This paper investigates the application of Deep Reinforcement Learning (DRL) to address motion control challenges in drones for additive manufacturing (AM). Dronebased additive manufacturing offers a flexible and autonomous solution for material deposition in large-scale or hazardous environments. However, achieving robust real-time control of a multi-rotor aerial robot under varying payloads and potential disturbances remains challenging. Traditional controllers like PID often require frequent parameter re-tuning, limiting their applicability in dynamic scenarios. We propose a DRL framework that learns adaptable control policies for multirotor drones performing waypoint navigation in AM tasks. We compare Deep Deterministic Policy Gradient (DDPG) and Twin Delayed Deep Deterministic Policy Gradient (TD3) within a curriculum learning scheme designed to handle increasing complexity. Our experiments show TD3 consistently balances training stability, accuracy and success, particularly when mass variability is introduced. These findings provide a scalable path toward robust, autonomous drone control in additive manufacturing.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Motion Control in Multi-Rotor Aerial Robots Using Deep Reinforcement Learning


    Beteiligte:
    Shetty, Gaurav (Autor:in) / Ramezani, Mahya (Autor:in) / Habibi, Hamed (Autor:in) / Voos, Holger (Autor:in) / Sanchez-Lopez, Jose Luis (Autor:in)


    Erscheinungsdatum :

    14.05.2025


    Format / Umfang :

    646761 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Intercepting Unauthorized Aerial Robots in Controlled Airspace Using Reinforcement Learning

    Giral, Francisco / Gómez, Ignacio / Clainche, Soledad Le | ArXiv | 2024

    Freier Zugriff

    NAVIGATING AERIAL VEHICLES USING DEEP REINFORCEMENT LEARNING

    CANDIDO SALVATORE J / GONG JUN / GENDRON-BELLEMARE MARC et al. | Europäisches Patentamt | 2021

    Freier Zugriff


    Multi-agent Deep Reinforcement Learning for Countering Uncrewed Aerial Systems

    Pierre, Jean-Elie / Sun, Xiang / Novick, David et al. | Springer Verlag | 2024