High Degree Cubature Kalman Filters for Nonlinear Systems with Correlated Noises

Abstract

To solve the accuracy degeneracy of traditional high-degree cubature Kalman filter (HCKF) with cross-correlation between process noises and measurement noises at the same time, this paper proposes the improved HCKFs. Through fifth-degree cubature rule and different decor related principles, the frames of proposed filters in the approximated minimum mean square error sense are derived. The air-traffic maneuvering target tracking simulations are performed among the improved filters and traditional HCKF. Simulations results demonstrate the proposed filters not only can achieve almost the same accuracy as the traditional HCKF with independent white noise sequences, but also have superior performance to traditional HCKF while the noises are cross-correlated at the same time.