Skip to main content

A Dynamic Spatial-temporal Deep Learning Framework for Traffic Speed Prediction on Large-scale Road Networks.

Zheng, G., Chai, W. K. and Katos, V., 2022. A Dynamic Spatial-temporal Deep Learning Framework for Traffic Speed Prediction on Large-scale Road Networks. Expert Systems with Applications, 195, 116585.

Full text available as:

[img] PDF (Pre-print)
XSWA-Paper.pdf - Accepted Version
Restricted to Repository staff only until 7 February 2023.
Available under License Creative Commons Attribution Non-commercial.

5MB

DOI: 10.1016/j.eswa.2022.116585

Abstract

Traffic prediction plays a crucial role in an intelligent transportation system (ITS) for enabling advanced transportation management and services. In this paper, we address the problem of multi-step traffic speed prediction, includ- ing both short- and long-term predictions. We assert that it is important to consider not just the fixed spatial dependency of the road network (i.e., the connections between road segments) but also the dynamic spatial dependency of traffic within the static topology that intertwines with the temporal evolu- tion of traffic condition across the entire network. We propose a novel deep learning model, named Self-Attention Graph Convolutional Network with Spa- tial, Sub-spatial and Temporal blocks (SAGCN-SST) model, that specifically capture such complex dynamic spatial-temporal processes. In SAGCN-SST, we integrate self-attention mechanism into graph convolutional networks in a novel framework design while using a sequence-to-sequence model in an encoder- decoder architecture for extracting long-temporal dependency of traffic speed. Two real-world datasets with frequent traffic congestion and accidents from large-scale road networks (i.e., Seattle and Los Angeles) are used to train and test our model. Our experiment results indicate that the proposed deep learn- ing model consistently achieves the most accurate predictions (higher than 98% accuracy on both datasets for the short- and long-term predictions) when com- pared against well-known existing models in recent literature. The results also indicate that SAGCN-SST is robust against emergent traffic situations.

Item Type:Article
ISSN:0957-4174
Additional Information:Funded by Information Resilience in Information-Centric Networking. Funded by Smart Transportation Network for Smart City Planning.
Uncontrolled Keywords:Intelligent transportation system,; traffc prediction; deep learning; large-scale road networks
Group:Faculty of Science & Technology
ID Code:36582
Deposited By: Symplectic RT2
Deposited On:04 Feb 2022 16:54
Last Modified:14 Mar 2022 14:32

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -