A Kude Optical Path Debugging Fixture and Method
By using the Kude optical path debugging tooling and method, and by independently debugging the reflectors group by group using the beam splitting and detection units, the problem of error accumulation was solved, the optical path accuracy and production efficiency were improved, and the cost was reduced.
Patent Information
- Application Number
- CN202611437657.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-16
- Publication Date
- 2026-11-17
AI Technical Summary
In the existing Kuder optical path debugging method, the installation error of the reflector will accumulate step by step, making it difficult to guarantee the final optical path accuracy. In addition, the debugging cycle is long and the rework rate is high. Existing equipment is complex and costly.
The Kude optical path debugging fixture includes a support frame, a beam splitter, a first detection unit, and a second detection unit. The beam is split into two paths by the beam splitter. The first detection unit detects the angle information of the beam, and the second detection unit detects the position information of the beam. The reflectors are debugged independently one by one with the global reference optical axis as the sole reference.
It effectively prevents the accumulation of reflector errors, improves the efficiency of optical path assembly and adjustment and the finished product qualification rate, and reduces rework costs and debugging cycle.
Smart Images

Figure CN123113167A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of optical system assembly and adjustment technology, and in particular to a Kuder optical path adjustment fixture and method. Background Technology
[0002] The Coude optical path is widely used in large-aperture telescopes, laser communication terminals, optoelectronic tracking systems, and laser rangefinders. Its function is to guide an incident light beam, which rotates with the turntable, through multiple sets of reflectors to a stationary output end, thereby achieving stable transmission of the optical signal. Since the Coude optical path typically contains four or more reflectors, each mounted on a different mechanical axis, its optical path adjustment is one of the key procedures in the equipment assembly and adjustment process.
[0003] Existing methods for adjusting the Kuder optical path typically involve sequentially adjusting each Kuder mirror. Taking four Kuder mirrors as an example, the current method first adjusts Kuder mirror I and Kuder mirror II to align with mechanical axis I, then installs and adjusts Kuder mirror III and Kuder mirror IV in sequence to align with mechanical axis II. The installation position and angle of each mirror rely on the optical path established by the previous mirror as a reference. However, this sequential installation and step-by-step adjustment method has a significant technical drawback: the installation and adjustment errors of the previous mirror will inevitably be superimposed on the adjustment results of the subsequent mirrors, and the errors accumulate and amplify step by step along the optical path, ultimately making it difficult for the accuracy of the emitted beam to meet the design requirements.
[0004] Specifically, the installation angle deviation of Couder I will directly affect the incident light direction of Couder II, causing the adjustment reference of Couder II to deviate from the ideal optical axis; by analogy, Couder III and Couder IV have already carried all the errors of the previous stage when establishing their respective adjustment references, and the cumulative deviation of the final output light path is much greater than the algebraic sum of the errors of a single stage mirror.
[0005] This technical problem of error accumulating step by step has two main drawbacks. First, the final optical path accuracy heavily depends on the adjustment accuracy of each stage of the mirror assembly. Even minute deviations at any stage cannot be eliminated in subsequent adjustments and can only be passively accumulated. Second, when a defective outgoing optical path is finally detected, operators find it difficult to pinpoint the source of the error. Often, it is necessary to disassemble all mirror assemblies and readjust each stage, resulting in long adjustment cycles and high rework rates. While existing technologies employ multiple autocollimators and reference mirrors for simultaneous monitoring, these methods are complex, costly, and fail to fundamentally change the inherent pattern of error propagation along the optical path. Summary of the Invention
[0006] In view of this, embodiments of the present invention provide a Couder optical path debugging fixture and method to solve the problem that in the traditional method of debugging Couder mirrors one by one in sequence, the installation and debugging error of the previous mirror will inevitably be superimposed on the debugging results of the subsequent mirrors, making it difficult to guarantee the final optical path accuracy.
[0007] In a first aspect, embodiments of the present invention provide a Kuder optical path debugging fixture, comprising: A support frame, and a beam splitter, a first detection unit, and a second detection unit disposed on the support frame; The beam splitter is disposed on the outgoing optical path of the Couder optical path and is used to split the outgoing beam into a first beam and a second beam. The first detection unit is disposed in the optical path of the first beam and is used to detect the angle information of the emitted beam. The second detection unit is disposed in the optical path of the second beam and is used to detect the position information of the emitted beam; The first detection unit includes an image acquisition device and an imaging lens disposed in front of the image acquisition device, while the second detection unit is an image acquisition device without an imaging lens.
[0008] Secondly, a method for debugging a Kuder optical path is provided, including: S1: A reference element is set at the incident end of the Kud optical path, and the center of the reference element is aligned with the optical axis center of the Kud optical path; S2: A collimated beam is emitted to the reference element using a collimating light source, and the collimating light source is adjusted so that its optical axis center is aligned with the center of the reference element to achieve self-collimation; S2: Use a collimating light source to emit a collimated beam to the reference element, and adjust the collimating light source so that its collimating optical axis center is aligned with the center of the reference element to achieve self-collimation and establish a global reference optical axis; S3: Set the Kude optical path debugging fixture at the output end of the Kude optical path; S4: Divide the multiple Couder mirrors in the Couder optical path into several groups according to their corresponding mechanical axes, and each group includes at least a first optical adjustment element and a second optical adjustment element; S5: Perform the following debugging steps on the current group: S51: Rotate the mechanical axis corresponding to the current group, observe the trajectory of the emitted beam during the rotation process through the first detection unit, adjust the first optical adjustment element in the group, so that the angular offset of the beam during the rotation process is not greater than the first preset threshold, and complete the angular domain locking of the current group. S52: The second detection unit obtains the position coordinates of the emitted beam before and after the mechanical shaft rotates, and calculates the position offset in each coordinate axis direction; S53: While keeping the angular offset no greater than the first preset threshold, adjust the second optical adjustment element in the group to make the beam offset along a certain coordinate axis direction, and then adjust the first optical adjustment element to make the beam return to the reference position. Repeat this process until the position offset in the coordinate axis direction is no greater than the second preset threshold. S54: Adjust the position offset in the other coordinate axis direction according to the method of S53 until the position offset in the coordinate axis direction is not greater than the second preset threshold. S55: When the mechanical shaft rotates back and forth, the angular offset detected by the first detection unit is not greater than the first preset threshold and the position offset of each coordinate axis direction detected by the second detection unit is not greater than the second preset threshold, the current group debugging is completed; S6: Lock the pose of all optical adjustment elements in the current group, and then use the global reference optical axis as the only reference to independently debug the next group in the manner of S5 until all groups are debugged; during the debugging of any group, the state of the outgoing beam corresponding to that group does not participate in the reference establishment of the next group.
