WO2024090126A1 - Procédé de commande, dispositif de commande, dispositif de lithographie et procédé de fabrication d'article - Google Patents

Procédé de commande, dispositif de commande, dispositif de lithographie et procédé de fabrication d'article Download PDF

Info

Publication number
WO2024090126A1
WO2024090126A1 PCT/JP2023/035347 JP2023035347W WO2024090126A1 WO 2024090126 A1 WO2024090126 A1 WO 2024090126A1 JP 2023035347 W JP2023035347 W JP 2023035347W WO 2024090126 A1 WO2024090126 A1 WO 2024090126A1
Authority
WO
WIPO (PCT)
Prior art keywords
control
disturbance
signal
sequence
learning
Prior art date
Application number
PCT/JP2023/035347
Other languages
English (en)
Japanese (ja)
Inventor
拓海 橋本
智康 畑
博一 草柳
康伸 朝倉
Original Assignee
キヤノン株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by キヤノン株式会社 filed Critical キヤノン株式会社
Publication of WO2024090126A1 publication Critical patent/WO2024090126A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G03—PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
    • G03F—PHOTOMECHANICAL PRODUCTION OF TEXTURED OR PATTERNED SURFACES, e.g. FOR PRINTING, FOR PROCESSING OF SEMICONDUCTOR DEVICES; MATERIALS THEREFOR; ORIGINALS THEREFOR; APPARATUS SPECIALLY ADAPTED THEREFOR
    • G03F7/00—Photomechanical, e.g. photolithographic, production of textured or patterned surfaces, e.g. printing surfaces; Materials therefor, e.g. comprising photoresists; Apparatus specially adapted therefor
    • G03F7/20—Exposure; Apparatus therefor
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Definitions

