This article analyzes the stability-related properties of long short-term mem-ory (LSTM) networks and investigates their use as the model of the plant in the design of model predictive controllers (MPC). First, sufficient conditions guaranteeing the Input-to-State stability (ISS) and Incremental Input-to-State stability (deltaISS) of LSTM are derived. These properties are then exploited to design an observer with guaranteed convergence of the state estimate to the true one. Such observer is then embedded in a MPC scheme solving the tracking problem. The resulting closed-loop scheme is proved to be asymptotically stable. The training algorithm and control scheme are tested numerically on the simulator of a pH reactor, and the reported results confirm the effectiveness of the proposed approach.

Learning model predictive control with long short-term memory networks

F. Bonassi;M. Farina;R. Scattolini
2021-01-01

Abstract

This article analyzes the stability-related properties of long short-term mem-ory (LSTM) networks and investigates their use as the model of the plant in the design of model predictive controllers (MPC). First, sufficient conditions guaranteeing the Input-to-State stability (ISS) and Incremental Input-to-State stability (deltaISS) of LSTM are derived. These properties are then exploited to design an observer with guaranteed convergence of the state estimate to the true one. Such observer is then embedded in a MPC scheme solving the tracking problem. The resulting closed-loop scheme is proved to be asymptotically stable. The training algorithm and control scheme are tested numerically on the simulator of a pH reactor, and the reported results confirm the effectiveness of the proposed approach.
2021
learning-based control, long short-term memory neural networks, machine learning, nonlinear model predictive control, output feedback predictive control
File in questo prodotto:
File Dimensione Formato  
rnc.5519.pdf

Accesso riservato

Descrizione: Paper
: Publisher’s version
Dimensione 728.98 kB
Formato Adobe PDF
728.98 kB Adobe PDF   Visualizza/Apri
11311-1169896_Bonassi.pdf

accesso aperto

: Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione 462.96 kB
Formato Adobe PDF
462.96 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/11311/1169896
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 40
  • ???jsp.display-item.citation.isi??? 30
social impact