Fuzzy Adaptive Stabilization of Higher Order Kalman Filters in Application to Precision Kinematic GPS

K. S. Mostov and A. A. Soloviev

Abstract:	Kinematic GPS measurements are used to estimate absolute coordinates and velocities of moving vehicles. Particularly, when GPS receivers are integrated into navigation systems, they serve as sensors providing input to the navigation computer. The major objective of data processing algorithms implemented in the computer is to separate useful signals from measurement noise. Currently, the most effective tool for reduction of random noise in real-time systems is the Kahnan filter. However, the uncertainty in the specification of the vehicle dynamics model can lead to instability of the Kalman filter and substantial navigation errors. Proposed in this paper is a method for Kalman filter stabilization, based on joining the least mean squares (LMS) method and Kahnan filtering. Our approach allows one to incorporate the simple method of weight function construction for the LMS method with the recursive Kahnan filtering approach. We propose an attractive computational method for the construction of weight functions on the basis of fuzzy theory. The algorithm has been simulated using MATLAB Fuzzy Toolbox and tested on several examples.
Published in:	Proceedings of the 9th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 1996) September 17 - 20, 1996 Kansas City, MO
Pages:	1451 - 1456
Cite this article:	Mostov, K. S., Soloviev, A. A., "Fuzzy Adaptive Stabilization of Higher Order Kalman Filters in Application to Precision Kinematic GPS," Proceedings of the 9th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GPS 1996), Kansas City, MO, September 1996, pp. 1451-1456.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In