[0009] The Cod optical path debugging fixture and method provided by this invention have the following beneficial effects: When the traditional method is used to adjust the Couder mirrors one by one in sequence, the installation angle deviation of Couder I directly causes the incident light direction of Couder II to deviate from the ideal optical axis, so that the adjustment reference of Couder II itself already carries errors; by analogy, Couder III and Couder IV have accumulated all the errors of the previous stage when establishing their respective adjustment references, and the deviation of the final output light path is much greater than the algebraic sum of the errors of a single mirror body.
[0010] In this invention, each Couder mirror is divided into several groups according to its corresponding mechanical axis, and is adjusted in a way that is "independently adjusted group by group, with the global reference optical axis as the sole reference between groups". After the adjustment of each group of mirrors is completed, its pose is locked. The adjustment of subsequent groups is no longer based on the actual optical axis of the emitted beam of that group, but on the global reference optical axis established at the incident end as the sole reference. This ensures that the residual error of the previous group of mirrors is completely blocked within the group and does not participate in the establishment of the reference for the adjustment of any subsequent group. The error of each group of mirrors converges independently within the tolerance range of its corresponding mechanical axis. Finally, the total error of the emitted beam is equal to the envelope of the errors of each group, rather than the algebraic sum of the errors of each group.
[0011] Meanwhile, this invention employs a domain-specific debugging strategy within each group: "first lock the angle domain, then iteratively converge the position domain." By separating and independently feeding back angle detection and position detection, and by real-time interlocking verification of angle offset during position iteration, it avoids the repeated oscillation problem caused by the coupling of angle error and position error in traditional methods. This ensures that the deviations in both dimensions within each group can converge independently and efficiently to the qualified range, significantly reducing rework costs and debugging cycles, and significantly improving the production efficiency and finished product qualification rate of Kude Optical Path Assembly and Adjustment. Attached Figure Description
[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, and these are all within the protection scope of the present invention.
[0013] Figure 1 This is a schematic diagram of the overall debugging process of a Kude optical path debugging fixture; Figure 2 This is a schematic diagram of the overall debugging process of a Kude optical path debugging fixture from another angle; Figure 3 This is a cross-sectional schematic diagram of the overall debugging of a Kude optical path debugging fixture; Parts and component numbers in the diagram: 100-Bearing frame, 110-First support arm, 111-First mounting part, 120-Second support arm, 121-Second mounting part, 122-First type of fixing ear, 130-Mounting frame, 131-Receiving groove, 132-Second type of fixing ear, 133-Pressure block; 200 - beam splitter, 211 - first beam, 212 - second beam; 310 - First detection unit, 320 - Imaging lens, 330 - Second detection unit; 410-Collimating light source, 420-Reference element, 421-Scale, 431-First optical adjustment element, 432-Second optical adjustment element, 441-Collimating optical axis, 442-Couder optical path. Detailed Implementation
[0014] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. In the description of the present invention, it should be understood that the terms "center," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements, but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, the element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. Where there is no conflict, embodiments of the present invention and the various features thereof can be combined with each other, all of which are within the scope of protection of the present invention.
[0015] Example 1 Please see Figures 1-3 This invention provides a debugging fixture for the Kude Optical Path 442, comprising: The support frame 100, and the beam splitting element 200, the first detection unit 310 and the second detection unit 330 disposed on the support frame 100; The beam splitter 200 is disposed on the outgoing optical path of the Couder optical path 442 and is used to split the outgoing beam into a first beam 211 and a second beam 212. The first detection unit 310 is disposed in the optical path of the first beam 211 and is used to detect the angle information of the emitted beam; The second detection unit 330 is disposed in the optical path of the second beam 212 and is used to detect the position information of the emitted beam; The first detection unit 310 includes an image acquisition device and an imaging lens 320 disposed in front of the image acquisition device, and the second detection unit 330 is an image acquisition device without an imaging lens 320.
[0016] like Figures 1-3As shown, the Kuder optical path 442 debugging fixture is set at the output end of the Kuder optical path 442 to receive the beam emitted after being deflected by the Kuder reflector group, and to detect the angle and position information of the emitted beam in real time. The debugging fixture mainly includes a support frame 100, and a beam splitter 200, a first detection unit 310, and a second detection unit 330 fixedly mounted on the support frame 100. The support frame 100 serves as the mounting base for the entire debugging fixture, providing a unified mounting reference and stable structural support for the beam splitter 200, the first detection unit 310, and the second detection unit 330, ensuring that the relative positional relationship between the components remains unchanged during the debugging process, thereby ensuring the stability and repeatability of the detection data. The support frame 100 is preferably made of rigid material (such as aluminum alloy or stainless steel), and its bottom is provided with a positioning mounting surface for detachably fixing the entire debugging fixture to the equipment base at the output end of the Kuder optical path 442.
[0017] Furthermore, the beam splitter 200 is fixedly mounted at the foremost end of the support frame 100, meaning the outgoing beam from the Coulomb optical path 442 first enters the beam splitter 200. The beam splitter 200 is located on the optical axis of the outgoing beam and is used to split an incident outgoing beam into two paths, a first beam 211 and a second beam 212, according to a certain energy ratio. In this embodiment, the beam splitter 200 is preferably a beam splitter. The beam splitter is mounted on the support frame 100 with its surface coated with a beam splitting film facing the direction of the incident beam. This beam splitting film is coated according to different beam splitting ratio requirements. In this embodiment, a beam splitter with a transmission to reflection energy ratio of 1:1 is preferably used, meaning that after the incident beam passes through the beam splitter, 50% of the energy is transmitted to form the first beam 211, and 50% of the energy is reflected to form the second beam 212. The angle between the normal of the beam splitter and the optical axis of the incident beam is preferably set to 45°. At this angle, the transmitted first beam 211 continues to propagate in a straight line along the original incident optical axis, while the reflected second beam 212 turns 90° in a direction perpendicular to the incident optical axis, thereby forming a sufficient spatial distance between the two beams. This facilitates the arrangement of the first detection unit 310 and the second detection unit 330 within the limited installation space of the support frame 100, avoids optical path interference between the two beams, and makes the overall tooling structure layout more compact and reasonable.
