Applying ICARUS to Task Scheduling of the Air Force's Satellite Control Network

We will develop a proof-of-concept system that implements an innovative methodology to diagnose and repair unknown satellite anomalies. Our system is based on a novel approach that integrates Case-Based Reasoning (CBR) with Utility Theory. The resulting system deals with unknown and missing information, and selects from a case base the plans or actions that will either fix the satellite or will generate additional information to assist in the diagnosis. The actions are ranked based on criteria selected by the ground operator such as "minimum use of power" or "minimum risk to the spacecraft." The system provides the operator with alternative actions and their ramifications.The main goal of the Phase I effort is to determine how to represent the cases of satellite anomalies, how to implement utility evaluation rules based on user-defined criteria, and to adapt previous research in integrating CBR and utility theory to the satellite anomaly resolution domain. We will also develop metrics of the quality of our technique and will test and evaluate our finished prototype demonstration.We have ensured the collaboration of Hughes Corporation which will provide us with cases and expertise.