A real-time error compensation method for a precision air-floating motion platform

Abstract

The application relates to the technical field of precision air-floating motion tables, in particular to a precision air-floating motion table error real-time compensation method. The method comprises a deep learning dynamic prediction algorithm: an abnormal trajectory of a throttle hole region is taken as a label, temperature in a motion process, air supply pressure of the throttle hole region and a load current are taken as inputs, and a prediction model is trained; in the process that a target object moves along an air-floating motion table, current temperature, current air supply pressure of the throttle hole region and a current load current are detected; the detected data are input into the prediction model to obtain a predicted trajectory; a difference between the predicted trajectory and a reference trajectory is calculated; a sum of the unevenness and the difference is taken as an error, and the error is input into a proportional-integral-derivative control equation to obtain an air pressure control amount of the throttle hole region, so that motion compensation of the target object is realized. The application can realize accurate and real-time motion error compensation.

Description

Technical Field

[0001] This application relates to the field of precision air-bearing motion stage technology, and more specifically, to a method for real-time error compensation of a precision air-bearing motion stage. Background Technology

[0002] A precision air-bearing motion stage is an ultra-precision positioning platform that utilizes a high-pressure gas film to achieve frictionless motion. Its core technology lies in the formation of a stable gas film, approximately 5-20 micrometers thick, between the moving parts and the stationary base using aerostatic bearings. This unique working principle eliminates the stick-slip effect and wear problems of traditional mechanical guideways, making it an indispensable core equipment in fields such as semiconductor manufacturing, optical inspection, and space simulation. The air-bearing motion stage mainly consists of a base structure, air-bearing guideways, a linear motor, displacement sensors (such as optical scales and laser interferometers), and a control system. The air-bearing bearings inject compressed gas (usually dry air or inert gas) into the bearing surface through throttling orifices, forming a uniformly distributed pressure field to support the load.

[0003] The systematic geometric errors of the motion platform are obtained through precision measuring equipment, a mathematical model is established, and the model is incorporated into the control system to achieve real-time compensation. The patented technology proposed by Wuhan Huazhiyang Technology demonstrates the complete process of this method: First, a global geometric error model considering the coupling of translation and rotation errors is constructed, with precise modeling specifically for the Abbe and cosine errors of the gantry structure; next, a laser interferometer system is used to measure the geometric errors of the platform in 6 degrees of freedom, obtaining raw data at the sub-micron and sub-arcsecond levels; then, the least squares method is used to perform polynomial fitting on discrete error points to establish a functional relationship between error and displacement command; finally, the error function is imported into the servo controller to implement real-time compensation based on position loop feedforward.

[0004] However, the mathematical model has errors compared to the actual precision air-bearing motion platform, resulting in poor compensation effect.

[0005] In view of the above, this application is hereby submitted. Summary of the Invention

[0006] The purpose of this application is to provide a real-time error compensation method for a precision air-bearing motion stage, so as to achieve accurate and real-time motion error compensation.

[0007] To achieve the above objectives, this application adopts the following technical solution:

[0008] This application provides a method for real-time error compensation of a precision air-bearing motion stage, including:

[0009] Control multiple identical spherical objects to move multiple times along the throttle orifice of the air-floating motion platform;

[0010] During the movement, the three-dimensional motion trajectory of the spherical object is collected;

[0011] The three-dimensional motion trajectory is compared with the reference trajectory to identify abnormal trajectories and the throttle orifice regions corresponding to the abnormal trajectories.

[0012] The abnormal trajectory in the orifice region is used as a label, and the temperature, air supply pressure in the orifice region, and load current during the movement are used as inputs to train the prediction model.

[0013] During the movement of the target object along the air-bearing motion platform, the current temperature, the current air supply pressure in the throttling orifice area, and the current load current are detected; and the detected current temperature, current air supply pressure in the throttling orifice area, and current load current are input into the prediction model to obtain the predicted trajectory;

[0014] Calculate the difference between the predicted trajectory and the baseline trajectory;

[0015] Detect the unevenness of the target object as it moves through the orifice region;

[0016] The sum of the unevenness and the difference is used as the error and input into the proportional-integral-derivative control equation to obtain the air pressure control quantity of the throttle orifice region, thereby realizing the motion compensation of the target object.