  • the present invention relates to a control method, a control device, a lithography apparatus, and an article manufacturing method.
  • Patent Document 1 In recent years, there has been a growing demand for improved control accuracy, and conventional feedback control alone is sometimes unable to achieve the required accuracy. As a result, efforts have been made to configure controllers including neural networks in parallel in addition to conventional controllers (compensators) (Patent Document 1). The parameters of neural networks are adjusted by machine learning, but in this case, it is necessary to acquire behavioral rules that are robust against environmental fluctuations. To address this issue, a method has been proposed in which a disturbance is applied to the controlled object or the environment (Patent Document 2).
  • the present invention aims to provide an advantageous technique for improving control accuracy in controlling a control object using a neural network.
  • a control method for controlling a control object using a control signal generated from a first signal generated by a first compensator based on a control deviation of the control object and a second signal generated by a second compensator by a neural network based on the control deviation, the control method comprising: a first control step for controlling the control object based on a target value defined in a first sequence; a second control step for controlling the control object based on a target value defined in a second sequence while applying to the control object a pseudo disturbance set based on a first index obtained in the first control step; and a determination step for determining parameter values of the neural network by learning based on the second index obtained in the second control step.
  • the present invention provides an advantageous technique for improving control accuracy, for example, in controlling a control target using a neural network.
  • FIG. 1 is a diagram showing an example of the configuration of a system according to a first embodiment
  • FIG. 1 is a diagram showing an example of the configuration of a system according to a first embodiment
  • FIG. 2 is a diagram showing an example of the configuration of a sequence unit in the system of the first embodiment
  • FIG. 2 is a diagram showing an example of the configuration of a controller in the system of the first embodiment
  • FIG. 2 is a diagram showing an example of the configuration of a disturbance generator in the system of the first embodiment
  • 1 is a flowchart showing an example of the operation of the system according to the first embodiment
  • 1 is a flowchart showing an example of the operation of the system according to the first embodiment
  • FIG. 13 is a diagram showing an example of the configuration of a stage control device according to a second embodiment
  • FIG. 13 is a diagram showing an example of the configuration of a stage control device according to a second embodiment
  • FIG. 13 is a diagram showing an example of the configuration of an exposure apparatus according to a third embodiment
  • FIG. 13 is a diagram showing an example of the configuration of an exposure apparatus according to a third embodiment
  • FIG. 13 is a diagram showing an example of the configuration of an exposure apparatus according to a third embodiment
  • FIG. 1 shows an example of the configuration of a system SS according to the first embodiment.
  • the system SS can be applied to, for example, a manufacturing apparatus for manufacturing an article.
  • the manufacturing apparatus can include, for example, a processing apparatus for processing a material or member of an article or a part constituting a part of the article.
  • the processing apparatus can be, for example, any one of a lithography apparatus for transferring a pattern to a material or member, a film forming apparatus for forming a film on a material or member, an apparatus for etching a material or member, and a heating apparatus for heating a material or member.
  • the system SS may include, for example, a sequence unit 101, a control device 100, and a controlled object 103.
  • the control device 100 may include a controller 102.
  • the control device 100 or the controller 102 may generate an operation amount (operation amount signal) MV as a control signal for controlling the controlled object 103.
  • a production sequence (first sequence) may be provided to the sequence unit 101.
  • the production sequence may specify a procedure for production (operation), specifically, time-series data of a target value (target control amount) R for controlling the controlled object 103 in production.
  • the sequence unit 101 may generate a target value R for controlling the controlled object 103 based on the production sequence, and provide the target value R to the control device 100 or the controller 102.
  • the control device 100 or the controller 102 can perform feedback control of the controlled object 103. Specifically, the control device 100 or the controller 102 can control the controlled object 103 so that the controlled amount CV follows the target value R based on a control deviation, which is the difference between the target value R provided by the sequence unit 101 and the controlled amount CV provided by the controlled object 103.
  • the controlled object 103 can have a sensor that detects the controlled amount CV, and the controlled amount CV detected by the sensor can be provided to the controller 102.
  • the target value R, the operating amount MV, and the controlled amount CV can be time-series data whose values change over time.
  • the system SS may incorporate a learning unit 201 and a disturbance generator 202.
  • the learning unit 201 and the disturbance generator 202 may be configured as part of the control device 100 or may be configured as an external device of the control device 100.
  • the learning unit 201 and the disturbance generator 202 may be disconnected from the control device 100 after learning is completed.
  • the learning unit 201 may be configured to send a learning sequence (second sequence) to the sequence unit 101.
  • a target value (target control amount) R for controlling the control object 103 in learning is specified as time-series data.
  • the learning sequence may be set (configured) so that the time for controlling the control object 103 by the learning sequence is shorter than the time for controlling the control object 103 by the production sequence.
  • the sequence unit 101 may generate a target value R according to the learning sequence and provide it to the controller 102.
  • the disturbance generator 202 may be configured to send a disturbance signal N to the controller 102.
  • the disturbance signal N may be generated based on an operation history HP provided by the controller 102 during execution of a production sequence.
  • the controller 102 may generate an operation amount MV based on a control deviation E, which is the difference between a target value R generated and provided by the sequence unit 101 according to a production sequence or a learning sequence and a control amount CV provided by the controlled object 103.
  • the controller 102 has a neural network and may generate an operation amount MV using the neural network.
  • the operation amount MV generated by the controller 102 is provided to the controlled object 103, and the controlled object 103 may operate according to the operation amount MV.
  • the control amount CV as a result of this operation may be provided to the controller 102.
  • the controller 102 may provide the learning unit 201 with an operation history HL indicating the history of the operation of the controller 102 based on the target value R.
  • the learning unit 201 may determine parameter values of the neural network of the controller 102 based on the operation history HL and set the parameter values in the neural network.
  • the parameter values may be determined by machine learning such as reinforcement learning, for example.
  • FIG. 3 shows an example of one configuration of the sequence unit 101.
  • the sequence unit 101 may include a first target value generator 301 that generates a first target value R1 based on a production sequence, and a second target value generator 302 that generates a second target value R2 based on a learning sequence.
  • the first target value R1 may be a target value (target control amount) of the control object 103 during execution of the production sequence.
  • the second target value R2 is a target value (target control amount) of the control object 103 during execution of the learning sequence, and may take a value different from the first target value R1.
  • the learning sequence may, for example, specify a target value over time for keeping the control object 103 in a stopped state, and in this case, the second target value R2 may be a stop command indicating a target value for keeping the control object 103 stopped, but is not limited thereto.
  • the sequence unit 101 may also include a target value switcher 303 that generates a target value R based on the first target value R1 and the second target value R2.
  • the target value R is a selection of either the first target value R1 or the second target value R2.
  • the target value switch 303 can output the target value R by selectively switching between the first target value R1 and the second target value R2 depending on whether the production sequence or the learning sequence is to be executed.
  • FIG. 4 shows one example of the configuration of the controller 102.
  • the controller 102 may include a subtractor 405, a first compensator 401, a second compensator 402, and a calculator 406.
  • the subtractor 405 generates a control deviation E, which is the difference between the target value R and the controlled variable CV.
