Traffic flow prediction based on dynamic spatial enhancement

Shumei Bao; Cheng Chen; Boyan Huo; Xiaoyi Lv; Chen Chen

doi:10.1117/12.3039888

28 August 2024 Traffic flow prediction based on dynamic spatial enhancement

Shumei Bao, Cheng Chen, Boyan Huo, Xiaoyi Lv, Chen Chen

Author Affiliations +

Proceedings Volume 13251, Ninth International Conference on Electromechanical Control Technology and Transportation (ICECTT 2024); 132516A (2024) https://doi.org/10.1117/12.3039888
Event: 9th International Conference on Electromechanical Control Technology and Transportation (ICECTT 2024), 2024, Guilin, China

Abstract

Traffic flow prediction is a spatiotemporal prediction task. The biggest difference between traffic flow and ordinary time series prediction problems is that it is restricted by the topology of the road network. Therefore, it is necessary to accurately mine the spatial information in the road network. This paper introduces a spatial block designed to obtain spatial correlation, incorporating both Graph Convolutional Networks (GCN) and a novel spatial transformer module for handling spatial properties. By leveraging spatial transformers, the model can extract dynamic spatial correlations among individual nodes. The dynamic spatial information and static spatial information are integrated through the gating mechanism. In addition, this chapter also adds the Informer module based on GRU to mine long-term sequence information to improve the forgetting problem of using the GRU module. The final experimental results were verified on the London M25 highway data set and compared with other benchmark methods and found to achieve good results.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Shumei Bao, Cheng Chen, Boyan Huo, Xiaoyi Lv, and Chen Chen "Traffic flow prediction based on dynamic spatial enhancement", Proc. SPIE 13251, Ninth International Conference on Electromechanical Control Technology and Transportation (ICECTT 2024), 132516A (28 August 2024); https://doi.org/10.1117/12.3039888

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

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
8 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Data modeling

Performance modeling

Roads

Machine learning

Deep learning

Education and training

Mathematical optimization

Show All Keywords

Keywords/Phrases

Search In:

Publication Years