Kalman filters predictive steps comparison for vehicle localization

摘要

The aim of this paper is to perform a comparison among several different Kalman filters algorithms designed for nonlinear systems. After presenting the most popular of them and showing its limitations, we introduce some new Kalman filters in order to compare them in the vehicle localization context. This comparison is based on the sole use of their predictive steps that corresponds to the worst case that it can occur in vehicle localization (corrective data are unavailable).