The application of deep reinforcement learning (DRL) techniques in intelligent transportation systems garners significant attention. In this field, reward function design is a crucial factor for DRL performance. Current research predominantly relies on a trial-and-error approach for designing reward functions, lacking mathematical support and necessitating extensive empirical experimentation. Our research uses vehicle velocity control as a case study, build training and test sets, and develop a DRL framework for speed control. This framework examines both single-objective and multi-objective optimization in reward function designs. In single-objective optimization, we introduce “expected optimal velocity” as an optimization objective and analyze how different reward functions affect performance, providing a mathematical perspective on optimizing reward functions. In multi-objective optimization, we propose a reward function design paradigm and validate its effectiveness. Our findings offer a versatile framework and theoretical guidance for developing and optimizing reward functions in DRL, particularly for intelligent transportation systems.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Exploring the design of reward functions in deep reinforcement learning-based vehicle velocity control algorithms


    Weitere Titelangaben:

    Y. HE ET AL.
    TRANSPORTATION LETTERS


    Beteiligte:
    He, Yixu (Autor:in) / Liu, Yang (Autor:in) / Yang, Lan (Autor:in) / Qu, Xiaobo (Autor:in)

    Erschienen in:

    Transportation Letters ; 16 , 10 ; 1338-1352


    Erscheinungsdatum :

    25.11.2024


    Format / Umfang :

    15 pages




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch





    Reinforcement learning reward function in unmanned aerial vehicle control tasks

    Tovarnov, Mikhail S. / Bykov, Nikita V. | ArXiv | 2022

    Freier Zugriff

    Adaptive traffic signal control system using composite reward architecture based deep reinforcement learning

    Jamil, Abu Rafe Md / Ganguly, Kishan Kumar / Nower, Naushin | Wiley | 2020

    Freier Zugriff


    Adaptive traffic signal control system using composite reward architecture based deep reinforcement learning

    Jamil, Abu Rafe Md / Ganguly, Kishan Kumar / Nower, Naushin | IET | 2021

    Freier Zugriff