A COMBINED APPROACH TO SYSTEM IDENTIFICATION OF A CLASS OF HYBRID SYSTEM

Hybrid Systems consist of continuous time and/or discrete time processes interfaced with some logical or decision making process. In this paper, a class of hybrid systems - switched linear systems is considered. It is shown that for this class of hybrid systems, it is possible to combine subspace methods with mixed integer programming. While most approaches are based on an input-output framework, we a state space identification approach is advocated. The states of the system are extracted from input-output data using sub-space methods. Once these states are known, the switched system is re-written as a mixed logical dynamical (MLD) system and the model parameters are solved for via mixed integer programming. An example is reported at the end of this paper.