An Adaptive Control Method for Precision Motion Platform Based on Grating Ruler Feedback

By analyzing the signal characteristics of the grating ruler to perform system state hierarchical identification and dynamic compensation, the problem that the feedback of the grating ruler in the prior art failed to effectively utilize the system state information was solved, and the high precision and stability of the precision motion platform were achieved, avoiding the use of additional sensors.

CN122816080BActive Publication Date: 2026-11-10HOPU TECH (NINGBO) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202611307774.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-08-27
Publication Date
2026-11-10
Estimated Expiration
2046-08-27

AI Technical Summary

Technical Problem

Existing precision motion platform control technology based on grating ruler feedback fails to effectively utilize the system state information in the original sine and cosine signals output by the grating ruler, resulting in error drift that is difficult to adapt to dynamic changes during long-term operation of the equipment, and the reliance on additional sensors increases system complexity and cost.

Method used

By analyzing the original signal characteristics of the grating ruler, the system state is hierarchically identified, and the compensation and control strategies are dynamically adjusted to achieve multi-source error adaptive compensation without additional sensors. This includes online identification of changes in the reading head installation gap, temperature drift of the signal conditioning circuit, and sway angle of the motion platform, and triggering a subdivision self-calibration mode for online correction.

Benefits of technology

It improves the long-term operating accuracy and stability of the precision motion platform, and enables online identification and dynamic compensation of the system status without adding additional sensors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122816080B_ABST
    Figure CN122816080B_ABST
Patent Text Reader

Abstract

This invention discloses an adaptive control method for a precision motion platform based on grating ruler feedback. The method obtains the original signal feature vector of the grating ruler by analyzing the original sine and cosine signals. This feature vector is then used for hierarchical identification of the system state, estimating the change in the reading head mounting gap, the temperature drift state of the signal conditioning circuit, and the yaw angle component of the motion platform. Based on this yaw angle component, a dynamic estimate of the Abbe error is calculated. A subdivision self-calibration mode is triggered based on the temperature drift state, and the subdivision compensation table is corrected online to obtain the corrected position feedback value. The dynamic estimate of the Abbe error and the corrected position feedback value are superimposed as the position loop feedback input. This invention enables online identification of the system state and dynamic compensation of multi-source errors without adding additional sensors, improving the long-term operating accuracy and stability of the precision motion platform.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of precision motion control technology, and specifically to an adaptive control method for a precision motion platform based on grating ruler feedback. Background Technology

[0002] Precision motion platforms are core components of high-end manufacturing equipment such as semiconductor lithography and precision inspection, and their positioning accuracy directly determines the processing quality and yield of products. As a key sensing element for closed-loop position feedback, the accuracy and stability of the measurement signal from the grating ruler are fundamental guarantees of the motion platform's performance. Existing control technologies based on grating ruler feedback typically only use the final position value calculated by the grating ruler as the feedback signal, while the rich system state information contained in the original sine and cosine signals output by the grating ruler, such as changes in signal amplitude and phase, is not effectively utilized. Furthermore, to address error drift caused by factors such as temperature changes and component wear, existing technologies often rely on additional sensors (such as temperature sensors and laser interferometers) for offline calibration and static compensation. This not only increases system complexity and cost, but also makes it difficult for the compensation model to adapt to the dynamically changing system state during long-term equipment operation, leading to a decline in accuracy maintenance. Summary of the Invention

[0003] In view of the above problems, the present invention provides an adaptive control method for a precision motion platform based on grating ruler feedback. By analyzing the original signal characteristics of the grating ruler to identify the physical state of the system online, and dynamically adjusting the compensation and control strategies accordingly, the method achieves multi-source error adaptive compensation without the need for additional sensors.

[0004] To achieve the above objectives, this application provides an adaptive control method for a precision motion platform based on grating ruler feedback, comprising: Obtain the original sine and original cosine signals output by the grating ruler reading head; DC bias estimation and AC component extraction are performed on the original sine and cosine signals. The synthesized amplitude and instantaneous phase difference of the original signals are calculated. The comparison results of the synthesized amplitude of the original signals with the preset standard amplitude and the comparison results of the instantaneous phase difference with the preset standard phase difference are used as the feature vector of the original signal of the grating ruler. The amplitude attenuation coefficient and phase offset are input to the system state layer identification module. The change in the reading head installation gap is estimated and output based on the changing trend of the amplitude attenuation coefficient. The temperature drift status of the signal conditioning circuit is estimated and output based on the change in the phase offset. The yaw angle component of the motion platform is estimated and output based on the difference in amplitude attenuation coefficients output by the first and second grating rulers symmetrically arranged on both sides of the motion axis. The Abbe error dynamic estimate is calculated and output based on the yaw angle component. The temperature drift status is input to the subdivision error online correction module. The subdivision self-calibration mode is triggered based on the temperature drift status. In the subdivision self-calibration mode, the motion platform is controlled to perform uniform reciprocating scanning. Abnormal jump intervals are identified based on the sampling point density distribution of each phase interval within one grid pitch period. The abnormal jump intervals are locally smoothed to update the subdivision compensation table. The uncompensated position values ​​are corrected based on the updated subdivision compensation table, and the corrected position feedback value is obtained and output. The Abbe error dynamic estimate is superimposed with the corrected position feedback value to obtain the compensated position feedback value, which is then used as the position loop feedback input for adaptive control of the precision motion platform.

[0005] In some embodiments, DC bias estimation and AC component extraction are performed on the original sine and cosine signals, the synthesized amplitude and instantaneous phase difference of the original signals are calculated, and the comparison results of the synthesized amplitude and the preset standard amplitude, as well as the comparison results of the instantaneous phase difference and the preset standard phase difference, are used as the feature vector of the original signal of the grating ruler, including: The original sine signal and the original cosine signal are respectively subjected to moving average filtering to obtain the DC bias of the sine channel and the DC bias of the cosine channel; Subtract the DC bias of the sine channel from the original sine signal to obtain the sine AC component, and subtract the DC bias of the cosine channel from the original cosine signal to obtain the cosine AC component. The original signal composite amplitude is obtained by performing square root operations on the sine AC component and the cosine AC component; Perform arctangent operation on the sinusoidal AC component and the cosine AC component to obtain the instantaneous phase relationship. Compare the instantaneous phase relationship with the orthogonal phase reference to obtain the instantaneous phase difference. The amplitude attenuation coefficient is obtained by comparing the original signal synthesized amplitude with the preset standard amplitude. The phase offset is obtained by comparing the instantaneous phase difference with the preset standard phase difference. The DC offset of the sine channel is obtained by comparing the DC offset of the sine channel with the preset standard DC offset. The DC offset of the cosine channel is obtained by comparing the DC offset of the cosine channel with the preset standard DC offset. The amplitude attenuation coefficient, phase offset, DC bias drift of the sine channel, and DC bias drift of the cosine channel are used together as the original signal feature vector of the grating ruler.

[0006] In some embodiments, estimating and outputting the change in the reading head mounting gap based on the changing trend of the amplitude attenuation coefficient includes: Obtain the first amplitude attenuation coefficient output by the first grating ruler and the second amplitude attenuation coefficient output by the second grating ruler, which are symmetrically arranged on both sides of the motion axis. The average amplitude attenuation coefficient is obtained by averaging the first amplitude attenuation coefficient and the second amplitude attenuation coefficient. Monitor the trend of the average amplitude attenuation coefficient over time to determine whether the average amplitude attenuation coefficient is increasing or decreasing. Based on the pre-calibrated mapping relationship between the air gap between the grating ruler reading head and the main ruler and the signal amplitude, the change in the average amplitude attenuation coefficient is converted into the corresponding change in the overall installation gap. The difference between the first amplitude attenuation coefficient and the second amplitude attenuation coefficient is obtained by performing a difference operation on the amplitude attenuation coefficient. Based on the difference in amplitude attenuation coefficient, and combined with the installation offset distance between the first and second grating rulers relative to the motion axis, the difference in local installation gap caused by the yaw motion is calculated. The overall installation gap change is superimposed with the local installation gap difference to obtain the comprehensive installation gap change, which is then output as the reading head installation gap change.

[0007] In some embodiments, estimating and outputting the temperature drift state of the signal conditioning circuit based on the change in phase offset includes: Obtain the phase offset time series within a preset time window; The phase shift time series is subjected to low-pass filtering to remove the fast random fluctuation component caused by mechanical vibration and electromagnetic interference, and retain the slow trend component caused by heat accumulation to obtain the phase shift trend component. The phase offset trend component is differentially calculated with the phase offset trend component at a preset time reference point to obtain the phase offset trend change. Query the pre-calibrated temperature-phase offset mapping table, which records the corresponding phase offset reference values ​​at different temperature points; The phase shift trend change is compared with the phase shift baseline value in the temperature-phase shift mapping table, and the corresponding temperature change is obtained through linear interpolation. The temperature change is compared with a preset temperature drift threshold. When the temperature change exceeds the preset temperature drift threshold, the signal conditioning circuit is determined to be in a significant temperature drift state. When the temperature change does not exceed the preset temperature drift threshold, the signal conditioning circuit is determined to be in a normal temperature drift state. The temperature change and the temperature drift determination result are used together as the temperature drift status output of the signal conditioning circuit.

[0008] In some embodiments, estimating and outputting the yaw angle component of the motion platform based on the difference in amplitude attenuation coefficients output by the first and second grating rulers symmetrically arranged on both sides of the motion axis includes: Obtain the first amplitude attenuation coefficient output by the first grating ruler and the second amplitude attenuation coefficient output by the second grating ruler; Monitor the direction of change of the first amplitude attenuation coefficient and the second amplitude attenuation coefficient. When the first amplitude attenuation coefficient and the second amplitude attenuation coefficient change in opposite directions, it is determined that the motion platform is yawing. The difference between the first amplitude attenuation coefficient and the second amplitude attenuation coefficient is obtained by performing a difference operation on the amplitude attenuation coefficient. Query the pre-calibrated amplitude attenuation coefficient difference and yaw angle mapping table. The amplitude attenuation coefficient difference and yaw angle mapping table records the amplitude attenuation coefficient difference values ​​corresponding to different yaw angles. The amplitude attenuation coefficient difference value is compared with the amplitude attenuation coefficient difference value in the yaw angle mapping table, and the corresponding yaw angle component is obtained through interpolation. Output the yaw angle component.

