Robustifying the Kalman filter
Abstract
Kalman filters are tracking and prediction algorithms based on Gaussian measurement errors and structural models. The Kalman filter performance may degrade if the measurement errors come from a thicker-tailed-than Gaussian distribution. In this report non-linear procedures are described which are based on Kalman-type models, but work with student-t measurement errors. Keywords: Kalman filter; Student-t measurement errors; Iterative reweighting procedure; Nonlinear filter; Biweight; Robust estimation
Rights
This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.NPS Report Number
NPS55-87-014Related items
Showing items related by title, author, creator and subject.
-
A comparative study of discrete time filtering for a non-stationary random input
Fletcher, Harold Grove, Jr. (Monterey, California. U.S. Naval Postgraduate School, 1967-12);This thesis is concerned with a comparative study of discrete time filters using the theories of Wiener-Kolmogorov, Bode-Shannon, and Kalman, applied to the filtering of a non-stationary random signal to the presence of ... -
USING REINFORCEMENT LEARNING TO SPOOF A MONITORED KALMAN FILTER
Bonitz, Dylan A. (Monterey, CA; Naval Postgraduate School, 2022-06);Modern hardware systems rely on state estimators such as Kalman filters to monitor key variables for feedback and performance monitoring. The performance of the hardware system can be monitored using a chi-squared fault ... -
Statistical post processes for the improvement of the results of numerical wave prediction models
Galanis, G.; Kallos, G.; Chu, Peter C. (2011);A new mathematical technique for adaptation of the results of numerical wave prediction models to local conditions is proposed.The main aim is to reduce the systematic part of the prediction error in the direct model ...