In this master thesis project a LiDAR-based, depth image-based and semantic segmentation image-based reinforcement learning agent is investigated and compared forlearning in simulation and performing in real-time. The project utilize the Deep Deterministic Policy Gradient architecture for learning continuous actions and was designed to control a RC car. One of the first project to deploy an agent in a real scenario after training in a similar simulation. The project demonstrated that with a proper reward function and by tuning driving parameters such as restricting steering, maximum velocity, minimum velocity and performing input data scaling a LiDAR-based agent could drive indefinitely on a simple but completely unseen track in real-time.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Domain Transfer for End-to-end Reinforcement Learning


    Beteiligte:
    Olsson, Anton (Autor:in) / Rosberg, Felix (Autor:in)

    Erscheinungsdatum :

    2020-01-01


    Medientyp :

    Hochschulschrift


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Klassifikation :

    DDC:    629



    Reinforcement Learning in the Multi-Robot Domain

    Mataric, M. J. | British Library Online Contents | 1997


    Learning and Transfer of Movement Gaits Using Reinforcement Learning

    Waidner, David / Strand, Marcus | TIBKAT | 2022



    Transfer of reinforcement learning for a robotic skill

    Gómez Rosal, Dulce Adriana | BASE | 2018

    Freier Zugriff

    Modular reinforcement learning : a case study in a robot domain

    Kalmár, Zsolt / Szepesvári, Csaba / Lőrincz, András | BASE | 2000

    Freier Zugriff