[0009] In some embodiments, calculating and outputting a dynamic estimate of the Abbe error based on the yaw angle component includes: Obtain the installation offset distance of the first and second grating rulers relative to the ideal motion axis, as well as the pitch and roll components of the motion platform in the motion direction; The Abbe error component caused by the yaw angle component is obtained by multiplying the installation offset distance. The pitch angle component is multiplied by the structural offset distance of the motion platform in the vertical direction to obtain the Abbe error component caused by pitch. The roll angle component is multiplied by the structural offset distance of the motion platform in the horizontal direction to obtain the Abbe error component caused by roll. The Abbe error components caused by yaw, pitch and roll are summed to obtain a comprehensive dynamic estimate of the Abbe error. Output the dynamic estimate of the combined Abbe error.

[0010] In some embodiments, the temperature drift status is input to the subdivision error online correction module, and the subdivision self-calibration mode is triggered based on the temperature drift status, including: Acquire the temperature change during the temperature drift state, and simultaneously acquire the position feedback data of the motion platform within a preset historical time window prior to the current moment; The position feedback data is sliced ​​by grid pitch period to obtain the historical position deviation sequence for each phase interval; Perform periodic consistency analysis on the historical position deviation sequence of each phase interval, and calculate the deviation difference between each phase interval; When the temperature change exceeds the preset temperature drift threshold and the deviation difference exceeds the preset difference threshold, the current motion task queue of the motion platform is further obtained. Determine if there are any high-priority motion tasks in the motion task queue that have a higher priority than the subdivision self-calibration; When there are no high-priority motion tasks in the motion task queue, generate a subdivision self-calibration trigger instruction; The subdivision self-calibration trigger command is output to the subdivision error online correction module to trigger the subdivision self-calibration mode.

[0011] In some embodiments, abnormal transition intervals are identified based on the sampling point density distribution of each phase interval within a grating period. Local smoothing corrections are performed on the abnormal transition intervals to update the subdivision compensation table. The uncompensated position values ​​are then corrected based on the updated subdivision compensation table to obtain and output the corrected position feedback value, including: The uncompensated position values ​​collected during the uniform reciprocating scanning process are obtained. The uncompensated position values ​​are divided into phase intervals according to the grid pitch period. The number of sampling points in each phase interval is counted to obtain the sampling point density of each phase interval. Calculate the ratio of the sampling point density of each phase interval to the sampling point density of the adjacent phase interval to obtain the density ratio sequence; The density ratio sequence is compared with a preset density ratio threshold. When the density ratio in the density ratio sequence exceeds the preset density ratio threshold, the corresponding phase interval is determined to be an abnormal transition interval. Obtain the subdivision compensation table correction amount of the abnormal jump interval and its adjacent normal interval, perform cubic spline interpolation on the subdivision compensation table correction amount of the abnormal jump interval to obtain the smoothed correction amount, and update the subdivision compensation table with the smoothed correction amount to obtain the online corrected subdivision compensation table. Determine whether the current motion mode is in subdivision self-calibration mode. When in subdivision self-calibration mode, query the online correction subdivision compensation table to obtain the final subdivision correction amount. When not in subdivision self-calibration mode, query the pre-stored offline subdivision compensation table to obtain the initial subdivision correction amount. The final subdivision correction or the initial subdivision correction is superimposed on the uncompensated position value to obtain the corrected position feedback value, and the corrected position feedback value is output.

[0012] In some embodiments, the ratio of the sampling point density of each phase interval to the sampling point density of adjacent phase intervals is calculated to obtain a density ratio sequence, including: Obtain the sampling point density of each phase interval, and arrange the sampling point density of each phase interval according to the phase interval order to obtain the sampling point density sequence; A sliding window process is applied to the sampling point density sequence. Taking the current phase interval as the center, the current phase interval and a preset number of adjacent phase intervals to its left and right are selected to form a local density window. Calculate the mean density of sampling points in all phase intervals within the local density window to obtain the local density mean. Divide the sampling point density of the current phase interval by the local density mean to obtain the local density ratio of the current phase interval; Traverse all phase intervals, repeat sliding window processing and local density ratio calculation to obtain the local density ratios corresponding to all phase intervals, forming a density ratio sequence. The density ratio sequence is compared with a preset density ratio threshold. When the density ratio in the density ratio sequence exceeds the preset density ratio threshold, the corresponding phase interval is determined to be an abnormal transition interval, including: Obtain each local density ratio in the density ratio sequence, and compare each local density ratio with a preset upper limit threshold and a preset lower limit threshold for density ratios, respectively; When the local density ratio is greater than the preset upper limit threshold of the density ratio, the phase interval corresponding to the local density ratio is determined to be an abnormal interval with high density, indicating that the subdivision compensation table correction amount of the phase interval is too large, resulting in excessive concentration of position sampling. When the local density ratio is less than the preset lower limit threshold, the phase interval corresponding to the local density ratio is determined to be an abnormal interval with low density, indicating that the subdivision compensation table correction amount of the phase interval is too small, resulting in excessively sparse position sampling. The high-density abnormal intervals and the low-density abnormal intervals are used together as the abnormal transition intervals.

[0013] In some embodiments, the Abbe error dynamic estimate is superimposed with the corrected position feedback value to obtain the compensated position feedback value, and the compensated position feedback value is used as the position loop feedback input, including: Obtain the dynamic estimate of Abbe error, the corrected position feedback value, the amplitude attenuation coefficient, and the temperature drift status; The first confidence weight of the Abbe error dynamic estimate is calculated based on the degree of deviation between the amplitude attenuation coefficient and the preset standard amplitude. The second confidence weight of the corrected position feedback value is calculated based on the degree of closeness between the temperature drift level in the temperature drift state and the preset temperature drift threshold. Multiply the Abbe error dynamic estimate by the first confidence weight to obtain the weighted Abbe error compensation, and multiply the corrected position feedback value by the second confidence weight to obtain the weighted position feedback value. The weighted Abbe error compensation amount and the weighted position feedback value are summed to obtain the confidence-weighted position feedback value; Determine whether the current motion phase is a high-speed scanning phase. When it is a high-speed scanning phase, use the confidence-weighted compensation position feedback value as the position loop feedback input. When it is a low-speed positioning phase, use the corrected position feedback value directly as the position loop feedback input.

[0014] Unlike existing technologies, the above-mentioned technical solution provides an adaptive control method for a precision motion platform based on grating ruler feedback. It obtains the original signal feature vector of the grating ruler by analyzing the original sine and cosine signals. This feature vector is then used for hierarchical identification of the system state, estimating the change in the reading head mounting gap, the temperature drift state of the signal conditioning circuit, and the yaw angle component of the motion platform. Based on this yaw angle component, a dynamic estimate of the Abbe error is calculated. A subdivision self-calibration mode is triggered based on the temperature drift state, and the subdivision compensation table is corrected online to obtain the corrected position feedback value. The dynamic estimate of the Abbe error and the corrected position feedback value are superimposed as the position loop feedback input. This invention can achieve online identification of the system state and dynamic compensation of multi-source errors without adding additional sensors, improving the long-term operating accuracy and stability of the precision motion platform.

[0015] The above description of the invention is merely an overview of the technical solution of this application. In order to enable those skilled in the art to better understand the technical solution of this application and to implement it based on the description and drawings, and to make the above-mentioned objectives and other objectives, features and advantages of this application easier to understand, the following description is provided in conjunction with the specific embodiments and drawings of this application. Attached Figure Description

[0016] The accompanying drawings are only used to illustrate the principles, implementation methods, applications, features, and effects of specific embodiments of the present invention and other related contents, and should not be considered as limitations on this application.

[0017] In the accompanying drawings of the instruction manual: Figure 1 This is a schematic diagram illustrating steps S101 to S105 of the method described in a specific embodiment. Figure 2 This is a schematic diagram illustrating steps S201 to S206 of the method described in a specific implementation. Figure 3 The diagram illustrates steps S301 to S307 of the method described in the specific implementation. Detailed Implementation

[0018] To illustrate the possible application scenarios, technical principles, implementable specific solutions, and achievable objectives and effects of this application in detail, the following description, in conjunction with the listed specific embodiments and accompanying drawings, provides a detailed explanation. The embodiments described herein are merely illustrative of the technical solutions of this application and are therefore intended to limit the scope of protection of this application.

[0019] In this document, the term "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The term "embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment, nor does it specifically limit its independence or connection with other embodiments. In principle, in this application, as long as there are no technical contradictions or conflicts, the technical features mentioned in each embodiment can be combined in any way to form corresponding implementable technical solutions.

[0020] Unless otherwise defined, the technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the use of related terms herein is merely for the purpose of describing particular embodiments and is not intended to limit this application.

[0021] In the description of this application, the term "and / or" is used to describe the logical relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A exists, B exists, and A and B exist simultaneously. Additionally, the character " / " in this document generally indicates that the preceding and following objects have an "or" logical relationship.

[0022] In this application, terms such as “first” and “second” are used only to distinguish one entity or operation from another, and do not necessarily require or imply any actual quantity, hierarchy or order relationship between these entities or operations.

[0023] Without further limitations, the use of terms such as “comprising,” “including,” “having,” or other similar open-ended expressions in this application is intended to cover non-exclusive inclusion, which does not exclude the presence of additional elements in a process, method, or product that includes the stated elements, such that a process, method, or product that includes a list of elements may include not only those defined elements but also other elements not expressly listed, or elements inherent to such a process, method, or product.

[0024] The processor described in the embodiments of this application can be implemented by hardware, firmware, software, or a combination thereof. It can be a circuit, one or more of an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field-programmable gate array (FPGA), a central processing unit (CPU), a controller, a microcontroller, or a microprocessor. It also includes other physical, biological, or chemical structures that can implement the same or equivalent functions as the processors listed above, such as biological neurons, quantum computing units, DNA computing units, etc., so that the processor can execute some or all of the steps in the computer program or method involved in the various embodiments of this application, or any combination of the steps mentioned therein.

