Online clustering of switching models based on a subspace framework

This paper deals with the modelling of switching systems and focuses on the characterization of the local functioning modes using the online clustering approach. The system considered is represented as a weighted sum of local linear models where each model could have its own structure. This implies that the parameters and the order of the switching system could change when the system switches. Moreover, possible constants of the local models are also unknown. The method presented consists of two steps. First, an online estimation method of the Markov parameters matrix of the local linear models is established. Secondly, the labelling of these parameters is done using a dynamical decision space worked out with learning techniques; each local model being represented by a cluster. The paper ends with an example and a discussion with an aim of illustrating the method’s performance.