[0018] Further, the first detection unit 310 is fixedly mounted on the support frame 100 and located in the optical path of the first beam 211 (i.e., the transmitted beam), used to detect the angle information of the emitted beam. The first detection unit 310 includes an image acquisition device and an imaging lens 320 disposed in front of the image acquisition device. The image acquisition device and the imaging lens 320 are fixedly connected through a standard optical interface (such as a C-interface or a CS-interface). The image acquisition device and the imaging lens 320 are arranged sequentially on the support frame 100 along the optical axis of the transmitted beam, with the imaging lens 320 closer to the beam splitter and the image acquisition device located behind the imaging lens 320. The transmitted beam passes through the imaging lens 320 sequentially and converges on the photosensitive surface (target surface) of the image acquisition device to form a light spot image. The image acquisition device preferably uses an industrial camera (such as a CCD camera or a CMOS camera), whose photosensitive surface is composed of an array of pixel units. Each pixel unit has a defined physical size and is used to convert the received light signal into an electrical signal and transmit it to an external display device or computer processing system via a data line. When the emitted beam deviates angularly from the collimating optical axis 441 (ideal optical axis), this angular deviation, after being imaged by the imaging lens 320, manifests as a shift in the position of the light spot on the photosensitive surface of the image acquisition device. By reading the pixel coordinates of the light spot on the photosensitive surface, the magnitude and direction of the angular deviation of the emitted beam can be quantitatively characterized. The focal length of the imaging lens 320 determines the sensitivity of angle detection. The longer the focal length, the greater the light spot displacement caused by a unit angular deviation, and the higher the detection resolution. In this embodiment, the focal length of the imaging lens 320 is preferably 90mm.
[0019] The second detection unit 330 is fixedly mounted on the support frame 100 and located in the optical path of the second beam 212 (i.e., the reflected beam), used to detect the position information of the emitted beam. The second detection unit 330 is an image acquisition device without an imaging lens 320. This image acquisition device and the image acquisition device in the first detection unit 310 preferably use the same type of industrial camera to ensure that the pixel size of the two detections is consistent, which facilitates data comparison and calculation. The image acquisition device of the second detection unit 330 is fixed on the support frame 100, and its photosensitive surface is set directly facing the direction of the reflected beam of the beam splitter. Its optical axis coincides with the propagation direction of the reflected beam. The reflected beam directly enters the photosensitive surface of the second detection unit 330 without passing through any imaging optical elements to form a light spot. Since there is no imaging lens 320, the position of the light spot on the photosensitive surface of the reflected beam directly reflects the translational position of the emitted beam in space. That is, when the emitted beam has a parallel displacement (positional deviation) relative to the collimating optical axis 441, this displacement is directly manifested as the pixel coordinate offset of the light spot on the photosensitive surface of the second detection unit 330. Unlike the imaging principle of the first detection unit 310 which detects angular deviation, the second detection unit 330 does not carry a lens. Therefore, the change in the position of the light spot on its photosensitive surface is basically unaffected by the change in the angle of the outgoing beam (the amount of light spot displacement caused by the angle change is much smaller in a lensless system than in a lens-equipped system), thus realizing independent detection of the beam position information.
[0020] Both the first detection unit 310 and the second detection unit 330 are electrically connected to an external display device or data processing terminal. Operators can observe the light spot images and their real-time coordinates acquired by the two detection units on the display screen. The first detection unit 310, equipped with an imaging lens 320, is sensitive to changes in beam angle but not to beam translation; the second detection unit 330, without a lens, is sensitive to beam translation but not to changes in beam angle. Both work collaboratively on the support frame 100, separating and independently outputting information on the "angle deviation" and "position deviation" of the emitted beam. Operators adjust the corresponding reflector angle based on the angle deviation data and adjust the corresponding reflector translation position based on the position deviation data. These adjustments do not interfere with each other, thus decoupling the complex multi-dimensional optical path adjustment problem into an independent single-dimensional adjustment problem. The support frame 100 serves as a unified installation reference, ensuring that the relative positions of the beam splitter 200, the first detection unit 310, and the second detection unit 330 remain consistent during each use. Even with repeated disassembly and reassembly of the debugging fixture, there is no need to recalibrate the relative positions of the components, greatly improving the practicality and reliability of the debugging fixture. Furthermore, without departing from the principles of this invention, the beam splitting ratio of the beam splitter 200 can be adjusted according to actual needs. For example, when higher precision angle detection is required, the energy ratio of the transmitted beam can be increased (e.g., adjusted to 2:1), allowing more light energy to enter the first detection unit 310 with the imaging lens 320 to improve the image signal-to-noise ratio and detection accuracy; when higher precision position detection is required, the energy ratio of the reflected beam can be increased.
[0021] Furthermore, the support frame 100 includes a first arm 110 and a second arm 120 integrally formed, and the first arm 110 and the second arm 120 are respectively provided with a first mounting part 111 and a second mounting part 121, which are respectively used to install the first detection unit 310 and the second detection unit 330. A mounting frame 130 is provided at the connection between the first arm 110 and the second arm 120, and the mounting frame 130 is provided with a receiving groove 131 for accommodating the beam splitter 200.
[0022] like Figures 1-3 As shown, the support frame 100 serves as the mounting base for the entire debugging fixture, and its structure must simultaneously meet the spatial layout requirements of the beam splitting element 200, the first detection unit 310, and the second detection unit 330.
