Abstract:Practical Bayes filters often assume the state distribution of each time step to be Gaussian for computational tractability, resulting in the so-called Gaussian filters. When facing nonlinear systems, Gaussian filters such as extended Kalman filter (EKF) or unscented Kalman filter (UKF) typically rely on certain linearization techniques, which can introduce large estimation errors. To address this issue, this paper reconstructs the prediction and update steps of Gaussian filtering as solutions to two distinct optimization problems, whose optimal conditions are found to have analytical forms from Stein's lemma. It is observed that the stationary point for the prediction step requires calculating the first two moments of the prior distribution, which is equivalent to that step in existing moment-matching filters. In the update step, instead of linearizing the model to approximate the stationary points, we propose an iterative approach to directly minimize the update step's objective to avoid linearization errors. For the purpose of performing the steepest descent on the Gaussian manifold, we derive its natural gradient that leverages Fisher information matrix to adjust the gradient direction, accounting for the curvature of the parameter space. Combining this update step with moment matching in the prediction step, we introduce a new iterative filter for nonlinear systems called \textit{N}atural Gr\textit{a}dient Gaussia\textit{n} Appr\textit{o}ximation filter, or NANO filter for short. We prove that NANO filter locally converges to the optimal Gaussian approximation at each time step. Furthermore, the estimation error is proven exponentially bounded for nearly linear measurement equation and low noise levels through constructing a supermartingale-like property across consecutive time steps.

A Novel Gaussian Nonlinear Iterated Update Filtering

A Gaussian Nonlinear Iterated Update Filter

Maneuvering Target Tracking with an Adaptive UKF Filter

Nonlinear Bayesian Filtering with Natural Gradient Gaussian Approximation

A Novel Adaptive Unscented Kalman Filter for Nonlinear Estimation

A Novel Extended Unscented Kalman Filter is Designed Using the Higher-Order Statistical Property of the Approximate Error of the System Model

The Geometric Unscented Kalman Filter

Information Geometric Approach to Recursive Update in Nonlinear Filtering.

Gaussian Filter for Nonlinear Systems with One-Step Randomly Delayed Measurements.

Nonlinear filtering of SINS/GPS integrated navigation

An Improved Gaussian Filter with Asynchronously Correlated Noises

Comparing of Tracking Performances of Several Filter Algorithms

Algorithm of Gaussian Sum Based Cubature Kalman Filter for Non-Gaussian Systems

An improved Gaussian filter with randomly delayed measurements and synchronously correlated noises

A Novel Kalman Filter for Nonlinear Non-Gaussian Systems and Its Applications in Self-localization

A Novel Adaptive Unscented Kalman Filter

Gaussian non-linear filter algorithm based on integrated navigation system

Gaussian Filter Estimation Framework for Nonlinear Systems with Multiple Non-Ideal Scenarios

Gaussian Sum Filtering Methods for Nonlinear Non-Gaussian Models

Accurate Gaussian Sum-filter for Continuous-discrete Nonlinear Systems with Non-Gaussian Noise

A New Nonlinear State Estimator Using the Fusion of Multiple Extended Kalman Filters