Multi-Source Relational Situational Intelligence (MRSI)

Timely and effective decision making to best protect and employ our space-based resources in complex evolving situations depends on rapid and perceptive multi-INT space situational awareness (SSA) in support of Battle Management Command and Control (BMC2). To achieve satisfactory SSA, the space community is increasingly turning to machine learning (ML) technology enabled by rapidly growing computing capacity and quantities of data. However, system developers face inherent challenges in applying ML technology that robustly satisfies current and future SSA needs. These include constantly changing sensors and other data sources, unforeseen contingencies, expanded SSA needs including assessing intentional behaviors, and continued growth in data volume and complexity. To address these requirements and provide a robust adaptive architecture for ML, we propose to design an innovative Multi-source Relational Situational Intelligence (MRSI) system. MRSI constructs and maintains relational models of evolving multi-domain situations from dynamic multi-INT evidence for SSA. Its ability to learn to recognize situations and their components creates opportunities for conveying multiple results to analysts and decision-makers as well as other systems in the form of actionable characterizations of the situation to aid overall SSA.