The invention relates to the technical field of aerospace, in particular to application in space orbit game, and discloses a pulse type orbit pursuit game method based on a PRD-MADDPG algorithm, modeling is carried out through a pulse type orbit pursuit game problem, and aiming at task targets of spacecrafts of two parties in the pulse type orbit pursuit game, the pulse type orbit pursuit game problem is solved. The method comprises the following steps: designing a game model and a reward function of a pulse type track pursuit game, designing a predictive reward detection training framework based on the designed game model and the reward function, and finishing pursuit game intelligent control strategy network training based on the designed predictive reward detection training framework in combination with an MADDPG algorithm. A control instruction is output according to observation information of the spacecraft on the environment, and intelligent control over the pulse type track pursuit game of the spacecraft is achieved.

    本发明涉及航空航天技术领域,特别是涉及在空间轨道博弈的应用,公开了一种基于PRD‑MADDPG算法的脉冲式轨道追逃博弈方法,通过脉冲式轨道追逃博弈问题建模,并针对脉冲式轨道追逃博弈中双方航天器的任务目标,设计脉冲式轨道追逃博弈双方的奖励函数,基于所设计的博弈模型和奖励函数,设计预测奖励检测训练框架,基于所设计的预测奖励检测训练框架,结合MADDPG算法完成追逃博弈智能控制策略网络训练,航天器利用训练好的策略网络,根据自身对环境的观测信息输出控制指令,实现航天器脉冲式轨道追逃博弈的智能控制。


    Access

    Download


    Export, share and cite



    Title :

    Pulse type track pursuit game method based on PRD-MADDPG algorithm


    Additional title:

    一种基于PRD-MADDPG算法的脉冲式轨道追逃博弈方法


    Contributors:
    ZHAO LIRAN (author) / DANG CHAOHUI (author) / TANG SHENGYONG (author) / WEI GUONING (author) / XU XUSHENG (author)

    Publication date :

    2022-11-11


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    B64G Raumfahrt , COSMONAUTICS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen





    Research on autonomous formation of Multi-UAV based on MADDPG algorithm

    Zhang, Yaozhong / Wu, Zhuoran / Ma, Yunhong et al. | IEEE | 2022



    Multi-vehicle joint longitudinal control method based on improved MADDPG

    LI ZHIBIN / LI GEYANG | European Patent Office | 2023

    Free access