Interval estimation for discrete-time switched linear systems based on $L_\infty$ observer and ellipsoid analysis

Monia Dkhil (1) , Thach Ngoc Dinh (2) , Zhenhua Wang (3) , Tarek Raissi (2) , Messaoud Amairi (1)

1 Modélisation, Analyse et Commande de Systèmes - MACS (Gabès, Tunisie)
2 CEDRIC - LAETITIA - CEDRIC. Traitement du signal et architectures électroniques
3 HIT - Harbin Institute of Technology

This paper proposes a two-step interval estimation method for discrete-time switched linear systems with unknown but bounded uncertainties. Based on the ellipsoidal analysis, the proposed estimator provides upper and lower bounds of the system state with high computational analysis. The size of the obtained ellipsoids is minimized using the trace criterion. The idea of introducing an $L_\infty$ performance is also employed in order to improve the estimation accuracy. Its design conditions are given in terms of Linear Matrix Inequalities (LMIs). Finally, a numerical example emphasizes the effectiveness of the contribution.

Mots clés: Interval estimation; ellipsoid; switched linear systems; L1 performance.

Domaines: Automatique / Robotique

Monia Dkhil, Thach Ngoc Dinh, Zhenhua Wang, Tarek Raissi, Messaoud Amairi. Interval estimation for discrete-time switched linear systems based on $L_\infty$ observer and ellipsoid analysis. 59th IEEE Conference on Decision and Control, Dec 2020, Jeju Island, South Korea. ⟨hal-02900676⟩