  • the first compensator 401 is, for example, a PID compensator, and generates a first signal S1 based on the control deviation E.
  • the second compensator 402 generates a second signal S2 by a neural network based on the control deviation E.
  • the calculator 406 generates a control signal S based on the first signal S1 and the second signal S2.
  • the control signal S is the sum of the first signal S1 and the second signal S2, and the calculator 406 may be configured as an adder.
  • the control signal S is a signal obtained by correcting the first signal S1 based on the second signal S2.
  • the controller 102 may further include a calculator 407 that generates a manipulated variable MV based on the control signal S and the disturbance signal N.
  • the disturbance signal N is a signal for applying a pseudo disturbance (vibration) to the controlled object 103 during execution of the learning sequence, and is generated by the disturbance generator 202 based on the operation history HP obtained from the controller 102 during execution of the production sequence.
  • the disturbance signal N may be a signal for applying a pseudo disturbance simulating a disturbance generated during execution of the production sequence to the controlled object 103.
  • the disturbance signal N may be composed of a single frequency or a combination of multiple frequencies.
  • the disturbance signal N may be a sine wave, random noise, or impulse disturbance, but is not limited to these.
  • the operation history HP may be a history (time series data) of the first index obtained from the controller 102 during execution of the production sequence.
  • the first index may be, for example, a control deviation E and/or a control signal S, but is not limited to these.
  • the manipulated variable MV is the sum of the control signal S and the disturbance signal N, and the calculator 407 can be configured as an adder.
  • the controller 102 may further include a first recording unit 404.
  • the learning unit 201 may be configured to perform learning to determine parameter values of the neural network of the second compensator 402.
  • the first recording unit 404 may record (store) an operation history HL required for learning by the learning unit 201 and provide the recorded operation history HL to the learning unit 201.
  • the operation history HL may be a history (time series data) of a second index obtained from the controller 102 during execution of the learning sequence.
  • the second index may be, for example, a control deviation E which is input data to the second compensator 402, and/or a second signal S2 which is output data from the second compensator 402, but may alternatively or additionally include other data or signals.
  • the disturbance generator 202 may include a reference disturbance generating unit 502, a second recording unit 501, and a third recording unit 504.
  • the reference disturbance generating unit 502 generates a reference disturbance signal NO used as a reference disturbance signal.
  • the second recording unit 501 records (stores) the operation history HP obtained from the controller 102 during the execution of the production sequence.
  • the operation history HP may be a history of the first index (e.g., control deviation E and/or control signal S) obtained from the controller 102 during the execution of the production sequence.
  • the third recording unit 504 records (stores) the operation history HA obtained from the controller 102 during disturbance adjustment (during learning) in order to adjust and set the pseudo disturbance.
  • the operation history HA may be a history (time series data) of the first index (e.g., control deviation E and/or control signal S) obtained from the controller 102 during disturbance adjustment.
  • the disturbance generator 202 may also include a disturbance adjustment unit 503 that adjusts the disturbance signal N based on the reference disturbance signal NO, the operation history HP during execution of the production sequence, and the operation history HA during disturbance adjustment.
  • the disturbance signal N may be the reference disturbance signal NO whose amplitude has been corrected based on the operation history HP during execution of the production sequence and the operation history HA during disturbance adjustment.
  • the disturbance adjustment unit 503 corrects the amplitude of the reference disturbance signal NO by multiplying the reference disturbance signal NO by a gain based on the operation history HP during execution of the production sequence and the operation history HA during disturbance adjustment, and provides the signal obtained thereby to the calculator 407 as the disturbance signal N.
  • FIGS. 6A and 6B show an example of the operation of the system SS when the system SS of the first embodiment is applied to a production device.
  • FIG. 6A shows a first control process (S601-S602) for controlling the control object 103 according to a production sequence, and a setting process (S603-S612) for adjusting and setting a pseudo disturbance.
  • FIG. 6B shows a second control process (S613-S617) for controlling the control object 103 according to a learning sequence while applying a pseudo disturbance, and a determination process (S618) for determining parameter values of the neural network of the second compensator 402 by learning.
  • step S601 the control device 100 executes a production sequence.
  • the sequence unit 101 may generate a first target value R1 defined in a given production sequence as a target value R, and provide the target value R to the control device 100 (controller 102).
  • the control device 100 may then control the control target 103 based on the target value R provided from the sequence unit 101.
  • the control device 100 (second recording unit 501) may record the history of the first index (e.g., control deviation E) obtained in the production sequence of step S601 as an operation history HP. Note that step S602 may be performed in parallel with step S601.
  • step S603 the control device 100 (the disturbance adjustment unit 503 included in the disturbance generator 202) may adjust (correct) the amplitude of the reference disturbance signal NO by multiplying the reference disturbance signal NO by an arbitrary gain, and generate the resulting disturbance signal N.
  • step S604 the control device 100 (controller 102) may start vibrating the control object 103 based on the disturbance signal N (application of a pseudo disturbance).
  • the disturbance generator 202 starts providing the disturbance signal N to the calculator 407. This starts the application of a pseudo disturbance to the control object 103 in accordance with the disturbance signal N.
  • step S605 the control device (controller 102) executes the learning sequence.
  • the sequence unit 101 may generate the second target value R2 defined in the learning sequence provided by the learning unit 201 as the target value R, and provide the target value R to the control device 100 (controller 102).
  • the control device 100 may then control the control object 103 based on the target value R provided by the sequence unit 101.
  • the control object 103 is controlled based on the learning sequence while a pseudo disturbance is provided to the control object 103.
  • the pseudo disturbance may be provided to the control object 103 continuously or intermittently.
  • the learning sequence may be set so that the time during which the control object 103 is controlled by the learning sequence is shorter than the time during which the control object 103 is controlled by the production sequence.
  • step S606 the control device 100 (controller 102) may stop applying vibration to the control object 103 based on the disturbance signal N. Specifically, the provision of the disturbance signal N from the disturbance generator 202 to the calculator 407 is stopped. This stops the application of the pseudo disturbance corresponding to the disturbance signal N to the control object 103.
  • the control device 100 third recording unit 504 may record the history of the first index (e.g., control deviation E) obtained in step S605 as an operation history HA. Note that step S607 may be performed in parallel with step S605.
  • step S608 the control device 100 may determine whether the difference between the first index of the operation history HP recorded in step S602 and the first index of the operation history HA recorded in step S607 is less than a threshold value. For example, when the control deviation E is used as the first index, the control device 100 may determine whether the difference between the amplitude of the control deviation E in the operation history HP recorded in step S602 and the amplitude of the control deviation E in the operation history HA recorded in step S607 is less than a threshold value. The maximum value, average value, or median value of the amplitude may be used as the amplitude of the control deviation E.
  • the control device 100 decides to use the currently used disturbance signal N (pseudo disturbance) in the second control step (S613 to S617) and proceeds to step S613. That is, the control device 100 sets the currently used disturbance signal N as the disturbance signal N to be used in the second control step. On the other hand, if the difference is equal to or greater than the threshold value, the control device 100 proceeds to step S609.
  • step S609 the control device 100 determines whether the first index (e.g., the amplitude of the control deviation E) is greater for the operation history HA than for the operation history HP. If the first index is greater for the operation history HA than for the operation history HP, the process proceeds to step S610; if not, the process proceeds to step S611.
  • the control device 100 may decrease the gain value by which the reference disturbance signal NO is multiplied by one step.
