Samenvatting
This work aims at developing methods for estimating nonlinear state-space models of the form
x(t +1) = f (x(t);u(t)) (1)
y(t) = g(x(t))
based on a combination of ideas from the statistical learning community used to solve nonlinear regression problems on one hand, see e.g. [1] and [2], and methods to handle dynamics from the system identification community on the other hand. The approach targets systems which are fairly well approximated by linear models. The proposed method consists of the following steps:
- model the dynamics of the system
- estimate the nonlinear states
- model the nonlinearities
x(t +1) = f (x(t);u(t)) (1)
y(t) = g(x(t))
based on a combination of ideas from the statistical learning community used to solve nonlinear regression problems on one hand, see e.g. [1] and [2], and methods to handle dynamics from the system identification community on the other hand. The approach targets systems which are fairly well approximated by linear models. The proposed method consists of the following steps:
- model the dynamics of the system
- estimate the nonlinear states
- model the nonlinearities
| Originele taal-2 | English |
|---|---|
| Titel | 29th Benelux Meeting on Systems and Control, Heeze, The Netherlands, March 30-1 April, 2010 |
| Status | Published - 30 mrt. 2010 |
| Evenement | Unknown - Stockholm, Sweden Duur: 21 sep. 2009 → 25 sep. 2009 |
Conference
| Conference | Unknown |
|---|---|
| Land/Regio | Sweden |
| Stad | Stockholm |
| Periode | 21/09/09 → 25/09/09 |