Federico Lozano‑Cuadra

Ph.D. Researcher in AI for Space Communications (Fulbright)

Federico works on machine learning for space communications, with a focus on Low Earth Orbit satellite constellations, autonomous lunar rover swarms, and decentralized intelligent networks. His work spans multi-agent reinforcement learning, foundational models, and communication-aware AI for autonomous space systems.

Malaga, Spain · California, USA · Baltimore, USA

University of Malaga · NASA Jet Propulsion Laboratory · Johns Hopkins University ACE Lab

Multicast traffic optimization in mesh satellite constellations

Joint routing and downlink resource allocation for multicast traffic in mesh non-terrestrial networks, aiming to improve delivery efficiency across interconnected LEO constellations. Scalable optimization methods for shared traffic demands in dynamic satellite mesh architectures.

Side-by-side comparison of shortest-path and Steiner-tree multicast routing across a satellite mesh, with sources and destinations highlighted and links colored by traffic load

Journals

  • Continual deep reinforcement learning for decentralized satellite routing

    F. Lozano-Cuadra, B. Soret, I. Leyva-Mayorga, and P. Popovski. IEEE Transactions on Communications, 2025. DOI: 10.1109/TCOMM.2025.3562522.

  • Routing and downlink resource allocation for multicast traffic in LEO satellite constellations

    F. Lozano-Cuadra, I. Leyva-Mayorga, J. J. Nielsen, and B. Soret. International Journal of Satellite Communications and Networking, 2026. DOI: 10.1002/sat.70086.

Conferences

  • Learning decentralized routing policies via graph attention-based multi-agent reinforcement learning in lunar delay-tolerant networks

    F. Lozano-Cuadra, B. Soret, M. Sanchez Net, A. Cauligi, and F. Rossi. Accepted for presentation at the International Symposium on Planetary Robotics (iSpaRo), Sendai, Japan, 2025.

  • An open source multi-agent deep reinforcement learning routing simulator for satellite networks

    F. Lozano-Cuadra, M. Thorsager, I. Leyva-Mayorga, and B. Soret. In Proceedings of SPAICE2024: The First Joint European Space Agency / IAA Conference on AI in and for Space, pp. 420-424, Sep. 17-19, 2024, ESA European Centre for Space Applications and Telecommunications (ECSAT), Oxford, UK. DOI: 10.5281/zenodo.13885645.

  • Multi-Agent Deep Reinforcement Learning for Distributed Satellite Routing

    F. Lozano-Cuadra and B. Soret. In 2024 IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN), Stockholm, Sweden, 2024, pp. 1-2. DOI: 10.1109/ICMLCN59089.2024.10624767.

  • Q-learning for distributed routing in LEO satellite constellations

    B. Soret, I. Leyva-Mayorga, F. Lozano-Cuadra, and M. D. Thorsager. In 2024 IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN), Stockholm, Sweden, 2024, pp. 208-213. DOI: 10.1109/ICMLCN59089.2024.10624807.

Under review

  • Decentralized graph attention-based multi-agent reinforcement learning for communications in autonomous lunar rover swarms

    F. Lozano-Cuadra, B. Soret, M. Sanchez Net, A. Cauligi, and F. Rossi. Under review, 2026.