Several key scenarios, such as intersection navigation, lane changing, and ramp merging, are active areas of research in autonomous driving. In order to properly navigate these scenarios, autonomous vehicles must implicitly negotiate with human drivers. Prior work in driving behaviors presents reinforcement learning as a promising technique, as it can leverage data as well as the underlying decision-making structure of driving with interaction. We apply a hierarchical approach to decision-making, where we train a high-level policy using reinforcement learning, and execute the policy's output on a low-level controller. This hierarchical structure helps increase the policy's overall safety, and allows the learning component to be agnostic to the low-level control scheme. We validate our approach on a simulation using real-world highway data and find improved results compared to prior work in ramp merging.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Learning Highway Ramp Merging Via Reinforcement Learning with Temporally-Extended Actions


    Beteiligte:
    Triest, Samuel (Autor:in) / Villaflor, Adam (Autor:in) / Dolan, John M. (Autor:in)


    Erscheinungsdatum :

    2020-10-19


    Format / Umfang :

    393773 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Deep Multi-Agent Reinforcement Learning for Highway On-Ramp Merging in Mixed Traffic

    Chen, Dong / Hajidavalloo, Mohammad R. / Li, Zhaojian et al. | IEEE | 2023



    Anti-Jerk On-Ramp Merging Using Deep Reinforcement Learning

    Lin, Yuan / McPhee, John / Azad, Nasser L. | IEEE | 2020


    A Deep Reinforcement Learning Approach for Automated On-Ramp Merging

    Zhao, Ruibin / Sun, Zhanbo / Ji, Ang | IEEE | 2022


    Combining Reinforcement Learning with Model Predictive Control for On-Ramp Merging

    Lubars, Joseph / Gupta, Harsh / Chinchali, Sandeep et al. | IEEE | 2021