Adaptive High-degree Cubature Kalman Filter with Unknown Noise Statistics

Abstract

This paper is concerned with the state estimation problem for nonlinear systems with unknown covariance of process noise. The advantages of recently developed High-degree Cubature Kalman Filter (HCKF) are signiflcant with its easy to implement and better estimation accuracy. However, it has bad robustness on modeling uncertainty for practical applications. To overcome the limitations of the HCKF, an Adaptive HCKF (AHCKF) is proposed by combing strong tracking flltering and Sage-Husa estimator. In the proposed state estimator, a fading factor is used to correct one state prediction covariance while the SageHusa estimator is adopted to recursively estimate the unknown process noise statistics. Therefore, the AHCKF can obtain better robustness and accuracy comparing with the conventional HCKF. Simulation examples on target tracking are demonstrated the validity of the proposed algorithms.