  • the control device 100 may increase the gain value by which the reference disturbance signal NO is multiplied by one step.
  • the amount by which the gain is decreased by one step in step S610 and the amount by which the gain is increased by one step in step S611 may be arbitrary, but are preferably set so as to reduce the difference in the first index (e.g., the amplitude of the control deviation E).
  • step S612 the control device 100 (disturbance adjustment unit 503) may adjust the amplitude of the reference disturbance signal NO by multiplying the reference disturbance signal NO by the gain adjusted in step S610 or step S611, and generate the resulting disturbance signal N.
  • step S612 ends, the process proceeds to step S604, and steps S604 to S608 may be executed again.
  • the control device 100 can start applying vibration to the control object 103 based on the disturbance signal N set through steps S603 to S612.
  • the disturbance generator 202 starts providing the disturbance signal N to the calculator 407. This starts the application of a pseudo disturbance to the control object 103 in accordance with the disturbance signal N.
  • step S614 the control device 100 (controller 102) executes a learning sequence. Specifically, similar to step S605 described above, the sequence unit 101 may generate the second target value R2 defined in the learning sequence provided by the learning unit 201 as a target value R, and provide the target value R to the control device 100 (controller 102). The control device 100 (controller 102) may then control the control object 103 based on the target value R provided by the sequence unit 101. In step S614, the control object 103 is controlled based on the learning sequence while a pseudo disturbance is provided to the control object 103. Here, the pseudo disturbance may be provided to the control object 103 continuously or intermittently.
  • step S615 the control device 100 (controller 102) may stop applying vibration to the control object 103 based on the disturbance signal N. Specifically, the provision of the disturbance signal N from the disturbance generator 202 to the calculator 407 is stopped. This stops the application of the pseudo disturbance to the control object 103 in response to the disturbance signal N.
  • step S616 the control device 100 (first recording unit 404) may record the history of the second index (e.g., control deviation E) obtained in step S614 as the operation history HL. Note that step S616 may be performed in parallel with step S614.
  • step S617 the control device 100 may determine whether the second index of the operation history HL recorded in step S616 is less than a specified value. For example, when the control deviation E is used as the second index, the control device 100 may determine whether the control deviation E recorded in step S616 is less than a specified value. As the second index, the maximum value, average value, or median value of the amplitude of the control deviation E may be used, or the second signal S2 output from the second compensator 402 may be used. If the second index recorded in step S616 is less than the specified value, the process ends. In this case, the control device 100 determines the parameter value of the neural network of the second compensator 402 currently being used as the parameter value to be used in the future. On the other hand, if the second index recorded in step S616 is equal to or greater than the specified value, the process proceeds to step S618.
  • step S618 the control device 100 (learning unit 201) performs machine learning based on the operation history HL recorded in step S616 to determine (update) the parameter values of the neural network of the second compensator 402.
  • step S618 ends, the process proceeds to step S613, where the parameter values determined in step S618 are set in the neural network of the second compensator 402, and steps S613 to S617 are executed again.
  • the control object 103 is controlled according to the learning sequence while applying a pseudo disturbance to the control object 103 by the disturbance signal N. Then, by learning based on the second index (e.g., control deviation E) obtained by this control, the parameter value of the neural network of the second compensator 402 is determined so that the control deviation of the control object 103 is reduced.
  • the first embodiment by applying a pseudo disturbance similar to the disturbance generated during the execution of the production sequence, learning for determining the parameter value of the neural network is effectively performed. Furthermore, the control accuracy is improved in the control of the control object 103 using the neural network. Also, according to the first embodiment, a pseudo disturbance is applied to the control object 103 in the learning sequence.
  • the time for controlling the control object 103 is made shorter than the production sequence, it is possible to make the time for applying the disturbance to the control object 103 (effective time) longer than the production sequence. Therefore, the time required for learning to determine the parameter value of the neural network of the second compensator 402 can be shortened, and the decrease in productivity of the manufacturing device to which the system SS is applied can be suppressed.
  • learning to determine parameter values of a neural network is performed by controlling the control object 103 in a sequence similar to the production sequence.
  • the control accuracy of the control object 103 is indefinite, and therefore production by the device must be stopped. In other words, this can lead to a decrease in productivity of the device.
  • learning is performed using a learning sequence in which the control object is controlled for a short period of time while applying a pseudo disturbance to the control object 103 so as to simulate a disturbance environment that occurs during the execution of the production sequence.
  • control device 100 of the first embodiment may be configured by a computer (information processing device) having a processor such as a CPU (Central Processing Unit) and a storage unit such as a memory.
  • the control device 100 may be configured by, for example, a PLD (Programmable Logic Device) such as an FPGA (Field Programmable Gate Array), or an ASIC (Application Specific Integrated Circuit), or a general-purpose computer with a built-in program, or a combination of all or part of these.
  • the sequence unit 101 may be incorporated in the computer that constitutes the control device 100, or may be configured by a computer different from the control device 100.
  • Second Embodiment A second embodiment of the present invention will be described.
  • the second embodiment basically follows the first embodiment, and can follow the first embodiment except for the matters mentioned below.
  • FIG. 7 shows an example in which the system SS or the control device 100 of the first embodiment is applied to a stage control device 700.
  • the stage control device 700 is configured to control a stage 704.
  • the stage control device 700 may include, for example, a control board 701, a current driver 702, a motor 703, a stage 704, and a sensor 705.
  • the control board 701 corresponds to the control device 100 or the controller 102 in the system SS of the first embodiment.
  • the current driver 702 may be incorporated into the control board 701.
  • the stage control device 700 may include the sequence unit 101, learning unit 201, and disturbance generator 202 described in the first embodiment.
  • the control board 701 can be supplied with a position target value PR as a target value for the position of the stage 704 from the sequence unit 101.
  • the control board 701 can generate a current command as a control signal or an operation amount (operation amount command) based on the position target value PR supplied from the sequence unit 101 and position information (current position) of the stage 704 supplied from the sensor 705, and supply it to the current driver 702.
  • the control board 701 can also supply an operation history HL to the sequence unit 101.
  • the current driver 702 can supply a current according to a current command to the motor 703.
  • the motor 703 can be an actuator that converts the current supplied from the current driver 702 into thrust and drives the stage 704 with that thrust.
  • the stage 704 can hold an object such as a plate (substrate) or a mask (original).
  • the sensor 705 can detect the current position of the stage 704 and supply the position information obtained thereby to the control board 701.
  • FIG. 8 shows an example of the configuration of the control board 701 as a block diagram.
  • the control board 701 may include a first compensator 401 that generates a first signal S1 based on a position control deviation E of a stage 704 as a control target, and a second compensator 402 that generates a second signal S2 by a neural network based on the position control deviation E.
  • the control board 701 may also include a calculator 406 that generates a control signal S based on the first signal S1 and the second signal S2, and a calculator 407 that generates a current command as a manipulated variable signal based on the control signal S and a disturbance signal N.
  • the control board 701 may also include a subtractor 405 that generates a position control deviation E, which is the difference between the position target value PR and the position information.
  • the stage control device 700 of the second embodiment may include the learning unit 201 and the disturbance generator 202 described in the first embodiment.
  • the learning unit 201 may be configured to perform learning to determine parameter values of the neural network of the second compensator 402.
