For the reinforcement learning of agents in virtual environment, a flight simulation model architecture is designed by combining the integration method, flight algorithm, external force calculation, earth model and other factors in flight simulation. This architecture realizes the separation of data and algorithm in the process of flight computation at the software level, and adopts inheritance and pointer registration mechanism to achieve unified and hierarchical management of data related to flight computation. The state transfer model of aircraft is designed to realize the smooth connection of flight states between different stages, which can realize efficient large-scale parallel computation in reinforcement learning. The architecture proposed in this paper provides a flexible and efficient interaction mechanism for enhancing reinforcement learning of aircraft agents.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    A Flight Simulation Model Architecture for Reinforcement Learning


    Weitere Titelangaben:

    Lect. Notes Electrical Eng.


    Beteiligte:
    Chu, Shu-Chuan (Herausgeber:in) / Lin, Jerry Chun-Wei (Herausgeber:in) / Li, Jianpo (Herausgeber:in) / Pan, Jeng-Shyang (Herausgeber:in) / Wang, KaiXuan (Autor:in) / Shen, YuTing (Autor:in) / Zhang, FuQuan (Autor:in) / Zhao, Nan (Autor:in) / Yang, Lijie (Autor:in)

    Kongress:

    International Conference on Genetic and Evolutionary Computing ; 2021 ; Jilin City, China October 21, 2021 - October 23, 2021



    Erscheinungsdatum :

    2022-01-04


    Format / Umfang :

    11 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    DEEP REINFORCEMENT LEARNING BASED AGGRESSIVE FLIGHT TRAJECTORY TRACKER

    Shadeed, Omar / Hasanzade, Mehmet / Koyuncu, Emre | TIBKAT | 2021


    Modular Reinforcement Learning for Autonomous UAV Flight Control

    Jongkwan Choi / Hyeon Min Kim / Ha Jun Hwang et al. | DOAJ | 2023

    Freier Zugriff

    Deep Reinforcement Learning based Aggressive Flight Trajectory Tracker

    Shadeed, Omar / Hasanzade, Mehmet / Koyuncu, Emre | AIAA | 2021


    Learning-to-Dispatch: Reinforcement Learning Based Flight Planning under Emergency

    Zhang, Kai / Yang, Yupeng / Xu, Chengtao et al. | IEEE | 2021


    An HLA based flight simulation architecture

    Huiskamp, Wim / Jense, Hans / Janssen, Henk | AIAA | 2000