[0017] Optionally, a light-emitting point is provided at the center of the spherical object;

[0018] During the motion, the three-dimensional motion trajectory of the spherical object is acquired, including:

[0019] The movement of the light-emitting point is captured by multiple infrared cameras, and the three-dimensional trajectory of the light-emitting point is calculated by combining stereo vision and triangulation principles.

[0020] Optionally, multiple identical spherical objects can be controlled to move multiple times along the throttle orifice of the air-floating motion stage, including:

[0021] When a spherical object is controlled to move multiple times along the target column or the target row or the target column ...

[0022] The target column throttling orifice can be 1 column, 2 columns, or 3 columns; the target row throttling orifice can be 1 row, 2 rows, or 3 rows.

[0023] Optionally, the three-dimensional motion trajectory is compared with the reference trajectory to determine the abnormal trajectory and the throttle orifice region corresponding to the abnormal trajectory, including:

[0024] The three-dimensional motion trajectories corresponding to all target column throttle orifices and the three-dimensional motion trajectories corresponding to all target row throttle orifices are compared with the baseline trajectory to identify abnormal trajectories.

[0025] The row and column intersection area corresponding to the abnormal trajectory is taken as the throttle orifice area corresponding to the abnormal trajectory.

[0026] Optionally, after inputting the detected current temperature, current supply gas pressure in the orifice region, and current load current into the prediction model to obtain the predicted trajectory, the model further includes:

[0027] The predicted trajectory is classified to obtain anomaly categories; the anomaly categories include low air pressure, high air pressure, and unstable air pressure.

[0028] The anomaly categories have a one-to-one correspondence with the proportional-integral-derivative control equations.

[0029] Optionally, after inputting the sum of the unevenness and the difference as an error into the proportional-integral-derivative control equation to obtain the air pressure control quantity of the orifice region, the method further includes:

[0030] The air pressure in the orifice region is controlled using the aforementioned air pressure control quantity;

[0031] Return to the operation of detecting the current temperature, current air supply pressure in the throttling orifice area, and current load current during the movement of the target object along the air-bearing motion platform.

[0032] Optionally, detecting the unevenness of the target object as it moves in the orifice region includes:

[0033] The unevenness of the target object as it moves in the orifice region is detected using a laser interferometer.

[0034] Optionally, the prediction model is a long short-term memory network.

[0035] Compared with the prior art, the beneficial effects of this application are as follows:

[0036] Due to various factors such as air supply pressure, load, and temperature, the air supply pressure can become unstable, causing instability in the movement of the target object on the air-bearing motion platform, requiring real-time compensation. This application pre-trains a prediction model by extracting data from experiments on the movement of a spherical object on the air-bearing motion platform. This model can then predict the trajectory of the spherical object based on the current temperature, the current air supply pressure in the orifice region, and the current load current. The difference between the predicted trajectory and the reference trajectory, as well as the unevenness of the target object moving in the orifice region, are then calculated. The air pressure control quantity is generated from both the air supply pressure angle of the orifice (microscopic) and the overall unevenness of the target object (macroscopic). This application uses a proportional-integral-derivative control equation to adjust the air pressure control quantity, thus eliminating the need to consider the physical relationship between the air supply pressure and the object's motion, making it simple and effective for industrial implementation. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0038] Figure 1 This is a flowchart of a real-time error compensation method for a precision air-bearing motion stage provided in an embodiment of this application;

[0039] Figure 2 This is a schematic diagram of the orifice movement region provided in the embodiments of this application;

[0040] Figure 3 This is a schematic diagram of the throttle orifice region where a malfunction has occurred, as provided in an embodiment of this application. Detailed Implementation

[0041] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0042] The present application will be further described in detail below with reference to the embodiments.

[0043] Figure 1This is a flowchart of a real-time error compensation method for a precision air-bearing motion stage provided in an embodiment of this application. The application scenario of this application is that, during the movement of a target object on a precision air-bearing motion stage (hereinafter referred to as the air-bearing motion stage), the faulty throttling orifice area is identified, and the air supply pressure of the throttling orifice area is controlled to compensate for the movement of the target object, thereby ensuring the stability of the target object's movement.

