An investigation into Kalman filter target tracking algorithms and their real time parallel transputer implementation

This thesis reviews the applications of Kalman filtering estimation to the problem of target tracking. Both linear and nonlinear forms of Kalman filter are reviewed and models of target manoeuvre discussed. Manoeuvre adaptation schemes are examined to detect the onset and completion of manoeuvres. Target manoeuvre coordinates are also examined and a new target model proposed which significantly improves tracking performance. The new model includes turn rate estimation. The real-time implementation of tracking Kalman filters is also studied both for a simple processor and a multiple processor architecture. Tracking algorithms are coded in Parallel C and evaluated for speed and efficiency