Bayesian Troop-Target Detection, Tracking, and Prediction Using FOPEN-GMTI Radar
Department of Defense
Defense Advanced Research Projects Agency
Agency Tracking Number:
Solicitation Topic Code:
Small Business Information
SCIENTIFIC SYSTEMS CO., INC.
500 West Cummings Park - Ste 3000, Woburn, MA, 01801
Socially and Economically Disadvantaged:
Associate Group Leader
Associate Group Leader
AbstractThe "troop-target" problem of automatic detection, tracking, and prediction of the formations of dismounted ground troops concealed in foliage, using FOPEN-GMTI radars introduces challenges well beyond those encountered in more conventional group-target problems. AD HOC troop formations are considerably more amorphous and unpredictable and they must be deduced from the constituent individuals. Yet these individuals may be small in number, may rapidly change direction and speed, can be widely dispersed, and will usually form no predictable pattern. We address this problem with a theoretically rigorous generalization of the constant-gain Kalman filter to the multitarget domain. The resulting PHD filter propagates a first-order multitarget moment of the multitarget posterior distribution, and first detects and tracks the over-all bulk behavior of a formation. What is being tracked at any time is an estimate of the geographical shape and density of the troop formation. Phase I showed the feasibility assessment for this PHD "bulk" tracking, and demonstrated: (1) successful bulk troop tracking, (2) considerable improvements in CROSS-RANGE accuracy when using two sensors with suitable triangulation geometry, (3) successful inference of troop parameters, and (4) successful formation typing (patrol, skirmish, bivouac, ambush, etc.) inferred from the estimated troop parameters. Phase II emphasis will be on more realistic implementations of the algorithms that include realistic sensor models, higher clutter densities, and hybrid tracking of an individual target and troop-target. The Phase II results will be examined by LM-Orincon with the goal of a Phase III delivery to FORESTER, providing a realistic and concrete Phase III technology transition path into FORESTER and potentially in the Army Future Combat Systems (FCS) program. The project team includes Dr. Ronald Mahler of Lockheed Martin. Lockheed Martin will provide both technical and commercialization support in the application of detection/tracking technologies.
* information listed above is at the time of submission.