Design of a Data-Driven Control System Based on the Abnormality Using Kernel Density Estimation

The data-driven control scheme has been proposed to control nonlinear systems by adaptively tuning controller parameters based on the database. In the data-driven control scheme, the fixed number of neighbors' data are selected to calculate controller parameters. Therefore, the control performance cannot be improved when the inappropriate neighbors' data are chosen. In other words, the inappropriate controller parameters are calculated when query data is not included in the database. In this study, the data-driven control scheme using the kernel density estimation is proposed. The kernel density estimation can calculate the similarity between query data and database. According to the proposed scheme, controller parameters are calculated based on the abnormality which is obtained by the similarity mentioned above. The effectiveness of the proposed scheme is evaluated by a numerical example.