A differential geometric approach to nonlinear filtering: The projection filter

被引：56

作者：

Brigo, D ^{[1
]}

Hanzon, B

LeGland, F

机构：

[1] CARIPLO Bank, I-20121 Milan, Italy

[2] Free Univ Amsterdam, Dept Econometr, NL-1081 HV Amsterdam, Netherlands

[3] Inst Natl Rech Informat & Automat, IRISA, F-35042 Rennes, France

来源：

IEEE TRANSACTIONS ON AUTOMATIC CONTROL | 1998年 / 43卷 / 02期

关键词：

differential geometry and statistics; exponential families; finite-dimensional filter; nonlinear filtering; projection filter;

D O I：

10.1109/9.661075

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper presents a new and systematic method of approximating exact nonlinear filters with finite dimensional filters, using the differential geometric approach to statistics, The projection filter is defined rigorously in the case of exponential families. A convenient exponential family is proposed which allows us to simplify the projection filter equation and to define an a posteriori measure of the local error of the projection filter approximation. Finally, simulation results are discussed for the cubic sensor problem.

引用

页码：247 / 252

页数：6

共 16 条

[1] Amari S., 1985, LECT NOTES STAT, V28, DOI DOI 10.1007/978-1-4612-5056-2
[2] [Anonymous], 1992, APPL MATH
[3] Brigo D, 1995, PROCEEDINGS OF THE 34TH IEEE CONFERENCE ON DECISION AND CONTROL, VOLS 1-4, P4006, DOI 10.1109/CDC.1995.479232
[4] New results on the Gaussian protection filter with small observation noise
Brigo, D
[J]. SYSTEMS & CONTROL LETTERS, 1996, 28 (05) : 273 - 279
[5] On the nice behaviour of the Gaussian projection filter with small observation noise
Brigo, D
[J]. SYSTEMS & CONTROL LETTERS, 1995, 26 (05) : 363 - 370
[6] BRIGO D, 1996, THESIS VRIJE U AMSTE
[7] Brigo D., 1996, P 2 PORTUGUESE C AUT, P251
[8] BRIGO D, 1996, 914 IRISA
[9] BRIGO D, 1996, 79618 TI TINB I
[10] DAVIS MHA, 1981, NATO ADV STUDY I SER

← 1 2 →