No label defined (Q3801455)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 4068756
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 4068756 |
Statements
1988
0 references
nonlinear filtering
0 references
Kalman filter
0 references
robustness
0 references
Zakaï equation
0 references
Wei- Norman theory
0 references
finite-dimensional recursive optimal filter
0 references
estimation Lie algebra
0 references
weak cubic sensor
0 references
0.8444057703018188
0 references
0.832609236240387
0 references
[For the entire collection see Zbl 0642.00023.] The aim of these lectures is to introduce recent developments of nonlinear filtering theory. The problem under study is the continuous- time nonlinear filtering problem where the signal and observation noises are Brownian and independent. First, the author reviews some classical sights of the theory: the Kalman filter for the linear case, the robustness, the Zakaï equation. Then he enters the heart of the problem; after an introduction to the Wei-Norman theory, he explains how the existence and the construction of a finite-dimensional recursive optimal filter is related to the algebraic properties of some Lie algebra, the estimation Lie algebra. This problem has motivated much work in the last decade. In these lectures, several examples are worked out. However, situations in which finite-dimensional filters exist are very special. So, for other cases, the author introduces a method for building approximate finite-dimensional filters by means of the estimation Lie algebra. This method permits the study of small perturbations of solvable problems, such as the so-called weak cubic sensor; it raises many open problems. [For an introduction to differentiable manifolds, see ibid. 303, 317-340 (1988; Zbl 0654.58002).]
0 references