Sensor Exploitation by Adaptive/Learning Systems (SEALS)

Signal Innovations Group (SIG) and its Fast Track partner, Integrian (www.integrian.com), propose development of an adaptive video sensing system, with a focus on intelligence, surveillance, and reconnaissance (ISR). The proposed adaptive system is applicable for ISR in general environments, addressing base and port security, as well as urban and suburban sensing during wartime and peace-keeping operations. Of significant importance for current and anticipated DoD activities, the proposed system will focus on detecting asymmetric threats, with the goal of recognizing unusual behavior or activities. Specifically, we must detect unusual events that may never have been observed previously. Therefore, rather then addressing the impossible task of training an algorithm on events that we may never observe a priori, the ISR system will focus on learning and modeling the characteristics of normal or typical behavior. This motivates development of graphical statistical models, such as hidden Markov models (HMMs), based on measured data characteristic of normal behavior. The graphical models will be employed within a partially observable Markov decision process (POMDP) algorithm, to develop an optimal policy for multi-sensor control and management. In this Phase II effort, SIG will develop and demonstrate real-time event detection algorithms on a multi-sensor Integrian video platform.