As hundreds of Unmanned Aircraft System (UAS) operate within urban airspaces, automated and decentralized UAS traffic management (UTM) will be critical to maintain safe and efficient operations. In this work, we present Learning-to-Fly with Reinforcement Learning (L2F-RL), a decentralized, on-demand Collision Avoidance (CA) framework that systematically combines machine learning with cooperative model predictive control for UAS collision avoidance while retaining satisfaction of higher-level mission objectives. L2F-RL consists of: 1) RL-based policy for conflict resolution (CR) with discrete-decision making, 2) decentralized, cooperative model predictive control for CA. To accelerate training with RL, we utilize reward shaping and curriculum learning. Our approach outperforms baseline approaches with a 99.10% separation rate (ratio of success to total test cases) in the worst case, improving to 100% in the best case with a 1000X improvement in computation time compared to centralized methods. Our results demonstrate the potential of combining learning approaches with optimization-based control, making it a significant contribution towards scalable, decentralized UTM.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Learning-to-Fly RL: Reinforcement Learning-based Collision Avoidance for Scalable Urban Air Mobility


    Beteiligte:
    Jang, Kuk (Autor:in) / Pant, Yash Vardhan (Autor:in) / Rodionova, Alena (Autor:in) / Mangharam, Rahul (Autor:in)


    Erscheinungsdatum :

    11.10.2020


    Format / Umfang :

    1151179 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Learning-to-Fly: Learning-based Collision Avoidance for Scalable Urban Air Mobility

    Rodionova, Alena / Pant, Yash Vardhan / Jang, Kuk et al. | IEEE | 2020


    REINFORCEMENT LEARNING-BASED MID-AIR COLLISION AVOIDANCE

    OSIPYCHEV DENIS / MARGINEANTU DRAGOS D | Europäisches Patentamt | 2023

    Freier Zugriff

    Pedestrian Collision Avoidance Using Deep Reinforcement Learning

    Rafiei, Alireza / Fasakhodi, Amirhossein Oliaei / Hajati, Farshid | Springer Verlag | 2022


    ICAS2014_0892: A REINFORCEMENT LEARNING BASED UAVS AIR COLLISION AVOIDANCE

    Yang, J. / Yin, D. / Xie, H. | British Library Conference Proceedings | 2014


    Vision-guided Collision Avoidance Through Deep Reinforcement Learning

    Song, Sirui / Zhang, Yuanhang / Qin, Xi et al. | IEEE | 2021