By interpreting a traffic scene as a graph of interacting vehicles, we gain a flexible abstract representation which allows us to apply Graph Neural Network (GNN) models for traffic prediction. These naturally take interaction between traffic participants into account while being computationally efficient and providing large model capacity. We evaluate two state-of-the art GNN architectures and introduce several adaptations for our specific scenario. We show that prediction error in scenarios with much interaction decreases by 30 % compared to a model that does not take interactions into account. This suggests that interaction is important, and shows that we can model it using graphs. This makes GNNs a worthwhile addition to traffic prediction systems.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Graph Neural Networks for Modelling Traffic Participant Interaction


    Beteiligte:
    Diehl, Frederik (Autor:in) / Brunner, Thomas (Autor:in) / Le, Michael Truong (Autor:in) / Knoll, Alois (Autor:in)


    Erscheinungsdatum :

    2019-06-01


    Format / Umfang :

    432349 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    ACTIVE TRAFFIC PARTICIPANT

    WEINSTEIN-RAUN BENJAMIN | Europäisches Patentamt | 2018

    Freier Zugriff

    ACTIVE TRAFFIC PARTICIPANT

    WEINSTEIN-RAUN BENJAMIN | Europäisches Patentamt | 2018

    Freier Zugriff

    Weak traffic participant early warning method and weak traffic participant early warning device

    JI XIAODONG / HU HAILONG | Europäisches Patentamt | 2024

    Freier Zugriff

    Determining I2X traffic-participant criticality

    ADIREDDY GANESH / BOLDUC ANDREW PHILLIP | Europäisches Patentamt | 2021

    Freier Zugriff

    Synthesizing Traffic Datasets Using Graph Neural Networks

    Rodriguez-Criado, Daniel / Chli, Maria / Manso, Luis J. et al. | IEEE | 2023