Abstract:Recently, the divergence-based minimax approach was proposed and introduced to robust Kalman filtering under model uncertainty and Gaussian noise, and it significantly outperformed the standard Kalman filter. In this study, we consider a new Wasserstein-distribution based Kalman filter in which model uncertainty and non-Gaussian noise exist simultaneously. First, we construct a joint Gaussian distribution of states and observations in state space. Then, we set the equivalent estimator joint Gaussian distribution as a Wasserstein ambiguity set allowable neighborhood centered on the true distribution. Thus, the estimator distribution problem is transformed into a minimax problem, and the model uncertainty is tolerated accordingly. Subsequently, we introduce minimum error entropy to optimize the minimax problem based on Wasserstein ambiguity sets, so as to handle the influence of non-Gaussian noise. The minimax problem constrained by the minimum error entropy is transformed into a semi-positive definite convex optimization problem. By constructing two iterative sub-problems that are mutually conditional, the nonlinear semi-definite program finite convex optimization problem is solved. Finally, the minimum error entropy Wasserstein distribution Kalman filter algorithm is proposed. Additionally, the convergence of the proposed algorithm is clarified, and its effectiveness verified by comparing a series of algorithms in typical simulation scenarios.

Minimum Entropy Filtering for Multivariate Stochastic Systems with Non-Gaussian Noises

A Markovian system approach to distributed H ∞ filtering for sensor networks with stochastic sampling

Entropy optimization based filtering for non-Gaussian stochastic systems

Improved Minimum Entropy Filtering for Continuous Nonlinear Non-Gaussian Systems Using a Generalized Density Evolution Equation

Novel robust minimum error entropy wasserstein distribution kalman filter under model uncertainty and non-gaussian noise

Entropy Optimization Filtering for Fault Isolation of Nonlinear Non-Gaussian Stochastic Systems

Fault detection for non-linear non-Gaussian stochastic systems using entropy optimization principle

H∞ Filtering for Discrete Fuzzy Stochastic Systems with Randomly Occurred Sensor Nonlinearities

A kernel recursive minimum error entropy adaptive filter

Minimum Total Error Entropy Method for Parameter Estimation

Filtering for Stochastic Systems Driven by Poisson Processes

Numerically stable minimum error entropy Kalman filter

Minimum-Variance Recursive Filtering over Sensor Networks with Stochastic Sensor Gain Degradation: Algorithms and Performance Analysis

Probability-Guaranteed Distributed Filtering for Nonlinear Systems on Basis of Nonuniform Samplings Subject to Envelope Constraints

On Mutual Information, Mmse And H-Infinity Entropy In Stochastic Linear Feedback Systems

Mixtures of Gaussians and Minimum Relative Entropy Techniques for Modeling Continuous Uncertainties

Complex-Valued Filtering Based on the Minimization of Complex-Error Entropy.

Generalized Minimum Error Entropy for Adaptive Filtering

Quantized Filtering of Linear Stochastic Systems

Recursive Quantized State Estimation of Discrete-Time Linear Stochastic Systems

Approximate MMSE Estimator for Linear Dynamic Systems With Gaussian Mixture Noise