Improved Mixture Correntropy Cubature Kalman Filter for Attitude Estimation

This work proposes an algorithm based on improved mixture correntropy cubature Kalman filtering to address the issues of low accuracy and susceptibility to complex non-Gaussian noise and outlier interference in inertial navigation attitude estimation. First, a combination of Gaussian kernel and Cauchy kernel is proposed to construct the mixture correntropy, aiming to address the issue of single kernel-based correntropy being inadequate when dealing with complex non-Gaussian noise. Second, the objective function is established utilizing the model fitting loss based on mean square error and measurement fitting loss based on mixture correntropy. The maximum correntropy criterion is utilized to replace the minimum mean square error criterion, and the fixed-point iteration method is used to solve the objective function. This process derives the mixture correntropy matrix, which adjusts the measurement noise covariance. Finally, a membership function is used to determine the mixture correntropy coefficient. Accordingly, the algorithm can adaptively select the proportions of each kernel function based on the respective noise interference scenario. Simulations and dynamic–static experiments have been conducted. The algorithm has been compared with single kernel-based correntropy algorithms and other robust algorithms to confirm its superior precision and stability under complex non-Gaussian noise and outlier interference conditions.