Agent-based Satellite Attack Analysis System (ASAAS)

Recent military operations demonstrate the heavy reliance of U.S. military operations on space assets for intelligence, surveillance, communications, and targeting. Using anti-satellite methods such as lasers or kinetic energy vehicles, potential adversaries have the capability to attack space assets and severely degrade the U.S.'s military superiority. To maintain dominance, defensive counterspace techniques and attack detection algorithms are aggressively being developed. Due to the complexities and limitations in current techniques and algorithms, improved user work-centered interface systems are needed to assist operators in analyzing the provided information (e.g. interpreting neural network outputs) in order to decrease decision and reporting time, thereby by enabling operators to intelligently and efficiently filter data to verify the output of the detection algorithm. To address this problem, we propose to construct a prototype Agent-based Satellite Attack Analysis System (ASAAS) for satellite operators or national asset managers who are in off-nominal situations due to anti-satellite attacks. Our approach is to use a work-centered support system design that combines cognitive task analysis with intelligent agents. We see considerable potential for this approach in enhancing current and planned Air Force Rapid Attack Identification, Detection, and Reporting Systems (RAIDRS) approaches and technologies.