A parameter identification based on tuning of a controller with one-shot experimental data

In this paper, we propose a new identification method that so as to yeild parameters of a plant by tuning a controller including a nominal plant model. Some of the authors previously have proposed fictitious reference iterative tuning (which is abbreviated to as FRIT[1]) as one of the useful and rational tuning methods of controllers. It enables us to obtain the optimal parameter of a controller based on only one-shot closed loop experiment data and an off-line nonlinear optimization. We apply FRIT to identify the dynamics of a plant in the closed loop by tuning the controller including the parameters of a plant. As a result, it is possible to obtain the parameters of a plant as well as the optimal parameters of a controller. In order to show the simplicity and. usefulness of this method in practical sense, we apply such a new method to identify parameters of a cart system.