[0025] The computer program involved in the embodiments can be stored in a computer device readable storage medium, which includes, but is not limited to, disks, magnetic tapes, magnetic cards, floppy disks, flash memory, optical disks, optical cards, read-only memory (ROM), random access memory (RAM), erasable programmable ROM (EPROM), and electrically erasable programmable ROM (EEPROM), etc., and also includes other biological, physical, or chemical structures that can achieve the same or equivalent functions as the storage media listed above, such as DNA, RNA, proteins, and other units with information storage capabilities. In specific embodiments, the storage medium involved can be one of the above-mentioned media types, or a combination of the above-mentioned media types. In different embodiments, the computer program involved in the embodiments can be centrally stored in a single medium, or distributed and stored in multiple media. The memory containing the computer device readable storage medium can be non-volatile memory or random access memory. These computer device readable storage media can be built into the device, or can be connected to the device involved in the embodiments as an external device or part of an external device. In some embodiments, the memory having a computer device readable storage medium is deployed locally; in other embodiments, the memory may be deployed remotely from the processor, for example, as a network-attached memory accessed via RF circuitry or an external port and a communication network, wherein the communication network may be the Internet, one or more intranets, a local area network (LAN), a wide area network (WLAN), a storage area network (SAN), or a suitable combination thereof, as long as computer device access to the memory is enabled. Furthermore, the computer program involved in the embodiments may be stored in plaintext / ciphertext form, or it may be designed as training data, integrated and recombined through model training and implicitly stored in the parameter states of a deep neural network or other machine learning model.

[0026] Please see Figure 1 This embodiment provides an adaptive control method for a precision motion platform based on grating ruler feedback, including: S101. Obtain the original sine and original cosine signals output by the grating ruler reading head; S102. Estimate the DC bias and extract the AC component of the original sine signal and the original cosine signal, calculate the synthesized amplitude and instantaneous phase difference of the original signal, and use the comparison results of the synthesized amplitude of the original signal with the preset standard amplitude and the comparison results of the instantaneous phase difference with the preset standard phase difference as the feature vector of the original signal of the grating ruler. S103. Input the amplitude attenuation coefficient and phase offset to the system state layer identification module. Estimate and output the change in the reading head installation gap based on the changing trend of the amplitude attenuation coefficient. Estimate and output the temperature drift status of the signal conditioning circuit based on the change in the phase offset. Estimate and output the yaw angle component of the motion platform based on the difference in amplitude attenuation coefficients output by the first and second grating rulers symmetrically arranged on both sides of the motion axis. Calculate and output the dynamic estimate of the Abbe error based on the yaw angle component. S104. Input the temperature drift status to the subdivision error online correction module. Determine whether to trigger the subdivision self-calibration mode based on the temperature drift status. In the subdivision self-calibration mode, control the motion platform to perform uniform reciprocating scanning. Identify abnormal jump intervals based on the sampling point density distribution of each phase interval within one grid pitch period. Perform local smoothing correction on the abnormal jump intervals to update the subdivision compensation table. Correct the uncompensated position values ​​based on the updated subdivision compensation table. Obtain and output the corrected position feedback value. S105. The Abbe error dynamic estimate is superimposed with the corrected position feedback value to obtain the compensated position feedback value, and the compensated position feedback value is used as the position loop feedback input to perform adaptive control of the precision motion platform.

[0027] In step S101, the grating ruler reading head converts the engraving information on the grating main scale into two electrical signals with a 90-degree phase difference, namely the original sine signal and the original cosine signal, through photoelectric conversion. The motion controller acquires these two signals in a synchronous sampling manner through a high-speed analog-to-digital converter module to ensure the signals are aligned in time, providing the original data basis for subsequent signal analysis.

[0028] In step S102, DC bias estimation is used to determine the fixed component in the original signal that deviates from zero, and AC component extraction removes this DC bias from the original signal to obtain a clean AC waveform. The synthesized amplitude of the original signal is obtained by combining the sine and cosine AC components, characterizing the overall signal strength; the instantaneous phase difference reflects the degree of orthogonal deviation between the two AC signals. The preset standard amplitude and preset standard phase difference are recorded and stored under standard operating conditions during the equipment factory calibration stage. The amplitude attenuation coefficient is obtained by comparing the real-time calculated synthesized amplitude of the original signal with the preset standard amplitude, and the phase offset is obtained by comparing the instantaneous phase difference with the preset standard phase difference. The two together constitute the feature vector of the original signal of the grating ruler.

[0029] In step S103, the system state layer identification module, based on the changing trend of the amplitude attenuation coefficient, utilizes a pre-established mapping relationship between the grating ruler signal amplitude and the reading head mounting gap to convert the real-time change of the amplitude attenuation coefficient into the corresponding change of the reading head mounting gap. This mapping relationship reflects the physical law that the amplitude attenuates as the gap increases, and its specific form can be obtained through factory calibration. Based on the changing trend of the phase offset, the temperature drift state of the signal conditioning circuit is estimated using a pre-calibrated mapping relationship between temperature and phase offset. This mapping relationship is obtained through temperature chamber calibration of the signal conditioning circuit. This temperature drift state reflects the electrical characteristic drift of circuit components caused by temperature changes.

[0030] By utilizing the difference in amplitude attenuation coefficients between the first and second grating rulers symmetrically arranged on both sides of the motion axis, a pre-calibrated amplitude difference-sway angle mapping relationship can be queried based on this difference to estimate the sway angle component of the motion platform. This mapping relationship is calibrated by combining a laser interferometer with synchronous measurement of the dual grating rulers. Based on the geometric relationship between this sway angle component and the grating ruler installation offset distance, the dynamic estimate of the Abbe error is calculated. This step, by establishing a quantitative mapping model between signal characteristics and physical states, enables sensorless online identification of various system states.

[0031] In step S104, the online subdivision error correction module determines whether the current subdivision compensation table is mismatched due to temperature drift based on the temperature drift status, and decides whether to trigger the subdivision self-calibration mode accordingly. The subdivision compensation table is pre-generated and stored through offline calibration. This table records the correction amount corresponding to the inherent position deviation caused by signal non-ideality at different phase positions within a complete grating period, and is used to correct the uncompensated position values ​​calculated in real time point by point. After triggering the subdivision self-calibration mode, the motion platform is controlled to perform uniform reciprocating scanning. During the scanning process, a large number of uncompensated position values ​​are collected. These position values ​​are divided into multiple phase intervals according to the grating period of the grating ruler, and the sampling point density distribution is formed by statistically analyzing the number of sampling points in each interval. When the sampling point density of a certain phase interval deviates significantly from the adjacent intervals, it is identified as an abnormal jump interval, indicating that there is a deviation in the subdivision compensation table correction amount at that point. This step realizes the online adaptive update of the subdivision error compensation table without the need for machine shutdown or external reference.

[0032] In step S105, the compensated position feedback value simultaneously incorporates the effects of dynamic Abbe error compensation and subdivision error correction. It is directly used as the feedback input signal of the position loop to participate in the closed-loop control calculation, enabling the control system to adjust based on more accurate position information, thereby achieving multi-source error adaptive and coordinated control of the precision motion platform.

[0033] This embodiment analyzes the original signal of the grating ruler and extracts its feature vector. Without adding additional sensors, it achieves online identification of system states such as the reading head installation gap, signal conditioning circuit temperature drift, and motion platform sway angle. Based on this, it dynamically estimates the Abbe error and online calibration subdivision error. Finally, it integrates the multi-source compensation results into the position loop feedback input, enabling the control system to adaptively adjust the feedback accuracy according to its real-time changes in state. This effectively improves the positioning accuracy maintenance capability and dynamic response performance of the precision motion platform during long-term operation.

[0034] Please see Figure 2 In some embodiments, DC bias estimation and AC component extraction are performed on the original sine and cosine signals, the synthesized amplitude and instantaneous phase difference of the original signals are calculated, and the comparison results of the synthesized amplitude and the preset standard amplitude, as well as the comparison results of the instantaneous phase difference and the preset standard phase difference, are used as the feature vector of the original signal of the grating ruler, including: S201. Perform moving average filtering on the original sine signal and the original cosine signal respectively to obtain the DC bias of the sine channel and the DC bias of the cosine channel; S202. Subtract the DC bias of the sine channel from the original sine signal to obtain the sine AC component, and subtract the DC bias of the cosine channel from the original cosine signal to obtain the cosine AC component. S203. Perform square root operation on the sine AC component and the cosine AC component to obtain the original signal synthesis amplitude. S204. Perform arctangent operation on the sinusoidal AC component and the cosine AC component to obtain the instantaneous phase relationship. Compare the instantaneous phase relationship with the orthogonal phase reference to obtain the instantaneous phase difference. S205. The amplitude attenuation coefficient is obtained by comparing the synthesized amplitude of the original signal with the preset standard amplitude. The phase offset is obtained by comparing the instantaneous phase difference with the preset standard phase difference. The DC offset of the sine channel is obtained by comparing the DC offset of the sine channel with the preset standard DC offset. The DC offset of the cosine channel is obtained by comparing the DC offset of the cosine channel with the preset standard DC offset. S206. The amplitude attenuation coefficient, phase offset, DC bias drift of the sine channel, and DC bias drift of the cosine channel are used together as the original signal feature vector of the grating ruler.

[0035] In step S201, the moving average filtering process calculates the arithmetic mean of multiple sampling points collected within a preset time window to suppress high-frequency noise in the original sine and cosine signals, thereby extracting the slowly varying DC component. The length of this preset time window needs to be set according to the period of the grating ruler signal, typically covering several complete signal periods to ensure that the average result can stably reflect the true level of the DC bias and avoid errors introduced by periodic signal fluctuations.

[0036] In step S202, the corresponding DC bias values ​​obtained in step S201 are subtracted from the original sine and cosine signals respectively by a subtraction operation, thereby obtaining a signal component containing only AC components. This operation eliminates the fixed offset introduced by circuit reference drift or light source fluctuations, enabling subsequent operations to be performed based on a pure AC waveform.

[0037] In step S203, the square root operation calculates the vector sum and magnitude of the two orthogonal signals by summing the instantaneous values ​​of the sinusoidal and cosine AC components and then taking the square root. This magnitude is the composite amplitude of the original signals. The result of this operation is independent of the instantaneous phase of the signals and only reflects the combined intensity of the two signals. Therefore, it can stably characterize the light intensity signal level received by the grating ruler reading head.

