This paper presents an air traffic management framework to enable multiple fleets of unmanned aerial vehicles to traverse dense, omni-directional air traffic safely and efficiently. The main challenge addressed here is separation assurance in the absence of full coordination and communication. In this framework, each fleet is independently managed by a routing agent, which progressively plans the non-overlapping move-ahead corridors for vehicles in the fleet by solving a nonlinear optimization model. The model is artfully designed so that agents of different fleets need not engage in complicated multilateral communications or make guesses about external vehicles’ flight intents to maintain effective inter-vehicle separation. For a complex routing problem, the framework is able to support centralized fleet routing, decentralized vehicle self-routing, and any other agent-vehicle configuration in between, allowing for customized trade-off between response time and traffic efficiency. Innovative algorithmic enhancements for solving the agent’s nonconvex routing problem are prescribed with detailed annotation. The effectiveness and noteworthy properties of the framework are demonstrated by several simulation experiments.


    Access

    Download


    Export, share and cite



    Title :

    A multi-agent semi-cooperative unmanned air traffic management model with separation assurance


    Contributors:
    Yanchao Liu (author)


    Publication date :

    2021




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    Unknown




    Autonomous Separation Assurance with Deep Multi-Agent Reinforcement Learning

    Brittain, Marc W. / Yang, Xuxi / Wei, Peng | AIAA | 2021





    A multi-agent approach to cooperative traffic management and route guidance

    Adler, Jeffrey L. / Satapathy, Goutam / Manikonda, Vikram et al. | Elsevier | 2004