Optimal Model-Based MTI Generation for Dim, Agile, Closely-Spaced Objects

The U.S. is currently working to develop and deploy multiple pLEO satellite constellations providing global persistence in the detection and tracking of aerial threats including aircraft, hypersonic missiles, and conventional cruise missiles. A key challenge to this critical mission is the generation of highly-accurate and timely Moving Target Indicators, especially given our adversaries' ability to field highly-agile, exceedingly dim, and closely-spaced airborne threats. To overcome this challenge, Numerica proposes the development of an optimization and model-based advanced detection and tracking algorithm that allows for the generation of super-resolved MTIs even for dim, agile and closely-spaced targets. This approach leverages a physics-based formulation rather then utilizing large quantities of training data, and is not sensitive to the characteristics of the EO/IR sensors. Further, the proposed approach jointly considers background and target state estimation, significantly reducing the false alarm rate suffered by current state-of-the-art approaches. Numerica is ideally suited to execute on this effort due to its support of recent pLEO efforts as well as our broad expertise in developing and deploying advanced algorithms to operational missile defense systems for MDA. The proposed effort will directly leverage Numerica's existing dim target detection and OPIR distributed tracking algorithms.