[0023] In this embodiment, the support frame 100 preferably adopts an integrated molding structure, including a first arm 110 and a second arm 120 integrally formed. The first arm 110 and the second arm 120 meet and connect at the front end of the support frame 100, forming a certain angle space between them. This angle space is adapted to the propagation direction of the two beams after the beam is split into a first beam 211 and a second beam 212 by the beam splitting element 200. The first arm 110 extends along the propagation direction of the first beam 211 to provide mounting support for the first detection unit 310; the second arm 120 extends along the propagation direction of the second beam 212 to provide mounting support for the second detection unit 330. The integrated arrangement of the first arm 110 and the second arm 120 allows both detection units to use the same support frame 100 as a positioning reference, maintaining a constant spatial positional relationship between them without the need for separate independent adjustments. The first support arm 110 is provided with a first mounting part 111, which is used to fix the first detection unit 310; the second support arm 120 is provided with a second mounting part 121, which is used to fix the second detection unit 330. The first mounting part 111 and the second mounting part 121 can be threaded holes, positioning holes or slot structures opened on the support arm. The first detection unit 310 and the second detection unit 330 are respectively fixed to the corresponding mounting parts by fasteners or quick-release structures, thereby realizing the detachable installation of each detection unit on the support frame 100.
[0024] Furthermore, a mounting frame 130 is provided at the connection between the first arm 110 and the second arm 120. This mounting frame 130 is located at the foremost end inside the support frame 100, meaning that the emitted beam from the Kuder optical path 442 first enters the position of the mounting frame 130. The mounting frame 130 is provided with a receiving groove 131, the shape of which is adapted to the outer contour of the beam splitter 200, for accommodating and positioning the beam splitter 200. After the beam splitter 200 is embedded in the receiving groove 131, it is fixed in the receiving groove 131 by means of a pressure plate, elastic buckle, or adhesive. The opening angle of the receiving groove 131 is determined according to the preset beam splitting angle requirements. When the beam splitter 200 is installed in the receiving groove 131, the optical working surface of the beam splitter 200 automatically forms the required angle relative to the optical axis of the emitted beam, without the need to adjust the angle of the beam splitter 200 during the assembly process.
[0025] Furthermore, the relative positional relationship between the first mounting part 111 on the first support arm 110 and the mounting frame 130, and the relative positional relationship between the second mounting part 121 on the second support arm 120 and the mounting frame 130, are precisely determined during the integrated machining of the support frame 100. This ensures that the optical axis of the first detection unit 310 mounted on the first mounting part 111 automatically coincides with the optical path of the first beam 211 transmitted by the beam splitter 200 mounted in the receiving groove 131, and the optical axis of the second detection unit 330 mounted on the second mounting part 121 automatically coincides with the optical path of the second beam 212 reflected by the beam splitter 200. This allows the entire debugging fixture to be used directly without any on-site calibration, significantly reducing the difficulty of using the fixture. The support frame 100 can be integrally formed from aluminum alloy material through CNC machining, ensuring structural accuracy while also meeting lightweight requirements. In applications with higher precision requirements, materials with low thermal expansion coefficients, such as Invar alloy, can also be used to reduce the impact of temperature changes on the fixture's accuracy.
[0026] Furthermore, the second support arm 120 is provided with a first type of fixing ear 122 on the outside, and the mounting frame 130 is provided with a second type of fixing ear 132 on both sides of the receiving groove 131; the beam splitter 200 is restricted in the receiving groove 131 by a number of pressure blocks 133.
[0027] Furthermore, a first type of fixing lug 122 is disposed on the outer wall surface of the second support arm 120, for detachably fixing the entire support frame 100 to the equipment base at the output end of the Kude optical path 442. There may be one or more first type fixing lugs 122, each with screw through holes or locating pin holes, which securely connect the support frame 100 to the equipment base using screws or locating pins, ensuring the entire debugging fixture remains stable during operation.
[0028] The second type of fixing ear 132 is disposed on the mounting frame 130 and located on both sides of the receiving groove 131. The second type of fixing ear 132 is used to install the pressure block 133. One end of the pressure block 133 is fixed to the second type of fixing ear 132 by fasteners (such as screws), and the other end of the pressure block 133 extends to the top of the receiving groove 131. After the beam splitter 200 is placed in the receiving groove 131, the pressure block 133 is pressed against the non-optical working surface (such as the edge area) of the beam splitter 200, and the fasteners are tightened. The beam splitter 200 can be reliably confined in the receiving groove 131 by the pressure block 133, preventing the beam splitter 200 from coming out of the receiving groove 131 due to vibration or external force during the commissioning process. It should be noted that the beam splitter 200 is confined within the receiving groove 131 by several pressure blocks 133. However, the pressure blocks 133 only apply a clamping force to the beam splitter 200 along the depth direction of the receiving groove 131. The positioning of the beam splitter 200 within the plane of the receiving groove 131 is still achieved by the sidewall of the receiving groove 131. The angular accuracy of the beam splitter 200 depends on the machining accuracy of the receiving groove 131 rather than the assembly accuracy of the pressure blocks 133. Therefore, even repeated disassembly and reassembly of the pressure blocks 133 will not affect the angular position of the beam splitter 200, ensuring the angular consistency and repeatable positioning accuracy of the beam splitter 200. In this embodiment, the number of the second type of fixing ears 132 and pressure blocks 133 is preferably four, with the four pressure blocks 133 pressing the beam splitter 200 from its four corners. In an alternative embodiment, the number of the second type of fixing ears 132 and pressure blocks 133 can also be two, pressing the beam splitter 200 from opposite sides, which can also achieve a reliable fixing effect.
[0029] Furthermore, the beam splitting element 200 is a beam splitter; the angle between the beam splitter and the optical axis of the emitted beam is 45°.
[0030] Furthermore, the beam splitter, as a precision optical element, preferably uses K9 optical glass or fused silica as its substrate material. A beam-splitting film is deposited on one surface of the substrate. This film is prepared using a vacuum deposition process, enabling a predetermined proportion of the incident light energy to be transmitted and the remaining proportion to be reflected. The uncoated surface of the beam splitter is typically coated with an anti-reflection film to reduce light energy reflection loss during transmission and improve light energy utilization. The beam splitter's outline is adapted to the groove shape of the receiving slot 131. After the beam splitter is installed in the receiving slot 131, its optical working surface faces the incident direction of the beam emitted from the Couder optical path 442.
