Identification of Task Model Using Piecewise Affine Systems with Variable Switching Points and its Application

Abstract

This paper considers a human-operated task model based on identification techniques using switched systems expressed by mixed logical dynamical systems. In the system, each subsystem is expressed by a piecewise affine system. For a model to be identified, it is not easy to determine a number of subsystems in advance. In this paper, we propose a simultaneous identification technique which takes into account both a number of subsystems and parameters so that a piecewise affine model of a human operation can be obtained. In addition, the obtained model is applied to automation. The effectiveness of the proposed method is illustrated through simulations and experiments.