Unscented Kalman Filter and Attitude Errors of Low-Cost Inertial Navigation Systems

Eun-Hwan Shin and Naser El-Sheimy

Peer Reviewed

Abstract:	This paper deals with the design and implementation of the unscented Kalman filter (UKF) for aided inertial navigation. A zero integrated heading rate measurement model is also developed to fix the heading drift during zero velocity updates. The UKF is different from the extended Kalman filter (EKF), a currently used approach, in dealing with nonlinearities which occur when large attitude errors exist. Hence, the behavior of the two filters was tested with large initial attitude errors. It is shown in the tests that the UKF outperforms the EKF with a standard small error model; the UKF converged faster than the EKF. Therefore, the UKF can be a preferable choice for low-cost inertial navigation systems, where the heading error can grow rapidly.
Published in:	NAVIGATION: Journal of the Institute of Navigation, Volume 54, Number 1
Pages:	1 - 9
Cite this article:	Shin, Eun-Hwan, El-Sheimy, Naser, "Unscented Kalman Filter and Attitude Errors of Low-Cost Inertial Navigation Systems", NAVIGATION: Journal of The Institute of Navigation, Vol. 54, No. 1, Spring 2007, pp. 1-9.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In