[0031] The angle between the beam splitter and the optical axis of the emitted beam is 45°. Setting the angle between the beam splitter and the optical axis of the emitted beam to 45° has the following structural advantages: when the emitted beam reaches the optical working surface of the beam splitter at an incident angle of 45°, the transmitted first beam 211 continues to propagate in a straight line along the original optical axis of the emitted beam, while the reflected second beam 212 deflects 90° in a direction perpendicular to the original optical axis. This 90° deflection allows the first beam 211 and the second beam 212 to form mutually perpendicular propagation paths in space, providing a 90° angled layout space for the first arm 110 and the second arm 120 on the support frame 100. This not only helps to reduce the overall space occupied by the debugging fixture, but also ensures that the optical paths between the first detection unit 310 and the second detection unit 330 do not interfere with each other. Meanwhile, the 45° incident angle ensures a relatively balanced transmission and reflection characteristic of the beam splitter for P-polarized and S-polarized light, which helps maintain the consistency of the polarization state of the transmitted and reflected beams and avoids significant differences in light intensity or spot quality between the two beams due to polarization effects. At this 45° angle, the normal direction of the beam splitter is at 45° to the optical axis of the emitted beam. A corresponding angular relationship exists between the normal of the beam splitter's optical working surface and its external reference surface. Once the beam splitter is placed in the receiving groove 131, the machining accuracy of the receiving groove 131 automatically ensures that the 45° angle between the beam splitter and the optical axis of the emitted beam meets the requirements, eliminating the need for on-site adjustment. It is understood that although 45° is the preferred angle used in this embodiment, under certain special spatial layout requirements, the angle between the beam splitter and the optical axis of the emitted beam can be set to other angles, and the corresponding angle between the first arm 110 and the second arm 120 can also be adjusted accordingly. These alternative solutions are all equivalent embodiments of the present invention.
[0032] Furthermore, the focal length of the imaging lens 320 is 90mm; the pixel size of the image acquisition device is 2.4μm; and the image acquisition device is a camera.
[0033] Furthermore, the imaging lens 320 is one of the core optical components constituting the first detection unit 310, and its focal length directly determines the detection sensitivity of the first detection unit 310 to the angular deviation of the emitted beam. According to the optical imaging formula y=F·tanθ, where y is the spot displacement caused by the beam angular deviation on the photosensitive surface of the image acquisition device, F is the focal length of the imaging lens 320, and θ is the actual angular deviation of the emitted beam, it can be seen from this formula that, for the same angular deviation θ, the larger the focal length F, the larger the spot displacement y on the photosensitive surface, i.e., the higher the detection sensitivity. In this embodiment, an imaging lens 320 with a focal length of 90mm is selected. Combined with the image acquisition device of a specific pixel size involved in subsequent discussions, it can convert the angular deviation at the micro-radian level into a measurable displacement of several pixels, achieving high-resolution detection of the 442 angular deviation of the Couder optical path. The 90mm focal length imaging lens 320 can be a fixed-focus industrial lens, with an aperture value preferably between F2.8 and F5.6 to ensure sufficient image plane illuminance and good imaging contrast under normal indoor lighting conditions. Of course, those skilled in the art will understand that in other embodiments, depending on different detection accuracy requirements and spatial layout limitations, the focal length of the imaging lens 320 can also be selected as 50mm, 75mm or 100mm or other specifications, all of which can achieve the angle detection function with the corresponding detection sensitivity. These alternative solutions are all equivalent embodiments of the present invention.
[0034] The pixel size of the image acquisition device is 2.4 μm. Pixel size refers to the physical size of a single pixel unit on the photosensitive surface of the image acquisition device, which determines the spatial resolution and detection accuracy of the image acquisition device. In this embodiment, the pixel size of the image acquisition device is 2.4 μm, meaning the distance between the centers of two adjacent pixel units on the photosensitive surface is 2.4 μm. When a beam angle deviation causes displacement of the light spot on the photosensitive surface, this displacement is measured in pixels, with each pixel corresponding to an actual physical displacement of 2.4 μm. Combined with the 90mm focal length imaging lens 320, the corresponding angle detection resolution for each pixel can be calculated to be approximately 27 μrad / pixel (i.e., 2.4 μm ÷ 90mm). (27μrad) The detection resolution can be further improved through sub-pixel positioning algorithms, thereby meeting the precision detection requirements of the 442-angle deviation of the Couder optical path. In this embodiment, the use of an image acquisition device with a pixel size of 2.4μm is a preferred solution, which ensures sufficient detection resolution while also taking into account the dynamic range and signal-to-noise ratio of the photosensitive surface. In other embodiments, image acquisition devices with other specifications such as 1.85μm, 3.45μm, or 4.8μm pixel sizes can also be selected, and these are all alternative implementation schemes.
[0035] Specifically, a camera is a concrete implementation of an image acquisition device. In this embodiment, the image acquisition devices in both the first detection unit 310 and the second detection unit 330 preferably use industrial cameras. Industrial cameras have advantages such as small size, high frame rate, stable image data transmission, and easy communication with computers, making them suitable for precision optical inspection scenarios. The camera in the first detection unit 310 is preferably a monochrome industrial camera to obtain higher sensitivity and resolution, avoiding the resolution loss caused by Bayer filters in color cameras. The camera in the second detection unit 330 can use the same model of industrial camera as the first detection unit 310 to facilitate unified driving and image acquisition parameter settings. The camera's data interface preferably uses a USB 3.0 or GigE interface to achieve high-speed real-time transmission of image data, ensuring that operators can observe the real-time position and trajectory of the light spot on the photosensitive surface without delay during debugging.
[0036] Example 2 Please see Figures 1-3 This invention provides a method for debugging the Kuder optical path 442. The debugging method described in this embodiment uses the Kuder optical path 442 debugging fixture described in the previous embodiment. This method achieves independent convergence of errors of each mechanical axis by grouping the Kuder reflectors into pairs and debugging them independently axis by axis, thereby avoiding the accumulation and transmission of errors during the debugging process.
[0037] In this embodiment, the debugging method of the Kuder optical path 442 includes the following steps: S1: A reference element 420 is provided at the incident end of the Kuder optical path 442, such that the center of the reference element 420 is aligned with the optical axis center of the Kuder optical path 442. like Figure 3 As shown, the Kuder optical path 442 is generally composed of an internal focusing telescope, a collimating mirror (as a reference element 420), Kuder mirror I, Kuder mirror II, and Kuder adjustment fixture arranged sequentially along the optical path.
