Robust Integrated Navigation Signal Processing Method Using Variational Bayesian Cubature Kalman Filter with Chi-Square Test

This paper considers a robust integrated navigation system that copes with the time-varying measurement noise in a hostile environment. The Chi-square test is used to get rid of the disturbance of measurement outlier. And the estimation of measurement noise is provided by the variational Bayesian method to perform an accuracy state estimation within a challenged navigation environment. With regard to a practical system, the computation burden of the proposed algorithm is vital for the embedded system resource. The detailed numerical simulations demonstrate that the proposed filter has an advantage over the traditional filter in global performance.