  • the first recording unit 404 may record the operation history HL required for learning by the learning unit 201 and provide the recorded operation history HL to the learning unit 201.
  • the operation history HL may be a history (time series data) of the second index obtained during execution of the learning sequence.
  • the second index may be, for example, the position control deviation E which is input data to the second compensator 402, and/or the second signal S2 which is output data of the second compensator 402, but may alternatively or additionally include other data or signals.
  • the disturbance generator 202 may generate a disturbance signal N based on the operation history HP during execution of the production sequence and provide it to the control board 701.
  • the operation history HP may be a history (time series data) of the first index obtained during the execution of the production sequence.
  • the first index may be, for example, a position control deviation E or a control signal S, but is not limited to these.
  • the learning unit 201 and the disturbance generator 202 may be configured as components of the control board 701.
  • step S601 the stage control device 700 executes a production sequence. Specifically, the sequence unit 101 may generate a position target value PR based on a given production sequence and provide the position target value PR to the stage control device 700 (control board 701). The stage control device 700 (control board 701) may then control the position of the stage 704 based on the position target value PR provided from the sequence unit 101. In step S602, the stage control device 700 (second recording unit 501) may record the history of the first index (e.g., control deviation E) obtained in the production sequence of step S601 as an operation history HP.
  • the first index e.g., control deviation E
  • the stage control device 700 can generate a disturbance signal N for applying a pseudo disturbance to the stage 704 by adjusting the amplitude of the reference disturbance signal NO based on the operation history HP recorded in the second recording unit 501.
  • the stage control device 700 may start vibrating the stage 704 based on the disturbance signal N.
  • the stage control device 700 executes a learning sequence. Specifically, the sequence unit 101 may generate a position target value PR based on the learning sequence and provide it to the stage control device 700 (control board 701). The stage control device 700 (control board 701) may then control the stage 704 based on the target value R provided by the sequence unit 101.
  • the stage 704 is controlled based on the learning sequence while a pseudo disturbance is applied to the stage 704.
  • the pseudo disturbance may be applied to the stage 704 continuously or intermittently.
  • step S615 the stage control device 700 (control board 701) may stop applying vibration to the stage 704 based on the disturbance signal N.
  • the stage control device 700 (first recording unit 404) may record the history of the second index (e.g., position control deviation E) obtained in step S614 as the operation history HL.
  • step S617 the stage control device 700 may determine whether the second index of the operation history HL recorded in step S616 is less than a specified value. For example, when the position control deviation E is used as the second index, the stage control device 700 may determine whether the position control deviation E recorded in step S616 is less than a specified value. If the second index recorded in step S616 is less than the specified value, the process ends.
  • the stage control device 700 determines the parameter values of the neural network of the second compensator 402 currently being used as the parameter values to be used in the future. On the other hand, if the position control deviation E recorded in step S616 is equal to or greater than the specified value, the process proceeds to step S618. In step S618, the stage control device 700 (learning unit 201) determines (updates) the parameter values of the neural network of the second compensator 402 by performing machine learning based on the operation history HL recorded in step S616.
  • the stage 704 is controlled according to a learning sequence while applying a pseudo disturbance to the stage 704 by the disturbance signal N. Then, by learning based on the second index (e.g., position control deviation E) obtained by this control, the parameter values of the neural network of the second compensator 402 are determined so as to reduce the control deviation of the stage 704.
  • the second index e.g., position control deviation E
  • learning for determining the parameter values of the neural network is performed effectively, and the control accuracy is improved in the control of the stage 704 using the neural network.
  • the time required for learning for determining the parameter values of the neural network of the second compensator 402 can be shortened, and a decrease in productivity of the manufacturing device to which the stage control device 700 is applied can be suppressed.
  • the third embodiment basically follows the first and second embodiments, and can follow the first and second embodiments except for the matters mentioned below.
  • FIG. 9 shows a schematic configuration example of the exposure apparatus EXP of the third embodiment.
  • the exposure apparatus EXP can be configured as a scanning exposure apparatus.
  • the exposure apparatus EXP can include, for example, an illumination light source 900, an illumination optical system 901, a mask stage 903, a projection optical system 904, and a plate stage 906.
  • the illumination light source 900 can include, but is not limited to, a mercury lamp, an excimer laser light source, or an EUV light source.
  • the exposure light 910 from the illumination light source 900 is shaped by the illumination optical system 901 into the shape of the irradiation area of the projection optical system 904 with uniform illuminance.
  • the exposure light 910 can be shaped into a rectangle that is long in the X direction, which is an axis perpendicular to the plane defined by the Y axis and the Z axis.
  • the exposure light 910 can be shaped into an arc shape.
  • the shaped exposure light 910 is irradiated onto the pattern of the mask (original) 902, and the exposure light 910 that passes through the pattern of the mask 902 forms an image of the pattern of the mask 902 on the surface of the plate (substrate) 905 via the projection optical system 904.
  • the mask 902 is held by a mask stage 903 using vacuum suction or the like, and a pattern to be transferred to a plate 905 is formed on the mask 902.
  • the plate 905 is held by a chuck 907 of a plate stage 906 using vacuum suction or the like.
  • the positions of the mask stage 903 and the plate stage 906 can be controlled by a multi-axis position control device equipped with position sensors 930a-930b such as laser interferometers or laser scales, driving mechanisms 931a-931b such as linear motors, and a controller 932.
  • the position of the plate stage 906 (plate 905) is controlled by a position sensor 930a, a drive mechanism 931a, and a controller 932.
  • the position measurement value of the plate stage 906 output from the position sensor 930a can be provided to the controller 932.
  • the controller 932 generates a control signal (operation amount signal) based on a position control deviation, which is the difference between the position target value and the position measurement value, and provides it to the drive mechanism 931a to drive the plate stage 906.
  • the position of the mask stage 903 (mask 902) is controlled by a position sensor 930b, a drive mechanism 931b, and a controller 932.
  • the position measurement value of the mask stage 903 output from the position sensor 930b can be provided to the controller 932.
  • the controller 932 generates a control signal (operation amount signal) based on a position control deviation, which is the difference between the position target value and the position measurement value, and provides it to the drive mechanism 931b to drive the mask stage 903.
  • the mask stage 903 and the plate stage 906 are driven synchronously in the Y direction while scanning and exposing the plate 905, so that the pattern of the mask 902 is transferred to the plate 905 (the photosensitive material thereon).
  • control device 100 of the first embodiment or the stage control device 700 of the second embodiment is used to control the position of the mask stage 903 (mask 902) and/or the position of the plate stage 906 (plate 905). That is, the controller 932 that controls the mask stage 903 (mask 902) and/or the plate stage 906 (plate 905) as the control object can be configured by the control device 100 of the first embodiment or the stage control device 700 of the second embodiment.
  • the exposure apparatus EXP or the controller 932 may be configured to include the sequence unit 101, learning unit 201 and/or disturbance generator 202 described in the first embodiment.
  • FIG. 10 shows one example of the configuration of the controller 932.
  • the controller 932 can switch between single-axis control and synchronous control depending on multiple control targets and multiple controllers.
  • the plate stage 906 is the master axis and the mask stage 903 is the slave axis, and master-slave synchronous control is shown in which the slave axis follows the master axis.
  • the controller 932 may include a first compensator 1001 for plate position control that generates a first signal S1 based on a position control deviation E1 of the plate stage 906 as a first control object, and a calculator 407 that generates a manipulated variable MV1 based on the first signal S1a and a disturbance signal N.
  • the controller 932 may also include a subtractor 405a that generates a position control deviation E1, which is the difference between a position target value PR1 of the plate stage 906 and a controlled variable (position measurement value) CV1.
