A polarization state decoupling laser collimation reference correction method

By setting a polarization separation element in the laser collimation optical path, the reference and polarization components are detected and measured separately, and the coordinate difference is calculated and corrected. This solves the reference point positioning error caused by polarization coupling and improves the reference accuracy and stability of the laser collimation system.

CN123065385APending Publication Date: 2026-11-10迅衡(天津)光电科技有限公司
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202611555327.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-09
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing technologies cannot effectively solve the problem of systematic deviation of the reference point coordinates in laser collimation systems caused by polarization component coupling, and existing methods have failed to identify and correct the reference point positioning error caused by polarization leakage.

Method used

By setting a polarization state separation element in the laser collimation optical path, the reference polarization component and the measurement polarization component are spatially separated and guided to independent photodetectors, the coordinate difference is calculated and the reference is corrected, thus eliminating the influence of polarization coupling.

Benefits of technology

It significantly improves the reference accuracy of the laser collimation system, eliminates systematic deviations caused by polarization leakage, and enhances the stability and accuracy of reference point positioning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123065385A_ABST
    Figure CN123065385A_ABST
Patent Text Reader

Abstract

The application provides a polarization state decoupling laser collimation reference correction method, relates to the technical field of collimation measurement, and comprises the following steps: a polarization state separation element is arranged in a laser collimation light path, so that a reference polarization component transmitted by a reference light path and a measurement polarization component transmitted by a reflection light path are separated in space to obtain a reference polarization beam and a measurement polarization beam which are independent of each other; the reference polarization beam and the measurement polarization beam are guided to a first photoelectric detector and a second photoelectric detector which are independent of each other respectively, reference spot images and measurement spot images are obtained; reference point coordinates and measurement point coordinates are calculated respectively; a coordinate difference value of the two is calculated, and the coordinate difference value is determined as a reference point positioning systematic deviation amount containing a polarization coupling effect; and the collimation reference is corrected based on the deviation amount, so that the influence of polarization component coupling on reference point positioning is eliminated. The method avoids the centroid deviation of a synthesized spot caused by the superposition of a polarization leakage component and a main component from the source, so that the reference precision of a laser collimation system is significantly improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of collimation measurement technology, specifically relating to a laser collimation reference correction method based on polarization state decoupling. Background Technology

[0002] In the field of laser collimation measurement, laser collimation systems typically employ polarization beam splitters to separate the reference beam from the reflected beam, thereby acquiring the reference beam and the measurement beam separately. However, the extinction ratio of polarization beam splitters is not infinitely large, and actual devices exhibit polarization leakage: the measurement polarization component in the reflected beam partially leaks into the reference beam, and the reference polarization component in the reference beam partially leaks into the reflected beam. The leaked polarization component and the principal polarization component in the corresponding beam are superimposed on the photodetector, forming a composite spot. Because the polarization directions of the leaked component and the principal component are orthogonal, and their spot positions on the detector differ, the centroid of the composite spot deviates from the centroid position when the principal component acts alone, resulting in a systematic deviation in the coordinates of the reference point calculated based on the composite spot. This deviation is not random noise, but a systematic error that varies with the polarization leakage ratio and the relative position of the spot, directly affecting the reference accuracy of the laser collimation system.

[0003] In the prior art, invention patent CN102176087A proposes a two-dimensional photoelectric self-collimation method and device for compensating for the common optical path of a polarized light plane mirror reference. This method utilizes a polarizing beam splitter to separate a reference beam, and a controller dynamically controls a two-dimensional beam deflection device based on the drift reflected by the reference beam to compensate for beam angular drift during measurement. However, the technical logic of this scheme is: separating the reference beam → monitoring dynamic drift → closed-loop control of beam deflection. Its compensation target is the angular drift coupled within the beam during propagation, failing to identify the systematic deviation of the reference point coordinates caused by polarization leakage, and also failing to correct the reference point coordinates. Therefore, the prior art cannot solve the problem of systematic deviation in reference point positioning caused by polarization component coupling. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, a laser collimation reference correction method for polarization state decoupling is provided, comprising: Polarization state decoupling: A polarization state separation element is set in the laser collimation optical path to spatially separate the reference polarization component transmitted by the reference optical path from the measurement polarization component transmitted by the reflected optical path, resulting in mutually independent reference polarization beams and measurement polarization beams; the reference polarization component and the measurement polarization component are in a mutually coupled state before separation; Independent detection: The reference polarized beam and the measurement polarized beam are respectively guided to a first photodetector and a second photodetector that are independent of each other, so as to obtain the reference spot image and the measurement spot image respectively; Reference coordinate calculation: Calculate the coordinates of the reference point corresponding to the reference polarization component based on the reference spot image, and calculate the coordinates of the measurement point corresponding to the measurement polarization component based on the measurement spot image; Systematic deviation determination: Calculate the coordinate difference between the coordinates of the reference point and the coordinates of the measurement point, and determine the coordinate difference as the systematic deviation of the reference point positioning including the polarization coupling effect; Reference correction: Based on the systematic deviation, the alignment reference is corrected to eliminate the influence of the mutual coupling between the polarization component of the reflected optical path and the polarization component of the reference optical path on the positioning of the reference point.

[0005] According to the technical solution provided in this application, the determination of the systematic deviation also includes decomposing the coordinate difference into polarization coupling contribution. The first light intensity value sequence obtained by the reference polarized beam through the first photodetector and the second light intensity value sequence obtained by the measurement polarized beam through the second photodetector are acquired. Based on the first light intensity value sequence and the second light intensity value sequence, calculate the proportion of the residual reference polarization component in the measured polarization beam and the proportion of the residual measured polarization component in the reference polarization beam. The residual reference polarization component ratio and the residual measurement polarization component ratio are respectively correlated with the coordinate difference to obtain the first deviation component introduced by polarization component crosstalk and the second deviation component introduced by optical path geometric asymmetry. The square root of the sum of the squares of the first deviation component and the second deviation component is taken as the systematic deviation.

[0006] According to the technical solution provided in this application, the correction of the alignment reference includes: Based on the systematic deviation, a compensation control signal is generated for driving the polarization rotation element; The compensation control signal is applied to a polarization rotation element coupled to the polarization state separation element. The polarization rotation element is configured to adjust the polarization state angle between the reference polarization component and the measured polarization component before separation in response to the compensation control signal, thereby increasing the spatial separation between the reference polarization component and the measured polarization component after separation, thereby suppressing the generation of the systematic deviation at the optical path level.

[0007] According to the technical solution provided in this application, the benchmark correction further includes time-domain stability determination and correction mode switching for the systematic deviation: Obtain the numerical sequence of the systematic deviation over M consecutive measurement periods, and calculate the time-domain standard deviation and time-domain mean of the numerical sequence; If the time-domain standard deviation is less than the first stability threshold, then the time-domain mean is used as the steady-state systematic deviation, and the collimation reference is periodically corrected with the first correction period. If the time-domain standard deviation is greater than or equal to the first stability threshold, the systematic deviation is determined to be in a non-steady-state fluctuation state. The correction period is shortened to a second correction period, which is less than the first correction period. In each measurement period, the instantaneous systematic deviation obtained in the current measurement period is used to correct the alignment reference in real time.

[0008] According to the technical solution provided in this application, the systematic deviation is taken as the norm of the coordinate difference; in the polarization coupling contribution decomposition, the determination methods of the first deviation component and the second deviation component include: A polarization coupling contribution matrix is ​​constructed, wherein each row of the polarization coupling contribution matrix corresponds to a measurement period, and the columns include the first light intensity value, the second light intensity value, and the components of the coordinate difference in the two-dimensional plane along two mutually perpendicular coordinate axes. The polarization coupling contribution matrix is ​​decomposed to obtain the polarization component crosstalk contribution direction and the optical path geometric asymmetry contribution direction in the coordinate difference component space. The polarization component crosstalk contribution direction and the optical path geometric asymmetry contribution direction are orthogonal to each other. The coordinate difference is projected onto the crosstalk contribution direction of the polarization component and the optical path geometric asymmetry contribution direction, respectively, to obtain the first deviation component and the second deviation component.

[0009] According to the technical solution provided in this application, the matrix decomposition is a singular value decomposition, and the polarization component crosstalk contribution direction and the optical path geometric asymmetry contribution direction are determined by the following methods: The polarization coupling contribution matrix is ​​subjected to singular value decomposition to obtain a singular value diagonal matrix and a right singular vector matrix; The direction of the right singular vector in the right singular vector matrix whose singular value is greater than a preset singular value threshold in the coordinate difference component space is taken as the polarization component crosstalk contribution direction; the direction of the right singular vector whose singular value is less than or equal to the preset singular value threshold in the coordinate difference component space is taken as the optical path geometric asymmetry contribution direction.

