This paper deals with discrete-time linear systems affected by an additive stochastic disturbance and subject to input and state constraints. We consider the problem of designing a linear state-feedback control law so as to minimize a cost function while guaranteeing the satisfaction in probability of the constraints by the steady state solution. The problem can be formulated as a chance-constrained optimization program with constraint satisfaction depending on the stationary state distribution, which in turn depends on the controller gain. A data-driven solution is proposed where constraints are imposed on a finite number of disturbance realizations of finite length. By leveraging the new wait-and-judge paradigm of the scenario approach to address the intrinsic nonconvexity of the resulting optimization problem and by explicitly accounting for the introduced approximation error when using finite length disturbance realizations to approximate the steady state, chance-constrained feasibility of the proposed solution is proven.

A scenario solution to state-feedback controller design for discrete-time linear systems subject to probabilistic constraints

Salizzoni, Giulio;Falsone, Alessandro;Prandini, Maria;Garatti, Simone
2022-01-01

Abstract

This paper deals with discrete-time linear systems affected by an additive stochastic disturbance and subject to input and state constraints. We consider the problem of designing a linear state-feedback control law so as to minimize a cost function while guaranteeing the satisfaction in probability of the constraints by the steady state solution. The problem can be formulated as a chance-constrained optimization program with constraint satisfaction depending on the stationary state distribution, which in turn depends on the controller gain. A data-driven solution is proposed where constraints are imposed on a finite number of disturbance realizations of finite length. By leveraging the new wait-and-judge paradigm of the scenario approach to address the intrinsic nonconvexity of the resulting optimization problem and by explicitly accounting for the introduced approximation error when using finite length disturbance realizations to approximate the steady state, chance-constrained feasibility of the proposed solution is proven.
2022
Proceedings of the 61th Conference on Decision and Control
978-1-6654-6761-2
File in questo prodotto:
File Dimensione Formato  
OneShotSF.pdf

Accesso riservato

: Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione 950.57 kB
Formato Adobe PDF
950.57 kB Adobe PDF   Visualizza/Apri

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: https://hdl.handle.net/11311/1228387
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact