Dynamic Stochastic Modeling of Inertial Sensors for INS/GNSS Navigation

M. Wis, I. Colomina

Peer Reviewed

Abstract:	Alan Variance is a sophisticated method that allows a detailed characterization of the error model of a determined IMU under static conditions. However, it is known that this static characterization fails with low cost IMUs when a set of dynamics is applied to the sensor. In order to understand this behavior a series of experiments with a low cost inertial device rigidly attached to a navigation grade reference IMU has been performed. Once the synchronization and alignment of both sensors is corrected, it is possible to compare the measurements of the two sensors and to compute the error of the low-cost IMU related to the dynamics measured by the reference IMU. In this comparison it was found that there is a direct relationship between the low-cost IMU errors and the high order dynamics. . This relationship has been modeled through an equation in which a coefficient is applied to a series of n-th order derivatives used to model the high order dynamics. These coefficients can be estimated through a least-squares adjustment when reference measurements are available. The result of this adjustment can be transformed into a set of improved IMU observations that can be processed with an ordinary INS/GNSS navigator. After comparing the results of processing the “original” observations with the “improved” observations it is observed that there is a noticeable improvement in the results of the navigation solution. In addition to this, these coefficients can also be implemented as calibration states that can be estimated with a Kalman Filter. These “extended calibration parameters” have been implemented and tested in a loosely coupled INS/GNSS, and it has also been found that trajectory performance is improved too. These preliminary results suggest a modelling first approach of low-cost sensor modeling that may help in reducing some of the errors inherent to the dynamics applied to the sensors.
Published in:	Proceedings of the 28th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2015) September 14 - 18, 2015 Tampa Convention Center Tampa, Florida
Pages:	2013 - 2019
Cite this article:	Wis, M., Colomina, I., "Dynamic Stochastic Modeling of Inertial Sensors for INS/GNSS Navigation," Proceedings of the 28th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2015), Tampa, Florida, September 2015, pp. 2013-2019.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In