[0010] According to the technical solution provided in this application, the method further includes, after the polarization coupling contribution decomposition, adaptively adjusting the extinction ratio configuration of the polarization state separation element based on the proportional relationship between the first deviation component and the second deviation component: The ratio of the absolute value of the first deviation component to the systematic deviation is calculated and used as the polarization crosstalk contribution weight. If the polarization crosstalk contribution weight is greater than the first weight threshold, an extinction ratio enhancement command is generated. The extinction ratio enhancement command is used to drive the polarization state pre-alignment element disposed on the incident side of the polarization state separation element, so that the angle between the polarization direction of the reference polarization component and the transmission axis of the polarization state separation element is reduced, and the angle between the polarization direction of the measured polarization component and the reflection axis of the polarization state separation element is reduced. If the polarization crosstalk contribution weight is less than or equal to the first weight threshold, the current extinction ratio configuration is maintained.

[0011] According to the technical solution provided in this application, the polarization rotation angle of the polarization state pre-alignment element is determined in the following way: Obtain the current extinction ratio of the polarization state separation element and the polarization crosstalk contribution weight; The target extinction ratio is determined based on the polarization crosstalk contribution weight. The difference between the target extinction ratio and the current extinction ratio is calculated. The preset extinction ratio-polarization rotation angle mapping table is consulted to obtain the polarization rotation angle required for the polarization state pre-alignment element.

[0012] According to the technical solution provided in this application, the adjustment of the polarization state angle by the polarization rotation element and the correction of the collimation reference form a coordinated control, specifically including: Obtain the sequence of changes in the systematic deviation of the polarization rotation element over P consecutive measurement periods after adjusting the polarization angle. Based on the sequence of changes, determine whether the systematic deviation converges as the polarization angle is adjusted; If the systematic deviation converges, the current adjustment state of the polarization rotation element is maintained, and the converged systematic deviation is taken as the residual deviation. The correction period for the collimation reference is determined based on the time-domain standard deviation of the residual deviation. If the systematic deviation does not converge or shows a divergent trend, it is determined that there is an uncontrollable component introduced by the optical path geometric asymmetry in the systematic deviation. An optical path geometric calibration control signal is generated and output to drive the displacement adjustment mechanism, which is located below the first photodetector and / or the second photodetector.

[0013] According to the technical solution provided in this application, determining the correction period for the collimation reference based on the time-domain standard deviation of the residual deviation includes: If the time-domain standard deviation is less than the first stability threshold, the collimation reference is periodically corrected with a first correction period. If the time-domain standard deviation is greater than or equal to the first stability threshold, the collimation reference is corrected in real time with a second correction period that is less than the first correction period.

[0014] Compared with the prior art, the beneficial effects of this application are as follows: First, by decoupling polarization states, a polarization separation element is set in the laser collimation optical path to spatially separate the reference polarization component from the measurement polarization component, resulting in independent reference and measurement polarization beams. Consequently, the reference and measurement polarization beams are guided to independent first and second photodetectors, respectively, preventing the leakage polarization component from superimposing with the main polarization component on the detector to form a composite spot. This eliminates the influence of the composite spot's centroid shift on the calculation of the reference point coordinates at the source.

[0015] Second, based on the reference spot image and the measured spot image obtained by independent detection, the coordinates of the reference point and the measured point are calculated respectively. Then, the coordinate difference between the two is calculated, and this coordinate difference is determined as the systematic deviation of the reference point positioning including the polarization coupling effect. This systematic deviation can accurately reflect the degree of influence of polarization component coupling on the reference point positioning, providing a clear quantitative basis for subsequent corrections.

[0016] Third, the alignment reference is corrected based on the systematic deviation to eliminate the influence of the coupling between the polarization components of the reflected light path and the polarization components of the reference light path on the positioning of the reference point. Through this correction, the coordinates of the reference point of the laser collimation system are no longer affected by the systematic deviation caused by polarization leakage, and the reference accuracy is significantly improved. Moreover, this correction does not rely on physical modification of the extinction ratio of the polarization beam splitter, and is engineering feasible. Attached Figure Description

[0017] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 A flowchart illustrating the steps of the laser collimation reference correction method for polarization state decoupling provided in this application; Figure 2 The optical path diagram in the laser collimation reference correction method for polarization state decoupling provided in this application; The text labels in the image represent: 1. Laser source; 2. Polarization pre-alignment element; 3. Polarization separation element; 4. First photodetector; 5. Signal acquisition and processing unit; 6. Second photodetector; 7. Displacement adjustment mechanism. Detailed Implementation

[0018] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] As mentioned in the background section, this application proposes a method for correcting a laser collimation reference by decoupling polarization states, such as... Figure 1-2 As shown, it includes: S1. Polarization state decoupling: A polarization state separation element 3 is set in the laser collimation optical path to separate the reference polarization component transmitted by the reference optical path from the measurement polarization component transmitted by the reflected optical path in space, so as to obtain a reference polarization beam and a measurement polarization beam that are independent of each other; the reference polarization component and the measurement polarization component are in a mutually coupled state before separation; S2. Independent detection: The reference polarized beam and the measurement polarized beam are respectively guided to the first photodetector 4 and the second photodetector 6, which are independent of each other, to obtain the reference spot image and the measurement spot image respectively; S3. Reference coordinate calculation: Calculate the reference point coordinates corresponding to the reference polarization component based on the reference spot image, and calculate the measurement point coordinates corresponding to the measurement polarization component based on the measurement spot image; S4. Determination of systematic deviation: Calculate the coordinate difference between the coordinates of the reference point and the coordinates of the measurement point, and determine the coordinate difference as the systematic deviation of the reference point positioning including the polarization coupling effect; S5. Reference Correction: Based on the systematic deviation, the alignment reference is corrected to eliminate the influence of the mutual coupling between the polarization component of the reflected optical path and the polarization component of the reference optical path on the positioning of the reference point.

[0021] Specifically, this embodiment is applicable to optical measurement equipment that requires the establishment of a highly stable collimation reference, such as laser collimators, laser trackers, and photoelectric autocollimators. The laser collimation optical path typically includes a laser source 1, a collimating objective lens, a reference beam splitter, a reflective target, and a photoelectric detection unit. The reference optical path refers to the optical path from the laser source 1 through the reference beam splitter directly to the detection unit; its polarization component is called the reference polarization component. The reflected optical path refers to the optical path of the laser beam returning from the reflective target to the detection unit; its polarization component is called the measurement polarization component. Without the polarization separation element 3, the reference polarization component and the measurement polarization component are superimposed and transmitted in the same optical path. Their polarization directions are usually orthogonal or approximately orthogonal, but due to the limited extinction ratio of the polarization beam splitter, they are not completely separated in space, forming a mutually coupled state. This mutually coupled state means that the reference polarization component and the measurement polarization component coexist in the same spatial optical path before separation. Ordinary photoelectric detectors cannot distinguish between them solely based on light intensity distribution, causing the centroid of the synthesized spot to deviate from the centroid position when the principal component acts alone.

[0022] The polarization state separation element 3 can be any of the following: a polarization beam splitter prism, a Glan Taylor prism, a Wollaston prism, a thin-film polarization beam splitter, a birefringent crystal polarization beam splitter, or a polarization grating. The function of the polarization state separation element 3 is to spatially split the incident beam into two beams based on the polarization direction. One beam corresponds to the reference polarization component, and the other corresponds to the measurement polarization component. Taking a polarization beam splitter prism as an example, its transmission axis allows the reference polarization component to pass through and be directed towards the first photodetector 4, while its reflection axis reflects the measurement polarization component to the second photodetector 6. The higher the extinction ratio of the polarization state separation element 3, the fewer residual cross-polarization components remain in the two beams after separation, resulting in higher spatial separation.

