Driver behavior modeling is an important task for predicting or simulating the evolution of traffic situations. We investigate the use of Adversarial Inverse Reinforcement Learning (AIRL), an IRL-based method, to learning a driving policy from a dataset of real-world trajectories. Compared to the commonly used direct Behavioral Cloning (BC), IRL aims to reconstruct the rewards of drivers, e.g., driving fast but with minimal accelerations. Simultaneously, a policy that maximizes these rewards is learned using standard Reinforcement Learning (RL) methods. This indirection enables us to train AIRL in fictional situations, for which no training trajectories exist. In our experiments, we find that this advantage enables AIRL to produce policies that are significantly more robust than the two competing approaches Generative Adversarial Imitation Learning (GAIL) and BC.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Modeling Driver Behavior using Adversarial Inverse Reinforcement Learning


    Beteiligte:
    Sackmann, Moritz (Autor:in) / Bey, Henrik (Autor:in) / Hofmann, Ulrich (Autor:in) / Thielecke, Jorn (Autor:in)


    Erscheinungsdatum :

    2022-06-05


    Format / Umfang :

    854743 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Predicting Driver Behavior on the Highway with Multi-Agent Adversarial Inverse Reinforcement Learning

    Radtke, Henrik / Bey, Henrik / Sackmann, Moritz et al. | IEEE | 2023



    Multiagent modeling of pedestrian-vehicle conflicts using Adversarial Inverse Reinforcement Learning

    Nasernejad, Payam / Sayed, Tarek / Alsaleh, Rushdi | Taylor & Francis Verlag | 2023


    Autonomous UAV Interception via Augmented Adversarial Inverse Reinforcement Learning

    Wang, Huan / Liu, Xiaofeng / Zhou, Xu | Springer Verlag | 2022