[0038] In step S204, the arctangent operation calculates the ratio of the sinusoidal AC component to the cosine AC component and obtains their arctangent angle to determine the instantaneous phase relationship between the two signals at the current sampling moment. This instantaneous phase relationship is then compared with the orthogonal phase reference, which is the standard phase value when the sinusoidal signal leads the cosine signal by 90 degrees under ideal orthogonal conditions. The resulting difference is the instantaneous phase difference, used to quantify the degree of orthogonality deviation between the two signals.

[0039] In step S205, the preset standard amplitude, preset standard phase difference, and preset standard DC bias are all reference values ​​measured and stored by the equipment during the factory calibration phase under standard operating conditions such as standard temperature and standard reading head gap. The real-time raw signal synthesized amplitude and instantaneous phase difference calculated in steps S203 and S204 are compared with the corresponding preset standard values ​​to obtain the amplitude attenuation coefficient and phase offset. The DC bias of the sine channel and the DC bias of the cosine channel obtained in step S201 are compared with the preset standard DC bias to obtain the DC bias drift of the dual channels. The above comparison calculation can be implemented using difference or ratio methods, and the output directly reflects the degree of deviation of the current signal quality from the factory reference state.

[0040] In step S206, the amplitude attenuation coefficient, phase offset, sine channel DC bias drift, and cosine channel DC bias drift obtained in step S205 are combined in a fixed order into a multi-dimensional data array, which is the feature vector of the original signal of the grating ruler. This feature vector completely encapsulates the various quality indicators of the original signal of the grating ruler at the current moment in a structured data form, and can be directly used as the input of the subsequent system state hierarchical identification module.

[0041] This embodiment utilizes digital signal processing techniques such as moving average filtering, subtraction, square root operations, and arctangent operations to progressively decompose and quantize the original sine and cosine signals output by the grating ruler into multiple characteristic indicators with clear physical meanings, including amplitude attenuation coefficient, phase offset, and dual-channel DC bias drift. These indicators are ultimately combined to form the original signal feature vector of the grating ruler. This feature vector comprehensively characterizes the current quality state of the original signal, providing a standardized data foundation for subsequent system state identification and error compensation based on signal characteristics.

[0042] Please see Figure 3 In some embodiments, estimating and outputting the change in the reading head mounting gap based on the changing trend of the amplitude attenuation coefficient includes: S301. Obtain the first amplitude attenuation coefficient output by the first grating ruler and the second amplitude attenuation coefficient output by the second grating ruler, which are symmetrically arranged on both sides of the motion axis. S302. The average amplitude attenuation coefficient is obtained by averaging the first amplitude attenuation coefficient and the second amplitude attenuation coefficient. S303. Monitor the trend of the average amplitude attenuation coefficient over time to determine whether the average amplitude attenuation coefficient is increasing or decreasing. S304. Based on the pre-calibrated mapping relationship between the air gap between the grating ruler reading head and the main ruler and the signal amplitude, the change in the average amplitude attenuation coefficient is converted into the corresponding change in the overall installation gap. S305. Perform a difference operation between the first amplitude attenuation coefficient and the second amplitude attenuation coefficient to obtain the amplitude attenuation coefficient difference value. S306. Based on the difference in amplitude attenuation coefficient, and combined with the installation offset distance between the first and second grating rulers relative to the motion axis, calculate the difference in local installation gap caused by the yaw motion. S307. The overall installation gap change is superimposed with the local installation gap difference to obtain the comprehensive installation gap change, and the comprehensive installation gap change is output as the reading head installation gap change.

[0043] In steps S301 to S302, the first amplitude attenuation coefficient and the second amplitude attenuation coefficient are obtained by feature analysis of the original signals of the first and second grating rulers symmetrically arranged on both sides of the motion axis. Both are directly read from the data interface of the system state hierarchical identification module and used as inputs for subsequent calculations. The mean value calculation eliminates the asymmetric components caused by the yaw motion and extracts the signal intensity change information shared by the two grating rulers.

[0044] In step S303, the average amplitude attenuation coefficient is continuously recorded over multiple consecutive sampling periods to form a time series of the coefficient. By comparing the current value with the average value at the previous time or within a historical sliding window, it is determined whether the average amplitude attenuation coefficient is increasing or decreasing over time. This direction of change directly corresponds to the increasing or decreasing trend of the air gap between the reading head and the main scale.

[0045] In step S304, the pre-calibrated mapping relationship between the air gap between the grating ruler reading head and the main scale and the signal amplitude is obtained by changing the gap distance between the reading head and the main scale using a precision displacement stage before the equipment leaves the factory, synchronously recording the signal amplitude under different gaps, and fitting the experimental data to obtain a functional relationship or lookup table. Substituting the change in the average amplitude attenuation coefficient monitored in step S303 into this mapping relationship, the overall installation gap change caused by non-slip factors can be calculated.

[0046] In steps S305 and S306, the first amplitude attenuation coefficient is subtracted from the second amplitude attenuation coefficient, eliminating the signal intensity variation component shared by the two grating rulers and extracting the information on the asymmetric signal variation caused by the yaw motion. The magnitude and direction of the difference in amplitude attenuation coefficients are directly related to the amplitude and direction of the yaw motion. Combining the installation offset distance of the first and second grating rulers relative to the motion axis, i.e., the vertical distance from the measurement position of the two grating rulers to the ideal motion axis, based on the geometric relationship between the yaw angle and the gap change, the local installation gap difference caused by the yaw motion is calculated. This difference characterizes the asymmetric change amplitude of the gap increase on one side and the gap decrease on the other side caused by the yaw.

[0047] In step S307, for the first grating ruler, the actual change in its mounting gap is the sum of the overall change and the local change; for the second grating ruler, the actual change in its mounting gap is the difference between the overall change and the local change. The combined change in mounting gap obtained after superposition is the overall change in the air gap between the reading head and the main ruler at each grating ruler position at the current moment relative to the factory reference state, and is output as the change in the mounting gap of the reading head.

[0048] This embodiment extracts the common-mode and differential-mode information from the attenuation coefficient of the dual-grating scale value through mean and difference operations, respectively. It uses a pre-calibrated mapping relationship to convert the signal change into a physically meaningful gap change, and separates the overall gap change and the local gap change caused by sway based on geometric relationships. Finally, it obtains the comprehensive installation gap change through superposition operation, realizing accurate online estimation of the reading head installation gap change, and providing key state parameters for subsequent Abbe error dynamic compensation and thermal expansion correction.

[0049] In some embodiments, estimating and outputting the temperature drift state of the signal conditioning circuit based on the change in phase offset includes: Obtain the phase offset time series within a preset time window; The phase shift time series is subjected to low-pass filtering to remove the fast random fluctuation component caused by mechanical vibration and electromagnetic interference, and retain the slow trend component caused by heat accumulation to obtain the phase shift trend component. The phase offset trend component is differentially calculated with the phase offset trend component at a preset time reference point to obtain the phase offset trend change. Query the pre-calibrated temperature-phase offset mapping table, which records the corresponding phase offset reference values ​​at different temperature points; The phase shift trend change is compared with the phase shift baseline value in the temperature-phase shift mapping table, and the corresponding temperature change is obtained through linear interpolation. The temperature change is compared with a preset temperature drift threshold. When the temperature change exceeds the preset temperature drift threshold, the signal conditioning circuit is determined to be in a significant temperature drift state. When the temperature change does not exceed the preset temperature drift threshold, the signal conditioning circuit is determined to be in a normal temperature drift state. The temperature change and the temperature drift determination result are used together as the temperature drift status output of the signal conditioning circuit.

[0050] In this embodiment, the phase offset time series is composed of phase offsets continuously collected by the system state hierarchical identification module within a preset time window, arranged in the order of sampling time. The length of the preset time window is set according to the thermal time constant of the signal conditioning circuit, and it needs to cover a sufficiently long thermal accumulation process to ensure the accuracy of trend analysis, while also taking into account real-time requirements.

[0051] Low-pass filtering of the phase shift time series can be achieved using a digital filter. The cutoff frequency is set in a range below the frequencies of mechanical vibration and electromagnetic interference but above the frequency of thermal accumulation, thus preserving the slow-changing trend component caused by thermal accumulation while effectively suppressing rapid random fluctuations. The phase shift trend component output after filtering is the gradual change component reflecting the thermal accumulation effect.

[0052] The preset time reference point is selected from the cold start time of the equipment or the stable operation time after the most recent calibration. The phase offset trend component corresponding to this time is used as the reference value for subsequent comparisons. The phase offset trend component at the current time is differentially calculated with the phase offset trend component at the preset time reference point. The difference is the change in phase offset trend. This change excludes the initial state differences and only reflects the phase offset increment caused by temperature changes since the reference time.

[0053] A pre-calibrated temperature-phase offset mapping table is generated by placing the signal conditioning circuit in a temperature chamber and measuring and recording the corresponding phase offset reference values ​​at multiple temperature nodes. This mapping table is stored in the non-volatile memory of the motion controller in the form of discrete data points or fitted curves. The phase offset trend change is compared with the phase offset reference value in the temperature-phase offset mapping table, and the temperature change that precisely corresponds to the current phase offset trend change is calculated between adjacent temperature nodes through linear interpolation.

[0054] The preset temperature drift threshold is set based on the device's tolerance for temperature drift. This tolerance is determined by the sensitivity of the subdivision error to temperature and the positioning accuracy requirements. The calculated temperature change is compared with the preset temperature drift threshold. When the temperature change exceeds the preset threshold, the signal conditioning circuit is determined to be in a significant temperature drift state, indicating that the current temperature drift may have a non-negligible impact on the subdivision accuracy. When the temperature change does not exceed the preset threshold, the signal conditioning circuit is determined to be in a normal temperature drift state, indicating that the current temperature drift is within an acceptable range.

[0055] The temperature change and the temperature drift status determination result are used together as the temperature drift status output of the signal conditioning circuit. This output is encapsulated in the form of structured data, providing a quantitative basis for triggering the self-calibration mode for the subdivision error online correction module.