[0044] See Figure 1 The method provided in this application includes the following operations:

[0045] S110: Control multiple identical spherical objects to move multiple times along the throttle orifice of the air-floating motion platform.

[0046] The surface of the air-bearing motion platform is provided with multiple throttling orifices arranged in a matrix. Gas is ejected from each throttling orifice under the pressure of the supplied air, forming an air film on the surface of the air-bearing motion platform to prevent direct contact between the object and the platform. This embodiment does not limit the orifice diameter, spacing, or number.

[0047] This application does not limit the material or size of the spherical object. See also Figure 2 The outermost rectangle represents the outline of the air-bearing motion platform, viewed from above. For example, five spherical objects move along a row of orifices, and five spherical objects move along a column of orifices. Depending on their radius, the spherical objects contact one, two, or three orifices. Figure 2 The red border indicates the region where a spherical object moves along two rows of orifices, meaning the object moves under the combined pressure of the air pressure from the two orifices. The green border indicates the region where another spherical object moves along two rows of orifices. It should be noted that the air-bearing motion platform has many orifices, requiring a large number of spherical objects to move. For ease of description and illustration, this application uses only 10 spherical objects.

[0048] For example, multiple identical spherical objects are controlled to move along the orifices of the air-bearing motion platform at 1-minute intervals. For each spherical object, there are two rows of orifices on the air-bearing motion platform. The spherical object is controlled to move along the two rows of orifices once at 1-minute intervals, thus achieving multiple movements. In this way, the three-dimensional motion trajectory of the spherical object on the fixed orifices is obtained for multiple time periods.

[0049] In actual operation, the air-bearing motion platform can be tilted slightly so that the spherical object moves along the throttling orifice under the action of gravity.

[0050] Optionally, to ensure that the spherical object moves along certain columns of throttling orifices without exceeding those columns, the following measures are adopted: When controlling a spherical object to move multiple times along the target column or row of throttling orifices on the air-bearing motion platform, the air supply pressure of adjacent throttling orifices is controlled to be greater than the air supply pressure of the target column orifices, thereby constraining the trajectory of the spherical object. The target column of throttling orifices consists of one, two, or three columns; the target row of throttling orifices consists of one, two, or three rows. See also Figure 2 When the spherical object moves along the red area, the air pressure supplied by the adjacent throttle orifice in the yellow area is higher, forming a barrier at the edge of the red area and constraining the trajectory of the spherical object within the red area.

[0051] S120. During the motion, the three-dimensional motion trajectory of the spherical object is collected.

[0052] To accurately capture the trajectory of a spherical object, a light-emitting point, such as a fluorescent marker, is placed at the center of the object. Multiple infrared cameras capture the movement of this light-emitting point, and the three-dimensional trajectory is calculated using stereo vision and triangulation principles. The triangulation principle involves calculating the three-dimensional coordinates of the light-emitting point using the parallax angles and camera parameters (intrinsic and extrinsic parameters) of two or more infrared cameras.

[0053] S130. Compare the three-dimensional motion trajectory with the reference trajectory to determine the abnormal trajectory and the throttle orifice region corresponding to the abnormal trajectory.

[0054] The baseline trajectory is the three-dimensional motion trajectory of a spherical object under the condition that the orifice pressure is correct and stable.

[0055] The 3D motion trajectories corresponding to all target column throttle orifices and all target row throttle orifices are compared with the baseline trajectory to identify abnormal trajectories. The row-column intersection region corresponding to the abnormal trajectory is taken as the throttle orifice region corresponding to the abnormal trajectory.

