Identification of switched systems with unknown switch points and its application

System identification covers a very wide range of techniques for obtaining a system model from its input-output data. In this paper, we address the identification problem of switched systems, by focusing our attention on hinging hyperplane autoregressive exogenous models (HHARX), which can be identified efficiently via mixed-integer programming. As an example, the problem of extracting operation modes in a driving system of a small toy vehicle and their switch conditions is formulated. The formulation of the identification problem of switched systems with unknown switch points is based on the hybrid dynamical system theory. The effectiveness of HHARX model approach in switched system identification is verified by simulation.