I-Corps: Translation Potential of a Multi-Modal Sensor Fusion Platform that Provides Precision Drone Swarm Tracking

This I-Corps project is based on the development of a portable tethered drone platform that may be used to monitor large, visually obstructed sites. Many environments suffer from coverage blind spots created by physical obstructions, such as rigging at temporary outdoor events and stacked containers and steel structures at industrial and maritime sites. Ground-level cameras cannot see past these occlusions, leaving operators unable to maintain safety and security. The scale is significant as unauthorized drone invasions over public venues alone now number in the thousands each year, and illegal activity in the maritime sector is estimated to cost tens of billions of dollars annually. This technology provides an uninterrupted view and automatically identifies problems that need attention using elevated sensing devices on a tethered platform. In addition, the technology is continuously powered and deploys in minutes without fixed infrastructure. This may lower the cost of safety monitoring for events and venues, industrial and port operations, and maritime security, and benefit the hospitality, logistics, and insurance sectors while strengthening public safety and protecting critical supply chains. This I-Corps project utilizes experiential learning coupled with first-hand investigation of the industry ecosystem to assess the translation potential of a multi-sensor aerial platform that provides situational awareness in large, obstructed environments. This technology is based on an edge-Artificial Intelligent (AI) architecture that splits computation between air and ground where the airborne unit is a lightweight sensor carrier, while heavy perception, fusion, and reasoning run on a powered ground station with a dedicated Graphics Processing Unit (GPU) connected by a high-bandwidth tether. It combines several sensing modalities into a single, continuously updated picture of activity, with calibrated confidence in each detection. Drawing on prior research in partitioned edge AI and multi-modal sensor fusion, the system maintains real-time awareness and automatically focuses its attention where it is needed. Rather than requiring continuous monitoring of raw video, it surfaces only the events that warrant a human decision. In addition, this technology may provide an advantage over satellite snapshots, costly patrols, and fixed ground sensors as it is designed for continuous, occlusion-resistant coverage from an elevated vantage point. This may give users accurate, real-time tracking and timely alerts where existing options are blocked, intermittent, or prohibitively expensive. 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.