In this paper, the derivation of multi-step-ahead prediction models from sampled input-output data of a linear system is considered. Specifically, a dedicated prediction model is built for each future time step of interest. Each model is linearly parametrized in a suitable regressor vector, composed of past output values and past and future input values. In addition to a nominal model, the set of all models consistent with data and prior information is derived as well, making the approach suitable for robust control design within a Model Predictive Control framework. The resulting parameter identification problem is solved through a sequence of convex programs. Convergence of the identified error bounds to their theoretical minimum is demonstrated, under suitable assumptions on the measured data, and features like worst-case accuracy computation are illustrated in a numerical example.

Learning multi-step prediction models for receding horizon control

TERZI, ENRICO;L. Fagiano;M. Farina;R. Scattolini
2018

Abstract

In this paper, the derivation of multi-step-ahead prediction models from sampled input-output data of a linear system is considered. Specifically, a dedicated prediction model is built for each future time step of interest. Each model is linearly parametrized in a suitable regressor vector, composed of past output values and past and future input values. In addition to a nominal model, the set of all models consistent with data and prior information is derived as well, making the approach suitable for robust control design within a Model Predictive Control framework. The resulting parameter identification problem is solved through a sequence of convex programs. Convergence of the identified error bounds to their theoretical minimum is demonstrated, under suitable assumptions on the measured data, and features like worst-case accuracy computation are illustrated in a numerical example.
Proceedings of the European Control Conference
AUT
File in questo prodotto:
File Dimensione Formato  
2018 - ECC - TerziEtAl1.pdf

Accesso riservato

Descrizione: Paper
: Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione 191.64 kB
Formato Adobe PDF
191.64 kB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11311/1062919
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 10
  • ???jsp.display-item.citation.isi??? 9
social impact