[0056] This embodiment extracts the slow change trend caused by thermal accumulation from the phase offset time series through low-pass filtering, eliminates the initial state difference by combining differential operation, and accurately converts the phase offset change into the temperature change by linear interpolation using a pre-calibrated temperature-phase offset mapping table. Finally, the quantitative result and judgment result of the temperature drift state are output by threshold comparison, realizing the online accurate estimation of the temperature drift state of the signal conditioning circuit, and providing reliable data support for the self-calibration trigger of the subdivision error online correction module.

[0057] In some embodiments, estimating and outputting the yaw angle component of the motion platform based on the difference in amplitude attenuation coefficients output by the first and second grating rulers symmetrically arranged on both sides of the motion axis includes: Obtain the first amplitude attenuation coefficient output by the first grating ruler and the second amplitude attenuation coefficient output by the second grating ruler; Monitor the direction of change of the first amplitude attenuation coefficient and the second amplitude attenuation coefficient. When the first amplitude attenuation coefficient and the second amplitude attenuation coefficient change in opposite directions, it is determined that the motion platform is yawing. The difference between the first amplitude attenuation coefficient and the second amplitude attenuation coefficient is obtained by performing a difference operation on the amplitude attenuation coefficient. Query the pre-calibrated amplitude attenuation coefficient difference and yaw angle mapping table. The amplitude attenuation coefficient difference and yaw angle mapping table records the amplitude attenuation coefficient difference values ​​corresponding to different yaw angles. The amplitude attenuation coefficient difference value is compared with the amplitude attenuation coefficient difference value in the yaw angle mapping table, and the corresponding yaw angle component is obtained through interpolation. Output the yaw angle component.

[0058] In this embodiment, the first amplitude attenuation coefficient and the second amplitude attenuation coefficient are synchronously read from the data bus of the system state hierarchical identification module to ensure time alignment. By continuously recording the values ​​of the first and second amplitude attenuation coefficients over multiple consecutive sampling periods, their respective time series are formed. For each time series, the instantaneous change direction of each coefficient is determined by calculating the sign of the difference between the current value and the previous value. When the signs of the difference between the first and second amplitude attenuation coefficients are opposite, i.e., one coefficient shows an upward trend while the other shows a downward trend, it is determined that the motion platform has produced a yaw motion. This judgment logic is based on the physical mechanism that the yaw motion causes a decrease in the gap between one side of the reading head and an increase in the gap between the other side, thereby causing opposite attenuation changes in the amplitudes of the two grating ruler signals.

[0059] The sign of the amplitude attenuation coefficient difference indicates the direction of the yaw motion, and its absolute value is positively correlated with the yaw angle amplitude. A pre-calibrated mapping table of amplitude attenuation coefficient difference to yaw angle is generated by placing the motion platform on a precision turntable and simultaneously measuring the amplitude attenuation coefficient difference of the dual grating rulers at multiple known yaw angles. During calibration, the yaw angle is gradually increased from zero degrees in fixed angular steps, and the stable amplitude attenuation coefficient difference value at each angular step is recorded, forming a discrete set of calibration data points. This data point set is stored in the non-volatile memory of the motion controller in the form of a lookup table, and adjacent nodes can be continuous through linear or polynomial interpolation.

[0060] The amplitude attenuation coefficient difference value is compared with the amplitude attenuation coefficient difference value in the yaw angle mapping table. First, the two calibration nodes adjacent to the current difference value are located. Then, interpolation is performed between these two nodes to calculate the yaw angle component that precisely corresponds to the current difference value. The choice of interpolation method depends on the density of data points in the mapping table. When the calibration step size is small, linear interpolation is sufficient to meet the accuracy requirements. When the calibration step size is large, polynomial interpolation can be used to improve the interpolation accuracy. The output yaw angle component is used as one of the input parameters for calculating the dynamic estimate of the Abbe error.

[0061] This embodiment identifies the occurrence of yaw motion by monitoring the change direction sign of the amplitude attenuation coefficient of the dual grating scale. It uses differential operation to quantify the asymmetric signal change caused by yaw. Combined with the amplitude attenuation coefficient difference and yaw angle mapping table obtained in advance through angle step calibration, the yaw angle component is accurately estimated through node positioning and interpolation operation, realizing sensorless online estimation of the yaw angle of the motion platform, and providing angle information for subsequent dynamic compensation of Abbe error.

[0062] In some embodiments, calculating and outputting a dynamic estimate of the Abbe error based on the yaw angle component includes: Obtain the installation offset distance of the first and second grating rulers relative to the ideal motion axis, as well as the pitch and roll components of the motion platform in the motion direction; The Abbe error component caused by the yaw angle component is obtained by multiplying the installation offset distance. The pitch angle component is multiplied by the structural offset distance of the motion platform in the vertical direction to obtain the Abbe error component caused by pitch. The roll angle component is multiplied by the structural offset distance of the motion platform in the horizontal direction to obtain the Abbe error component caused by roll. The Abbe error components caused by yaw, pitch and roll are summed to obtain a comprehensive dynamic estimate of the Abbe error. Output the dynamic estimate of the combined Abbe error.

[0063] In this embodiment, the installation offset distance of the first and second grating rulers relative to the ideal motion axis is a fixed parameter during the equipment design phase. This parameter describes the vertical distance from the grating ruler measurement axis to the ideal motion axis. After the equipment is assembled, it is determined by precise measurement and stored in the non-volatile memory of the motion controller. The pitch and roll components of the motion platform in the motion direction characterize the rotation angle of the platform about the horizontal transverse axis and the horizontal longitudinal axis, respectively. These components can be indirectly obtained through the motion platform's own attitude calculation module or the multi-axis grating ruler configuration, or determined as known fixed system parameters during the equipment design phase.

[0064] The multiplication of the yaw angle component and the installation offset distance is based on the fundamental definition of Abbe error, which states that the product of the angular error and the offset distance equals the additional positional deviation at the measurement point. The Abbe error component caused by yaw reflects the horizontal positional deviation of the grating ruler measurement point caused by the yaw motion, which deviates from the ideal motion axis.

[0065] The multiplication of the pitch angle component and the vertical structural offset distance of the motion platform is used to calculate the vertical position deviation caused by the pitch motion. The vertical structural offset distance of the motion platform refers to the vertical distance between the measuring point of the grating ruler and the ideal motion plane, which is also a fixed parameter in the equipment design phase. The Abbe error component caused by pitch characterizes the additional position deviation caused by the vertical offset of the measuring point due to the pitch motion.

[0066] The multiplication of the roll angle component and the horizontal structural offset distance of the motion platform is used to calculate the horizontal positional deviation caused by the roll motion. The horizontal structural offset distance of the motion platform refers to the horizontal distance between the measuring point of the grating ruler and the ideal motion axis; this parameter is determined by the mechanical structure of the equipment. The Abbe error component caused by roll characterizes the additional positional deviation caused by the horizontal offset of the measuring point due to the roll motion.

[0067] The Abbe error components caused by yaw, pitch, and roll are summed, and the additional position deviations caused by the three rotational degrees of freedom are vector-superimposed in space to comprehensively reflect the total Abbe error generated by the attitude change of the motion platform at the current moment. The dynamic estimate of the comprehensive Abbe error is output to the position loop feedback input, which is then superimposed with the corrected position feedback value to compensate for the position measurement deviation caused by attitude changes.

[0068] This embodiment introduces attitude information from three rotational degrees of freedom—yaw, pitch, and roll—and combines it with geometric parameters such as the offset distance of the grating ruler installation and the offset distance of the platform structure. It calculates the Abbe error components corresponding to each attitude angle and performs vector summation, thereby realizing dynamic estimation of Abbe error in the three-dimensional space of the motion platform and providing a basis for accurate real-time compensation of position feedback.

[0069] In some embodiments, the temperature drift status is input to the subdivision error online correction module, and the subdivision self-calibration mode is triggered based on the temperature drift status, including: Acquire the temperature change during the temperature drift state, and simultaneously acquire the position feedback data of the motion platform within a preset historical time window prior to the current moment; The position feedback data is sliced ​​by grid pitch period to obtain the historical position deviation sequence for each phase interval; Perform periodic consistency analysis on the historical position deviation sequence of each phase interval, and calculate the deviation difference between each phase interval; When the temperature change exceeds the preset temperature drift threshold and the deviation difference exceeds the preset difference threshold, the current motion task queue of the motion platform is further obtained. Determine if there are any high-priority motion tasks in the motion task queue that have a higher priority than the subdivision self-calibration; When there are no high-priority motion tasks in the motion task queue, generate a subdivision self-calibration trigger instruction; The subdivision self-calibration trigger command is output to the subdivision error online correction module to trigger the subdivision self-calibration mode.

[0070] In this embodiment, the position feedback data of the motion platform within a preset historical time window prior to the current moment is continuously collected from the position ring data bus and stored in a ring buffer. Preferably, the length of the preset historical time window is set to cover at least dozens of complete grid pitch cycles to ensure a sufficient sample size for subsequent statistical analysis.

[0071] The position feedback data undergoes grating period slicing, mapping continuous position values ​​to relative phase positions within the grating ruler's grating period through modulo operations. A grating period is divided into a fixed number of phase intervals, each corresponding to a specific phase range within the grating period. Historical position deviation values ​​belonging to the same phase interval are arranged chronologically to form a historical position deviation sequence for each phase interval. This sequence records the trajectory of subdivision error accumulation over time within that phase interval.

[0072] Periodic consistency analysis of the historical position deviation sequences for each phase interval is performed. By calculating the variance of the deviation sequence within each phase interval, the fluctuation amplitude of the deviation within that interval is quantified. The variances of each phase interval are compared, and the ratio of the maximum to the minimum value is taken as the deviation difference degree. This difference degree directly reflects the consistency of the subdivision error between different phase intervals; a larger difference degree indicates a more severe mismatch between the subdivision compensation table and the current actual subdivision error.

[0073] The preset temperature drift threshold is calibrated based on the sensitivity of the subdivision error to temperature and the equipment positioning accuracy requirements, while the preset difference threshold is calibrated based on the allowable error range of the subdivision compensation table. When the temperature change exceeds the preset temperature drift threshold and the deviation difference exceeds the preset difference threshold, it indicates that the temperature drift has caused a significant mismatch in the subdivision compensation table. At this point, the current motion task queue of the motion platform is further obtained, which stores all motion tasks to be executed in priority order.

