Adaptive unscented Gaussian likelihood approximation filter
Journal article, 2015

This paper focuses on the update step of Bayesian nonlinear filtering. We first derive the unscented Gaussian likelihood approximation filter (UGLAF), which provides a Gaussian approximation to the likelihood by applying the unscented transformation to the inverse of the measurement function. The UGLAF approximation is accurate in the cases where the unscented Kalman filter (UKF) is not and the other way round. As a result, we propose the adaptive UGLAF (AUGLAF), which selects the best approximation to the posterior (UKF or UGLAF) based on the Kullback-Leibler divergence. This enables AUGLAF to outperform both the UKF and UGLAF.

Kalman filter

Bayes' rule

Nonlinear filtering

Gaussian approximation

Author

Angel Garcia

Technical University of Madrid

M. R. Morelande

RMIT University

J. Grajal

Technical University of Madrid

Lennart Svensson

Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering, Signal Processing

Automatica

0005-1098 (ISSN)

Vol. 54 166-175

Subject Categories

Signal Processing

DOI

10.1016/j.automatica.2015.02.005

More information

Latest update

9/25/2020