Robust control and data reconstruction for nonlinear epidemiological models using feedback linearization and state estimation

Csutak Balázs; Szederkényi Gábor: Robust control and data reconstruction for nonlinear epidemiological models using feedback linearization and state estimation.
MATHEMATICAL BIOSCIENCES AND ENGINEERING, 22 (1). pp. 109-137. ISSN 1547-1063 (2025)

Szöveg
10.3934_mbe.2025006.pdf - Megjelent verzió
Download (1MB)

Mű típusa:

Folyóiratcikk

Szerző azonosítók:

Név	ORCID	MTMT szerző azonosító
Csutak Balázs		10083853
Szederkényi Gábor	0000-0003-4199-6089	10000614

Absztrakt (kivonat):

It has been clearly demonstrated over the past years that control theory can provide an efficient framework for the solution of several complex tasks in epidemiology. In this paper, we present a computational approach for the state estimation based reference tracking control and historical data reconstruction using nonlinear compartmental epidemic models. The control model is given in nonlinear input-affine form, where the manipulable input is the disease transmission rate influenced by possible measures and restrictions, while the observed or computed output is the number of infected people. The control design is built around a simple SEIR model and relies on a feedback linearization technique. We examine and compare different control setups distinguished by the availability of state information, complementing the directly measurable data with an extended Kalman filter used for state estimation. To illustrate the capabilities and robustness of the proposed method, we carry out multiple case studies for output tracking and data reconstruction on Swedish and Hungarian data, all in the presence of serious model and parameter mismatch. Computation results show that a well-designed feedback, even in the presence of significant observation uncertainties, can sufficiently reduce the effect of modeling errors.

Folyóirat címe:

MATHEMATICAL BIOSCIENCES AND ENGINEERING

Megjelenés éve:

2025

Kötet:

Szám:

Oldalak:

pp. 109-137

ISSN:

1547-1063

Intézmény:

Pázmány Péter Katolikus Egyetem