  • the first compensator 1001 for plate position control may be, for example, a PID compensator.
  • the controller 932 may include a synchronization switch 1004 that switches between synchronous control of the plate stage 906 and the mask stage 903.
  • the synchronization switch 1004 may generate a position control deviation E based on a position control deviation E2 of the mask stage 903 as a second control target, and a synchronous position deviation Es that is the difference between the position control deviations E1 and E2.
  • the position control deviation E is a selection of either the position control deviation E2 or the synchronous position deviation Es.
  • the synchronization switch 1004 may output the position control deviation E by selectively switching between the position control deviation E2 and the synchronous position deviation Es.
  • the controller 932 may include a first compensator 1002 for mask position control that generates a first signal S1b based on the position control deviation E, and a second compensator 1003 for mask position control that generates a second signal S2 by a neural network based on the position control deviation E.
  • the controller 932 may include an arithmetic unit 406 that generates a manipulated variable MV2 based on the first signal S1b and the second signal S2.
  • the controller 932 may also include a subtractor 405b that generates a position control deviation E2, which is the difference between the position target value PR2 of the mask stage 903 and the control variable (position measurement value) CV2.
  • the exposure apparatus EXP of the third embodiment may include a learning unit 201 and a disturbance generator 202.
  • the learning unit 201 may be configured to perform learning to determine parameter values of the neural network of the second compensator 1003 for mask position control.
  • the first recording unit 404 may record the operation history HL required for learning by the learning unit 201 and provide the recorded operation history HL to the learning unit 201.
  • the operation history HL may be a history (time series data) of the second index obtained during execution of the learning sequence.
  • the second index may be, for example, a position control deviation E which is input data to the second compensator 1003 for mask position control, and/or a second signal S2 which is output data of the second compensator 1003, but may alternatively or additionally include other data or signals.
  • the disturbance generator 202 may also generate a disturbance signal N based on the operation history HP during execution of the production sequence, and provide it to the controller 932 (the calculator 407).
  • the operation history HP may be a history (time-series data) of the first index obtained during execution of the production sequence.
  • the first index may be, for example, a position control deviation E, but is not limited to this.
  • step S601 the exposure apparatus EXP (controller 932) executes a production sequence.
  • the sequence unit 101 may generate a position target value PR1 of the plate stage 906 and a position target value PR2 of the mask stage 903 based on a given production sequence, and provide these to the exposure apparatus EXP (controller 932).
  • the controller 932 (subtractor 405a) may acquire a control amount CV1 of the plate stage 906 detected (observed) by the position sensor 930a, and calculate the difference between the control amount CV1 and the position target value PR1 as a position control deviation E1.
  • controller 932 (subtractor 405b) can obtain the control amount CV2 of the mask stage 903 detected (observed) by the position sensor 930b, and calculate the difference between the control amount CV2 and the position target value PR2 as the position control deviation E2.
  • the controller 932 can control the position of the plate stage 906 based on the position control deviation E1.
  • the controller 932 can also control the position of the mask stage 903 based on the position control deviation E2 that is switched by the synchronization switch 1004, or the synchronous position deviation Es, which is the difference between the position control deviations E1 and E2.
  • step S602 the controller 932 (second recording unit 501) can record the history of the first index (e.g., position control deviation E) obtained in the production sequence in step S601 as an operation history HP.
  • the exposure apparatus EXP disurbance adjustment unit 503 can generate a disturbance signal N for applying a pseudo disturbance to the plate stage 906 by adjusting the amplitude of the reference disturbance signal NO based on the operation history HP recorded in the second recording unit 501.
  • step S613 the controller 932 may start vibrating the plate stage 906 based on the disturbance signal N.
  • step S614 the controller 932 executes a learning sequence. Specifically, the sequence unit 101 may generate position target values PR1, PR2 based on the learning sequence and provide them to the exposure apparatus EXP (controller 932). The controller 932 may control the plate stage 906 and the mask stage 903 based on the position target values PR1, PR2.
  • step S614 the plate stage 906 and the mask stage 903 are controlled based on the learning sequence while a pseudo disturbance is applied to the plate stage 906.
  • the pseudo disturbance may be applied to the plate stage 906 continuously or intermittently.
  • step S615 the controller 932 may stop vibrating the plate stage 906 based on the disturbance signal N.
  • the controller 932 (first recording unit 404) may record the history of the second index (e.g., position control deviation E) obtained in step S614 as the operation history HL.
  • step S617 the controller 932 may determine whether the second index of the operation history HL recorded in step S616 is less than a specified value. For example, when the position control deviation E is used as the second index, the controller 932 may determine whether the position control deviation E recorded in step S616 is less than a specified value. If the position control deviation E recorded in step S616 is less than the specified value, the process ends.
  • the controller 932 determines the parameter values of the neural network of the second compensator 1003 for mask position control currently being used as the parameter values to be used in the future. On the other hand, if the position control deviation E recorded in step S616 is equal to or greater than the specified value, the process proceeds to step S618. In step S618, the controller 932 (learning unit 201) performs machine learning based on the operation history HL recorded in step S616 to determine (update) the parameter values of the neural network of the second compensator 1003 for mask position control.
  • the plate stage 906 and the mask stage 903 are controlled according to the learning sequence while applying a pseudo disturbance to the plate stage 906 by the disturbance signal N. Then, by learning based on the second index (for example, the position control deviation E) obtained by this control, the parameter values of the neural network of the second compensator 1003 are determined so that the control deviation of the plate stage 906 and the mask stage 903 is reduced.
  • the third embodiment as in the first and second embodiments, learning for determining the parameter values of the neural network is effectively performed, and the control accuracy is improved in the control of the plate stage 906 using the neural network.
  • the time required for learning for determining the parameter values of the neural network of the second compensator 1003 can be shortened. Also, by arranging the disturbance generator 202 on the synchronization master side, it becomes easier to simulate the disturbance environment during synchronization. As a result, it is possible to suppress a decrease in the productivity of the exposure apparatus EXP.
  • the third embodiment can be applied not only to the control of a stage in an exposure apparatus, but also to the control of a stage in other lithography apparatuses such as an imprint apparatus and an electron beam drawing apparatus.
  • the first or second embodiment can also be applied to the control of a movable part in a transport mechanism that transports an object, such as a hand that holds the object.
  • the above-mentioned lithography apparatus can be used to implement an article manufacturing method for manufacturing various articles (semiconductor IC elements, liquid crystal display elements, MEMS, etc.).
  • the article manufacturing method can include a formation step of forming a pattern on a substrate using the above-mentioned lithography apparatus (a transfer step of transferring a pattern of an original to a substrate), a processing step of processing the substrate that has undergone the formation step, and a manufacturing step of manufacturing an article from the substrate that has undergone the processing step.
  • the formation step can include an exposure step of exposing the substrate through an original, and a development step of developing the substrate that has undergone the exposure step.
  • Such an article manufacturing method includes other well-known steps (baking, cooling, cleaning, oxidation, film formation, deposition, doping, planarization, etching, resist stripping, dicing, bonding, packaging, etc.).
  • the article manufacturing method of the present embodiment is advantageous in at least one of the performance, quality, productivity, and production cost of the article compared to conventional methods.
  • control device 101: sequence unit, 102: controller, 103: controlled object, 201: learning unit, 202: disturbance generator, 401: first compensator, 402: second compensator

