{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T14:14:45Z","timestamp":1778076885570,"version":"3.51.4"},"reference-count":0,"publisher":"University of Florida George A Smathers Libraries","issue":"1","license":[{"start":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T00:00:00Z","timestamp":1778025600000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["FLAIRS"],"abstract":"<jats:p>Autonomous multi-agent systems face significant challenges due to the computational inefficiencies of centralized control methods. As a result, distributed approaches have gained increasing attention, enabling decentralized decision-making when a single control point is undesirable or impractical. In the context of disaster-response, one prominent application is cooperative target tracking with Unmanned Aerial Vehicles (UAVs), where multiple UAVs must coordinate to monitor ground targets and ensure that no target remains unobserved for extended periods of time.\nIn this paper, we formalize the multi-agent target tracking problem and introduce a scalable Multi-Agent Reinforcement Learning (MARL) training environment for cooperative UAV swarm tracking. Each agent is equipped with a set of independent Kalman filters (KFs) and must coordinate with other agents to maintain continuous tracking of multiple ground targets. We propose a MARL-based approach to address this problem and provide a detailed experimental evaluation between state-of-the-art on-policy and off-policy algorithms. The results demonstrate the effectiveness and scalability of MARL approaches for decentralized cooperative tracking in complex and dynamic environments.<\/jats:p>","DOI":"10.32473\/flairs.39.1.141776","type":"journal-article","created":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T13:18:02Z","timestamp":1778073482000},"source":"Crossref","is-referenced-by-count":0,"title":["When to Measure"],"prefix":"10.32473","volume":"39","author":[{"given":"Alessandro","family":"Amato","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Raffaele","family":"Galliera","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"K. Brent","family":"Venable","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Niranjan","family":"Suri","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"17357","published-online":{"date-parts":[[2026,5,6]]},"container-title":["The International FLAIRS Conference Proceedings"],"original-title":[],"link":[{"URL":"https:\/\/journals.flvc.org\/FLAIRS\/article\/download\/141776\/147064","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.flvc.org\/FLAIRS\/article\/download\/141776\/147064","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T13:18:03Z","timestamp":1778073483000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.flvc.org\/FLAIRS\/article\/view\/141776"}},"subtitle":["A Multi-Agent Reinforcement Learning Approach for Efficient Tracking"],"short-title":[],"issued":{"date-parts":[[2026,5,6]]},"references-count":0,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,5,6]]}},"URL":"https:\/\/doi.org\/10.32473\/flairs.39.1.141776","relation":{},"ISSN":["2334-0762","2334-0754"],"issn-type":[{"value":"2334-0762","type":"electronic"},{"value":"2334-0754","type":"print"}],"subject":[],"published":{"date-parts":[[2026,5,6]]}}}