Adaptive Hybrid Standard/Unscented Kalman Filtering Mechanization for GPS/MEMS IMU Integration

D. An, D. Liccardo, and J. Rios

Abstract:	This paper proposes a new hybrid Kalman filter, which innovatively takes advantage of standard (/extended) and unscented Kalman filtering, and which is tailored for filtering applications with linear/simple (well-behaved non-linear or Jacobian matrix-ready) nonlinear system and nonlinear measurements, especially Global Position System (GPS)/Micro-Electro-Mechanical Systems (MEMS) Inertial Measurement Unit (IMU) integration. In this paper, based on characterization of filtering mechanization of various GPS/IMU integrations and the extended and unscented Kalman filters, an adaptive hybrid standard (/extended) /unscented Kalman filter is proposed to achieve higher accuracy, more flexibility, simple handling, and adaptive functionality of uncertainties of inertial sensor errors and parameters for sigma point generation. With the proposed filter mechanization, the time propagation of standard/extended Kalman filter is employed, while a sigma point based measurement update as well as a filtering adaptive strategy is applied. Besides the theoretical formulation of the hybrid Kalman filter, the performance of the new hybrid Kalman filter is illustrated by experimental simulation results on a tightly-coupled GPS/MEMS IMU integration problem.
Published in:	Proceedings of the 2006 National Technical Meeting of The Institute of Navigation January 18 - 20, 2006 Hyatt Regency Hotel Monterey, CA
Pages:	656 - 663
Cite this article:	An, D., Liccardo, D., Rios, J., "Adaptive Hybrid Standard/Unscented Kalman Filtering Mechanization for GPS/MEMS IMU Integration," Proceedings of the 2006 National Technical Meeting of The Institute of Navigation, Monterey, CA, January 2006, pp. 656-663.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In