DocumentCode :
3616295
Title :
Derivative observations used in predictive control
Author :
J. Kocijan;D.J. Leigth
Author_Institution :
Jozef Stefan Inst., Ljubljana, Slovenia
Volume :
1
fYear :
2004
fDate :
6/26/1905 12:00:00 AM
Firstpage :
379
Abstract :
Gaussian processes provide approach to probabilistic nonparametric modelling which allows a straightforward combination of measured data and local linear models in an empirical model. This is of particular importance in the identification of nonlinear dynamic systems from experimental data where usually more data are available far from equilibrium points. We illustrate the utility of such simple nonlinear predictive control example.
Keywords :
"Predictive control","Gaussian processes","Predictive models","Safety","Prediction algorithms","Random variables","Covariance matrix","Bayesian methods","Space stations"
Publisher :
ieee
Conference_Titel :
Electrotechnical Conference, 2004. MELECON 2004. Proceedings of the 12th IEEE Mediterranean
Print_ISBN :
0-7803-8271-4
Type :
conf
DOI :
10.1109/MELCON.2004.1346883
Filename :
1346883
Link To Document :
بازگشت