SBIR-STTR Award

We will develop a proof of concept design / prototype demonstration for refining the performance of a dynamic Air-Sea-Land-Space IAMD Linguistic Geometry (LG)-Hypergame based model. This demonstration will include a conceptual model and a top-level architecture of how a high resolution IAMD LG-Hypergame could be run stand-alone or integrated into a systems of systems simulation environment. Specifically, we will develop the IAMD LG-Hypergame using Linguistic Geometry (LG) and we will apply LG for solving higher-dimensional board games (ABG) for BLUFOR interceptor flyouts. We will investigate the theoretical and computational limits of the applicability of LG to missile defense engagement planning. Approved for Public Release | 19-MDA-10270 (18 Nov 19)

Manual multi-domain kill chain planning for Integrated Air and Missile Defense (IAMD) is labor intensive and too slow to effectively respond to time critical targets. STILMAN has developed an Artificial Intelligence (AI)-based planning and supervisory decision aid algorithms and software for resource allocations (sensors and weapons) across multiple domains in real-time. This AI approach is based on Linguistic Geometry (LG), a break-through in constructive game-theory solutions originally developed by Dr. Boris Stilman and further refined by the STILMAN company. This LG AI capability is ideally suited to quickly solve extraordinarily complex IAMD resource planning and allocation problems. As the frequency and size of raids escalates, the urgency of the response and the complexity of resource allocation decisions increases. Within a multi-domain (near-Earth Space, Air, Land, and Sea) battlespace, exacerbated by presence of both conventional weapons and ballistic missiles/interceptors, this complexity can quickly overwhelm the human commander’s ability to successfully plan and respond. STILMAN’s AI-based algorithms and software provide dynamic kill chain planning and preview capability enabling the commander to deal with such complexity. The LG AI-based analysis is fast, flexible, scalable, and explainable – and it applies to multi-domain and multi-tier sensor and weapon platforms. Dynamic, reactive resource allocation recommendations are displayed via 2D or 3D animations that provide the human commander with an on-the-loop visualization that prioritizes resource allocations in faster than real-time. This is particularly advantageous for servicing pop-up targets that suddenly appear outside of the pre-planned mission profile. Approved for Public Release | 20-MDA-10643 (3 Dec 20) ---------- STILMAN has developed an Explainable Artificial Intelligence (XAI)-based planning and supervisory decision aid for resource allocations (sensors and weapons) across multiple platforms and domains in real-time. This quantitative analysis capability is ideally suited to quickly solve extraordinarily complex IAMD resource allocation problems. As raid frequency and size escalates, the need for a synchronized response increases the complexity of resource allocation decisions. STILMAN has developed an XAI-based dynamic kill chain planning and preview capability. Unlike many other Artificial Intelligence (AI) approaches, this LG-based XAI analysis is fast, flexible, scalable, and explainable – and it applies to multi-mission, multi-domain and multi-tier sensor and weapon platforms. Dynamic, “i.e., reactive” resource allocation recommendations are displayed via 2D or 3D animations. The animations and quantitative metrics provide the human commander with an on-the-loop visualization that prioritizes resource allocations in faster than real-time. Approved for Public Release | 22-MDA-11339 (13 Dec 22)