[0056] Taking a three-dimensional motion trajectory A as an example, the similarity between the three-dimensional motion trajectory A and the reference trajectory is calculated using Dynamic Time Warping (DTW). Dynamic Time Warping (DTW) is an algorithm used to calculate the similarity between two time series (i.e., the three-dimensional coordinate sequence that changes over time in this application), and is particularly suitable for handling sequences with inconsistent lengths or time axis offsets. Its core idea is to find the optimal path through dynamic programming to align the two sequences on the time axis, thereby minimizing the distance between matching points. If the obtained similarity is greater than a set threshold, it indicates that the two trajectories are similar, meaning the three-dimensional motion trajectory is normal. If the obtained similarity is less than or equal to the set threshold, it indicates that the two trajectories are not similar, meaning the three-dimensional motion trajectory is an abnormal trajectory.

[0057] See Figure 3 The three-dimensional motion trajectory corresponding to the target row throttle orifice in the red area and the three-dimensional motion trajectory corresponding to the target column throttle orifice in the green area are abnormal trajectories. The part where the red area and the green area intersect (represented by shading), that is, the part that is covered by both, is the throttle orifice area where the fault occurred.

[0058] S140. Using the abnormal trajectory of the throttling orifice region as a label, and the temperature, air supply pressure and load current of the throttling orifice region during the movement as inputs, train the prediction model.

[0059] Temperature changes affect the mechanical structure and positioning accuracy of the motion table through material thermal expansion and thermal stress, which in turn affects the stability of the air supply pressure. The magnitude of the air supply pressure directly affects the performance of the air bearing, including the stability and support capacity of the air film, which directly affects the trajectory of spherical objects. Load current affects the motion performance and responsiveness of the motion table through the drive system (such as a motor). Sudden changes in load current can directly reflect the fault condition in the throttling orifice area.

[0060] To simplify the calculation, the abnormal trajectory in the orifice region was sampled at equal intervals, resulting in 10 sampling points, each with three-dimensional coordinates. That is, a 3*10 matrix with labels.

[0061] The air-bearing motion platform is equipped with a temperature sensor to measure the temperature of the spherical object during its motion; this temperature can be a constant value. Assuming the orifice region includes 10 orifices, there are 10 supply air pressures. The load current can be measured via circuitry. The temperature, supply air pressure, and load current are combined into a row vector as the input sample, meaning the input dimension is 1*12.

[0062] Optionally, the prediction model can be a long short-term memory network, or other networks used for time series prediction can be selected.

[0063] By controlling a spherical object to move multiple times along the same orifice in S110, multiple abnormal trajectories and corresponding abnormal regions can be obtained, thereby constructing multiple labels and input samples. Increasing the number of samples is beneficial for training the prediction model more accurately.

[0064] The above steps, by controlling the movement of multiple spherical objects on an air-bearing motion platform and collecting data to train a predictive model, identified the faulty throttling orifice region on the air-bearing motion platform. The predictive model can then predict the abnormal trajectory within the throttling orifice region at the current moment based on temperature, air supply pressure in the throttling orifice region, and load current.

[0065] S150. During the movement of the target object along the air-floating motion platform, the current temperature, the current air supply pressure in the throttling orifice area, and the current load current are detected; and the detected current temperature, the current air supply pressure in the throttling orifice area, and the current load current are input into the prediction model to obtain the predicted trajectory.

[0066] The target object differs from the aforementioned spherical object; it is an actual object in operation, such as a wafer. The target object's volume is much larger than that of the spherical object, making it difficult to detect faulty orifices at the microscopic (orifice) level. Therefore, the orifice region is located by observing the movement trajectory of the aforementioned spherical object. In actual operation, the air pump provides the same supply air pressure to each orifice, but the air pressure ejected from the orifice may differ; thus, the faulty orifice region is located through the aforementioned steps.

[0067] The current temperature, the current gas supply pressure in the orifice region, and the current load current are combined into a row vector and input into the prediction model to obtain the predicted trajectory output by the prediction model. This predicted trajectory is the trajectory of a spherical object assuming it is moving in the orifice region.

[0068] Optionally, after S150, the predicted trajectory is classified to obtain anomaly categories, including low pressure, high pressure, and unstable pressure. For example, if the predicted trajectory includes spikes, it is classified as high pressure; if the predicted trajectory has low values, it is classified as low pressure; and if the predicted trajectory exhibits severe fluctuations, it is classified as unstable pressure. This anomaly category has a one-to-one correspondence with the following proportional-integral-derivative (PID) control equation. That is, the matching PID control equation needs to be selected based on the anomaly category to calculate the pressure control quantity.

