Structural Recurrent Neural Network for Traffic Speed Prediction

TL;DR

This paper introduces a structural RNN that leverages road network topology to improve traffic speed prediction, reducing model complexity and outperforming existing methods on real-world data.

Contribution

The novel integration of road network topology into a spatio-temporal RNN significantly enhances traffic prediction accuracy with fewer parameters.

Findings

Outperforms state-of-the-art methods in traffic speed prediction

Requires fewer parameters to achieve high accuracy

Effectively captures spatio-temporal dependencies in traffic data

Abstract

Deep neural networks have recently demonstrated the traffic prediction capability with the time series data obtained by sensors mounted on road segments. However, capturing spatio-temporal features of the traffic data often requires a significant number of parameters to train, increasing computational burden. In this work we demonstrate that embedding topological information of the road network improves the process of learning traffic features. We use a graph of a vehicular road network with recurrent neural networks (RNNs) to infer the interaction between adjacent road segments as well as the temporal dynamics. The topology of the road network is converted into a spatio-temporal graph to form a structural RNN (SRNN). The proposed approach is validated over traffic speed data from the road network of the city of Santander in Spain. The experiment shows that the graph-based method outperforms the state-of-the-art methods based on spatio-temporal images, requiring much fewer parameters to train.