Sellami, L.; Zidi, S.; Abderrahim, K. Active mode estimation via clustering algorithm for switched linear systems. (English) Zbl 1467.93074 Derbel, Nabil (ed.) et al., Systems, automation, and control. Selected extended papers from the international conference on systems, analysis and automatic control, Mahdia, Tunisia in 2015. Berlin: De Gruyter/Oldenbourg. Adv. Syst. Signals Devices 5, 89-104 (2018). Summary: The work presented in this paper deals with the active mode identification problem for switched linear systems based on a measurement data set. This problem is an issue closely related to the classification problem of input-output measure data. Indeed, each data group associated with their most appropriate sub-model presents a mode (discrete state) of operation. Therefore, we propose a method for discrete state estimation based on a clustering algorithm combined with a decision mechanism. The clustering algorithm provides the class centers which will be exploited by the decision mechanism in order to identify the discrete state. Simulation results are presented to illustrate the performance of the proposed method.For the entire collection see [Zbl 1417.93035]. MSC: 93B30 System identification 93C30 Control/observation systems governed by functional relations other than differential equations (such as hybrid and switching systems) 93C05 Linear systems in control theory Keywords:switched linear systems; system identification; mode estimation; clustering algorithm PDFBibTeX XMLCite \textit{L. Sellami} et al., Adv. Syst. Signals Devices 5, 89--104 (2018; Zbl 1467.93074) Full Text: DOI