4. Estimator for a nonlinear or non-Gaussian dynamic state model

4.1 Introduction

The optimality of the Kalman filter depends on the linearity of the state-dynamic system. In practice, few systems are linear. Two solutions are therefore possible. The first is to approximate the system formed by the equations (5) and (7) by a linear state-dynamic system. By linearizing the state and measurement equations, the Kalman filter can be used to estimate the state from past and current observations. This linearization...