INTELLISENSE SYSTEMS INC — Department of Defense SBIR Phase I: HQ085021S0001-07

Description

To address the Space Development Agency’s (SDA’s) need for emerging algorithms capable of recognizing and discriminating fast-moving, maneuverable targets from decoys and clutter in complex scenes that overhead persistent infrared (OPIR) sensors capture from space, Intellisense Systems, Inc. proposes to develop a new Persistent Infrared Tracker for Clustered Hypersonic Missiles (PITCH-M) based on an enhanced Bayesian multi-target tracking (MTT) filter for the detection and tracking of maneuverable high-velocity (i.e., hypersonic) vehicles from an OPIR satellite and the discrimination of these vehicles from decoys. Specifically, the use of a novel target motion model based on the physical characteristics of hypersonic vehicle flight and a new extrapolation algorithm to estimate the aimpoint based on measured characteristics and a priori knowledge will enable the PITCH-M system to detect and track closely spaced, maneuverable hypersonic vehicles in complex environments and distinguish warheads from decoys and clutter. As a result, this system enables the rapid detection, accurate tracking, and successful discrimination of fast-moving, maneuverable objects from OPIR sensors by leveraging mature and efficient algorithms that are implementable within a four-year timespan, which directly address the SDA’s requirements. In Phase I, Intellisense will develop and implement the PITCH-M algorithms in an open-framework architecture, conduct robust simulation and modeling to develop the upgraded Bayesian MTT filter algorithm and evaluate the trade space with respect to sensor attributes. Intellisense will demonstrate the feasibility of PITCH-M by quantifying PITCH-M’s performance (assuming nominal infrared system specifications) and by detailing risks and trade-offs in implementing the algorithms. In Phase II, Intellisense plans to integrate the PITCH-M algorithms with government-furnished payload image processing software and validate the algorithms with actual infrared scene data from a future SDA flight program.