Multiconstrained Real -Time Interceptor Guidance Algorithm for Maneuvering Targets

OUSD (R&E) CRITICAL TECHNOLOGY AREA(S): Advanced Computing and Software; Hypersonics; The technology within this topic is restricted under the International Traffic in Arms Regulation (ITAR), 22 CFR Parts 120-130, which controls the export and import of defense-related material and services, including export of sensitive technical data, or the Export Administration Regulation (EAR), 15 CFR Parts 730-774, which controls dual use items. Offerors must disclose any proposed use of foreign nationals (FNs), their country(ies) of origin, the type of visa or work permit possessed, and the statement of work (SOW) tasks intended for accomplishment by the FN(s) in accordance with the Announcement. Offerors are advised foreign nationals proposed to perform on this topic may be restricted due to the technical data under US Export Control Laws. OBJECTIVE: Develop real-time, nontraditional trajectory optimization algorithm for hypersonic interceptors. DESCRIPTION: In a modern threat environment and with today's missile systems, the ability to respond to dynamic threats is vital for successful defense. Thus, to enhance the hit to kill precision and robustness in trajectory optimization, there is a need for a novel, non-traditional hypersonic interceptor guidance optimization algorithm. The benefited outcome would ultimately produce a higher probability of a kill due to optimized use of kinetic energy and fuel efficiency, greater resilience and responsiveness against moving targets, improved target prioritization, and enhanced predictive capabilities. Trajectory finding algorithms can be roughly dichotomized into (1) sampling and (2) gradient variants. Sampling methods use techniques such as Monte Carlo to "guess and check" for the best solution by generating thousands of predicted paths in parallel and using predetermined cost functions to find the most effective path. Gradient methods, on the other hand, use an iterative process that repeats until it converges to a local/global minimum or maximum. The drawbacks of the two methods respectively are (1) that it can take a long time to perform sampling and (2) that it gets stuck at local extrema and only uses one predictive path at a time, but can have faster convergence than sampling algorithm. Thus, to improve the hit-to-kill probability, there is a need for a non-traditional interceptor guidance optimization algorithm that is able to handle complex, non-linear dynamics, different uncertainty sources, non-convexity, and the non-stationarity of the environment. Advancements in computational power through parallel computing and graphics processing unit (GPU) enable non-traditional algorithms that lie at the nexus of sampling and gradient-based methods. The resulting outcome would improve the interceptors' robustness, effectiveness, and capability of handling modern threats with increased efficiency and reliability. PHASE I: Implement in a single agent system, such as an individual interceptor. Develop algorithm and demonstrate a feasibility against a ballistic or limited maneuverability threat. PHASE II: Implement in a single agent system, such as an individual interceptor. Further demonstrate the algorithm can be tested and successful against a highly maneuverable threat and complex environment. PHASE III DUAL USE APPLICATIONS: Demonstrate real time application, and how it can be applied to multi-agent systems, such as clusters or swarms, for example in a group of interceptors deployed at once. REFERENCES: 1. Kim, K.-P., & Lee, C.-H. (2024). Adaptive Weight Model Predictive Path Integral Control for Multiconstrained Missile Guidance. Journal of Aerospace Information Systems, 21(4), 305-322. 2. Li, W., Zhu, Y., & Zhao, D. (2021). Missile guidance with assisted deep reinforcement learning for head on interception of maneuvering target. Complex & Intelligent Systems, 7(5), 2273-2284. 3. Kim, K.-P., & Lee, C.-H. (2023). Fixed Range Horizon MPPI-based Missile Computational Guidance for Constrained Impact Angle. International Journal of Control, Automation, and Systems, 21(6), 1866-1884. 4. Liang, C., Wang, W., Liu, Z., Lai, C., & Zhou, B. (2019). Learning to Guide: Guidance Law Based on Deep Meta-learning and Model Predictive Path Control. IEEE Access, vol. 7, pp. 38051-38063. DOI:10.1109/ACCESS.2019.2909579. KEYWORDS: Hypersonic, Missile, Guidance, Maneuvering, Algorithm, Optimization, Parallel computing, Sampling, Gradient, GPU