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Automation & Control Theory (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Feedback Control In General (AREA)
  • Exposure And Positioning Against Photoresist Photosensitive Materials (AREA)

Abstract

Le but de la présente invention est d'effectuer un apprentissage de manière efficace afin de déterminer une valeur de paramètre d'un réseau neuronal et d'améliorer la précision de commande lors de la commande d'une cible de commande à l'aide du réseau neuronal.　Ce procédé de commande servant à commander une cible de commande utilise un premier signal (S1) généré par un premier compensateur (401) sur la base d'un écart de commande de la cible de commande ainsi qu'un second signal (S2) généré par un second compensateur (402) sur la base de l'écart de commande à l'aide d'un réseau neuronal. Le procédé de commande comprend : une première étape de commande consistant à commander la cible de commande sur la base d'une première valeur cible de séquence ; une seconde étape de commande consistant à commander la cible de commande sur la base d'une seconde valeur cible de séquence, tout en appliquant une pseudo-perturbation (N) à la cible de commande ; et une étape de détermination consistant à déterminer une valeur de paramètre du réseau neuronal (402) par apprentissage sur la base d'un second indicateur obtenu dans la seconde étape de commande.
PCT/JP2023/035347 2022-10-25 2023-09-28 Procédé de commande, dispositif de commande, dispositif de lithographie et procédé de fabrication d'article WO2024090126A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2022-170869 2022-10-25
JP2022170869A JP2024062786A (ja) 2022-10-25 2022-10-25 制御方法、制御装置、リソグラフィ装置、および物品製造方法

Publications (1)

Publication Number Publication Date
WO2024090126A1 true WO2024090126A1 (fr) 2024-05-02

