The invention discloses an aircraft real-time collaborative flight path planning method based on dual deep Q learning, which comprises the following steps: firstly, constructing a deep reinforcement learning model based on environment-agent interaction, and designing a dual deep Q learning algorithm with a reviewer network to train a flight path planning artificial intelligence body, so that the flight path planning artificial intelligence body can complete real-time collaborative flight path planning in a random dynamic cumulonage cloud scene; and the two aircrafts with any position, course and track intention cooperate with each other in real time to plan the track. And then designing a heuristic method to convert a multi-aircraft cooperative flight path planning problem in an airspace into a multi-time cooperative flight path planning problem of two aircrafts, obtaining a cooperative flight path planning algorithm of polynomial calculation time complexity, and performing cooperative flight path planning by the trained intelligent agent. The method aims at reducing the workload of a controller and improving the automation, collaboration and intelligence level of flight path planning in the tactical operation stage of the aircraft.

    本发明公开了一种基于双重深度Q学习的航空器实时协同航迹规划方法,首先构建基于环境‑智能体交互的深度强化学习模型,设计带评论者网络的双重深度Q学习算法训练航迹规划人工智能体,使其能够完成随机动态积雨云场景下,任意位置、航向、航迹意图的两架航空器实时协同航迹规划任务。然后设计启发式方法将空域内多航空器协同航迹规划问题转换为多次两架航空器协同航迹规划问题,获得多项式计算时间复杂度的协同航迹规划算法,并由训练后智能体进行协同航迹规划。该方法旨在降低管制员工作负荷,提升航空器战术运行阶段航迹规划自动化、协同化、智能化水平。


    Access

    Download


    Export, share and cite



    Title :

    Aircraft real-time collaborative flight path planning method based on dual deep Q learning


    Additional title:

    一种基于双重深度Q学习的航空器实时协同航迹规划方法


    Contributors:
    ZHANG HONGHAI (author) / ZHOU JINLUN (author) / LI YIKE (author) / SHI ZONGBEI (author) / HUA MINGZHUANG (author)

    Publication date :

    2024-04-05


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G01C Messen von Entfernungen, Höhen, Neigungen oder Richtungen , MEASURING DISTANCES, LEVELS OR BEARINGS / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS