Abstract
As a crucial part of the Intelligent Transportation System, traffic forecasting is of great help for traffic management and guidance. However, predicting short-term traffic conditions on a large-scale road network is challenging due to the complex spatio-temporal dependencies found in traffic data. Previous studies used Euclidean proximity or topological adjacency to explore the spatial correlation of traffic flows, but did not consider the higher-order connectivity patterns exhibited in a road network, which have a significant influence on traffic propagation. Meanwhile, traffic sequences display distinct multiple time-frequency properties, yet few researchers have made full use of this resource. To fill this gap, we propose a novel hybrid framework – Wavelet-based Higher-order Spatial-Temporal method (Wavelet-HST) to accurately predict network-scale traffic speeds. Wavelet-HST first uses discrete wavelet transform (DWT) to decompose raw traffic data into several components with different frequency sub-bands. Then a motif-based graph convolutional recurrent neural network (Motif-GCRNN) is proposed to learn the higher-order spatio-temporal dependencies of traffic speeds from low-frequency components, and auto-regressive moving average (ARMA) models are employed to simulate random fluctuations from the high-frequency components. We evaluate the framework on a traffic dataset collected in Chengdu, China, and experimental results demonstrate that Wavelet-HST outperforms six state-of-art prediction methods by an improvement of 7.8% ~10.5% in the root mean square error.
Highlights
Predicting short-term traffic conditions in urban areas is of paramount importance for travel planning and traffic control
With the increase of urban road vehicles and the construction of intelligent transportation system (ITS) [2], rich trajectory and flow data are collected by variety of sensors (e.g. GPS detectors equipped on floating vehicles and loop detectors fixed on roads), which provide information about the infrastructure and data environment enabling traffic prediction [3]
We integrated these road motifs with the graph convolution neural network (GCN) to extract higher-order spatial correlations of traffic speeds, which is different from existing deep learning methods that only consider the geographical proximity in Euclidean space or the low-order topological adjacency on general graphs
Summary
Predicting short-term traffic conditions in urban areas is of paramount importance for travel planning and traffic control. N. Zhang et al.: Wavelet-HST: A Wavelet-Based Higher-Order Spatio-Temporal Framework for Urban Traffic Speed Prediction one-dimensional sequences [4] or transformed road networks as two-dimensional images [5], and used convolutional neural network (CNN) to capture the geographic similarity in local areas. The main contributions of this paper can be summarized as follows: 1) We employ motifs (local sub-graph structures [8]) to define the higher-order connectivity patterns presented in road networks We integrated these road motifs with the graph convolution neural network (GCN) to extract higher-order spatial correlations of traffic speeds, which is different from existing deep learning methods that only consider the geographical proximity in Euclidean space or the low-order topological adjacency on general graphs. K is the number of successive filtering operations or convolutional layers and K localized convolution filters effectively exploit the information from the K-1-order neighborhood of a node
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
