Human driving depends on latent states, such as aggression and intent, that cannot be directly observed. In this work, we propose a method for learning driver models that can account for unobserved states. When trained on a synthetic dataset, our model is able to learn encodings for vehicle trajectories that distinguish between four distinct classes of driver behavior. Such encodings are learned without any knowledge of the number of driver classes or any objective that directly requires the model to learn encodings for each class. We show that latent state driving policies outperform baseline methods at replicating driver behavior. Furthermore, we demonstrate that the actions chosen by our policy are heavily influenced by its assigned latent state.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Simultaneous policy learning and latent state inference for imitating driver behavior


    Beteiligte:


    Erscheinungsdatum :

    2017-10-01


    Format / Umfang :

    208098 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Imitating driver behavior with generative adversarial networks

    Kuefler, Alex / Morton, Jeremy / Wheeler, Tim et al. | IEEE | 2017


    Trajectory planning for vehicle collision avoidance imitating driver behavior

    Yang, Bin / Song, Xuewei / Gao, Zhenhai et al. | SAGE Publications | 2022




    Bird-imitating ornithopter

    LI WANGPENG / CAO YUANPENG / XU JIE et al. | Europäisches Patentamt | 2024

    Freier Zugriff