One-step condensed forms for square-root maximum correntropy criterion Kalman filtering

Abstract

This paper suggests a few novel Cholesky-based square-root algorithms for the maximum correntropy criterion Kalman filtering. In contrast to the previously obtained results, new algorithms are developed in the so-called {\it condensed} form that corresponds to the {\it a priori} filtering. Square-root filter implementations are known to possess a better conditioning and improved numerical robustness when solving ill-conditioned estimation problems. Additionally, the new algorithms permit easier propagation of the state estimate and do not require a back-substitution for computing the estimate. Performance of novel filtering methods is examined by using a fourth order benchmark navigation system example.

References in corpus (4)

• Square-root algorithms for maximum correntropy estimation of linear discrete-time systems in presence of non-Gaussian noise
• Improved Discrete-Time Kalman Filtering within Singular Value Decomposition
• Chandrasekhar-based maximum correntropy Kalman filtering with the adaptive kernel size selection
• Factored-form Kalman-like implementations under maximum correntropy criterion