A Genetic Fuzzy and Kalman Filtering Model for MEMS-IMU/GPS Integration

Tamer Abdelazim, Walid Abdel-Hamid, Naser El-Sheimy

Abstract:	The Kalman Filter (KF) represents the optimal estimation that can be achieved for integrating signals from systems that have short-term stability systems with reference systems that provide long-term stability. This integration is usually performed utilizing two main steps namely prediction and updating. However, the Kalman Filter works well only under appropriately predefined accurate linear dynamic models. If the filter is exposed to input data that does not fit the dynamic model, it will not result in reliable estimates. Consequently, if the reference signal is absent or the Kalman Filter is working only in prediction mode, the corresponding state estimate errors will quickly drift with time causing a dramatic degradation in the overall accuracy of the integrated system. To overcome such problem, a new design model integrating a Genetic Fuzzy Logic system and KF is described in this paper. An offline Fuzzy model is first identified to simulate the IMU error dynamic characteristics. The identified fuzzy model is then tuned and updated online by a Genetic Algorithm (GA) using the output of KF during the availability of GPS signals. The online trained Genetic Fuzzy model is used to predict the position and velocity errors which are the inputs to the KF during GPS signal outages. The proposed model has been verified with test data of a MEMS inertial system in a land vehicle test. A number of 30 seconds GPS outages were simulated during data processing at different times and along different dynamics on the trajectory of the vehicle. The test results indicate that the proposed Genetic Fuzzy model can efficiently reimburse the GPS updates during GPS short outages.
Published in:	Proceedings of the 2009 International Technical Meeting of The Institute of Navigation January 26 - 28, 2009 Disney's Paradise Pier Hotel Anaheim, CA
Pages:	609 - 616
Cite this article:	Abdelazim, Tamer, Abdel-Hamid, Walid, El-Sheimy, Naser, "A Genetic Fuzzy and Kalman Filtering Model for MEMS-IMU/GPS Integration," Proceedings of the 2009 International Technical Meeting of The Institute of Navigation, Anaheim, CA, January 2009, pp. 609-616.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In