Hierarchical Graph Structures for Congestion and ETA Prediction. (arXiv:2211.11762v1 [cs.LG])

Traffic4cast is an annual competition to predict spatio temporal traffic
based on real world data. We propose an approach using Graph Neural Networks
that directly works on the road graph topology which was extracted from
OpenStreetMap data. Our architecture can incorporate a hierarchical graph
representation to improve the information flow between key intersections of the
graph and the shortest paths connecting them. Furthermore, we investigate how
the road graph can be compacted to ease the flow of information and make …

arxiv congestion graph hierarchical prediction