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.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    ReinforcementDriving: Exploring Trajectories and Navigation for Autonomous Vehicles


    Contributors:
    Liu, Meng (author) / Zhao, Fei (author) / Niu, Jianwei (author) / Liu, Yu (author)


    Publication date :

    2021-02-01


    Size :

    3687469 byte




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English



    GENERATING TRAJECTORIES FOR AUTONOMOUS VEHICLES

    DUPRE GUILLAUME / CHARROW BENJAMIN | European Patent Office | 2020

    Free access

    Generating trajectories for autonomous vehicles

    DUPRE GUILLAUME / CHARROW BENJAMIN | European Patent Office | 2022

    Free access

    GENERATING TRAJECTORIES FOR AUTONOMOUS VEHICLES

    DUPRE GUILLAUME / CHARROW BENJAMIN | European Patent Office | 2022

    Free access


    Selecting trajectories for controlling autonomous vehicles

    AHMED SAMI / O'DEA KEVIN A / PORTER AUDREY D et al. | European Patent Office | 2023

    Free access