[0023] In the independent detection step, the first photodetector 4 and the second photodetector 6 can be charge-coupled devices, complementary metal-oxide-semiconductor image sensors, four-quadrant photoelectric position sensors, position-sensitive detectors, or photodiode arrays. The first photodetector 4 and the second photodetector 6 are independent, meaning they receive different light beams; there is no situation where the same detector simultaneously receives a reference polarized beam and a measurement polarized beam. Focusing lenses can be respectively placed in front of the first photodetector 4 and the second photodetector 6 to focus the corresponding beams onto the photosensitive surface of the detector. The reference spot image is the intensity distribution image of the reference polarized beam acquired by the first photodetector 4. The measurement spot image is the intensity distribution image of the measurement polarized beam acquired by the second photodetector 6, and is transmitted to the signal acquisition and processing unit 5. During image acquisition, noise reduction, background subtraction, and threshold segmentation can be performed to improve the accuracy of spot center extraction. The reference spot image, also acquired by the first photodetector 4, is also transmitted to the signal acquisition and processing unit 5. The first photodetector 4 and the second photodetector 6 are connected to the signal acquisition and processing unit 5 via data lines. The signal acquisition and processing unit 5 synchronously acquires the two light spot images. The synchronization method adopts hardware trigger synchronization or software timestamp alignment to ensure that the coordinates of the reference point and the coordinates of the measurement point correspond within the same measurement cycle.

[0024] In the reference coordinate calculation step, the reference point coordinates refer to the two-dimensional coordinates of the center of the spot in the reference spot image, and the measurement point coordinates refer to the two-dimensional coordinates of the center of the spot in the measurement spot image. Coordinate calculation methods can include the gray-level weighted centroid method, Gaussian surface fitting method, ellipse fitting method, or four-quadrant difference method. Taking the gray-level weighted centroid method as an example, the gray-level value of each pixel in the spot image is weighted, with the weight being the gray-level value of that pixel, and the weighted average coordinates are calculated as the center of the spot. The reference point coordinates and the measurement point coordinates can be represented in the same coordinate system, for example, a two-dimensional rectangular coordinate system can be established based on the photosensitive surfaces of the first photodetector 4 and the second photodetector 6.

[0025] In the systematic deviation determination step, the coordinate difference between the reference point coordinates and the measured point coordinates is calculated. This coordinate difference can be represented as a two-dimensional vector, including the difference along the first coordinate axis and the difference along the second coordinate axis. This coordinate difference includes the polarization coupling effect; that is, due to the limited extinction ratio of the polarization beam splitter, the residual measured polarization component enters the reference optical path, and the residual reference polarization component enters the reflected optical path. The superposition of these two components with their respective principal components causes a shift in the centroid of the synthesized light spot, thus generating a systematic deviation between the reference point coordinates and the measured point coordinates. This deviation is not random noise, but a deterministic error that varies with the polarization leakage ratio and the relative position of the light spot. Determining this coordinate difference as the systematic deviation of the reference point positioning, including the polarization coupling effect, provides a quantitative basis for subsequent corrections.

[0026] In the reference correction step, the collimation reference refers to the reference axis, reference point, or reference coordinate system established by the laser collimation system. The correction method can be either subtracting the systematic deviation from the coordinates of the measurement point obtained in subsequent measurement cycles, or adjusting the reference point coordinates based on the systematic deviation. The purpose of the correction is to eliminate the influence of the coupling between the polarization components of the reflected light path and the polarization components of the reference light path on the positioning of the reference point. Let the coordinates of the reference point be (xb, yb), and the coordinates of the measurement point be (xm, ym). The systematic deviation (Δx, Δy) is defined as the reference point coordinates minus the measurement point coordinates, i.e., Δx = xb. xm, Δy=yb Taking the subtraction method as an example, the corrected coordinates of the measurement point are (xm+Δx, ym+Δy). Taking the adjustment of the reference point coordinates as an example, if the original reference point coordinates are (x0, y0), then the adjusted reference point coordinates are (x0+Δx, y0+Δy).

[0027] The hardware structure of the signal acquisition and processing unit 5 includes: an analog-to-digital converter (ADC) for converting the analog image signals output by the first photodetector 4 and the second photodetector 6 into digital signals; a field-programmable gate array (FPGA) or digital signal processor (DSP) for performing reference point coordinate calculation, measurement point coordinate calculation, coordinate difference calculation, and correction calculation; a digital-to-analog converter (DAC) for converting digital control quantities into analog drive signals; and a communication interface for outputting corrected measurement results or receiving external control commands. The signal acquisition and processing unit 5 is connected to the first photodetector 4 and the second photodetector 6 via data lines, and to the polarization pre-alignment element 2 and the displacement adjustment mechanism 7 via control lines.

[0028] The technical principle lies in the fact that polarization leakage causes the centroid of the synthesized spot to shift. By decoupling the polarization state, the reference polarization component and the measurement polarization component are detected separately, preventing the formation of the synthesized spot. Then, the residual coupling effect is quantified and corrected by coordinate difference, thereby improving the positioning accuracy of the reference point. The technical effect is to significantly reduce the systematic deviation caused by polarization coupling and improve the reference stability and measurement accuracy of the laser collimation system.

[0029] In a preferred embodiment, the determination of the systematic deviation further includes decomposing the coordinate difference into polarization coupling contribution. The first light intensity value sequence obtained by the reference polarized beam through the first photodetector 4 and the second light intensity value sequence obtained by the measurement polarized beam through the second photodetector 6 are acquired. Based on the first light intensity value sequence and the second light intensity value sequence, calculate the proportion of the residual reference polarization component in the measured polarization beam and the proportion of the residual measured polarization component in the reference polarization beam. The residual reference polarization component ratio and the residual measurement polarization component ratio are respectively correlated with the coordinate difference to obtain the first deviation component introduced by polarization component crosstalk and the second deviation component introduced by optical path geometric asymmetry. The square root of the sum of the squares of the first deviation component and the second deviation component is taken as the systematic deviation.

[0030] Specifically, polarization coupling contribution decomposition refers to the process of separating and quantifying the deviations caused by polarization component crosstalk and optical path geometric asymmetry in the coordinate difference. The first light intensity sequence refers to the total light intensity value collected by the first photodetector 4 over multiple consecutive measurement periods. This total light intensity value includes the principal light intensity of the reference polarization component and the light intensity of the residual measured polarization component. The second light intensity sequence refers to the total light intensity value collected by the second photodetector 6 over multiple consecutive measurement periods. This total light intensity value includes the principal light intensity of the measured polarization component and the light intensity of the residual reference polarization component. The first and second light intensity sequences can be synchronously acquired through an analog-to-digital converter, with the sampling period consistent with the measurement period.

[0031] The residual reference polarization component ratio refers to the proportion of the reference polarization component mixed in with the measurement polarization beam to the total light intensity of the measurement polarization beam. The residual ratio can be calculated using the analyzer rotation method. Taking the analyzer before the first photodetector 4 as an example, the analyzer is rotated and the maximum light intensity I is recorded. max and minimum value I min Let the intensity of the principal polarization component be I. main The intensity of the residual polarization component is I res The polarization directions of the two components are orthogonal. When the transmission axis of the analyzer is aligned with the polarization direction of the principal polarization component, the light intensity reaches its maximum value, I. max =I main When the analyzer's transmission axis is perpendicular to the polarization direction of the principal polarization component, the light intensity reaches its minimum value, I. min =I res Therefore, the residual proportion R can be calculated using the following formula: This formula provides a clear calculation path from the extreme value of light intensity to the residual ratio. For the first photodetector 4, the main polarization component is the reference polarization component, and the residual is the measured polarization component. The resulting residual ratio is denoted as R. m For the analyzer before the second photodetector 6, the proportion of the residual reference polarization component is calculated using the same method and denoted as R. b .

