SRM driven by a Traffic Anticipation and Response Engine (STARE)

Today’s threats tend to maneuver in commercial vehicles through dense, cluttered urban streets where detection and tracking become exceptionally difficult. First, the high volume of background traffic makes it difficult to maintain tracks within the clutter of large numbers of confusing targets. Second, urban structures present obscuration problems that create gaps in coverage. Both of these problems cause frequent track losses, motivating the need for two specific capabilities. First, in order to effectively task sensor assets such that their probability of reacquiring the target is optimal requires the use of algorithms that can fuse all available information into an anticipatory model that exploits traffic trends to predict the likelihood of future locations of vehicles of interest. This capability can be utilized in concert with sensor resource management (SRM) algorithms to improve the ability to reacquire lost targets. Second, novel techniques are needed to improve track correlation and continuity once new sensor observations are provided. The DECISIVE ANALYTICS Corporation will leverage their experience in developing inference-driven sensor management systems to achieve the goals of this SBIR. We will utilize unparalleled techniques in computational probability to provide anticipatory traffic modeling to drive our SRM, as well as to provide improved track correlation.