Fixed interval smoothing algorithm for an extended Kalman filter for over-the-horizon ship tracking

Abstract

The performance of an extended Kalman filter used to track a maneuvering surface target using HFDF lines-of-bearing is substantially improved by implementing a fixed interval smoothing algorithm and a maneuver detection method that uses a noise variance estimator process. This tracking routine is designed and implemented in a computer program developed for this thesis. The Hall noise model is used to accurately evaluate the performance of the tracking algorithm in a noisy environment. Several tracking scenarios are simulated and analyzed. The application of the Kalman tracker to a tropical storm tracking is investigated. Actual storm tracks obtained from the Joint Typhoon Warning Center in Guam, Mariana Islands are used for this research