MINES ParisTech CAS - Centre automatique et systèmes

Generalized Multiplicative Extended Kalman Filter for Aided Attitude and Heading Reference System

Authors: Philippe Martin, Erwan Salaün, 2010 AIAA Guidance, Navigation, and Control Conference, AIAA 2010-8300, 2 - 5 August 2010, Toronto, Ontario Canada
In this paper, we propose a “Generalized Multiplicative Extended Kalman Filter” (GMEKF) to estimate the position and velocity vectors and the orientation of a flying rigid body, using measurements from lowcost Earth-fixed position and velocity, inertial and magnetic sensors. Thanks to well-chosen state and output errors, the gains and covariance equations converge to constant values on a much bigger set of trajectories than equilibrium points as it is the case for the standard Multiplicative Extended Kalman Filter (MEKF). We recover thus the fundamental properties of the Kalman filter in the linear case, especially the convergence and optimality properties, for a large set of trajectories, and it should result in a better convergence of the estimation. We illustrate the good performance and the nice properties of the GMEKF on simulation and on experimental comparisons with a commercial system.
Download PDF
BibTeX:
@Proceedings{,
author = {Philippe Martin, Erwan Salaün},
editor = {},
title = {Generalized Multiplicative Extended Kalman Filter for Aided Attitude and Heading Reference System},
booktitle = {AIAA Guidance, Navigation, and Control Conference},
volume = {},
publisher = {},
address = {Toronto, Ontario Canada},
pages = {AIAA 2010-8300},
year = {2010},
abstract = {In this paper, we propose a “Generalized Multiplicative Extended Kalman Filter” (GMEKF) to estimate the position and velocity vectors and the orientation of a flying rigid body, using measurements from lowcost Earth-fixed position and velocity, inertial and magnetic sensors. Thanks to well-chosen state and output errors, the gains and covariance equations converge to constant values on a much bigger set of trajectories than equilibrium points as it is the case for the standard Multiplicative Extended Kalman Filter (MEKF). We recover thus the fundamental properties of the Kalman filter in the linear case, especially the convergence and optimality properties, for a large set of trajectories, and it should result in a better convergence of the estimation. We illustrate the good performance and the nice properties of the GMEKF on simulation and on experimental comparisons with a commercial system.},
keywords = {}}