[0069] S160. Calculate the difference between the predicted trajectory and the baseline trajectory.

[0070] The predicted trajectory is time-aligned with the baseline trajectory, and the distance between the two points on the predicted trajectory and the baseline trajectory that are farthest apart at the same time is selected as the difference between the predicted trajectory and the baseline trajectory.

[0071] S170. Detect the unevenness of the target object when it moves in the orifice region.

[0072] The unevenness of a target object moving in the orifice region is detected using a laser interferometer. The laser interferometer is used to non-contactly and with high precision measure the height change of the surface of a moving target object (e.g., a wafer) relative to an ideal reference plane (i.e., the surface of the air-bearing motion stage).

[0073] Specifically, once the target object moves to a predetermined position, it sends a trigger signal to the laser interferometer. Upon receiving the signal, the laser interferometer performs a "snapshot" measurement within an extremely short time (microseconds), capturing the interferogram at the current position. Due to the extremely short exposure time, it "freezes" the motion, effectively overcoming the effects of vibration and motion blur. This technique is called dynamic interferometry. A phase-shifting algorithm is used to calculate the precise height of the interferogram captured at each point relative to an ideal reference plane. The target object continues to move to the next point, repeating the above process until the entire target object surface is scanned, obtaining the height of the entire target object surface. Optionally, the highest height of the target object surface is subtracted from the lowest height to obtain the unevenness.

[0074] In this application, the surface of the target object is absolutely flat, while the unevenness reflects the slight amplitude caused by the uneven air pressure during the movement of the target object.

[0075] S180. The sum of the unevenness and the difference is used as the error and input into the proportional-integral-derivative control equation to obtain the air pressure control quantity of the throttle orifice region, thereby realizing the motion compensation of the target object.

[0076] The error e is obtained by directly summing the unevenness and difference detected at time t, or by weighted summation. The corresponding PID control equation is selected based on the anomaly category; different PID control equations have different proportional gain, derivative gain, and integral gain. A typical PID control equation is as follows:

[0077]

[0078] Where u(t) is the air pressure control quantity at time t, and the same air pressure control quantity is shared within a throttle orifice region. K p It is the proportional gain, K i It is the integral gain, K d It is the differential gain. e(t) is the error at time t.

[0079] By controlling the air supply pressure in the orifice region using air pressure control, the air supply pressure changes to compensate for minor vibrations of the target object. At the next time step t+1, the process of tracking the target object's movement along the air-bearing platform is repeated. The current temperature, current air supply pressure in the orifice region, and current load current are monitored to obtain the predicted trajectory at time t+1. The difference between the predicted trajectory and the reference trajectory is calculated. The unevenness of the target object in the orifice region at time t+1 is detected, and the sum of the unevenness and the difference is used as the error, which is input into the proportional-integral-derivative (PID) control equation to obtain the air pressure control quantity in the orifice region at time t+1. This achieves real-time compensation for the target object's motion.

[0080] Due to various factors such as air supply pressure, load, and temperature, the air supply pressure can become unstable, causing instability in the movement of the target object on the air-bearing motion platform, requiring real-time compensation. This application pre-trains a prediction model by extracting data from experiments on the movement of a spherical object on the air-bearing motion platform. This model can then predict the trajectory of the spherical object based on the current temperature, the current air supply pressure in the orifice region, and the current load current. The difference between the predicted trajectory and the reference trajectory, as well as the unevenness of the target object moving in the orifice region, are then calculated. The air pressure control quantity is generated from both the air supply pressure angle of the orifice (microscopic) and the overall unevenness of the target object (macroscopic). This application uses a proportional-integral-derivative control equation to adjust the air pressure control quantity, thus eliminating the need to consider the physical relationship between the air supply pressure and the object's motion, making it simple and effective for industrial implementation.

