Abstract:A stable square-root approach has been recently proposed for the unscented Kalman filter (UKF) and fifth-degree cubature Kalman filter (5D-CKF) as well as for the mixed-type methods consisting of the extended Kalman filter (EKF) time update and the UKF/5D-CKF measurement update steps. The mixed-type estimators provide a good balance in trading between estimation accuracy and computational demand because of the EKF moment differential equations involved. The key benefit is a consolidation of reliable state mean and error covariance propagation by using delicate discretization error control while solving the EKF moment differential equations and an accurate measurement update according to the advanced UKF and/or 5D-CKF filtering strategies. Meanwhile the drawback of the previously proposed estimators is an utilization of sophisticated numerical integration scheme with the built-in discretization error control that is, in fact, a complicated and computationally costly tool. In contrast, we design here the mixed-type methods that keep the same estimation quality but reduce a computational time significantly. The novel estimators elegantly utilize any MATLAB-based numerical integration scheme developed for solving ordinary differential equations (ODEs) with the required accuracy tolerance pre-defined by users. In summary, a simplicity of the suggested estimators, their numerical robustness with respect to roundoff due to the square-root form utilized as well as their estimation accuracy due to the MATLAB ODEs solvers with discretization error control involved are the attractive features of the novel estimators. The numerical experiments are provided for illustrating a performance of the suggested methods in comparison with the existing ones.

A Square-Root Kalman Filter Using Only QR Decompositions

The Square-Root Unscented and the Square-Root Cubature Kalman Filters on Manifolds

Approved Square Root Cubature Kalman Filtering and Its Application to POS

Improved Discrete-Time Kalman Filtering within Singular Value Decomposition

MATLAB-based general approach for square-root extended-unscented and fifth-degree cubature Kalman filtering methods

Square-root filtering via covariance SVD factors in the accurate continuous-discrete extended-cubature Kalman filter

Robust Square Root Unscented Kalman filter of graph signals

Continuous-discrete derivative-free extended Kalman filter based on Euler-Maruyama and Itô-Taylor discretizations: Conventional and square-root implementations

Numerically robust square root implementations of statistical linear regression filters and smoothers

Continuous–discrete derivative-free extended Kalman filter based on Euler–Maruyama and Itô–Taylor discretizations: Conventional and square-root implementations

Tensor network square root Kalman filter for online Gaussian process regression

Strong Tracking Adaptive Square-Root Cubatu re Kal man Filter Algorith m

Inverse Cubature and Quadrature Kalman filters

Continuous-discrete unscented Kalman filtering framework by MATLAB ODE solvers and square-root methods

Square Root Unscented Kalman Filter With Modified Measurement for Dynamic State Estimation of Power Systems

A quantum algorithm for the Kalman filter using block encoding

Constructing numerically stable Kalman filter-based algorithms for gradient-based adaptive filtering

A Quantum Extended Kalman Filter

Quaternion kernel recursive least-squares algorithm

Convergence of the Square Root Ensemble Kalman Filter in the Large Ensemble Limit

UltimateKalman: Flexible Kalman Filtering and Smoothing Using Orthogonal Transformations