Design of a Data-Oriented Evolutionary Controller

Abstract

Recently, the data-oriented control method that directly designs the control parameters from the closed-loop data has been attracting attention. However, the structure of the controller should be determined previously in order to design a controller. Therefore, it is difficult to get the desired control performances for some complicated systems. In this paper, a new controller design method that designs both the controller structure and the control parameters simultaneously is proposed. The controller structure is determined by a genetic algorithm, and the control parameters are computed by using the least squares method. Furthermore, an experimental example is examined in order to verify the effectiveness of the proposed control scheme.