Collaborative Research: VINES: Track 2: SEA-NET: Scalable Edge Analytics and Networking for Autonomous Maritime Emergency Response

Maritime emergency response relies on slow and manual coordination, leading to critical delays during crises. Since terrestrial networks fail offshore, reliable broadband is essential for vessel safety. This project serves the U.S. national interest by developing a resilient communication and edge computing system for autonomous maritime intelligence. By enabling faster disaster response, it seeks to save thousands of lives and billions of dollars annually. Additionally, the project will strengthen the American workforce by providing students and early-career researchers with hands-on training in high-impact experimentation and artificial intelligence (AI). The deliverables from this project directly impact the evolution of 6G by pioneering AI-native architectures that move beyond 5G's bit-centric limitations. The project goal is to demonstrate autonomous maritime operations by integrating next-generation networks, edge computing, and AI models. The project team will dynamically orchestrate wireless spectrum across heterogeneous platforms like Unmanned Aerial Vehicles (UAVs) and surface vessels. Technical methods include hierarchical semantic encoding and Age of Information-based scheduling to ensure data freshness for real-time edge inference. Specifically, semantic encoding compresses payloads delivering a goodput more than 90% of the required semantic payload size, while a software-defined networking controller ensures resilient and low-latency (less than 50ms) handovers across aggregated 5G and satellite links. A cryptographic provenance framework, called TRUSTEDBITS, is introduced to maintain data security in Denied, Degraded, Intermittent, and Limited (DDIL) environments. The project also incorporates explainable AI models to provide transparent rationales for human-AI teaming, ensuring a Time-to-Actionable Decision of under 60 seconds. Technology validation involves high-speed towing tank tests and at-scale Hudson River demonstrations to reduce incident response time by up to 30% compared to the current manual response. These trials will progressively advance the system to Technology Readiness Level (TRL) 6. 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.