[0038] The internal focusing telescope, acting as a collimating light source 410, is positioned at the very front of the optical path to emit a collimated beam and provide autocollimation observation functionality. The collimating lens, acting as a reference element 420, is positioned at the incident end of the Kud optical path 442 to provide angular and positional references for the entire adjustment process. Kud reflectors I and II form a pair, corresponding to the adjustment of mechanical axis I. Kud reflector I is the first optical adjustment element 431, primarily responsible for adjusting the beam deflection angle, while Kud reflector II is the second optical adjustment element 432, primarily responsible for adjusting the spatial position of the beam. The Kud adjustment fixture is positioned at the exit end of the Kud optical path 442 to detect the angle and position information of the emitted beam in real time.
[0039] It should be noted that, for ease of description and understanding, this embodiment uses the debugging of one mechanical axis (mechanical axis I) with two Kud mirrors (Kud mirror I and Kud mirror II) as an example. In practical applications, the Kud optical path 442 with more Kud mirrors and multiple mechanical axes can be debugged in the same way, that is, Kud mirrors III and IV debug mechanical axis II, Kud mirrors V and Kud mirror VI debug mechanical axis III, and so on. The principle and steps are completely consistent with the following embodiment.
[0040] Specifically, the reference element 420 is preferably a planar reflective element (hereinafter referred to as a planar mirror or collimating mirror) marked with graduations 421. The main body of the planar mirror is a high-precision optical planar reflector with a surface shape accuracy of high (preferably above λ / 10), capable of reflecting the incident light back along its original path. The central area of the planar mirror is engraved with fine graduations 421, which are preferably crosshairs or concentric rings. The width of the graduations 421 is preferably 0.1 mm, used to assist the operator in accurately aligning the center of the planar mirror with the optical axis center of the Kud optical path 442. In specific operation, the planar mirror is fixed to the incident end of the Kud optical path 442 by structural components (such as positioning brackets or adjustable mounting bases). The operator uses mechanical centering tools or the observation function of the internal focusing telescope to adjust the spatial position of the planar mirror so that the geometric center of the planar mirror is aligned with the mechanical center of the entrance of the Kud optical path 442, that is, the center of the planar mirror falls on the extension line of the optical axis of the Kud optical path 442. The mounting plane of the flat mirror should be perpendicular to the optical axis direction at the entrance of the Kuder optical path 442 to ensure that the reflecting surface can reflect the incident beam back along the original path. Once the reference element 420 is installed and adjusted, it will not be moved during the entire adjustment process, providing a fixed reference for all subsequent adjustment steps.
[0041] S2: The collimated light source 410 emits a collimated beam to the reference element 420, and the collimated light source 410 is adjusted to align the center of its collimated optical axis 441 with the center of the reference element 420 to achieve self-collimation and establish a global reference optical axis. In this embodiment, the collimating light source 410 is preferably an internally focused telescope. This instrument integrates collimating beam emission and autocollimation observation functions. It contains a light source, a reticle, a focusing lens group, and an eyepiece system. The position of the focusing lens group can be changed by the focusing handwheel, so that target objects at different distances can be clearly imaged on the reticle image plane. At the same time, the overall attitude and spatial position of the telescope can be finely adjusted by the external pitch adjustment mechanism, yaw (heading) adjustment mechanism, and two-dimensional translation mechanism.
[0042] The operator first aligns the internal focusing telescope with the installed reference element 420, and rotates the focusing handwheel to move the focusing lens group back and forth along the optical axis until the scale mark 421 on the surface of the reference element 420 appears as a clear image on the eyepiece reticle. At this point, both the crosshairs on the reticle and the crosshair image reflected back from the reference element 420 can be observed simultaneously through the eyepiece. After focusing, the operator proceeds to the position calibration step: adjusting the two-dimensional translation mechanism (up and down and left and right) of the internal focusing telescope so that the center of the crosshair image reflected back from the reference element 420 observed in the eyepiece coincides with the center of the crosshairs on the reticle. This ensures that the output beam of the internal focusing telescope is aligned with the central area of the reference element 420—at this point, the collimated beam illuminates the central crosshair mark 421 on the reflective surface of the reference element 420, and the position calibration is complete.
[0043] The operator then proceeds to the angle calibration step: While maintaining the position calibration state, the light source of the internal focusing telescope is activated, emitting a collimated beam forward. This beam illuminates the reflector of reference element 420, is reflected back along its original path, and re-enters the imaging system of the internal focusing telescope. The operator observes the reflected crosshair image through the eyepiece, simultaneously adjusting the pitch and yaw (heading) adjustment mechanisms of the internal focusing telescope to gradually bring the reflected crosshair image closer to the center of the reticle's crosshair in the eyepiece's field of view until it completely coincides with it. When the reflected crosshair image completely coincides with the reticle's crosshair, it indicates that the optical axis of the internal focusing telescope is perfectly aligned with the normal direction of the reflector of reference element 420; that is, the outgoing beam from the internal focusing telescope is incident perpendicularly to the reflector of reference element 420—at this point, the angle calibration is complete.
[0044] After both position and angle calibrations are completed, the reference element 420 (a planar reflective element marked with crosshair 421) and the internal focusing telescope jointly establish an absolutely straight optical axis in space with a defined direction and position: the outgoing beam from the internal focusing telescope passes through the center of the reference element 420 (position reference determined) and is incident perpendicularly to the reflective surface of the reference element 420 (angle reference determined). This optical axis is the global reference optical axis upon which all subsequent mechanical axis adjustments depend. It should be noted that although position calibration and angle calibration have different physical mechanisms, they may have overlapping effects in actual operation—for example, angle adjustment may slightly change the position of the optical axis aligned with the center. Therefore, operators usually need to perform one or two iterative fine adjustments between position and angle calibrations until both dimensions simultaneously meet the alignment requirements. Once this global reference optical axis is established, it remains unchanged throughout the entire adjustment process. The adjustment of each subsequent set of Coudé mirrors is based solely on this global reference optical axis, rather than on the outgoing beam state after the previous set of adjustments. This is the key to the gradual accumulation and propagation of cutoff errors along the optical path. In an alternative embodiment, a collimator can be used in conjunction with an external autocollimating eyepiece to achieve the above-mentioned autocollimation function. The core of this method is to make the light beam emitted by the collimating light source 410 both aligned with the center position of the reference element 420 and incident perpendicularly to its reflecting surface, thereby forming a global reference optical axis.