[0074] The system checks if there are any high-priority motion tasks in the motion task queue that have a higher priority than subdivision self-calibration. High-priority motion tasks include exposure scanning tasks or critical dimension measurement tasks. If there are no high-priority motion tasks in the motion task queue, it indicates that the current system allows the self-calibration process to be executed, and a subdivision self-calibration trigger command is generated. This trigger command is output to the subdivision error online correction module in the form of a status flag, triggering the subdivision self-calibration mode.

[0075] This embodiment identifies the mismatch condition of the subdivision compensation table by judging the dual threshold of temperature change and deviation difference. Combined with the priority judgment of the motion task queue, it ensures that the self-calibration process does not interfere with high-priority production tasks. The subdivision self-calibration mode is triggered only when the system is idle, realizing the intelligent triggering of online correction of subdivision error and automatically maintaining subdivision accuracy while ensuring the normal operation of the equipment.

[0076] In some embodiments, abnormal transition intervals are identified based on the sampling point density distribution of each phase interval within a grating period. Local smoothing corrections are performed on the abnormal transition intervals to update the subdivision compensation table. The uncompensated position values ​​are then corrected based on the updated subdivision compensation table to obtain and output the corrected position feedback value, including: The uncompensated position values ​​collected during the uniform reciprocating scanning process are obtained. The uncompensated position values ​​are divided into phase intervals according to the grid pitch period. The number of sampling points in each phase interval is counted to obtain the sampling point density of each phase interval. Calculate the ratio of the sampling point density of each phase interval to the sampling point density of the adjacent phase interval to obtain the density ratio sequence; The density ratio sequence is compared with a preset density ratio threshold. When the density ratio in the density ratio sequence exceeds the preset density ratio threshold, the corresponding phase interval is determined to be an abnormal transition interval. Obtain the subdivision compensation table correction amount of the abnormal jump interval and its adjacent normal interval, perform cubic spline interpolation on the subdivision compensation table correction amount of the abnormal jump interval to obtain the smoothed correction amount, and update the subdivision compensation table with the smoothed correction amount to obtain the online corrected subdivision compensation table. Determine whether the current motion mode is in subdivision self-calibration mode. When in subdivision self-calibration mode, query the online correction subdivision compensation table to obtain the final subdivision correction amount. When not in subdivision self-calibration mode, query the pre-stored offline subdivision compensation table to obtain the initial subdivision correction amount. The final subdivision correction or the initial subdivision correction is superimposed on the uncompensated position value to obtain the corrected position feedback value, and the corrected position feedback value is output.

[0077] In this embodiment, the uncompensated position values ​​acquired during the uniform reciprocating scan are directly read by the motion controller from the original signal processing module of the grating ruler at a fixed sampling period during the scanning motion. The scanning speed is set to a constant low speed to ensure that a sufficient number of sampling points can be acquired in each phase interval within one grating pitch period. The uncompensated position values ​​are divided into phase intervals according to the grating pitch period. One grating pitch period is divided into a fixed number of phase intervals, and each uncompensated position value is assigned to the corresponding phase interval according to its relative position within one grating pitch period. The number of sampling points falling into each phase interval is counted to obtain the sampling point density of each phase interval. This density distribution directly reflects the local correction effect of the subdivision compensation table: in phase intervals with a large correction amount, the sampling points will have a high density due to excessive aggregation of position values; in phase intervals with a small correction amount, the sampling points will have a low density due to excessive dispersion of position values.

[0078] The ratio of the sampling point density of each phase interval to the sampling point density of adjacent phase intervals is calculated. This ratio is obtained by taking the arithmetic mean of the density values ​​of the two adjacent intervals to the left and right of the current interval as the denominator, and using the density of the current interval as the numerator. The local density ratios of all phase intervals are arranged in phase order to form a density ratio sequence. The preset density ratio threshold is calibrated according to the maximum local deviation allowed by the subdivision compensation table. This threshold is determined by statistically analyzing the historical distribution range of the density ratio under normal conditions, and is usually set to several times the standard deviation of the historical mean.

[0079] Each ratio in the density ratio sequence is compared one by one with a preset density ratio threshold. When a ratio exceeds the threshold range, it indicates that the sampling point density of that phase interval has deviated significantly from its neighborhood in a statistically significant way, and the interval is determined to be an abnormal jump interval. This determination result directly locates the specific phase position that needs to be corrected in the subdivision compensation table.

[0080] The correction values ​​for the abnormal transition interval and its adjacent normal interval are obtained through a subdivision compensation table. Using the correction values ​​of the adjacent normal intervals as boundary nodes, cubic spline interpolation is performed on the correction values ​​within the abnormal transition interval. Cubic spline interpolation constructs a piecewise cubic polynomial, ensuring the continuity of the function values, first derivative, and second derivative at the boundary nodes. This generates a correction value curve that runs through the abnormal interval and smoothly connects to the normal intervals at both ends. The values ​​at each phase position within the abnormal interval on the curve are the smoothed correction values. The smoothed correction values ​​are then written into the corresponding phase positions in the subdivision compensation table, replacing the original abnormal correction values, thus obtaining the online correction subdivision compensation table.

[0081] The system determines whether the device is in subdivision self-calibration mode based on the current motion mode. This determination is achieved by reading the mode status flag of the motion controller. When in subdivision self-calibration mode, the final subdivision correction amount is obtained by consulting the online subdivision compensation table based on the phase position of the current uncompensated position value. When not in subdivision self-calibration mode, the initial subdivision correction amount is obtained by consulting the pre-stored offline subdivision compensation table. The offline subdivision compensation table is generated at the factory using a laser interferometer and serves as the default subdivision error compensation benchmark.

[0082] This embodiment acquires uncompensated position values ​​through uniform reciprocating scanning and counts the sampling point density of each phase interval. It identifies abnormal jump intervals by comparing the local density ratio with historical statistical thresholds. It uses cubic spline interpolation to smoothly update the subdivision compensation table correction amount of abnormal intervals with normal intervals as boundaries. It selects online or offline compensation table query for subdivision correction amount according to motion mode. Finally, it superimposes the correction amount onto the uncompensated position value and outputs the corrected position feedback value, realizing online adaptive updating of the subdivision error compensation table and real-time correction of position feedback.

[0083] In some embodiments, the ratio of the sampling point density of each phase interval to the sampling point density of adjacent phase intervals is calculated to obtain a density ratio sequence, including: Obtain the sampling point density of each phase interval, and arrange the sampling point density of each phase interval according to the phase interval order to obtain the sampling point density sequence; A sliding window process is applied to the sampling point density sequence. Taking the current phase interval as the center, the current phase interval and a preset number of adjacent phase intervals to its left and right are selected to form a local density window. Calculate the mean density of sampling points in all phase intervals within the local density window to obtain the local density mean. Divide the sampling point density of the current phase interval by the local density mean to obtain the local density ratio of the current phase interval; Traverse all phase intervals, repeat sliding window processing and local density ratio calculation to obtain the local density ratios corresponding to all phase intervals, forming a density ratio sequence. The density ratio sequence is compared with a preset density ratio threshold. When the density ratio in the density ratio sequence exceeds the preset density ratio threshold, the corresponding phase interval is determined to be an abnormal transition interval, including: Obtain each local density ratio in the density ratio sequence, and compare each local density ratio with a preset upper limit threshold and a preset lower limit threshold for density ratios, respectively; When the local density ratio is greater than the preset upper limit threshold of the density ratio, the phase interval corresponding to the local density ratio is determined to be an abnormal interval with high density, indicating that the subdivision compensation table correction amount of the phase interval is too large, resulting in excessive concentration of position sampling. When the local density ratio is less than the preset lower limit threshold, the phase interval corresponding to the local density ratio is determined to be an abnormal interval with low density, indicating that the subdivision compensation table correction amount of the phase interval is too small, resulting in excessively sparse position sampling. The high-density abnormal intervals and the low-density abnormal intervals are used together as the abnormal transition intervals.

[0084] In this embodiment, the sampling point density sequence is stored in array form, with array indices corresponding to phase interval numbers and array elements corresponding to the sampling point density values ​​of that interval. A sliding window process is applied to the sampling point density sequence, sliding an interval-by-interval window of a fixed radius across the sequence. The window radius is determined by a predetermined number of adjacent phase intervals on the left and right, based on the total number of phase intervals within the grating period and the desired degree of local smoothness. A predetermined number of adjacent phase intervals are selected to the left and right of the current phase interval as the center, collectively forming a local density window. For boundary intervals near the beginning or end of the sequence, missing adjacent intervals within the window are filled with their own density values ​​to ensure consistent window size.

[0085] The arithmetic mean of the sampling point densities across all phase intervals within the local density window is calculated to obtain the local density mean. This mean reflects the average sampling point density level within the neighborhood of the current phase interval and serves as a reference benchmark for determining whether the density of the current interval is abnormal.

[0086] Divide the sampling point density of the current phase interval by the local density mean to obtain the local density ratio of the current phase interval. A ratio greater than 1 indicates that the density of the current interval is higher than the average level of the neighborhood, and a ratio less than 1 indicates that it is lower than the average level of the neighborhood. The degree of deviation from 1 reflects the severity of the anomaly. Traverse all phase intervals, repeat the sliding window processing and local density ratio calculation for each interval, and obtain the local density ratios corresponding to all phase intervals. Arrange them in phase order to form a density ratio sequence.

[0087] The preset upper and lower thresholds for density ratios are calibrated based on the maximum local deviation allowed by the subdivision compensation table. The upper threshold is greater than 1, and the lower threshold is less than 1, together defining a normal density ratio range. Local density ratios are obtained from the density ratio sequence, and each local density ratio is compared with the preset upper and lower thresholds. When a local density ratio is greater than the preset upper threshold, the phase interval is determined to be a high-density anomalous interval, indicating that the subdivision compensation table correction amount in this interval is too large, resulting in over-concentration of position sampling. When a local density ratio is less than the preset lower threshold, the phase interval is determined to be a low-density anomalous interval, indicating that the subdivision compensation table correction amount in this interval is too small, resulting in over-sparse position sampling. The high-density and low-density anomalous intervals are combined as anomalous transition intervals and output to the subsequent cubic spline interpolation processing steps.

