SBIR-STTR Award

Battlespace awareness within the space domain is crucial for protection of high-value assets (HVAs). Counterspace measures such as direct-ascent anti-satellite weapons have been demonstrated as early as 1959 and continue to pose a threat to spaceborne objects. Co-orbital threats represent a subset of counterspace measures from objects already in-orbit and may be kinetic or non-kinetic. Development of a framework for co-orbital threat detection is compounded by several challenges including timely detection and prediction, and the existence of a large set of possible threats. The principal goal of this Phase I effort is to evaluate potential solutions for co-orbital threat identification and course of action (CoA) prediction. Concepts of Operations (CONOPS) for adversary threats will be developed by leveraging our team’s extensive experience in all facets of CoA determination. CoAs will then be generated using highly flexible orbit transfer algorithms developed by Texas A&M University and will serve as training data for learning and inference. Artificial Intelligence and Machine Learning (AI/ML) techniques will be explored as potential solutions to identify threats and predict CoAs. The fusion of a physics-informed data generation scheme with AI/ML methodologies provides a foundation for timely availability of indications and warnings for co-orbital threats.