[0081] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired means, such as coaxial cable, optical fiber, digital subscriber line (DSL), or wireless means, such as infrared, wireless, microwave, etc. The computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium, or a semiconductor medium, etc. It is worth noting that the computer-readable storage medium mentioned in the embodiments of this application can be a non-volatile storage medium, in other words, it can be a non-transient storage medium.

[0082] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application can be achieved, and this is not limited herein.

[0083] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for real-time error compensation of a precision air-bearing motion stage, characterized in that, include: Control multiple identical spherical objects to move multiple times along the throttle orifice of the air-floating motion platform; During the movement, the three-dimensional motion trajectory of the spherical object is collected; The three-dimensional motion trajectory is compared with the reference trajectory to identify abnormal trajectories and the throttle orifice regions corresponding to the abnormal trajectories. The abnormal trajectory in the orifice region is used as a label, and the temperature, air supply pressure in the orifice region, and load current during the movement are used as inputs to train the prediction model. During the movement of the target object along the air-bearing motion platform, the current temperature, the current air supply pressure in the throttling orifice region, and the current load current are detected; the detected current temperature, the current air supply pressure in the throttling orifice region, and the current load current are input into the prediction model to obtain the predicted trajectory; the predicted trajectory is the motion trajectory of the spherical object assuming it is moving in the throttling orifice region. Calculate the difference between the predicted trajectory and the baseline trajectory; Detect the unevenness of the target object as it moves through the orifice region; The sum of the unevenness and the difference is used as the error and input into the proportional-integral-derivative control equation to obtain the air pressure control quantity of the orifice region, thereby realizing the motion compensation of the target object. A light-emitting point is provided at the center of the spherical object; During the motion, the three-dimensional motion trajectory of the spherical object is acquired, including: The movement of the light-emitting point is captured by multiple infrared cameras, and the three-dimensional trajectory of the light-emitting point is calculated by combining stereo vision and triangulation principles.

2. The real-time error compensation method for a precision air-bearing motion stage according to claim 1, characterized in that, Controlling multiple identical spherical objects to move repeatedly along the throttle orifice of an air-floating motion platform includes: When a spherical object is controlled to move multiple times along the target column or the target row or the target column ... The target column throttling orifice can be 1 column, 2 columns, or 3 columns; the target row throttling orifice can be 1 row, 2 rows, or 3 rows.

3. The real-time error compensation method for a precision air-bearing motion stage according to claim 2, characterized in that, The three-dimensional motion trajectory is compared with the reference trajectory to identify abnormal trajectories and the corresponding throttle orifice regions, including: The three-dimensional motion trajectories corresponding to all target column throttle orifices and the three-dimensional motion trajectories corresponding to all target row throttle orifices are compared with the baseline trajectory to identify abnormal trajectories. The row and column intersection area corresponding to the abnormal trajectory is taken as the throttle orifice area corresponding to the abnormal trajectory.

4. The real-time error compensation method for a precision air-bearing motion stage according to claim 1, characterized in that, After inputting the detected current temperature, current supply gas pressure in the orifice region, and current load current into the prediction model to obtain the predicted trajectory, the following steps are also included: The predicted trajectory is classified to obtain anomaly categories; the anomaly categories include low air pressure, high air pressure, and unstable air pressure. The anomaly categories have a one-to-one correspondence with the proportional-integral-derivative control equations.

5. The real-time error compensation method for a precision air-bearing motion stage according to claim 1, characterized in that, The sum of the roughness and the difference is used as the error and input into the proportional-integral-derivative control equation. After obtaining the air pressure control quantity of the orifice region, the following steps are also included: The air pressure in the orifice region is controlled using the aforementioned air pressure control quantity; Return to the operation of detecting the current temperature, current air supply pressure in the throttling orifice area, and current load current during the movement of the target object along the air-bearing motion platform.

6. The real-time error compensation method for a precision air-bearing motion stage according to claim 1, characterized in that, Detecting the unevenness of the target object as it moves through the orifice region includes: The unevenness of the target object as it moves in the orifice region is detected using a laser interferometer.

7. The real-time error compensation method for a precision air-bearing motion stage according to any one of claims 1-6, characterized in that, The prediction model is a long short-term memory network.

Images