[0045] S3: Set the Kude optical path debugging fixture at the output end of the Kude optical path 442; After establishing the global reference optical axis, the Kuder optical path debugging fixture is fixedly installed at the output end of the Kuder optical path 442. Specifically, the operator uses screws or positioning pins to detachably fix the entire debugging fixture to the equipment base at the output end of the Kuder optical path 442 using the first type of fixing lug 122 on the support frame 100, ensuring that the debugging fixture remains stable during operation. During installation, the optical working surface of the beam splitter 200 must be located on the optical axis of the output beam, and the angle between the beam splitter 200 and the optical axis of the output beam must be 45°. At this time, the beam splitter 200 splits the output beam into two paths: a first transmitted beam 211 and a second reflected beam 212. The first detection unit 310 is located on the optical path of the first beam 211, and the second detection unit 330 is located on the optical path of the second beam 212. The structural precision of the integrated machining of the first arm 110 and the second arm 120 on the support frame 100 ensures that the optical paths of each detection unit and the corresponding beam automatically coincide, eliminating the need for on-site calibration. Once installed, the relative positions of the beam splitter 200, the first detection unit 310, and the second detection unit 330 are precisely fixed by the support frame 100. The entire debugging fixture is in a state of readiness for testing. It can collect and display the spot image of the beam emitted from the Kuder optical path 442 in real time through the two detection units, providing real-time feedback of angle and position information for subsequent group-by-group debugging.
[0046] S4: Divide the multiple Couder mirrors in the Couder optical path 442 into several groups according to their corresponding mechanical axes, and each group includes at least a first optical adjustment element 431 and a second optical adjustment element 432. After the installation of the debugging fixture is completed, the operators divide all the Coud mirrors into several groups according to the mechanical axis assignments of each Coud mirror in the Coud optical path 442. Taking a typical Coud optical path with four Coud mirrors and two mechanical axes as an example, Coud mirror I and Coud mirror II correspond to mechanical axis I and are divided into the first group; Coud mirror III and Coud mirror IV correspond to mechanical axis II and are divided into the second group. In each group, the mirror mainly responsible for adjusting the beam deflection angle is defined as the first optical adjustment element 431, and the mirror mainly responsible for adjusting the spatial position of the beam is defined as the second optical adjustment element 432; for example, in the first group corresponding to mechanical axis I, Coud mirror I is configured as the first optical adjustment element 431, and Coud mirror II is configured as the second optical adjustment element 432. In the Couder optical path 442, which has more Couder mirrors and more mechanical axes, the mirrors are grouped sequentially according to the same rules. Couder mirrors V and VI correspond to mechanical axis III and are grouped into the third group, and so on. Each group contains at least one first optical adjustment element 431 and one second optical adjustment element 432, and each group is bound to its corresponding mechanical axis. The mirrors in each group are arranged sequentially along the optical path in physical space, but are independent of each other in their debugging logic. Once the position and angle of the previous group of mirrors are adjusted, they are locked and no longer used as the debugging reference for the next group. The debugging of the next group of mirrors will directly use the global reference optical axis established in S2 as the sole reference, thus achieving complete decoupling between the groups in the debugging strategy. This grouping strategy ensures that the error of each mechanical axis is independently converged within its corresponding tolerance range. The debugging results of any group of mirrors will not have a coupling effect on another group. Operators do not need to pay attention to the cumulative amount of errors in the preceding stage; they only need to complete the debugging of each group according to a unified standard procedure.
[0047] S5: Perform the following debugging steps on the current group: S51: Rotate the mechanical axis corresponding to the current group, observe the trajectory of the emitted beam during the rotation process through the first detection unit 310, adjust the first optical adjustment element 431 in the group, so that the angular offset of the beam during the rotation process is not greater than the first preset threshold, and complete the angular domain locking of the current group. After grouping, the operator performs debugging for the currently selected group. Debugging of this group does not involve adjusting the angle and position simultaneously, but rather proceeds in stages, strictly following the order of angle first, then position. First, the angle domain locking step of S51 is executed. Once the angle deviation converges to the acceptable range and is locked, the subsequent position convergence stage begins. This staged strategy ensures that angle and position debugging do not interfere with each other, avoiding the coupling oscillation problem found in traditional methods.
[0048] Specifically, the operator slowly rotates the mechanical axis corresponding to the current group (e.g., mechanical axis I), causing the mechanical axis to rotate around its own axis, while continuously observing the display screen of the first detection unit 310 during the rotation. Since the first detection unit 310 consists of an imaging lens 320 with a focal length of 90mm and an image acquisition device with a pixel size of 2.4μm, it is extremely sensitive to changes in the beam angle. When the emitted beam deviates angularly from the global reference optical axis established in S2, this deviation, after being imaged by the imaging lens 320, manifests as a circular motion trajectory of the light spot as the mechanical axis rotates on the photosensitive surface of the image acquisition device. By observing the radius of this light spot trajectory, the operator can intuitively judge the degree of the current angular deviation; the larger the trajectory radius, the greater the angular deviation of the emitted beam from the global reference optical axis.
[0049] Subsequently, the operator adjusts the first optical adjustment element 431 (i.e., the Couder I) in the group, which is mainly responsible for changing the deflection angle of the light beam. During the adjustment process, the operator continuously observes the changes in the light spot trajectory displayed on the screen of the first detection unit 310, and repeatedly fine-tunes the pitch and yaw angles of the first optical adjustment element 431 until the radius of the light spot's trajectory in the screen does not exceed a first preset threshold during the entire rotation of the mechanical axis (preferably 180°). In this embodiment, the first preset threshold is preferably 2 pixels, that is, the position change of the light spot on the photosensitive surface does not exceed 2 pixels before and after the mechanical axis rotates 180°. Since the pixel size of the first detection unit 310 is 2.4μm and the focal length of the imaging lens 320 is 90mm, when the light spot offset does not exceed 2 pixels, the corresponding outgoing beam angle deviation is approximately 5.45μrad, which is better than the design requirement of 0.01mrad.
