Development and Investigation of Real-time Hybrid Navigation System Using a DCM Based Integration Method

J. Zhang, E. Edwan, J. Zhou, O. Loffeld

Abstract:	Costs and sizes are two important factors of commercial navigation devices apart from the navigation performance. Due to the reduced costs and smaller sizes brought by the Micro-Electro-Mechanical Systems (MEMS) technology, integrating the MEMS Inertial Measurement Units (IMUs) in the real-time hybrid navigation system is a hot topic today. However, handling large errors of the MEMS inertial sensors, such as the drifting bias is still the main difficulty which hinters the real-time performance of such MEMS-IMU/GPS integration. This paper applies a DCM based loosely-coupled integration approach for fusing the MEMS inertial sensors with GPS and validates its practical feasibility in real-time by implementing it on a real-time GPS/MEMS-IMU hybrid navigation system. Two important features from the integration are achieved. One is that the Direction Cosine Matrix (DCM) elements are estimated instead of the Euler angles. The other is that the total state Kalman filtering (KF) is adopted instead of the commonly used INS error model. By doing so, the great benefits obtained include the pseudo-linear process model for the DCM estimation, the ability to detect gyro bias. The last term would be very attractive to real-time applications with MEMS gyros. Though the computation complexity increased due to the enlarged state vectors, the well scheduled system flow guarantees that our real-time system works efficiently within the tight timing budget. Moreover, two improvements are proposed for the system. The dynamic measurement update of the filter, in which the quality of GPS solutions is taken into consideration, improves quality of filter update, while the orthonormalization of the DCM matrix corrects part of the computational error of the DCM and consequently improves the attitude estimation. Field experiments in a land vehicle are carried out with a selected trajectory in the city to test the developed hybrid system. The system feasibility is validated. Furthermore, the navigation performance is investigated and discussed with the real-time data.
Published in:	Proceedings of the 23rd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2010) September 21 - 24, 2010 Oregon Convention Center, Portland, Oregon Portland, OR
Pages:	2986 - 2994
Cite this article:	Zhang, J., Edwan, E., Zhou, J., Loffeld, O., "Development and Investigation of Real-time Hybrid Navigation System Using a DCM Based Integration Method," Proceedings of the 23rd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2010), Portland, OR, September 2010, pp. 2986-2994.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In