SHERLOC-Multi: Multi-Object Space Threat Assessment Using Data Fusion and Other AI Techniques

Space battle management and command and control depend heavily on maintaining space domain awareness (SDA) to protect and defend our space assets.  Hostile activities conducted by adversaries may involve more than a single event and a single object, when the underlying strategic purpose is concerned with effects on the operating capabilities of multiple satellites.  Stottler Henke proposes to build on the progress made in SHERLOC-Multi Phase I, with a Phase II effort to develop a fully operational SHERLOC tool with automated similarity-based inference and intent analysis to help operators avoid surprise and support faster decision-making timelines for predicting, detecting, classifying, and prioritizing multi-object threats, and reasoning about their potential impacts on the missions of defended assets.  The Phase I feasibility study resulted in an initial operational prototype, which lays the groundwork for the Phase II SHERLOC tool planned for integration into a battle management system under development in the SDA TAP Lab.  SHERLOC will process space data autonomously and support both machine-to-machine and human-machine operations to accelerate SDA decision-making in multiple kill chains.  Phase II efforts will be tailored to specific multi-object threat-related use cases, which form the basis for implementing the reasoning algorithms and data throughput.