[0032] Correlation operation refers to establishing mathematical relationships between the residual reference polarization component ratio and the residual measurement polarization component ratio, respectively, and the coordinate difference. Let the coordinate difference in the i-th measurement cycle be Δi, and the residual ratio vector be R. i =[R b,i ,R m,i ] T , where R b,i R represents the proportion of the residual reference polarization component. m,i Let G be the proportion of the residual measured polarization components. i Its components include the relative positional offset of the first photodetector 4 and the second photodetector 6, the focusing objective lens distortion coefficient, and the detector responsivity difference. A linear model is established: Δi = AR i +BG i +ε i Where A is the polarization crosstalk coefficient matrix, B is the geometric asymmetry coefficient matrix, and ε i Let A and B be the residuals. Using data from N measurement periods, solve for A and B using the least squares method. Let... If C = [A, B], then the model can be written as Δi = CZ. i +ε i The least squares solution is: C=EZ T (ZZ T ) 1 Where E is the coordinate difference matrix and Z is the combination matrix. After solving for A and B, the first deviation component Δ pol,i =AR i The second deviation component Δ geo,i =BG i The first deviation component is the portion of the coordinate difference caused by polarization crosstalk, and the second deviation component is the portion caused by optical path geometric asymmetry. Since the directions of influence of polarization crosstalk and geometric asymmetry on the coordinate difference are usually orthogonal or approximately orthogonal, the square root of the sum of the squares of the first and second deviation components is taken as the systematic deviation. Systematic deviation =

[0033] This calculation method is equivalent to taking the Euclidean norm of the coordinate difference, which can comprehensively characterize the total magnitude of the systematic deviation. The technical principle lies in the fact that polarization crosstalk and geometric asymmetry have different physical mechanisms and directional characteristics in their effects on the spot centroid. These two can be decoupled through multi-period intensity sequences and coordinate difference sequences. The technical effect is that it can clearly identify the main sources of systematic deviation, providing a basis for subsequent targeted corrections. If polarization crosstalk contributes significantly, the extinction ratio configuration of polarization state separation element 3 can be adjusted; if geometric asymmetry contributes significantly, the detector position or the calibrated optical path can be adjusted. This decomposition improves the targeting and efficiency of the correction, avoiding blind adjustments to optical components.

[0034] In a preferred embodiment, the correction of the alignment reference includes: Based on the systematic deviation, a compensation control signal is generated for driving the polarization rotation element; The compensation control signal is applied to the polarization rotation element coupled to the polarization state separation element 3. The polarization rotation element is configured to adjust the polarization state angle between the reference polarization component and the measured polarization component before separation in response to the compensation control signal, so as to increase the spatial separation between the reference polarization component and the measured polarization component after separation, thereby suppressing the generation of the systematic deviation at the optical path level.

[0035] Specifically, a polarization rotation element is a device capable of changing the polarization direction of an incident beam in response to a control signal. It can be a liquid crystal polarization rotator, a Faraday rotator, a combination of a half-wave plate and an electric rotary stage, an electro-optic crystal polarization modulator, or a magneto-optic polarization rotator. The polarization rotation element is positioned on the incident side of the polarization state separation element 3 and coupled to it. Coupling means that the emitted beam from the polarization rotation element enters the polarization state separation element 3, and the polarization rotation effect of the element changes the polarization direction of both the reference polarization component and the measured polarization component upon reaching the polarization state separation element 3.

[0036] The compensation control signal is a signal generated based on the systematic deviation to drive the polarization rotation element. The compensation control signal can be an analog voltage signal, an analog current signal, a digital pulse width modulation signal, or a digital bus command. Proportional-integral-derivative (PID) control can be used to generate the compensation control signal. For example, the systematic deviation is used as the control error, and the control quantity is calculated by the PID controller, which is then converted into the driving voltage for the polarization rotation element. If the systematic deviation exceeds a preset threshold, a compensation control signal is generated; if the systematic deviation is less than or equal to the preset threshold, the current polarization state can be maintained.

[0037] The polarization state angle refers to the angle between the polarization directions of the reference polarization component and the measurement polarization component before separation. Ideally, the polarization directions of the reference polarization component and the measurement polarization component are orthogonal to each other, with an angle of 90 degrees. The transmission axis and reflection axis of the polarization state separation element 3 are also orthogonal to each other. When the angle between the polarization direction of the reference polarization component and the transmission axis decreases, the reference polarization component outputs more from the transmission direction, and the energy leaking into the reflection direction decreases. When the angle between the polarization direction of the measurement polarization component and the reflection axis decreases, the measurement polarization component outputs more from the reflection direction, and the energy leaking into the transmission direction decreases. Therefore, the polarization rotation element adjusts the polarization state angle in response to the compensation control signal, thereby increasing the spatial separation degree of the reference polarization component and the measurement polarization component after separation. Spatial separation degree refers to the degree of spatial isolation between the two beams after separation, which can be characterized by extinction ratio, crosstalk power ratio, or beam spacing.

[0038] The compensation control signal is generated by the signal acquisition and processing unit 5. Specifically, the signal acquisition and processing unit 5 takes the systematic deviation as input, calculates the control quantity through a proportional-integral-derivative (PID) controller, and then converts the control quantity into the driving voltage of the polarization pre-alignment element 2. The proportional, integral, and derivative coefficients of the PID controller are obtained through experimental calibration. The calibration method is as follows: a step signal is applied to the driving voltage terminal of the polarization pre-alignment element 2, the response curve of the systematic deviation is recorded, and the three coefficients are adjusted according to the overshoot, rise time, and steady-state error of the response curve until the systematic deviation can quickly converge to near zero after the step response. If the absolute value of the systematic deviation exceeds a preset dead-zone threshold, the signal acquisition and processing unit 5 outputs the compensation control signal; if it does not exceed the dead-zone threshold, it outputs a zero signal to maintain the current polarization state. The compensation control signal is transmitted to the driving terminal of the polarization pre-alignment element 2 through a control line.

[0039] The technical principle lies in the fact that the extinction ratio of a polarization beam splitter is related to the angle between the incident polarization direction and the transmission or reflection axis. When the incident polarization direction is perfectly aligned with the corresponding axis, the polarization component is almost completely transmitted or completely reflected, minimizing cross-leakage. When an angle exists, some energy leaks into the other path, creating crosstalk. By actively adjusting the polarization angle using a polarization rotation element, leakage can be reduced at the optical path level, thereby suppressing the generation of systematic deviations. The technical effect is not only to correct existing systematic deviations but also to reduce polarization coupling effects at the source, improving long-term stability and environmental adaptability. Compared to methods relying solely on electrical corrections, optical path-level feedforward suppression offers faster response speeds and better long-term stability, making it particularly suitable for scenarios where polarization state drift occurs under conditions such as temperature changes and vibration.

[0040] In a preferred embodiment, the benchmark correction further includes determining the time-domain stability of the systematic deviation and switching the correction mode: Obtain the numerical sequence of the systematic deviation over M consecutive measurement periods, and calculate the time-domain standard deviation and time-domain mean of the numerical sequence; If the time-domain standard deviation is less than the first stability threshold, then the time-domain mean is used as the steady-state systematic deviation, and the collimation reference is periodically corrected with the first correction period. If the time-domain standard deviation is greater than or equal to the first stability threshold, the systematic deviation is determined to be in a non-steady-state fluctuation state. The correction period is shortened to a second correction period, which is less than the first correction period. In each measurement period, the instantaneous systematic deviation obtained in the current measurement period is used to correct the alignment reference in real time.

[0041] Specifically, time-domain stability determination refers to judging whether a system is in a steady state or a non-steady state based on the degree of fluctuation of the systematic deviation over multiple consecutive measurement periods. M in the M consecutive measurement periods is a positive integer greater than or equal to 3, typically ranging from 10 to 100. The measurement period refers to the time interval between the laser collimation system completing one acquisition of the coordinates of the reference point and the measurement point, and calculating the systematic deviation.

[0042] After obtaining the numerical sequence of systematic deviation over M consecutive measurement periods, the time-domain mean and time-domain standard deviation of the sequence are calculated. The time-domain mean is calculated by summing all values ​​in the sequence and dividing by the number of values ​​M to obtain the arithmetic mean. The time-domain standard deviation is calculated by subtracting the time-domain mean from each value in the sequence to obtain M deviation values; squaring each deviation value to obtain M squared values; summing these M squared values ​​and dividing by M minus 1 to obtain the variance; and taking the square root of the variance to obtain the time-domain standard deviation. Those skilled in the art are familiar with the above methods for calculating the mean and standard deviation, and no further explanation is needed.

[0043] The first stability threshold is a constant set according to the accuracy requirements of the collimation reference. For example, when the accuracy requirement of the collimation reference is 0.1 micrometers, the first stability threshold can be 0.1 micrometers; when the accuracy requirement is 0.5 arcseconds, the first stability threshold can be 0.5 arcseconds. Alternatively, five percent to fifteen percent of the time-domain mean can be used as the first stability threshold. If the time-domain standard deviation is less than the first stability threshold, the systematic deviation is considered to be in a steady state. In this case, the time-domain mean is used as the steady-state systematic deviation, and the collimation reference is periodically corrected with a first correction period. The first correction period can be 10 to 100 measurement periods. Periodic correction means that at the end of each first correction period, the correction amount of the collimation reference is updated with the steady-state systematic deviation. For example, if the first correction period is 50 measurement periods, after every 50 measurements, the currently calculated steady-state systematic deviation is used as the correction amount and subtracted from the coordinates of subsequent measurement points.

