Researchers have developed AERIS, an offline policy improvement framework designed for integrated sensing and communication (ISAC) in multi-unmanned aerial vehicle (UAV) systems. This approach learns from existing flight logs, enabling decentralized execution by individual UAVs while using global data for training to assess team-level impacts. The framework incorporates STAR-CRDT, an algorithm that refines local actions and distills only reliable improvements for decentralized actors, offering a guarantee for offline policy improvement. Experiments demonstrate significant gains in ISAC objectives, communication rates, sensing reliability, and a substantial reduction in collision risks, even on real-world maps. AI
RANK_REASON The cluster contains a research paper detailing a new framework and algorithm for UAV systems. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →