A Modular Framework for Optimization of Pursuit Evasion Strategies for Hypersonic Intercept

The overall goal of this SBIR project is to develop a framework for creating simulation-based differential games for evaluation and optimization of pursuit and evasion strategies for hypersonic missile defense. In Phase I, the key software modules within the tool were developed, and proof-of-concept was demonstrated via case studies considering differential games of interest. Phase II efforts will focus on 1) extending the modeling and simulation capability to support a broader range of vehicle with subsystems resolved at higher fidelity in more complex engagement scenarios; 2) refining the machine learning-based guidance techniques through provision of additional information, more robust training procedures, varying architectures, and built-in fail-safe mechanisms; 3) defining differential games in which threats are forced to compete with interceptors; 4) developing and tuning optimization techniques to better support scheduling decision variables, categorical decision variables, and adversarial optimization studies; 5) performing optimization using the differential games to optimize dynamic flight behavior of hypersonic interceptor and threat vehicles to identify improved pursuit and evasion strategies; 6) extracting knowledge through application of the tool to define requirements and to inform technology investment; 7) demonstrating the capability applied to MDA relevant scenarios and agents and 8) identifying beyond Phase II technology maturation and transition opportunities. Successful development of the proposed framework will provide a powerful tool for identifying optimized guidance, navigation and control strategies for hypersonic pursuit and evasion from simulation-based differential games. Approved for Public Release | 22-MDA-11340 (16 Dec 22)