
Accession Number : ADA328926
Title : Progress Report on the Project Automatic Target Recognition (N94124).
Descriptive Note : Progress rept.
Corporate Author : AEGIR SYSTEMS INC OXNARD CA
PDF Url : ADA328926
Report Date : 01 JUL 1997
Pagination or Media Count : 75
Abstract : Filtering, prediction and smoothing (FPS) are the three basic components of the data assimilation process in target tracking. An analytical solution of the FPS problem is possible only in a handful of particular cases, the most important of which is linear. In this case the solution is given by the Kalman filter. However, in many important cases, such as passive sonar, radar warning systems, infrared search and track, the systems are generically nonlinear. To date, the extended Kalman filter (EKF) has been the dominant algorithm technology in realtime estimation, tracking, and similar applications. A major reason for its success has been the fact that it has offered a reasonable compromise between realtime operation and satisfactory performance in some nonlinear problems. On the other hand, the EKF is a completely heuristic algorithm, requires readjustment to each particular problem, and is unstable in nonlinear problems which involve jumps, maneuvers, etc.
Descriptors : *TARGET RECOGNITION, *KALMAN FILTERING, *TRACKING, ALGORITHMS, REAL TIME, NONLINEAR SYSTEMS, OPERATION, RADAR EQUIPMENT, INFRARED RADIATION, ASSIMILATION, AUTOMATIC, WARNING SYSTEMS, ANALYTIC FUNCTIONS, PASSIVE SONAR.
Subject Categories : Target Direction, Range and Position Finding
Distribution Statement : APPROVED FOR PUBLIC RELEASE