[0044] If the time-domain standard deviation is greater than or equal to the first stability threshold, the systematic deviation is determined to be in a non-steady-state fluctuation state. In this case, the correction period is shortened to a second correction period, which is shorter than the first correction period, for example, 1 to 3 measurement periods. Within each measurement period, the collimation reference is corrected in real-time using the instantaneous systematic deviation obtained in the current measurement period. The instantaneous systematic deviation refers to the coordinate difference or the norm of the coordinate difference calculated in the current measurement period. Real-time correction ensures that the collimation reference always tracks the rapid changes in the systematic deviation. Because steady-state deviation changes slowly, periodic correction can meet the accuracy requirements and reduce the computational burden; non-steady-state deviation fluctuates rapidly, and real-time correction is necessary to suppress its influence.

[0045] In a preferred embodiment, the systematic deviation is taken as the norm of the coordinate difference; in the polarization coupling contribution decomposition, the determination of the first deviation component and the second deviation component includes: A polarization coupling contribution matrix is ​​constructed, wherein each row of the polarization coupling contribution matrix corresponds to a measurement period, and the columns include the first light intensity value, the second light intensity value, and the components of the coordinate difference in the two-dimensional plane along two mutually perpendicular coordinate axes. The polarization coupling contribution matrix is ​​decomposed to obtain the polarization component crosstalk contribution direction and the optical path geometric asymmetry contribution direction in the coordinate difference component space. The polarization component crosstalk contribution direction and the optical path geometric asymmetry contribution direction are orthogonal to each other. The coordinate difference is projected onto the crosstalk contribution direction of the polarization component and the optical path geometric asymmetry contribution direction, respectively, to obtain the first deviation component and the second deviation component.

[0046] Specifically, a polarization coupling contribution matrix is ​​first constructed. There are N measurement cycles, where N is a positive integer greater than or equal to 3. In each measurement cycle, a first light intensity value, a second light intensity value, and the components of the coordinate difference along two mutually perpendicular coordinate axes in a two-dimensional plane are collected: the first coordinate axis component and the second coordinate axis component. An N x 4 matrix is ​​constructed. Each row of this matrix corresponds to one measurement cycle. The four elements of each row are, in order: the first light intensity value, the second light intensity value, the component of the coordinate difference along the first coordinate axis, and the component of the coordinate difference along the second coordinate axis. Therefore, the first column of the matrix stores the first light intensity value for all measurement cycles, the second column stores the second light intensity value for all measurement cycles, the third column stores the first coordinate axis component of the coordinate difference for all measurement cycles, and the fourth column stores the second coordinate axis component of the coordinate difference for all measurement cycles.

[0047] Perform matrix decomposition on the matrix. Matrix decomposition can be achieved using singular value decomposition or principal component analysis. The decomposition yields several orthogonal directions. The directions related to the coordinate difference components reside in a two-dimensional subspace spanned by the third and fourth columns. Specifically, the decomposition result contains several direction vectors, each four-dimensional, corresponding to the four columns of the matrix. We are only concerned with the third and fourth dimensions, i.e., the dimensions corresponding to the coordinate difference components. Take the third and fourth components of each direction vector to form a two-dimensional vector. If the magnitude of this two-dimensional vector is not zero, divide it by its own magnitude to obtain a unit direction vector. This unit direction vector represents the projection direction of that direction into the coordinate difference space.

[0048] Matrix decomposition yields multiple directions, each corresponding to a singular value or eigenvalue, representing the degree of data variation along that direction. Based on the physical properties of polarization crosstalk and geometric asymmetry, the coordinate difference changes caused by polarization crosstalk are usually related to light intensity changes, thus their directions correspond to larger singular values; the coordinate difference changes caused by geometric asymmetry are usually unrelated to light intensity changes, corresponding to smaller singular values. Therefore, the spatial direction of the coordinate difference corresponding to the larger singular value is taken as the polarization component crosstalk contribution direction, and the direction corresponding to the smaller singular value is taken as the optical path geometric asymmetry contribution direction. These two directions are orthogonal because the right singular vectors or principal component directions obtained from matrix decomposition are orthogonal.

[0049] Since the coordinate difference space is a two-dimensional space, the two-dimensional vectors corresponding to the two directions with the largest singular values ​​can form an orthogonal basis for this two-dimensional space. The coordinate difference vector can be completely represented by these two directions, and there is no third independent component. Then, the coordinate difference vector is projected onto these two directions respectively. The coordinate difference vector consists of the first coordinate axis direction component and the second coordinate axis direction component. The projection operation uses a dot product: the first deviation component is equal to the dot product of the coordinate difference vector and the unit vector of the polarization component crosstalk contribution direction; the second deviation component is equal to the dot product of the coordinate difference vector and the unit vector of the optical path geometric asymmetry contribution direction. Since the two directions are orthogonal and form a complete orthogonal basis, the norm of the coordinate difference satisfies: the square of the norm of the coordinate difference is equal to the square of the first deviation component plus the square of the second deviation component. Therefore, the systematic deviation is taken as the norm of the coordinate difference, that is, the square root of the sum of the squares of the first and second deviation components.

[0050] The technical principle is that polarization crosstalk and geometric asymmetry have different statistical characteristics in affecting coordinate differences. Matrix decomposition can automatically find two orthogonal directions, corresponding to the two physical mechanisms respectively. The technical effect is that it can adaptively separate the two bias components from the data without prior knowledge of the crosstalk coefficient and geometric asymmetry coefficient, improving the accuracy and robustness of the decomposition.

[0051] In a preferred embodiment, the matrix decomposition is a singular value decomposition, and the polarization component crosstalk contribution direction and the optical path geometric asymmetry contribution direction are determined by the following methods: The polarization coupling contribution matrix is ​​subjected to singular value decomposition to obtain a singular value diagonal matrix and a right singular vector matrix; The direction of the right singular vector in the right singular vector matrix whose singular value is greater than a preset singular value threshold in the coordinate difference component space is taken as the polarization component crosstalk contribution direction; the direction of the right singular vector whose singular value is less than or equal to the preset singular value threshold in the coordinate difference component space is taken as the optical path geometric asymmetry contribution direction.

[0052] Specifically, singular value decomposition (SVD) is a concrete implementation of matrix decomposition. Performing SVD on the polarization coupling contribution matrix yields a left singular vector matrix, a singular value diagonal matrix, and a right singular vector matrix. The column vectors of the left singular vector matrix are called left singular vectors, and the column vectors of the right singular vector matrix are called right singular vectors. The diagonal elements of the singular value diagonal matrix are called singular values, arranged in descending order. The right singular vector matrix has four columns, each corresponding to a right singular vector. Each right singular vector has four components, corresponding to the four columns of the matrix: the first light intensity value, the second light intensity value, the first coordinate axis direction difference component, and the second coordinate axis direction difference component.

[0053] Take the third and fourth components of each right singular vector to form a two-dimensional vector. If the magnitude of this two-dimensional vector is not zero, divide it by its own magnitude to obtain the unit direction vector. This unit direction vector represents the direction of the right singular vector in the coordinate difference space. Since the column vectors of the right singular vector matrix are orthogonal, the unit direction vectors corresponding to different right singular vectors are also orthogonal to each other in the coordinate difference space.

[0054] A preset singularity threshold is used to distinguish between primary and secondary directions. The threshold can be set as a certain percentage of the sum of all singularities, such as 60% to 80%. Alternatively, singularities can be sorted from largest to smallest, and the average of the top two singularities can be used as the threshold. The unit direction vector corresponding to the right singular vector with a singularity greater than the threshold is taken as the polarization component crosstalk contribution direction, and the unit direction vector corresponding to the right singular vector with a singularity less than or equal to the threshold is taken as the optical path geometric asymmetry contribution direction. Typically, since polarization crosstalk is the primary source of deviation, its corresponding singularity is relatively large. Therefore, the unit direction vector corresponding to the largest singularity is selected as the polarization component crosstalk contribution direction, and the unit direction vector corresponding to the second largest singularity is selected as the optical path geometric asymmetry contribution direction. If multiple directions are greater than the threshold, the largest one can be taken as the polarization component crosstalk contribution direction, and the rest as geometric asymmetry directions, but usually there are only two primary directions.

