Bridging Traffic State and Trajectory for Dynamic Road Network and Trajectory Representation Learning

Chengkai Han, Jingyuan Wang, Wang Yongyao, Xie Yu, Hao Lin, Li Chao, Junjie Wu

Abstract: Effective urban traffc management is vital for sustainable city development, relying on intelligent systems with machine learning tasks such as traffc fow prediction and travel time estimation. Traditional approaches usually focus on static road network and trajectory representation learning, and overlook the dynamic nature of traffc states and trajectories, which is crucial for downstream tasks. To address this gap, we propose TRACK, a novel framework to bridge traffc state and trajectory data for dynamic road network and trajectory representation learning. TRACK leverages graph attention networks (GAT) to encode static and spatial road segment features, and introduces a transformer-based model for trajectory representation learning. By incorporating transition probabilities from trajectory data into GAT attention weights, TRACK captures dynamic spatial features of road segments. Meanwhile, TRACK designs a traffc transformer encoder to capture the spatial-temporal dynamics of road segments from traffc state data. To further enhance dynamic representations, TRACK proposes a co-attentional transformer encoder and a trajectory-traffc state matching task. Extensive experiments on real-life urban traffc datasets demonstrate the superiority of TRACK over state-of-the-art baselines. Case studies confrm TRACK’s ability to capture spatial-temporal dynamics effectively.