Feedback control device that suppresses disturbance vibration using machine learning, article manufacturing method, and feedback control method

An article manufacturing method using a lithography device,

• wherein the lithography device comprises a feedback control device;
• wherein the feedback control device takes information regarding a control deviation, measured at a predefined sampling frequency, between a measured value and a target value of a controlled object as input, and outputs a control amount for the controlled object;
• wherein the feedback control device comprises:
• at least one processor or circuit configured to function as:
• a first control unit configured to take information regarding the control deviation as input, and to output a first control amount for the controlled object;
• a second control unit configured to take information regarding the control deviation as input and to output a second control amount for the controlled object, and in which a parameter for calculating the second control amount is determined by machine learning;
• an operation unit configured to operate the controlled object using the first control amount output from the first control unit and the second control amount output from the second control unit;
• and
• a sampling unit configured to thin out, at a predetermined period, the information regarding the control deviation input to the second control unit so that a sampling frequency of the second control unit is lower than a sampling frequency in the first control unit;
• wherein the lithography device comprises a forming unit configured to form a pattern for lithography using a controlled object that is controlled by a feedback control device;
• and
• wherein the manufacturing method comprises:
• a step for forming a pattern on a substrate using the lithography device;
• a step for processing a substrate on which the pattern is formed;
• and
• a manufacturing step for manufacturing an article from a processed substrate.

A feedback control method that takes information regarding a control deviation, measured at a predefined sampling frequency, between a measured value and a target value of a controlled object as input, and outputs a control amount for the controlled object, the feedback control method comprising the following steps:

• a first control step for taking information regarding the control deviation as input, and for outputting a first control amount for the controlled object;
• a second control step for taking information regarding the control deviation as input, and for outputting the second control amount for the controlled object, and the parameters for calculating the second control amount are determined by machine learning;
• an operation step for operating the controlled object using the first control amount output by the first control step, and the second control amount output by the second control step;
• and
• a sampling step for thinning out, at a predetermined period, the information regarding the control deviation input to the second control step so that a sampling frequency of the second control step is lower than a sampling frequency in the first control step.