FA865023P1008

Small Business Innovation Research Program (SBIR) Phase I

Automated Framework for Target Enrichment and Activity Modeling (AFTEAM)

Automated/assisted target behavior model development

It is vital that commanders and planners have as complete of understanding of the battlespace as possible to base decisions on. As the Air Force implements the doctrine set forth by Agile Combat Employment (ACE), airmen will be required to work proactivity within threat timelines to help achieve that understanding. This will require analysts to translate Command and Control (C2) and Intelligence information into action at the speed and scale of relevance, regardless of the operational environment. To accomplish this, designers of Intelligence and Analytics systems are presented with a tradeoff between offering more data sources or offering a more streamlined workflow for the analyst. The option chosen nearly always seems to fall on the side of more data sources. If the probability of finding an intelligence golden nugget is proportional to the amount and diversity of data ingested, the design reasoning goes, then more data and more sources of data increase the probability of finding something important. JDM Solutions understands this perspective on intelligence analysis, but also acknowledges that a more-is-better approach doesn't necessarily lead to more insightful or timely analysis. Some analysis problems are matters of finding a nugget in a sea of data and connecting the dots. Analysts are often tasked with a Sherlock Holmes style job of finding patterns of behavior for a potential target or patterns between multiple potential targets. In these cases, the quantity of data ingested as a measure of a system's potential analytic power is secondary to the measure of the quantity of connections that can be made across disparate elements of data. Our vision for an Automated Framework for Target Enrichment and Activity Modelling (AFTEAM) will move from simple situational awareness to true understanding. This will require data from all sources relating to an entity or target to be processed and fused in real-time. The data processed includes traditional sources such as radar plots and tracking reports, as well as unstructured sources such as situation reports, engineering documents and intelligence assessments. It also includes information about a targets environment, such as geospatial and weather data. A major obstacle to reliably, quickly, and automatically finding the relationships that exist across these large, diverse data sets, is the difficulty in automating the processing of unstructured and semi-structured text. These are the places that true insights live. The AFTEAM will provide a solution that both reduces risk through use of proven capabilities while at the same time delivering more capabilities within a shortened timeline. The resulting microservices and containers will provide modular solutions that will readily support integration into the Insight program as well as any other application or system built using modern DevSecOps best practices.

The goal of phase I is to establish the technical merit, feasibility, and commercial potential of proposed R&D efforts and determine the quality of performance of the small business awardee organization.

AF222-0013

F222-0013-0052

22.2

Darren Woodruff