SSA Anomaly Characterization using Case-Based Reasoning (CBR) and other AI techniques

Defending space systems against intentional and unintentional threats is paramount, and requires space situational awareness, i.e., knowing what is going on, what the root problem is that is manifesting itself, and even predicting likely threats based on existing patterns. The ultimate goal of this proposed effort is to automate and thereby radically improve SSA threat assessment by correlating the current situation(s) with historic scenario cases and utilizing the knowledge from those past cases to dramatically shorten the associated response time to new events. Stottler Henke proposes to develop the Ssa HEuRistics using fuzzy LOgic and Cbr (SHERLOC) system, which uses case-based reasoning (CBR) and probabilistic reasoning from multi-domain evidence to interpret the current situation based on knowledge of past cases, and produce real-time threat assessments. The advantage of a CBR approach based on prior examples for categorical threat cases is that complex relationships between evidentiary factors and threat conditions can be captured. In fact, as new threat examples are added to the case base, a CBR system can adapt with correspondingly updated threat assessments. Building on the Phase I prototype, the fully operational Phase II SHERLOC prototype will be implemented in three spirals with end user demonstration and feedback.