Speedy UAV Swarms Detection, Identification, and Tracking using Deep Learning-Based Fusion Methodology for Radar and Infrared Imagers

R-DEX Systems, Inc. (R-DEX) proposes to leverage its unique and extensive expertise in Artificial Intelligence/Machine Learning (AI/ML), real time sensor processing, automatic target recognition (ATR), and target tracking to develop an innovative solution for speedy UAV swarm detection, identification, and tracking. The proposed solution will leverage Physics-Informed Neural Networks (PINNs) for the real-time sensor fusion of data from multiple sensor modalities. In Phase I, R-DEX proposes to work closely with its strategic partner to demonstrate R-DEX’s innovative PINN-based fusion methodology with a focus on radar and infrared sensor modalities. Specifically, R-DEX will conduct trade studies and engineering design necessary to define the operational system and associated technologies, leveraging mature key enabling technologies, including sensor hardware, sensor software, target tracking software, user interfaces, and data processing systems. In addition, R-DEX will demonstrate the technical risk elements and quantify the performance tradeoffs as a function of processing capability and environmental conditions to achieve real time performance, while meeting the performance benchmarks specified in the topic description. Specific attention will be paid to 1) ease of integration into existing pipelines, 2) interoperability with different sensor types and modalities, 3) scalability to a growing library of current and future-generation UAVs, and 4) a simple and intuitive user interface.