Human-Machine Collaboration for Automated Patterns of Life Analysis

A fundamental challenge in United States Air Force (USAF) intelligence processing, exploitation and dissemination (PED) is balancing the need to have human analysts in the PED process against the capacity limitations of human analysts. At the same time, military requirements demand an increase in quality & validity of metadata supporting AFPEO for Battle Management Operations C2 Division intelligence needs by providing timely assessments, enhanced battlespace awareness and threat warnings. To meet this dual challenge, the primary objective of this Phase II effort is to continue to research, develop and demonstrate enhancing situational awareness using Patterns of Life on multi-modal streaming data, reducing time and improving quality in their definition and application through interactive collaboration by USAF intelligence analysts and automation technologies. This objective continues to be pursued in this Phase II through the strategy of using “models as first class objects� in the PED process. These capabilities will ultimately be integrated into the Watchman for Defense analytics platform for delivery and deployment. As part of the proposed effort, an option period is proposed to bring these and related Phase II capabilities to maturity and transition them into the Open Architecture Distributed Common Ground System AN/GSQ-272 SENTINEL weapon system program.