System identification of a galvano scanner using input-output data obtained from positioning control

To achieve high-speed and high-precision control, an accurate plant model is required. Since the plant parameters may vary by an environmental change and so on, it is desirable to identify the plant parameters during positioning control. In the system identification, a pseudo random signal is in general used to fully excite the dynamics of the plant. Therefore, when the input-output data obtained from positioning control is used, the identification accuracy may be degraded. In this study, we propose methods to improve the identification accuracy by introducing appropriate pre-processes for the input-output data. Further, a method to identify the mechanical characteristic of the plant is proposed. The effectiveness of the proposed methods is evaluated by using the input-output data obtained from control experiments of a galvano scanner.