Affine-mapping based variational ensemble Kalman filter

Abstract We propose an affine-mapping based variational ensemble Kalman filter for sequential Bayesian filtering problems with generic observation models. Specifically, the proposed method is formulated as to construct affine mapping from prior posterior one, and computed via a formulation, i.e., by minimizing Kullback–Leibler divergence between transformed distribution through actual posterior. Some theoretical properties of resulting optimization problem are studied gradient descent scheme solve problem. With numerical examples we demonstrate that has competitive performance against existing methods.