A Review and Applications of the Nonlinear Filters to GNSS/INS Integrated Algorithms

S. Sugimoto, Y. Kubo, M. Tanikawara

Abstract:	In the positioning algorithms for GPS/INS navigation, the error models of positions and azimuths often have nonlin- ear nature such that the nonlinear recursive filters are play- ing important roles. Recently, many attentions focus on the nonlinear filters applying to GNSS positioning algo- rithms, specially to GNSS/INS integration [1]-[10]. This paper contains a review of many classical nonlinear fil- tering methods, beginning from famous Kushner’s nonlin- ear filters for stochastic differential equations in late 1960 [11]-[13], and many approximation methods of nonlinear filters [12], [14]-[16], which arrive to recent approxima- tion filters such as the unscented Kalman filter [17, 18], particle (or Monte Carlo) filters [19]-[21], and Gaussian filters [22]-[23]. Especially, the Gaussian sum filter [16] and the stochastic equivalent linearlization method [15] in the classical nonlinear filters are deeply reviewed so that a new Gaussian sum filter are derived. Finally we consider which nonlinear filters are favorable for GPS/INS position- ing from the aspects of the computational cost and estima- tion accuracy.
Published in:	Proceedings of the 22nd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2009) September 22 - 25, 2009 Savannah International Convention Center Savannah, GA
Pages:	3101 - 3113
Cite this article:	Sugimoto, S., Kubo, Y., Tanikawara, M., "A Review and Applications of the Nonlinear Filters to GNSS/INS Integrated Algorithms," Proceedings of the 22nd International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2009), Savannah, GA, September 2009, pp. 3101-3113.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In