T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction

Accurate and real-time traffic forecasting plays an important role in the intelligent traffic system and is of great significance for urban traffic planning, traffic management, and traffic control. However, traffic forecasting has always been considered an “open” scientific issue, owing to the constraints of urban road network topological structure and the law of dynamic change with time. To capture the spatial and temporal dependences simultaneously, we propose a novel neural network-based traffic forecasting method, the temporal graph convolutional network (T-GCN) model, which is combined with the graph convolutional network (GCN) and the gated recurrent unit (GRU). Specifically, the GCN is used to learn complex topological structures for capturing spatial dependence and the gated recurrent unit is used to learn dynamic changes of traffic data for capturing temporal dependence. Then, the T-GCN model is employed to traffic forecasting based on the urban road network. Experiments demonstrate that our T-GCN model can obtain the spatio-temporal correlation from traffic data and the predictions outperform state-of-art baselines on real-world traffic datasets. Our tensorflow implementation of the T-GCN is available at https://www.github.com/lehaifeng/T-GCN.

I. Introduction

With the development of Intelligent Traffic Systems, traffic forecasting has received more and more attention. It is a key part of an advanced traffic management system and is an important part of realizing traffic planning, traffic management, and traffic control. Traffic forecasting is a process of analyzing traffic conditions on urban roads, including flow, speed, and density, mining traffic patterns, and predicting the trends of traffic on roads. Traffic forecasting can provide not only a scientific basis for traffic managers to sense traffic congestion and limit vehicles in advance but also security for urban travelers to choose appropriate travel routes and improve travel efficiency [1]–[3]. However, traffic forecasting has always been a challenge task due to its complex spatial and temporal dependences: (1)

Spatial dependence. The change in traffic volume is dominated by the topological structure of the urban road network. The traffic status at upstream roads impact traffic status at downstream roads through the transfer effect, and the traffic status at downstream roads impact traffic status at upstream through the feedback effect [4]. As shown in Figure 1, due to the strong influence among adjacent roads, the short-term similarity is changed from state ① (the upstream road is similar to the midstream road) to state ② (the upstream road is similar to the downstream road).

(2)

Temporal dependence. The traffic volume changes dynamically over time and is mainly reflected in periodicity and trend. As shown in Figure 2(a), the traffic volume on the road shows a periodic change over a week. As shown in Figure 2(b), the traffic volume in one day changes over time; for example, the present traffic volume is affected by the traffic condition of the previous moment or even longer.

Fig. 1.

Spatial dependence is restricted by the topological structure of the road network. Due to the strong influence between adjacent roads, the short-term traffic flow similarity is changed from state ① to state ②.

Fig. 2.

(a) Periodicity. The traffic volume in the road changes periodically within one week. (b) Trend. The traffic volume in the road has tendency change within one day.