Traffic flow prediction based on residual neural network

Yanguo Huang; Xuan He; Luo Li; Zehao Rao

doi:10.1117/12.2686413

10 August 2023 Traffic flow prediction based on residual neural network

Yanguo Huang, Xuan He, Luo Li, Zehao Rao

Proceedings Volume 12759, International Conference on Automation Control, Algorithm, and Intelligent Bionics (ACAIB 2023); 1275932 (2023) https://doi.org/10.1117/12.2686413
Event: 2023 3rd International Conference on Automation Control, Algorithm and Intelligent Bionics (ACAIB 2023), 2023, Xiamen, China

Abstract

With the rapid economic development and the advancement of urbanization, the structure of urban road networks is becoming more and more complex. It is particularly important to provide real-time and accurate traffic flow forecasting for traffic management departments. Aiming at the importance of traffic flow parameter prediction in intelligent transportation systems, this paper proposes a traffic flow prediction method based on the Residual Network (ResNet) model. Through the processing of taxi GPS trajectory data, effective and complete traffic data is obtained. Considering characteristic factors such as weather and holidays, the spatial characteristics of traffic flow are modeled by residual convolution units for different time granularities. The different branches and regions are weighted to predict the final traffic flow for each region. The experimental results show that the design model considers the complex dynamic space-time characteristics of the region, and the predictability is improved compared with other models, which is an effective traffic flow forecasting method.

Citation Download Citation

Yanguo Huang, Xuan He, Luo Li, and Zehao Rao "Traffic flow prediction based on residual neural network", Proc. SPIE 12759, International Conference on Automation Control, Algorithm, and Intelligent Bionics (ACAIB 2023), 1275932 (10 August 2023); https://doi.org/10.1117/12.2686413

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available