A Comparison of Nonlinear Filtering Approaches for In-motion Alignment of SINS

Abstract

As for in-motion alignment of strapdown inertial navigation system (SINS), two groups of nonlinear filtering approaches are compared in this paper: Gaussian approximation and Monte Carlo simulation. The former group, consisting of the extended kalman filter (EKF) and unscented kalman filter (UKF), approximates probability densities of nonlinear systems using either single or multiple points in a state space, while the latter group, being particle filter (PF), estimate probability densities using random samples. This paper presents the in-motion alignment of SINS using different nonlinear filters, which allows large initial attitude error uncertainty. The results of an extensive simulation study are implemented in which PF is compared to EKF and UKF for handling the non-Gaussian problem. The comparison demonstrates that the convergence rate and alignment precision of the particle filter are superior to that of EKF and UKF under large misalignment uncertainty.