Distributed State Estimation for GPS Navigation: The Correntropy Extended Kalman Filter Approach

In this study, we propose a new filter for nonlinear systems called the Correntropy Extended Kalman Filter (CEKF). The correntropy-based cost function used in the CEKF is more effective at mitigating the effects of impulsive noise, making it a valuable tool for filtering in real-world applications with non-Gaussian noise. The CEKF algorithm is a promising solution for Global Positioning System (GPS)-based navigation applications as it outperforms the KF and EKF algorithms regarding accuracy and robustness. By incorporating correntropy into state estimation algorithms, the proposed algorithm could improve navigation performance in challenging environments, such as urban canyons or under heavy foliage, where traditional algorithms may struggle to provide accurate estimates. This work highlights the potential benefits of the CEKF algorithm for real-world applications and demonstrates its effectiveness in improving state estimation accuracy.