The Department of Defense (DoD) is seeking proposals for the development of machine learning algorithms for infrared search and track (IRST) applications. The United States Air Force specifically needs extended range passive air-to-air surveillance capabilities for contested environments. The current detection and tracking algorithms are computationally expensive and require large compute resources, limiting their deployment on platforms with cost, size, weight, and power constraints. The objective is to develop machine learning algorithms that can be implemented on low-cost, size, weight, and power processing hardware to aid detection and tracking processing for IRST applications. The algorithms should have similar or improved performance compared to current algorithms but require significantly less computational resources. The effort should explore and demonstrate the ability to train the machine learning algorithm using properly simulated target signatures and clutter plus noise and interference effects. The government will provide limited real-world IRST data, modeled target signatures, and IRST sensor characteristics. The project will be conducted in three phases: Phase I involves developing a modular machine learning architecture optimized for IRST; Phase II focuses on refining the architecture, training the machine learning algorithm using synthetic and real-world data, and comparing its performance with baseline algorithms; Phase III involves implementing the machine learning-enhanced IRST detection and tracking algorithm in a ruggedized low SWaP processor and demonstrating its performance and capability through mountaintop and/or flight testing. The project is restricted under the International Traffic in Arms Regulation (ITAR) and requires the ability to process and store classified data up to Secret//Collateral. The proposal submission deadline is October 18, 2023. For more information, visit the solicitation link.