State of the Art on Hybrid Systems Identification

Hybrid systems has attracted the attention of many researchers in the recent years because of their ability to capture the mixed continuous and discrete dynamics of many real systems. In the recent years many work has been devoted to the formalization of such models and to the control of systems presenting hybrid dynamics. But the identification of systems based on hybrid models has not coped as much attention as the control until recently. Still, some methods have been developed in the recent years which try to identify models of hybrid systems using diverse techniques. In the following sections we will try to present the problem addressed under the name of Hybrid Systems Identification, and do a brief review of the methods solving this problem currently available. A problem associated to the identification of hybrid linear systems is the formalization of the observability of such systems. This problem has also received much attention in the recent years and much literature [1, 2] is available for different models and assumptions on the systems under study. In this paper we will not include results in this area, but we will concentrate in the theory for identification/estimation of models and the algorithms to achieve that [3, 4, 5, 6, 7, 8, 9]. In Section 2 we present a formalization of a general class of Hybrid Systems, Section 3 is devoted to the presentation of the general identification problem as well as different solutions to particularizations of it. In Section 4 some current open issues on the topic can be found. Finally Section 5 concludes the document with a short summary and conclusions.