[0055] After determining the two directions, the coordinate difference vector is projected onto these two directions. Specifically, the calculation is as follows: the first deviation component is equal to the dot product of the coordinate difference vector and the unit vector contributing to the crosstalk of the polarization component; the second deviation component is equal to the dot product of the coordinate difference vector and the unit vector contributing to the optical path geometric asymmetry. The dot product is calculated by multiplying the corresponding components of the two vectors and then adding the products. For example, if the coordinate difference vector is a two-dimensional vector composed of the first and second coordinate axis direction components, and the unit direction vector is also a two-dimensional vector, multiplying the two first components and the two second components, and then adding the two products, the dot product is obtained.

[0056] The technical principle is that singular value decomposition (SVD) can sort the main directions of change in the data matrix according to their variance contribution. Polarization crosstalk, being strongly correlated with light intensity changes, will have its corresponding coordinate difference change direction occupying a large number of singular values; geometric asymmetry, being weakly correlated with light intensity changes, corresponds to smaller singular values. By using thresholding, the two physical mechanisms can be automatically identified.

[0057] In a preferred embodiment, the method further includes, after the polarization coupling contribution decomposition, adaptively adjusting the extinction ratio configuration of the polarization state separation element 3 according to the proportional relationship between the first deviation component and the second deviation component: The ratio of the absolute value of the first deviation component to the systematic deviation is calculated and used as the polarization crosstalk contribution weight. If the polarization crosstalk contribution weight is greater than the first weight threshold, an extinction ratio enhancement command is generated. The extinction ratio enhancement command is used to drive the polarization state pre-alignment element 2 disposed on the incident side of the polarization state separation element 3, so that the angle between the polarization direction of the reference polarization component and the transmission axis of the polarization state separation element 3 is reduced, and the angle between the polarization direction of the measured polarization component and the reflection axis of the polarization state separation element 3 is reduced. If the polarization crosstalk contribution weight is less than or equal to the first weight threshold, the current extinction ratio configuration is maintained.

[0058] Specifically, the extinction ratio configuration of the polarization separation element 3 is adaptively adjusted based on the ratio between the first and second deviation components. The polarization crosstalk contribution weight is an indicator that measures the proportion of polarization crosstalk in the systematic deviation. It is calculated by taking the absolute value of the first deviation component and dividing it by the systematic deviation. The first deviation component is the portion of the coordinate difference caused by polarization crosstalk, while the systematic deviation is the total magnitude of the entire systematic deviation. A larger ratio indicates a greater contribution of polarization crosstalk to the systematic deviation, necessitating adjustments to the extinction ratio to suppress crosstalk.

[0059] The first weight threshold is a preset constant used to determine whether adaptive extinction ratio adjustment needs to be initiated. The value of the first weight threshold ranges from 0.6 to 0.8. When the contribution weight of polarization crosstalk is greater than the first weight threshold, it indicates that polarization crosstalk has become the dominant factor of systematic deviation, and an extinction ratio enhancement command is generated. The extinction ratio enhancement command is used to drive the polarization state pre-alignment element 2, which is located on the incident side of the polarization state separation element 3. The polarization state pre-alignment element 2 can be a liquid crystal polarization rotator, a Faraday rotator, a combination of a half-wave plate and an electric rotary stage, or an electro-optic crystal polarization modulator. The function of the polarization state pre-alignment element 2 is to adjust the polarization direction of the reference polarization component and the measurement polarization component in the incident beam.

[0060] The specific adjustment method is as follows: the angle between the polarization direction of the reference polarization component and the transmission axis of the polarization state separation element 3 is reduced, and the angle between the polarization direction of the measured polarization component and the reflection axis of the polarization state separation element 3 is also reduced. Since the extinction ratio of the polarization state separation element 3 is related to the angle between the incident polarization direction and the transmission or reflection axis, when the angle decreases, more of the corresponding polarization component is output from the target port, reducing the energy leaked to the other port, thereby effectively increasing the extinction ratio and reducing polarization crosstalk. When the polarization crosstalk contribution weight is less than or equal to the first weight threshold, it indicates that polarization crosstalk is not the main source of deviation. In this case, the current extinction ratio configuration is maintained, and no adjustment is made.

[0061] The extinction ratio enhancement command is generated by the signal acquisition and processing unit 5. The generation method is as follows: After calculating the polarization crosstalk contribution weight, the signal acquisition and processing unit 5 compares the weight with a first weight threshold. If the weight is greater than the first weight threshold, the signal acquisition and processing unit 5 queries a preset weight-target extinction ratio correspondence table based on the weight to obtain the target extinction ratio. Then, based on the difference between the target extinction ratio and the current extinction ratio, it queries an extinction ratio-polarization rotation angle mapping table to obtain the polarization rotation angle required by the polarization state pre-alignment element 2. The signal acquisition and processing unit 5 converts this polarization rotation angle into a drive signal for the polarization state pre-alignment element 2 and transmits it to the polarization state pre-alignment element 2 via a control line. The polarization state pre-alignment element 2 responds to this drive signal by adjusting the polarization directions of the reference polarization component and the measurement polarization component in the incident beam.

[0062] The technical principle is that the extinction ratio of the polarization state separation element 3 is determined by both the characteristics of the device itself and the incident polarization direction. When polarization crosstalk becomes the dominant factor in systematic deviation, the incident polarization direction can be actively adjusted by the polarization state pre-alignment element 2, which can reduce cross-leakage and suppress polarization crosstalk at its source. The technical effect is that it achieves online adaptive adjustment of the extinction ratio, which can adapt to changes in polarization crosstalk under different operating conditions without replacing the device, thus improving the system's environmental adaptability and long-term stability.

[0063] In a preferred embodiment, the polarization rotation angle of the polarization state pre-alignment element 2 is determined in the following manner: Obtain the current extinction ratio of the polarization state separation element 3 and the polarization crosstalk contribution weight; The target extinction ratio is determined based on the polarization crosstalk contribution weight. The difference between the target extinction ratio and the current extinction ratio is calculated. The preset extinction ratio-polarization rotation angle mapping table is queried to obtain the polarization rotation angle required by the polarization state pre-alignment element 2.

[0064] Specifically, the polarization rotation angle refers to the angle by which the polarization direction of the incident beam needs to be rotated by the polarization state pre-alignment element 2. The determination process consists of four steps: obtaining the current extinction ratio and polarization crosstalk contribution weight, determining the target extinction ratio based on the weight, calculating the difference between the target extinction ratio and the current extinction ratio, and obtaining the required polarization rotation angle by querying the preset extinction ratio-polarization rotation angle mapping table.

[0065] The current extinction ratio refers to the actual extinction ratio of polarization state separation element 3 under the current incident polarization direction. It can be obtained by measuring the intensity of the main polarization component and the residual cross-polarization component at the output port of polarization state separation element 3, and dividing the intensity of the main polarization component by the intensity of the residual cross-polarization component to obtain the current extinction ratio. Alternatively, it can be calculated based on the factory calibration data of polarization state separation element 3 and the angle between the current incident polarization direction and the transmission axis. The polarization crosstalk contribution weight is obtained using the aforementioned calculation method.

[0066] The target extinction ratio is the desired extinction ratio determined based on the polarization crosstalk contribution weight. It is determined by pre-establishing a correspondence table between the polarization crosstalk contribution weight and the target extinction ratio. The physical basis of this correspondence is that a larger polarization crosstalk contribution weight indicates more severe crosstalk, requiring a higher extinction ratio to suppress it. For example, when the polarization crosstalk contribution weight is 0.6, the target extinction ratio can be set to 1000:1; when the weight is 0.8, the target extinction ratio can be set to 5000:1. Specific values ​​can be obtained through experimental calibration.

[0067] Calculate the difference between the target extinction ratio and the current extinction ratio. This difference represents the gap between the current extinction ratio and the target extinction ratio. Then, consult the preset extinction ratio-polarization rotation angle mapping table to obtain the polarization rotation angle required for the polarization state pre-alignment element 2. The extinction ratio-polarization rotation angle mapping table is pre-established through calibration experiments. The calibration experiment process is as follows: Under multiple known polarization rotation angles, measure the extinction ratio of the polarization state separation element 3 to obtain multiple sets of corresponding data of polarization rotation angle and extinction ratio; then fit these data to obtain the corresponding relationship between polarization rotation angle and extinction ratio. For example, when the polarization rotation angle is 0 degrees, the extinction ratio is 500:1; when the rotation angle is 2 degrees, the extinction ratio is 1000:1; when the rotation angle is 5 degrees, the extinction ratio is 3000:1. The above data is compiled into a table for real-time querying.