[0088] This embodiment calculates the local density mean using a sliding window process, and replaces the global ratio with the local density ratio to eliminate the influence of overall density fluctuations. It introduces a two-way judgment of upper and lower thresholds to distinguish between two different types of anomalies: the correction amount is too large and the correction amount is too small. This enables more accurate identification and classification of anomaly jump intervals and provides more accurate correction position information for subsequent cubic spline interpolation.

[0089] In some embodiments, the Abbe error dynamic estimate is superimposed with the corrected position feedback value to obtain the compensated position feedback value, and the compensated position feedback value is used as the position loop feedback input, including: Obtain the dynamic estimate of Abbe error, the corrected position feedback value, the amplitude attenuation coefficient, and the temperature drift status; The first confidence weight of the Abbe error dynamic estimate is calculated based on the degree of deviation between the amplitude attenuation coefficient and the preset standard amplitude. The second confidence weight of the corrected position feedback value is calculated based on the degree of closeness between the temperature drift level in the temperature drift state and the preset temperature drift threshold. The Abbe error dynamic estimate is multiplied by the first confidence weight to obtain the weighted Abbe error compensation amount, and the corrected position feedback value is multiplied by the second confidence weight to obtain the weighted position feedback value. This weighting operation dynamically adjusts the contribution ratio of the Abbe error compensation and subdivision error correction in the final feedback value according to the confidence of the current system state. The weighted Abbe error compensation amount and the weighted position feedback value are summed to obtain the confidence-weighted position feedback value. The confidence-weighted position feedback value integrates the compensation effects of dynamic Abbe error compensation and subdivision error correction, and the contribution ratio of each compensation is adaptively adjusted according to its current confidence level. Determine whether the current motion phase is a high-speed scanning phase. When it is a high-speed scanning phase, use the confidence-weighted compensation position feedback value as the position loop feedback input. When it is a low-speed positioning phase, use the corrected position feedback value directly as the position loop feedback input.

[0090] In this embodiment, the Abbe error dynamic estimate, the corrected position feedback value, the amplitude attenuation coefficient, and the temperature drift status are read synchronously from the output interface of the corresponding processing module to ensure that the four correspond to the same control cycle, providing a time-aligned data basis for subsequent confidence calculation and weighted fusion.

[0091] The first confidence weight is calculated based on the deviation of the amplitude attenuation coefficient from the preset standard amplitude. The preset standard amplitude is the reference amplitude recorded under standard operating conditions when the equipment leaves the factory. The closer the amplitude attenuation coefficient is to 1, the closer the signal strength of the grating ruler is to the factory condition, and the more reliable the sway angle estimation on which the Abbe error dynamic estimation depends is, and the first confidence weight approaches 1. The greater the deviation of the amplitude attenuation coefficient from 1, the more severe the signal attenuation, the lower the confidence of the sway angle estimation, and the lower the first confidence weight is accordingly. This weight is obtained by mapping through a monotonically decreasing function, the function form of which is pre-calibrated according to the influence characteristics of signal attenuation on the accuracy of sway angle estimation.

[0092] The second confidence weight is calculated based on the proximity of the temperature drift level in the temperature drift state to a preset temperature drift threshold. The temperature drift level is quantified by the temperature change, and the preset temperature drift threshold defines the maximum acceptable temperature drift range for subdivision error. When the temperature drift level is much lower than the preset temperature drift threshold, the subdivision compensation table remains within the effective range, the reliability of the corrected position feedback value is high, and the second confidence weight approaches 1. When the temperature drift level is close to or exceeds the preset temperature drift threshold, the subdivision compensation table may mismatch, the reliability of the corrected position feedback value decreases, and the second confidence weight decreases accordingly. This weight is obtained through a monotonically decreasing function mapping, the function form of which is pre-calibrated based on the characteristics of the influence of temperature drift on subdivision accuracy.

[0093] Determining whether the current motion phase is a high-speed scanning phase can be achieved by reading the motion command speed setting value of the motion controller. When the motion speed is higher than the preset speed threshold, it is determined to be a high-speed scanning phase. At this time, the position feedback value after confidence weighting compensation is used as the position loop feedback input to fully utilize the ability of Abbe error dynamic compensation to correct attitude errors under high-speed motion. When the motion speed is lower than the preset speed threshold, it is determined to be a low-speed positioning phase. At this time, the corrected position feedback value is directly used as the position loop feedback input to avoid weakening the Abbe error compensation effect due to the first confidence weight being too low caused by signal attenuation, while also avoiding additional noise that may be introduced by weighting calculation.

[0094] This embodiment introduces confidence weights based on amplitude attenuation coefficients and temperature drift states to achieve adaptive weighted fusion of Abbe error compensation and subdivision error correction. By switching the feedback signal source according to the motion stage, the position loop feedback input can maintain the optimal compensation effect under different working conditions, thereby improving the positioning accuracy and robustness of the precision motion platform across the entire working range.

[0095] By adopting the above technical solutions, this invention differs from existing technologies and has the following beneficial effects: By analyzing the original signal of the grating ruler to extract characteristic indicators such as amplitude attenuation coefficient and phase offset, a quantitative mapping model between signal characteristics and physical state is established, enabling sensorless online identification of system states such as reading head installation gap, signal conditioning circuit temperature drift, and motion platform sway angle, and dynamically estimating Abbe error and online calibration subdivision error accordingly; Based on confidence weight, adaptive weighted fusion of Abbe error compensation and subdivision error correction is achieved, and the feedback signal source is switched according to the motion stage, so that the position loop feedback input maintains the optimal compensation effect under different working conditions, thereby significantly improving the positioning accuracy maintenance capability and dynamic response performance of the precision motion platform during long-term operation without adding additional sensors.

[0096] Finally, it should be noted that although the above embodiments have been described in the text and drawings of this application, this should not limit the scope of patent protection of this application. Any technical solutions that are based on the essential concept of this application and utilize the content described in the text and drawings of this application, resulting in equivalent structural or procedural substitutions or modifications, as well as the direct or indirect application of the technical solutions of the above embodiments to other related technical fields, are all included within the scope of patent protection of this application.

Claims

1. An adaptive control method for a precision motion platform based on grating ruler feedback, characterized in that, include: Obtain the original sine and original cosine signals output by the grating ruler reading head; DC bias estimation and AC component extraction are performed on the original sine signal and the original cosine signal to obtain the DC bias of the sine channel and the DC bias of the cosine channel. The synthesized amplitude and instantaneous phase difference of the original signal are calculated. The comparison results of the synthesized amplitude of the original signal with the preset standard amplitude, the comparison results of the instantaneous phase difference with the preset standard phase difference, the comparison results of the DC bias of the sine channel with the preset standard DC bias, and the comparison results of the DC bias of the cosine channel with the preset standard DC bias are used as the feature vector of the original signal of the grating ruler. The amplitude attenuation coefficient is obtained by comparing the synthesized amplitude of the original signal with the preset standard amplitude, and the phase offset is obtained by comparing the instantaneous phase difference with the preset standard phase difference. The amplitude attenuation coefficient and the phase offset are input to the system state layer identification module. The change in the reading head installation gap is estimated and output based on the changing trend of the amplitude attenuation coefficient. The temperature drift state of the signal conditioning circuit is estimated and output based on the change in the phase offset. The yaw angle component of the motion platform is estimated and output based on the difference in amplitude attenuation coefficients output by the first and second grating rulers symmetrically arranged on both sides of the motion axis. The Abbe error dynamic estimate is calculated and output based on the yaw angle component. The temperature drift status is input to the subdivision error online correction module. The subdivision self-calibration mode is triggered based on the temperature drift status. In the subdivision self-calibration mode, the motion platform is controlled to perform uniform reciprocating scanning. Abnormal jump intervals are identified based on the sampling point density distribution of each phase interval within one grid pitch period. The abnormal jump intervals are locally smoothed to update the subdivision compensation table. The uncompensated position values ​​are corrected based on the updated subdivision compensation table, and the corrected position feedback value is obtained and output. The Abbe error dynamic estimate is superimposed with the corrected position feedback value to obtain the compensated position feedback value, which is then used as the position loop feedback input for adaptive control of the precision motion platform.

2. The adaptive control method for a precision motion platform based on grating ruler feedback according to claim 1, characterized in that, DC bias estimation and AC component extraction are performed on the original sine and cosine signals to obtain the DC bias of the sine channel and the cosine channel. The synthesized amplitude and instantaneous phase difference of the original signals are calculated. The comparison results of the synthesized amplitude of the original signals with a preset standard amplitude, the comparison results of the instantaneous phase difference with a preset standard phase difference, the comparison results of the DC bias of the sine channel with a preset standard DC bias, and the comparison results of the DC bias of the cosine channel with a preset standard DC bias are used as the feature vector of the original signal of the grating ruler, including: The original sine signal and the original cosine signal are respectively subjected to moving average filtering to obtain the DC bias of the sine channel and the DC bias of the cosine channel; Subtracting the DC bias of the sine channel from the original sine signal yields the sine AC component, and subtracting the DC bias of the cosine channel from the original cosine signal yields the cosine AC component. The original signal synthesis amplitude is obtained by performing square root operations on the sine AC component and the cosine AC component; Perform arctangent operation on the sinusoidal AC component and the cosine AC component to obtain the instantaneous phase relationship, and compare the instantaneous phase relationship with the orthogonal phase reference to obtain the instantaneous phase difference; The amplitude attenuation coefficient is obtained by comparing the synthesized amplitude of the original signal with the preset standard amplitude; the phase offset is obtained by comparing the instantaneous phase difference with the preset standard phase difference; the DC bias of the sine channel is obtained by comparing the DC bias of the sine channel with the preset standard DC bias; and the DC bias of the cosine channel is obtained by comparing the DC bias of the cosine channel with the preset standard DC bias. The amplitude attenuation coefficient, the phase offset, the DC bias drift of the sine channel, and the DC bias drift of the cosine channel are collectively used as the original signal feature vector of the grating ruler.

