EAGER: Service Composition with LLM-based Services for Pervasive Awareness

This research aims to make smart devices and applications aware of their surroundings in real time by combining advanced artificial intelligence (AI) with nearby connected devices and software services. Unlike today's AI systems, which primarily rely on past information, the proposed approach enables AI to analyze, reason and respond to events as they happen. This capability has the potential to transform disaster resilience and emergency services, healthcare, autonomous transportation, and smart manufacturing by enabling intelligent systems to understand and respond to rapidly changing situations in real time. It will also help develop the next generation of researchers and engineers by providing graduate and undergraduate students with interdisciplinary training in artificial intelligence, networking, sensing, and distributed systems. This research explores a new paradigm for integrating Large Language Model (LLM) services with traditional service composition to realize Pervasive Awareness as a Service (PASS). Proposed research tasks will develop novel methods for fusing live data, interpreting context, and combining AI capabilities with distributed services to enable real time awareness in dynamic environments. The project will also advance parallel and distributed computing techniques that support timely reliable decision making in resource constrained pervasive computing systems. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.