Position control apparatus including iterative learning circuit, exposure apparatus, method for manufacturing device, and iterative learning method for use in position control apparatus having iterative learning circuit including learning filter

A method for manufacturing a device, comprising:

• exposing a substrate to light through a pattern of an original using an exposure apparatus;
• and
• developing the exposed substrate;
• wherein the exposure apparatus includes a stage having one of the original and the substrate mounted thereon, the stage positioning the one of the original and the substrate, and wherein the stage includes a position control apparatus, and wherein the position control apparatus includes a detecting unit configured to detect a position of a control target, a subtracting unit configured to subtract an output of the detecting unit from a target value, an iterative learning control circuit including a time variant filter into which a deviation between the output of the detecting unit and the target value is input and a linear time invariant filter configured to block a predetermined frequency range, wherein the iterative learning control circuit feeds forward a control input to the control target, and a parameter computing unit configured to compute a variation in a parameter of the control target, wherein a characteristic of the time variant filter is changed in accordance with the variation in the parameter of the control target, and wherein the time variant filter includes two invariant filters having invariant characteristics and an interpolation computing unit that interpolates values between the two invariant filters.

A position control apparatus comprising:

• a detecting unit configured to detect a position of a control target;
• a deviation computing unit configured to compute a deviation between an output of the detecting unit and a target value;
• a feedback controller configured to generate a feedback signal to be input to the control target using the deviation;
• an iterative learning control circuit configured to generate a feedforward signal to be input to the control target, the iterative learning control circuit including a time variant filter that generates a control signal using the deviation obtained in a kth trial (k≧1) and a linear time invariant filter configured to block a predetermined frequency range of a signal obtained by summing the control signal generated by the time variant filter using the deviation computed in a kth trial (k≧1) and the feedforward signal generated for the kth trial, the iterative learning control circuit generating the feedforward signal for a (k+1)th trial using the control signal generated by the time variant filter using the deviation computed in the kth trial and the feedforward signal generated for the kth trial;
• a parameter computing unit configured to compute a variation in a parameter of the control target;
• a changing unit configured to change a characteristic of the time variant filter in accordance with the variation in the parameter of the control target;
• an adding unit configured to sum the feedback signal generated in the (k+1)th trial and the feedforward signal generated for the (k+1)th trial;
• and
• wherein the time variant filter includes two invariant filters having invariant characteristics and an interpolation computing unit that interpolates values between the two invariant filters.

A position control apparatus comprising:

• a detecting unit configured to detect a position of a control target;
• a filter configured to generate a control input using a deviation between an output of the detecting unit and a target value;
• an iterative learning control circuit configured to feed forward the control input to the control target and including a learning filter and a linear time invariant filter for blocking a predetermined frequency range;
• a parameter computing unit configured to compute a variation in a parameter of the control target;
• a changing unit configured to change a characteristic of the learning filter in accordance with the variation in the parameter of the control target;
• and
• wherein the learning filter includes two invariant filters having invariant characteristics and an interpolation computing unit that interpolates values between the two invariant filters.

An iterative learning method for use in a position control apparatus having an iterative learning control circuit including a learning filter, the method comprising:

• applying an iterative learning input obtained in a kth trial to a control target;
• computing the learning filter by acquiring a variation parameter of the control target;
• and
• computing the iterative learning input for a (k+1)th trial using a control input generated by the learning filter using a deviation between a position of the control target in a kth trial and a target value and an iterative learning input computed for the kth trial and a linear time invariant filter for blocking a predetermined frequency range;
• and
• wherein the learning filter includes two invariant filters having invariant characteristics and an interpolation computing unit that interpolates values between the two invariant filters.