Autonomous vehicles need to solve the road keeping problem and the existing solutions based on reinforcement learning are mainly implemented in the simulators. The key of transferring the well-trained models to the real world is bridging the gaps between the simulator scenarios and the real scenarios. In this paper, we propose a method called ReinforcementDriving which explores navigation skills and trajectories from simulator for full-sized road keeping. Based on the real scenario, a driving simulator is firstly established to train an intelligent driving agent. The well-trained ReinforcementDriving agent is evaluated in a real-world scenario. We compare our work with human driving, optimal control-based tracking methods and other reinforcement learning-based lane following methods. The results demonstrate that the ReinforcementDriving system can effectively achieve lane keeping in a realistic scenario with satisfactory running time and lateral accuracy.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    ReinforcementDriving: Exploring Trajectories and Navigation for Autonomous Vehicles


    Beteiligte:
    Liu, Meng (Autor:in) / Zhao, Fei (Autor:in) / Niu, Jianwei (Autor:in) / Liu, Yu (Autor:in)


    Erscheinungsdatum :

    01.02.2021


    Format / Umfang :

    3687469 byte




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Generating trajectories for autonomous vehicles

    DUPRE GUILLAUME / CHARROW BENJAMIN | Europäisches Patentamt | 2022

    Freier Zugriff


    Generating trajectories for autonomous vehicles

    DUPRE GUILLAUME / CHARROW BENJAMIN | Europäisches Patentamt | 2024

    Freier Zugriff

    GENERATING TRAJECTORIES FOR AUTONOMOUS VEHICLES

    DUPRE GUILLAUME / CHARROW BENJAMIN | Europäisches Patentamt | 2020

    Freier Zugriff

    GENERATING TRAJECTORIES FOR AUTONOMOUS VEHICLES

    DUPRE GUILLAUME / CHARROW BENJAMIN | Europäisches Patentamt | 2025

    Freier Zugriff