3. The adaptive control method for a precision motion platform based on grating ruler feedback according to claim 1, characterized in that, Estimate and output the change in the reading head installation gap based on the changing trend of the amplitude attenuation coefficient, including: Obtain the first amplitude attenuation coefficient output by the first grating ruler and the second amplitude attenuation coefficient output by the second grating ruler, which are symmetrically arranged on both sides of the motion axis. The average amplitude attenuation coefficient is obtained by averaging the first amplitude attenuation coefficient and the second amplitude attenuation coefficient. Monitor the trend of the average amplitude attenuation coefficient over time to determine whether the average amplitude attenuation coefficient is increasing or decreasing; Based on the pre-calibrated mapping relationship between the air gap between the grating ruler reading head and the main ruler and the signal amplitude, the change in the average amplitude attenuation coefficient is converted into the corresponding change in the overall installation gap. The difference between the first amplitude attenuation coefficient and the second amplitude attenuation coefficient is obtained by performing a difference operation on the amplitude attenuation coefficient. Based on the difference in amplitude attenuation coefficient, and combined with the installation offset distance between the first grating ruler and the second grating ruler relative to the motion axis, the difference in local installation gap caused by the yaw motion is calculated. The overall installation gap change is superimposed with the local installation gap difference to obtain the comprehensive installation gap change, and the comprehensive installation gap change is output as the reading head installation gap change.

4. The adaptive control method for a precision motion platform based on grating ruler feedback according to claim 1, characterized in that, The temperature drift state of the signal conditioning circuit is estimated and output based on the change in the phase offset, including: Obtain the phase offset time series within a preset time window; The phase offset time series is subjected to low-pass filtering to remove the fast random fluctuation components caused by mechanical vibration and electromagnetic interference, and retain the slow trend components caused by heat accumulation to obtain the phase offset trend components. The phase offset trend component is differentially calculated with the phase offset trend component at a preset time reference point to obtain the phase offset trend change. Query the pre-calibrated temperature-phase offset mapping table, which records the corresponding phase offset reference values ​​at different temperature points; The phase shift trend change is compared with the phase shift reference value in the temperature-phase shift mapping table, and the corresponding temperature change is obtained through linear interpolation. The temperature change is compared with a preset temperature drift threshold. When the temperature change exceeds the preset temperature drift threshold, the signal conditioning circuit is determined to be in a significant temperature drift state. When the temperature change does not exceed the preset temperature drift threshold, the signal conditioning circuit is determined to be in a normal temperature drift state. The temperature change and the temperature drift determination result are used together as the temperature drift status output of the signal conditioning circuit.

5. The adaptive control method for a precision motion platform based on grating ruler feedback according to claim 1, characterized in that, And based on the difference in amplitude attenuation coefficients output by the first and second grating rulers symmetrically arranged on both sides of the motion axis, the yaw angle component of the motion platform is estimated and output, including: Obtain the first amplitude attenuation coefficient output by the first grating ruler and the second amplitude attenuation coefficient output by the second grating ruler; Monitor the direction of change of the first amplitude attenuation coefficient and the second amplitude attenuation coefficient. When the first amplitude attenuation coefficient and the second amplitude attenuation coefficient change in opposite directions, it is determined that the motion platform is yawing. The difference between the first amplitude attenuation coefficient and the second amplitude attenuation coefficient is obtained by performing a difference operation on the amplitude attenuation coefficient. Query the pre-calibrated amplitude attenuation coefficient difference and yaw angle mapping table, which records the amplitude attenuation coefficient difference values ​​corresponding to different yaw angles; The amplitude attenuation coefficient difference value is compared with the amplitude attenuation coefficient difference value in the yaw angle mapping table, and the corresponding yaw angle component is obtained through interpolation. Output the yaw angle component.

6. The adaptive control method for a precision motion platform based on grating ruler feedback according to claim 1, characterized in that, The Abbe error dynamic estimate is calculated and output based on the yaw angle components, including: Obtain the installation offset distance of the first and second grating rulers relative to the ideal motion axis, as well as the pitch and roll components of the motion platform in the motion direction; The yaw angle component is multiplied by the installation offset distance to obtain the Abbe error component caused by the yaw. The pitch angle component is multiplied by the structural offset distance of the motion platform in the vertical direction to obtain the Abbe error component caused by pitch. The roll angle component is multiplied by the structural offset distance of the motion platform in the horizontal direction to obtain the Abbe error component caused by roll. The Abbe error components caused by yaw, pitch and roll are summed to obtain a comprehensive dynamic estimate of the Abbe error. Output the dynamic estimate of the integrated Abbe error.

7. The adaptive control method for a precision motion platform based on grating ruler feedback according to claim 1, characterized in that, The temperature drift status is input to the subdivision error online correction module, and the subdivision self-calibration mode is triggered based on the temperature drift status, including: The temperature change during the temperature drift state is obtained, and the position feedback data of the motion platform within a preset historical time window before the current moment is also obtained. The position feedback data is sliced ​​by grid pitch period to obtain the historical position deviation sequence for each phase interval; Perform periodic consistency analysis on the historical position deviation sequences of each phase interval, and calculate the deviation difference between each phase interval; When the temperature change exceeds a preset temperature drift threshold and the deviation difference exceeds a preset difference threshold, the current motion task queue of the motion platform is further obtained. Determine whether there are any high-priority motion tasks with a higher priority than the subdivision self-calibration in the motion task queue; When the high-priority motion task is not present in the motion task queue, a subdivision self-calibration trigger instruction is generated. The subdivision self-calibration trigger command is output to the subdivision error online correction module to trigger the subdivision self-calibration mode.

8. The adaptive control method for a precision motion platform based on grating ruler feedback according to claim 1, characterized in that, Abnormal transition intervals are identified based on the sampling point density distribution of each phase interval within a grating period. Local smoothing corrections are performed on these abnormal transition intervals to update the subdivision compensation table. Uncompensated position values ​​are then corrected based on the updated subdivision compensation table, resulting in and outputting corrected position feedback values, including: The uncompensated position values ​​collected during the uniform reciprocating scanning process are obtained. The uncompensated position values ​​are divided into phase intervals according to the grid pitch period. The number of sampling points in each phase interval is counted to obtain the sampling point density of each phase interval. Calculate the ratio of the sampling point density of each phase interval to the sampling point density of the adjacent phase interval to obtain the density ratio sequence; The density ratio sequence is compared with a preset density ratio threshold. When the density ratio in the density ratio sequence exceeds the preset density ratio threshold, the corresponding phase interval is determined to be an abnormal transition interval. Obtain the subdivision compensation table correction amount of the abnormal jump interval and its adjacent normal interval, perform cubic spline interpolation on the subdivision compensation table correction amount of the abnormal jump interval to obtain the smoothed correction amount, and update the subdivision compensation table with the smoothed correction amount to obtain the online corrected subdivision compensation table. Determine whether the current motion mode is in subdivision self-calibration mode. When in subdivision self-calibration mode, query the online correction subdivision compensation table to obtain the final subdivision correction amount. When not in subdivision self-calibration mode, query the pre-stored offline subdivision compensation table to obtain the initial subdivision correction amount. The final subdivision correction or the initial subdivision correction is superimposed on the uncompensated position value to obtain the corrected position feedback value, and the corrected position feedback value is output.

9. The adaptive control method for a precision motion platform based on grating ruler feedback according to claim 8, characterized in that, Calculate the ratio of the sampling point density in each phase interval to the sampling point density in the adjacent phase interval to obtain a density ratio sequence, including: Obtain the sampling point density of each phase interval, and arrange the sampling point density of each phase interval according to the phase interval order to obtain the sampling point density sequence; The sampling point density sequence is processed by a sliding window. Taking the current phase interval as the center, the current phase interval and a preset number of adjacent phase intervals to its left and right are selected to form a local density window. Calculate the mean density of sampling points in all phase intervals within the local density window to obtain the local density mean. Divide the sampling point density of the current phase interval by the local density mean to obtain the local density ratio of the current phase interval; Traverse all phase intervals, repeat the sliding window processing and the local density ratio calculation to obtain the local density ratios corresponding to all phase intervals, thus forming a density ratio sequence. The density ratio sequence is compared with a preset density ratio threshold. When the density ratio in the density ratio sequence exceeds the preset density ratio threshold, the corresponding phase interval is determined to be an abnormal transition interval, including: Obtain each local density ratio in the density ratio sequence, and compare each local density ratio with a preset upper limit threshold and a preset lower limit threshold for density ratios, respectively; When the local density ratio is greater than the preset upper limit threshold of the density ratio, the phase interval corresponding to the local density ratio is determined to be an abnormally high density interval, indicating that the subdivision compensation table correction amount of the phase interval is too large, resulting in excessive concentration of position sampling. When the local density ratio is less than the preset lower limit threshold of the density ratio, the phase interval corresponding to the local density ratio is determined to be an abnormal interval with low density, indicating that the subdivision compensation table correction amount of the phase interval is too small, resulting in excessively sparse position sampling. The high density anomaly interval and the low density anomaly interval are collectively referred to as the anomaly jump interval.

10. The adaptive control method for a precision motion platform based on grating ruler feedback according to claim 1, characterized in that, The Abbe error dynamic estimate is superimposed on the corrected position feedback value to obtain the compensated position feedback value, and the compensated position feedback value is used as the position loop feedback input, including: Obtain the dynamic estimate of the Abbe error, the corrected position feedback value, the amplitude attenuation coefficient, and the temperature drift state; Based on the degree of deviation between the amplitude attenuation coefficient and the preset standard amplitude, the first confidence weight of the Abbe error dynamic estimate is calculated. The second confidence weight of the corrected position feedback value is calculated based on the degree of closeness between the temperature drift level in the temperature drift state and the preset temperature drift threshold. The Abbe error dynamic estimate is multiplied by the first confidence weight to obtain the weighted Abbe error compensation amount, and the corrected position feedback value is multiplied by the second confidence weight to obtain the weighted position feedback value. The weighted Abbe error compensation amount and the weighted position feedback value are summed to obtain the confidence-weighted position feedback value; Determine whether the current motion phase is a high-speed scanning phase. When it is a high-speed scanning phase, use the confidence-weighted compensated position feedback value as the position loop feedback input. When it is a low-speed positioning phase, use the corrected position feedback value directly as the position loop feedback input.

Citation Information

Patent Citations

  • Grating Moire fringe signal error detection and correction method

    CN108151653A

  • Self-adaptive positioning error compensation control method and system for gantry machining center

    CN122284498A