SUPR
Decentralized Traffic Prediction
Dnr:

NAISS 2024/22-1383

Type:

NAISS Small Compute

Principal Investigator:

Jeannie He

Affiliation:

Kungliga Tekniska högskolan

Start Date:

2024-10-25

End Date:

2025-11-01

Primary Classification:

10201: Computer Sciences

Webpage:

Allocation

Abstract

As an important part of intelligent transport systems, traffic flow prediction has attracted substantial research interest. Here, one challenging yet essential problem is how to provide accurate real-time traffic predictions. This research is about the proposal of a new algorithm for decentralized traffic flow predictions based on edge computing, graph neural networks, and gated recurrent units.