Data Driven Dynamic Decoupling Control for MIMO Precision Mechatronic Systems

Decoupling control is extensively employed in precision mechatronic systems, particularly within the semiconductor field. In this context, the inputs and outputs are effectively decoupled through the utilization of a static decoupling matrix. However, the prevalent challenge arises from cross-talk induced by errors inherent in the static decoupling process. This challenge poses a significant obstacle in meeting the stringent demands of high-speed and high-precision motion, stemming from inaccuracies in mechanical parameters and variations in motor characteristics, particularly under conditions of high dynamics. To enhance the performance of precision mechatronic systems, it is recommended to replace the decoupling matrix with a dynamic decoupling controller. In this paper, a data driven dynamic decoupling control method is developed to improve the decoupling accuracy by calibrating the mechanical parameters and compensating the differences of motors. The effectiveness and practicality of the proposed method are illustrated through numerical simulations.