Family

ID=90830674

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2023/035347 WO2024090126A1 (fr) 2022-10-25 2023-09-28 Procédé de commande, dispositif de commande, dispositif de lithographie et procédé de fabrication d'article

Country Status (2)

Country Link
JP (1) JP2024062786A (fr)
WO (1) WO2024090126A1 (fr)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07187072A (ja) * 1993-12-27 1995-07-25 Ishikawajima Harima Heavy Ind Co Ltd Ｒｏｖの自動制御方法
JP2003264134A (ja) * 2002-03-08 2003-09-19 Nikon Corp ステージ制御装置、露光装置、及びデバイス製造方法
JP2009205641A (ja) * 2008-02-29 2009-09-10 Canon Inc 反復学習制御回路を備える位置制御装置
WO2018151215A1 (fr) * 2017-02-20 2018-08-23 株式会社安川電機 Dispositif et procédé de commande
JP2018186610A (ja) * 2017-04-25 2018-11-22 株式会社安川電機 システムおよび評価装置ならびに評価方法
JP2020060827A (ja) * 2018-10-05 2020-04-16 株式会社日立製作所 制御装置および制御方法
JP2020095352A (ja) * 2018-12-10 2020-06-18 富士電機株式会社 制御装置、制御方法及びプログラム
JP2022046317A (ja) * 2020-09-10 2022-03-23 キヤノン株式会社 制御装置およびその調整方法、リソグラフィー装置、ならびに、物品製造方法
WO2022124281A1 (fr) * 2020-12-11 2022-06-16 キヤノン株式会社 Dispositif de commande, procédé d'ajustement, dispositif de lithographie et procédé de fabrication d'article

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH07187072A (ja) * 1993-12-27 1995-07-25 Ishikawajima Harima Heavy Ind Co Ltd Ｒｏｖの自動制御方法
JP2003264134A (ja) * 2002-03-08 2003-09-19 Nikon Corp ステージ制御装置、露光装置、及びデバイス製造方法
JP2009205641A (ja) * 2008-02-29 2009-09-10 Canon Inc 反復学習制御回路を備える位置制御装置
WO2018151215A1 (fr) * 2017-02-20 2018-08-23 株式会社安川電機 Dispositif et procédé de commande
JP2018186610A (ja) * 2017-04-25 2018-11-22 株式会社安川電機 システムおよび評価装置ならびに評価方法
JP2020060827A (ja) * 2018-10-05 2020-04-16 株式会社日立製作所 制御装置および制御方法
JP2020095352A (ja) * 2018-12-10 2020-06-18 富士電機株式会社 制御装置、制御方法及びプログラム
JP2022046317A (ja) * 2020-09-10 2022-03-23 キヤノン株式会社 制御装置およびその調整方法、リソグラフィー装置、ならびに、物品製造方法
WO2022124281A1 (fr) * 2020-12-11 2022-06-16 キヤノン株式会社 Dispositif de commande, procédé d'ajustement, dispositif de lithographie et procédé de fabrication d'article

Also Published As

Publication number Publication date
JP2024062786A (ja) 2024-05-10

Similar Documents

Publication Publication Date Title
TWI418947B (zh) 包含反覆學習電路的位置控制設備，曝光設備，製造裝置的方法，及用於具有包含學習過濾器之反覆學習電路的位置控制設備中的反覆學習方法
JP5235707B2 (ja) 制御装置
JP2005354088A (ja) リソグラフィ運動制御システム及び方法
TWI598654B (zh) 光學裝置、投影光學系統、曝光裝置及製造物品的方法
JP2006128685A (ja) 運動システム用自己適応フィードフォワード制御調整、およびそのような運動システムを備えるリソグラフィ装置
JP2022188045A (ja) 制御装置、露光装置及び物品の製造方法
WO2024090126A1 (fr) Procédé de commande, dispositif de commande, dispositif de lithographie et procédé de fabrication d'article
WO2022124281A1 (fr) Dispositif de commande, procédé d'ajustement, dispositif de lithographie et procédé de fabrication d'article
WO2022054724A1 (fr) Dispositif de commande, procédé de réglage de dispositif de commande, dispositif de lithographie et procédé de fabrication d'article
KR102566715B1 (ko) 제어 장치, 리소그래피 장치, 및 물품 제조 방법
JP6760975B2 (ja) 制御システム、位置決めシステム、リソグラフィ装置及びデバイス製造方法
JP2019121656A (ja) 制御方法、制御装置、リソグラフィ装置、および物品の製造方法
JP6307101B2 (ja) リソグラフィ装置、および物品の製造方法
JP5473575B2 (ja) 露光装置及びデバイス製造方法
TW202418011A (zh) 控制方法、控制裝置、微影裝置及物品製造方法
KR102423011B1 (ko) 처리 장치를 관리하는 관리 방법, 관리 장치, 컴퓨터 프로그램 및 물품 제조 방법
JP7466403B2 (ja) 制御装置、リソグラフィー装置、制御方法および物品製造方法
US20220365454A1 (en) Mark detecting apparatus, mark learning apparatus, substrate processing apparatus, mark detecting method, and manufacturing method of article
JP2016099736A (ja) ステージ装置、リソグラフィ装置、物品の製造方法、および制御方法
CN115685692A (zh) 控制设备、光刻装置和物品制造方法
KR20220044652A (ko) 피드백 제어 장치, 리소그래피 장치, 측정 장치, 가공 장치, 평탄화 장치, 물품 제조 방법, 저장 매체 및 피드백 제어 방법
KR20230009303A (ko) 처리 장치, 관리 장치, 리소그래피 장치, 및 물품 제조 방법
JP2023013737A (ja) 処理装置および物品製造方法
JP2022002292A (ja) ステージ装置、パターン形成装置、及び物品の製造方法