Robust INS/GPS Coupled Navigation Based on Minimum Error Entropy Kalman Filtering
Hamza Benzerrouk, Rene Jr. Landry, Vladimir A. Nebylov, Alexander V. Nebylov · 2020
This paper addresses the results showing the expanded use or improvement of the accuracy, availability, and/or integrity performance of multisensory navigation systems. In addition, Processing algorithms and methods for multisensory systems are significantly improved when noises are non-Gaussian. In the literature, different modified linear and nonlinear Kalman filters (KFs) were derived under the Gaussian assumption and the well-known minimum mean square error (MMSE) criterion. In order to improve their robustness with respect to impulsive non-Gaussian noises, different algorithms and techniques based on Gaussian sum filtering, Huber based estimators and recently introduced maximum Correntropy criterion (MCC) have recently been used to counter the weakness of the MMSE criterion in developing different versions of robust Kalman filters.