[0068] The technical principle is that there is a definite correspondence between the polarization rotation angle of the polarization state pre-alignment element 2 and the extinction ratio of the polarization state separation element 3. By establishing a mapping table through pre-calibration, the required polarization rotation angle can be obtained by looking up the table based on the difference between the current extinction ratio and the target extinction ratio during actual operation, thus quickly determining the adjustment amount. The technical effect is that it achieves quantitative determination of the polarization rotation angle, avoids repeated trial and error, and improves adjustment efficiency and accuracy.

[0069] In a preferred embodiment, the adjustment of the polarization state angle by the polarization rotation element and the correction of the collimation reference are coordinated under control, specifically including: Obtain the sequence of changes in the systematic deviation of the polarization rotation element over P consecutive measurement periods after adjusting the polarization angle. Based on the sequence of changes, determine whether the systematic deviation converges as the polarization angle is adjusted; If the systematic deviation converges, the current adjustment state of the polarization rotation element is maintained, and the converged systematic deviation is taken as the residual deviation. The correction period for the collimation reference is determined based on the time-domain standard deviation of the residual deviation. If the systematic deviation does not converge or shows a divergent trend, it is determined that there is an uncontrollable component introduced by the optical path geometric asymmetry in the systematic deviation. An optical path geometric calibration control signal is generated and output to drive the displacement adjustment mechanism 7, which is located below the first photodetector 4 and / or the second photodetector 6.

[0070] Specifically, coordinated control refers to determining whether the systematic deviation has converged after the polarization rotation element adjusts the polarization angle, and then deciding on subsequent operations based on the convergence result.

[0071] The sequence of changes in the systematic deviation of the polarization rotation element over P consecutive measurement periods is obtained after adjusting the polarization angle. P is a positive integer greater than or equal to 3, typically between 5 and 20. The sequence of changes refers to the change in the systematic deviation relative to the initial value in each measurement period. The convergence of the systematic deviation with the adjustment of the polarization angle is determined based on this sequence. Convergence can be determined as follows: if the absolute value of the last change in the sequence is less than the absolute value of the first change, and the sequence shows a decreasing trend, then the systematic deviation is considered convergent; if the absolute value of the last change in the sequence is greater than or equal to the absolute value of the first change, or if the sequence shows an increasing or oscillating trend, then the systematic deviation is considered non-convergent or divergent.

[0072] If the systematic deviation converges, the current adjustment state of the polarization rotation element is maintained, and the converged systematic deviation is taken as the residual deviation. The residual deviation refers to the systematic deviation that still exists after the feedforward suppression by the polarization rotation element. The correction period for the alignment reference is determined based on the time-domain standard deviation of the residual deviation. Specifically, the time-domain standard deviation of the residual deviation is obtained over multiple consecutive measurement periods. If the time-domain standard deviation is less than a first stability threshold, the alignment reference is periodically corrected with a first correction period; if the time-domain standard deviation is greater than or equal to the first stability threshold, the alignment reference is corrected in real time with a second correction period shorter than the first correction period.

[0073] If the systematic deviation does not converge or shows a divergent trend, it is determined that there is an unsuppressible component introduced by optical path geometric asymmetry in the systematic deviation. An unsuppressible component refers to a deviation component that cannot be eliminated by adjusting the polarization angle; it originates from the geometric asymmetry between the first photodetector 4 and the second photodetector 6, such as detector mounting position deviation, focusing lens distortion, or optical path length difference. At this time, an optical path geometric calibration control signal is generated and output. This signal is used to drive the displacement adjustment mechanism 7, which is located below the first photodetector 4 and / or the second photodetector 6. The displacement adjustment mechanism 7 can be a piezoelectric micro-displacement stage or a precision ball screw displacement stage, with a displacement resolution smaller than the pixel size of the photodetector. By adjusting the position of the first photodetector 4 and / or the second photodetector 6 through the displacement adjustment mechanism 7, the optical path geometric symmetry can be changed, thereby reducing or eliminating the deviation component introduced by geometric asymmetry.

[0074] The optical path geometry calibration control signal is generated by the signal acquisition and processing unit 5. The generation method is as follows: when the systematic deviation does not converge or shows a divergent trend, the signal acquisition and processing unit 5 determines that an uncontrollable component introduced by geometric asymmetry exists. Based on the direction and magnitude of the geometric asymmetry component in the residual deviation, the signal acquisition and processing unit 5 generates a drive signal for the displacement adjustment mechanism 7. The drive signal is in the form of a number of pulses or a voltage value, which is proportional to the amplitude of the geometric asymmetry component and its direction is consistent with the direction of the geometric asymmetry component. After receiving the drive signal, the displacement adjustment mechanism 7 drives the first photodetector 4 and / or the second photodetector 6 to move along the corresponding direction until the geometric asymmetry component is reduced to a preset range. The signal acquisition and processing unit 5 and the displacement adjustment mechanism 7 are connected via a control line.

[0075] The displacement adjustment mechanism 7 is disposed below the first photodetector 4 and / or the second photodetector 6. The displacement adjustment mechanism 7 includes a fixed base, a moving platform, and a driving element. The fixed base is fixedly connected to the system base, the moving platform is fixedly connected to the first photodetector 4 or the second photodetector 6, and the driving element is disposed between the fixed base and the moving platform. In response to a geometric calibration control signal, the driving element drives the moving platform to displace relative to the fixed base. The displacement direction of the displacement adjustment mechanism 7 includes both along a first coordinate axis and along a second coordinate axis, thereby achieving position adjustment in a two-dimensional plane.

[0076] The technical principle is that feedforward suppression of the polarization rotation element can only solve the deviation caused by polarization crosstalk, but cannot solve the deviation caused by geometric asymmetry. By judging whether the systematic deviation converges with the adjustment of the polarization angle, the two sources of deviation can be distinguished. If they converge, it means that polarization crosstalk has been effectively suppressed, and the remaining deviation can be handled by reference correction; if they do not converge, it means that there is a geometric asymmetry component, and geometric calibration is required through the displacement adjustment mechanism 7. The technical effect is to achieve synergy between polarization feedforward suppression and geometric calibration, which not only ensures the effective suppression of polarization crosstalk, but also allows for timely switching of the processing method when geometric asymmetry becomes the dominant factor, avoiding blindly adjusting the polarization element while ignoring the fundamental geometric problem.

[0077] In a preferred embodiment, determining the correction period for the collimation reference based on the time-domain standard deviation of the residual deviation includes: If the time-domain standard deviation is less than the first stability threshold, the collimation reference is periodically corrected with a first correction period. If the time-domain standard deviation is greater than or equal to the first stability threshold, the collimation reference is corrected in real time with a second correction period that is less than the first correction period.

[0078] Specifically, the residual bias is the convergent value of the systematic bias after feedforward suppression by the polarization rotation element. The time-domain standard deviation is a measure of the fluctuation of the residual bias over multiple consecutive measurement periods. It is calculated as follows: obtain the numerical sequence of the residual bias over Q consecutive measurement periods, where Q is a positive integer greater than or equal to 3, typically between 10 and 50; calculate the time-domain mean and time-domain standard deviation of this sequence, using the same method as described above.

[0079] The first stability threshold is a constant set according to the accuracy requirements of the collimation reference. If the time-domain standard deviation is less than the first stability threshold, the residual deviation is considered to be in a steady state. At this point, the collimation reference is periodically corrected using the first correction cycle. The first correction cycle can be 10 to 100 measurement cycles. Periodic correction means that at the end of each first correction cycle, the correction amount of the collimation reference is updated using the time-domain mean of the residual deviation. For example, if the first correction cycle is 50 measurement cycles, then after every 50 measurements, the currently calculated time-domain mean of the residual deviation is used as the correction amount and subtracted from the coordinates of subsequent measurement points.

[0080] If the time-domain standard deviation is greater than or equal to the first stability threshold, the residual deviation is determined to be in a non-steady-state fluctuation state. In this case, real-time correction is performed on the alignment reference using the second correction period. The second correction period is shorter than the first correction period, for example, 1 to 3 measurement periods. Real-time correction refers to correcting the alignment reference within each measurement period using the instantaneous residual deviation obtained in the current measurement period. The instantaneous residual deviation refers to the residual deviation calculated in the current measurement period.

