream deep neural network to effectively learn the features and correlations between individual lanes, in the spatial–temporal dimensions, and between speed and volume. Accordingly, a new loss function that considers the volume impact in speed prediction is developed. A case study using 1-year data validates the TM-CNN model and demonstrates its superiority. This paper contributes to two research areas: (1) traffic speed prediction, and (2) multi-lane traffic flow study.

Authors:
 [1]; ORCiD logo [2];  [1];  [3]
  1. Univ. of Washington, Seattle, WA (United States)
  2. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
  3. Univ. of Washington, Seattle, WA (United States); Pacific U.S. DOT Univ. Transportation Center for Federal Region 10, Seattle, WA (United States)
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1648914
Grant/Contract Number:  
AC05-00OR22725
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Transportation Research Record
Additional Journal Information:
Journal Volume: 2674; Journal Issue: 4; Journal ID: ISSN 0361-1981
Publisher:
National Academy of Sciences, Engineering and Medicine
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING

Citation Formats

Ke, Ruimin, Li, Wan, Cui, Zhiyong, and Wang, Yinhai. Two-Stream Multi-Channel Convolutional Neural Network for Multi-Lane Traffic Speed Prediction Considering Traffic Volume Impact. United States: N. p., 2020. Web. doi:10.1177/0361198120911052.
Ke, Ruimin, Li, Wan, Cui, Zhiyong, & Wang, Yinhai. Two-Stream Multi-Channel Convolutional Neural Network for Multi-Lane Traffic Speed Prediction Considering Traffic Volume Impact. United States. doi:10.1177/0361198120911052.
Ke, Ruimin, Li, Wan, Cui, Zhiyong, and Wang, Yinhai. Sat . "Two-Stream Multi-Channel Convolutional Neural Network for Multi-Lane Traffic Speed Prediction Considering Traffic Volume Impact". United States. doi:10.1177/0361198120911052.
@article{osti_1648914,
title = {Two-Stream Multi-Channel Convolutional Neural Network for Multi-Lane Traffic Speed Prediction Considering Traffic Volume Impact},
author = {Ke, Ruimin and Li, Wan and Cui, Zhiyong and Wang, Yinhai},
abstractNote = {Traffic speed prediction is a critically important component of intelligent transportation systems. Recently, with the rapid development of deep learning and transportation data science, a growing body of new traffic speed prediction models have been designed that achieved high accuracy and large-scale prediction. However, existing studies have two major limitations. First, they predict aggregated traffic speed rather than lane-level traffic speed; second, most studies ignore the impact of other traffic flow parameters in speed prediction. To address these issues, the authors propose a two-stream multi-channel convolutional neural network (TM-CNN) model for multi-lane traffic speed prediction considering traffic volume impact. In this model, the authors first introduce a new data conversion method that converts raw traffic speed data and volume data into spatial–temporal multi-channel matrices. Then the authors carefully design a two-stream deep neural network to effectively learn the features and correlations between individual lanes, in the spatial–temporal dimensions, and between speed and volume. Accordingly, a new loss function that considers the volume impact in speed prediction is developed. A case study using 1-year data validates the TM-CNN model and demonstrates its superiority. This paper contributes to two research areas: (1) traffic speed prediction, and (2) multi-lane traffic flow study.},
doi = {10.1177/0361198120911052},
journal = {Transportation Research Record},
issn = {0361-1981},
number = 4,
volume = 2674,
place = {United States},
year = {2020},
month = {3}
}

Journal Article:
Free Publicly Available Full Text
This content will become publicly available on March 7, 2021
Publisher's Version of Record

Save / Share:

Works referenced in this record:

A behavioral theory of multi-lane traffic flow. Part II: Merges and the onset of congestion
journal, February 2002


Multilane traffic flow dynamics: Some macroscopic considerations
journal, August 1984

  • Michalopoulos, Panos G.; Beskos, Dimitrios E.; Yamauchi, Yasuji
  • Transportation Research Part B: Methodological, Vol. 18, Issue 4-5
  • DOI: 10.1016/0191-2615(84)90019-5

Deep learning
journal, May 2015

  • LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey
  • Nature, Vol. 521, Issue 7553
  • DOI: 10.1038/nature14539

A Hierarchy of Models for Multilane Vehicular Traffic I: Modeling
journal, January 1998


Short-term inter-urban traffic forecasts using neural networks
journal, March 1997


Multilane first-order traffic flow model with endogenous representation of lane-flow equilibrium
journal, October 2015

  • Shiomi, Yasuhiro; Taniguchi, Tomoki; Uno, Nobuhiro
  • Transportation Research Part C: Emerging Technologies, Vol. 59
  • DOI: 10.1016/j.trc.2015.07.002

A Cost-Effective Framework for Automated Vehicle-Pedestrian Near-Miss Detection Through Onboard Monocular Vision
conference, July 2017

  • Ke, Ruimin; Lutin, Jerome; Spears, Jerry
  • 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
  • DOI: 10.1109/CVPRW.2017.124

Traffic Prediction Using Multivariate Nonparametric Regression
journal, March 2003


Macroscopic dynamics of multilane traffic
journal, June 1999


Real-Time Bidirectional Traffic Flow Parameter Estimation From Aerial Videos
journal, April 2017

  • Ke, Ruimin; Li, Zhibin; Kim, Sung
  • IEEE Transactions on Intelligent Transportation Systems, Vol. 18, Issue 4
  • DOI: 10.1109/TITS.2016.2595526

Flexible and Robust Method for Missing Loop Detector Data Imputation
journal, January 2015

  • Henrickson, Kristian; Zou, Yajie; Wang, Yinhai
  • Transportation Research Record: Journal of the Transportation Research Board, Vol. 2527, Issue 1
  • DOI: 10.3141/2527-04

Traffic flow data compression considering burst components
journal, November 2017


Innovative method for traffic data imputation based on convolutional neural network
journal, April 2019


A comparison of the performance of artificial. neural networks and support vector machines for the prediction of traffic speed
conference, January 2004


Short Term Prediction of Traffic Parameters Using Support Vector Machines Technique
conference, November 2010

  • Theja, P. V. V. K.; Vanajakshi, L.
  • Third International Conference on Emerging Trends in Engineering and Technology (ICETET 2010), 2010 3rd International Conference on Emerging Trends in Engineering and Technology
  • DOI: 10.1109/ICETET.2010.37

Traffic Flow Prediction using Kalman Filtering Technique
journal, January 2017


Spatiotemporal Recurrent Convolutional Networks for Traffic Prediction in Transportation Networks
journal, June 2017

  • Yu, Haiyang; Wu, Zhihai; Wang, Shuqin
  • Sensors, Vol. 17, Issue 7
  • DOI: 10.3390/s17071501

Traffic speed prediction for urban transportation network: A path based deep learning approach
journal, March 2019

  • Wang, Jiawei; Chen, Ruixiang; He, Zhaocheng
  • Transportation Research Part C: Emerging Technologies, Vol. 100
  • DOI: 10.1016/j.trc.2019.02.002