Survey of a controller design method based on experimental data and a proposal of data conversion method

摘要: Several methods have been proposed for designing a controller using experimental data directly without identifying controlled plant. Among them, optimal design from one-time data, such as Virtual Reference Feedback Tuning (VRFT), Fictitious Iterative (FRIT), and Noniterative Correlation-based (NCbT), are especially expected to reduce the time cost of controller. VRFT FRIT make reference signal data. In contrast, NCbT is method that removes noise influence by determining parameters so not interrelation correlation function between noise. However, cannot deal with include noises. Although can be used an noise, it limited cases where there no this paper, we show summary conventional argue about problem We also propose conversion linearity after applying spline fitting instead directly. addition, discuss advantages method.