SBIR Phase II: Decentralized Evacuation Intelligence with Generative AI, Personalized Preparedness and Robust Real-Time Navigation

The broader/commercial impact of this Small Business Innovation Research (SBIR) Phase II project is to enhance public safety and community resilience in wildfire-prone areas through a personalized evacuation technology that integrates planning and real-time response. Evacuation remains one of the most urgent and time-critical aspects of wildfire disaster management, yet residents often lack access to trustworthy, actionable information when it matters most. This project will commercialize a novel digital platform that provides both proactive preparedness tools and dynamic evacuation guidance that functions even without internet connectivity. By tailoring evacuation strategies to individual needs and local hazards, this technology bridges the gap between emergency protocols and real-world decisions made by households under stress. Beyond saving lives, it reduces the burden on first responders, supports better allocation of emergency resources, and strengthens long-term community preparedness. The innovation also opens export and scaling opportunities to other disaster-prone geographies and hazards (i.e. floods, volcanic eruptions, hazardous materials release), contributing to the growing market for smart emergency management systems.

This Small Business Innovation Research (SBIR) Phase II project addresses the technical challenge of developing a robust, scalable, and user-centered wildfire evacuation platform powered by a Digital Twin (DT) of the built and natural environment. The proposed solution includes three core innovations: (1) a surrogate modeling approach that allows for fast, on-device evacuation route predictions based on pre-simulated wildfire scenarios, (2) an AI-driven interface for emergency managers to interactively validate and refine evacuation strategies, and (3) a Preparedness App that enables residents to plan for wildfire threats using personalized virtual drills and destination planning, supported by carefully designed user engagement strategies. Building on Phase I feasibility studies, Phase II will focus on integrating these components into a minimum viable product, field-testing it in pilot communities, and validating both user outcomes and technical performance. Anticipated results include increased evacuation readiness, faster community clearance times, and reduced reliance on emergency services during wildfires. The project draws on insights from traffic modeling, disaster risk reduction, human behavior in disasters, user-centered design, and geospatial analytics to deliver a next-generation preparedness solution for an escalating national challenge.

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.

Award Number: 2528445
Principal Investigator: Sarah Lindbergh
Funds Obligated: $1,231,473
State: DE