[0081] The technical principle is that the time-domain standard deviation of the residual deviation reflects its degree of fluctuation. A small standard deviation indicates that the residual deviation is stable, and periodic correction can meet the accuracy requirements, reducing the computational burden; a large standard deviation indicates that the residual deviation fluctuates drastically, and real-time correction is necessary to track its changes. The technical effect is that the correction period is adaptively selected based on the actual fluctuation characteristics of the residual deviation, optimizing the use of computational resources while ensuring correction accuracy.

[0082] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for correcting a laser collimation reference based on polarization state decoupling, characterized in that, include: Polarization state decoupling: A polarization state separation element (3) is set in the laser collimation optical path to separate the reference polarization component transmitted by the reference optical path from the measurement polarization component transmitted by the reflection optical path in space, so as to obtain a reference polarization beam and a measurement polarization beam that are independent of each other; the reference polarization component and the measurement polarization component are in a mutually coupled state before separation; Independent detection: The reference polarized beam and the measurement polarized beam are guided to the first photodetector (4) and the second photodetector (6) which are independent of each other, respectively, to obtain the reference spot image and the measurement spot image; Reference coordinate calculation: Calculate the coordinates of the reference point corresponding to the reference polarization component based on the reference spot image, and calculate the coordinates of the measurement point corresponding to the measurement polarization component based on the measurement spot image; Systematic deviation determination: Calculate the coordinate difference between the coordinates of the reference point and the coordinates of the measurement point, and determine the coordinate difference as the systematic deviation of the reference point positioning including the polarization coupling effect; Reference correction: Based on the systematic deviation, the alignment reference is corrected to eliminate the influence of the mutual coupling between the polarization component of the reflected optical path and the polarization component of the reference optical path on the positioning of the reference point.

2. The laser collimation reference correction method for polarization state decoupling according to claim 1, characterized in that, The determination of the systematic deviation also includes decomposing the coordinate difference into polarization coupling contribution. Obtain the first light intensity value sequence obtained by the reference polarized beam through the first photodetector (4), and the second light intensity value sequence obtained by the measurement polarized beam through the second photodetector (6); Based on the first light intensity value sequence and the second light intensity value sequence, calculate the proportion of the residual reference polarization component in the measured polarization beam and the proportion of the residual measured polarization component in the reference polarization beam. The residual reference polarization component ratio and the residual measurement polarization component ratio are respectively correlated with the coordinate difference to obtain the first deviation component introduced by polarization component crosstalk and the second deviation component introduced by optical path geometric asymmetry. The square root of the sum of the squares of the first deviation component and the second deviation component is taken as the systematic deviation.

3. The laser collimation reference correction method for polarization state decoupling according to claim 1, characterized in that, In the reference correction, the correction of the alignment reference includes: Based on the systematic deviation, a compensation control signal is generated for driving the polarization rotation element; The compensation control signal is applied to the polarization rotation element coupled to the polarization state separation element (3). The polarization rotation element is configured to adjust the polarization state angle between the reference polarization component and the measured polarization component before separation in response to the compensation control signal, so as to increase the spatial separation of the reference polarization component and the measured polarization component after separation, thereby suppressing the generation of the systematic deviation at the optical path level.

4. The laser collimation reference correction method for polarization state decoupling according to claim 1, characterized in that, The benchmark correction also includes time-domain stability determination and correction mode switching for the systematic deviation: Obtain the numerical sequence of the systematic deviation over M consecutive measurement periods, and calculate the time-domain standard deviation and time-domain mean of the numerical sequence; If the time-domain standard deviation is less than the first stability threshold, then the time-domain mean is used as the steady-state systematic deviation, and the collimation reference is periodically corrected with the first correction period. If the time-domain standard deviation is greater than or equal to the first stability threshold, the systematic deviation is determined to be in a non-steady-state fluctuation state. The correction period is shortened to a second correction period, which is less than the first correction period. In each measurement period, the instantaneous systematic deviation obtained in the current measurement period is used to correct the alignment reference in real time.

5. The laser collimation reference correction method for polarization state decoupling according to claim 2, characterized in that, The systematic deviation is taken as the norm of the coordinate difference; In the polarization coupling contribution decomposition, the determination methods for the first deviation component and the second deviation component include: A polarization coupling contribution matrix is ​​constructed, wherein each row of the polarization coupling contribution matrix corresponds to a measurement period, and the columns include the first light intensity value, the second light intensity value, and the components of the coordinate difference in the two-dimensional plane along two mutually perpendicular coordinate axes. The polarization coupling contribution matrix is ​​decomposed to obtain the polarization component crosstalk contribution direction and the optical path geometric asymmetry contribution direction in the coordinate difference component space. The polarization component crosstalk contribution direction and the optical path geometric asymmetry contribution direction are orthogonal to each other. The coordinate difference is projected onto the crosstalk contribution direction of the polarization component and the optical path geometric asymmetry contribution direction, respectively, to obtain the first deviation component and the second deviation component.

6. The laser collimation reference correction method for polarization state decoupling according to claim 5, characterized in that, The matrix decomposition is a singular value decomposition, and the polarization component crosstalk contribution direction and the optical path geometric asymmetry contribution direction are determined by the following methods: The polarization coupling contribution matrix is ​​subjected to singular value decomposition to obtain a singular value diagonal matrix and a right singular vector matrix; The direction of the right singular vector in the right singular vector matrix whose singular value is greater than a preset singular value threshold in the coordinate difference component space is taken as the polarization component crosstalk contribution direction; the direction of the right singular vector whose singular value is less than or equal to the preset singular value threshold in the coordinate difference component space is taken as the optical path geometric asymmetry contribution direction.

7. The laser collimation reference correction method for polarization state decoupling according to claim 2, characterized in that, The method further includes, after the polarization coupling contribution decomposition, adaptively adjusting the extinction ratio configuration of the polarization state separation element (3) according to the ratio of the first deviation component to the second deviation component: The ratio of the absolute value of the first deviation component to the systematic deviation is calculated and used as the polarization crosstalk contribution weight. If the polarization crosstalk contribution weight is greater than the first weight threshold, an extinction ratio enhancement command is generated. The extinction ratio enhancement command is used to drive the polarization state pre-alignment element (2) disposed on the incident side of the polarization state separation element (3) to reduce the angle between the polarization direction of the reference polarization component and the transmission axis of the polarization state separation element (3), and to reduce the angle between the polarization direction of the measured polarization component and the reflection axis of the polarization state separation element (3). If the polarization crosstalk contribution weight is less than or equal to the first weight threshold, the current extinction ratio configuration is maintained.

8. The laser collimation reference correction method for polarization state decoupling according to claim 7, characterized in that, The polarization rotation angle of the polarization state pre-alignment element (2) is determined in the following way: Obtain the current extinction ratio of the polarization state separation element (3) and the polarization crosstalk contribution weight; The target extinction ratio is determined based on the polarization crosstalk contribution weight. The difference between the target extinction ratio and the current extinction ratio is calculated. The preset extinction ratio-polarization rotation angle mapping table is queried to obtain the polarization rotation angle required by the polarization state pre-alignment element (2).

9. The laser collimation reference correction method for polarization state decoupling according to claim 3, characterized in that, The adjustment of the polarization state angle by the polarization rotation element and the correction of the collimation reference form a coordinated control, specifically including: Obtain the sequence of changes in the systematic deviation of the polarization rotation element over P consecutive measurement periods after adjusting the polarization angle. Based on the sequence of changes, determine whether the systematic deviation converges as the polarization angle is adjusted; If the systematic deviation converges, the current adjustment state of the polarization rotation element is maintained, and the converged systematic deviation is taken as the residual deviation. The correction period for the collimation reference is determined based on the time-domain standard deviation of the residual deviation. If the systematic deviation does not converge or shows a divergent trend, it is determined that there is an uncontrollable component introduced by the optical path geometric asymmetry in the systematic deviation, and an optical path geometric calibration control signal for driving the displacement adjustment mechanism (7) is generated and output. The displacement adjustment mechanism (7) is located below the first photodetector (4) and / or the second photodetector (6).

10. The laser collimation reference correction method for polarization state decoupling according to claim 9, characterized in that, The step of determining the correction period for the collimation reference based on the time-domain standard deviation of the residual deviation includes: If the time-domain standard deviation is less than the first stability threshold, the collimation reference is periodically corrected with a first correction period. If the time-domain standard deviation is greater than or equal to the first stability threshold, the collimation reference is corrected in real time with a second correction period that is less than the first correction period.

Citation Information

Patent Citations

  • Polarized light combined target common-path compensated two-dimensional photoelectric auto-collimation method and device

    CN102176087A