[0050] After the radius of the light spot trajectory meets the above threshold requirement, the operator fixes the pose of the current first optical adjustment element 431, that is, completing the angle domain locking of the current group. It should be particularly noted that the "locking" here means that the angle domain debugging result has become a fixed constraint for subsequent operations; in the subsequent position convergence stage (S53, S54), the iterative adjustment of position deviation shall not damage the locked angle accuracy, and the first optical adjustment element 431 is only used as a callback intervention means when the angle exceeds the limit during the position iteration process, rather than being used as the main adjustment element for position deviation. Through this domain-separated locking mechanism of "locking the angle first and then converging the position", the angle error and the position error always remain in a decoupled state during the debugging process, and the angle debugging will not be interfered by subsequent position adjustment actions after being completed, ensuring that each dimension of error can independently converge to its respective qualified range. The judgment of the first preset threshold can be quantitatively judged by the operator by directly reading the pixel coordinates of the light spot through the display software matched with the first detection unit 310, or the data processing terminal can automatically identify the centroid position of the light spot and calculate the offset in real time, and the operator only needs to judge whether the threshold requirement is met according to the reading in the display screen, without performing complex optical calculations.
[0051] S52: Acquire the position coordinates of the exit beam before and after the rotation of the mechanical axis through the second detection unit 330, and respectively calculate the position offset in each coordinate axis direction; After the angle domain locking of the current group is completed in S51, the operator acquires the position coordinates of the exit beam before and after the rotation of the mechanical axis through the second detection unit 330. The second detection unit 330 is an image acquisition device without an imaging lens 320, and its photosensitive surface directly receives the second light beam 212 reflected by the light splitting element 200. Since there is no imaging lens 320, the position of the light spot of the reflected light beam on the photosensitive surface directly reflects the translational position of the exit beam in space, and is not sensitive to the angle change of the light beam, that is, the angle deviation will not cause significant displacement on the photosensitive surface of the second detection unit 330, thereby realizing independent extraction of the position information of the light beam.
[0052] Specifically, before the mechanical axis rotates, the operator first observes the position of the current light spot on the photosensitive surface through the display device connected to the second detection unit 330. Using image processing algorithms (such as the centroid method or grayscale centroid method), the center position of the light spot is extracted, and the coordinate value of this center position is read using a preset coordinate system on the photosensitive surface as a reference. This coordinate value is recorded as the first position coordinate (X1, Y1). To ensure the accuracy of the coordinate readings, the operator preferably acquires multiple frames of images and calculates the average value of the light spot center coordinates to eliminate the influence of image sensor noise and random jitter on the coordinate readings. Subsequently, the operator rotates the mechanical axis corresponding to the current group 180° around its own axis. After the rotation is complete and the mechanical axis stabilizes, the coordinate value of the light spot on the photosensitive surface is read again through the second detection unit 330 and recorded as the second position coordinate (X2, Y2). It should be noted that during the rotation of the mechanical shaft, the entire debugging fixture remains fixed and must not be touched or moved. Any part of the beam splitter 200, the first detection unit 310, or the second detection unit 330 must be ensured to be acquired under the same reference system, thereby guaranteeing the authenticity and validity of the calculation results.
[0053] After obtaining the first and second position coordinates, the operator or data processing terminal calculates the positional offset of each coordinate axis before and after the mechanical axis rotation as follows: positional offset ΔX in the X-axis direction = X2 - X1, and positional offset ΔY in the Y-axis direction = Y2 - Y1. The sign of ΔX indicates the offset direction of the light spot in the X-axis direction, and the sign of ΔY indicates the offset direction of the light spot in the Y-axis direction. The absolute values of ΔX and ΔY represent the degree of positional offset in the two orthogonal directions, both in pixels. By decomposing the positional offset into independent components in the two orthogonal directions of the X and Y axes, subsequent iterations in S53 and S54 can be performed independently in both directions without interference—this is completely different from the traditional method of coupled adjustment of two-dimensional positional offset, where adjusting one direction often causes a change in the other, leading to repeated oscillations and difficulty in convergence during the adjustment process.
[0054] It is important to emphasize that the measurement operation in S52 is performed under the premise that the angle domain in S51 is locked. Since the second detection unit 330 is not sensitive to angle changes, the locked state of the angle domain will not substantially interfere with the position coordinate measurement results of S52. Conversely, the measurement results of S52 are also completely independent of the state of the angle domain, and the two have been completely decoupled at the physical detection level. This dual-domain separation detection architecture allows operators to focus only on the pixel displacement of the light spot on the X and Y axes in the display screen of the second detection unit 330 during the subsequent position iterations in S53 and S54, without worrying about crosstalk of angle deviations to the position readings, and without having to repeatedly switch detection equipment, which greatly reduces the difficulty of judgment and the risk of misoperation during the debugging process.
[0055] S53: While keeping the angular offset no greater than the first preset threshold, adjust the second optical adjustment element 432 in the group to make the beam offset along a certain coordinate axis direction, and then adjust the first optical adjustment element 431 to make the beam return to the reference position. Repeat this process until the position offset in the coordinate axis direction is no greater than the second preset threshold. After measuring the position offsets ΔX and ΔY in the X and Y axes in S52, the operator enters the position convergence phase. The core of this phase is: under the premise of the angle domain locked in S51, by alternately adjusting the second optical adjustment element 432 and the first optical adjustment element 431, the position offset in a certain coordinate axis direction is gradually approximated and eliminated in an iterative manner. The adjustment operation uses the display screen of the first detection unit 310 as the basis for real-time monitoring. Since the first detection unit 310 is sensitive to angle changes, the offset of the light spot in its screen relative to the reference zero position (i.e., the angle zero position corresponding to the global reference optical axis) reflects in real time whether the current angle locking state has been broken. When adjusting any optical adjustment element, the operator always prioritizes the screen of the first detection unit 310 to ensure that the angle offset never exceeds the first preset threshold.
[0056] Taking the initial adjustment of the X-axis direction as an example, the operator adjusts the second optical adjustment element 432 in this group (i.e., the Couder mirror mainly responsible for adjusting the spatial position of the beam). During the adjustment process, the operator continuously observes the screen of the first detection unit 310. After confirming that the angle offset is within the first preset threshold, the operator shifts attention to the screen of the second detection unit 330 and reads the displacement of the light spot along the X-axis. Subsequently, the operator adjusts the first optical adjustment element 431 in this group (i.e., Couder mirror I). In this step, the operator must closely monitor the screen of the first detection unit 310 for adjustment—because the zero position of the first detection unit 310 is the reference state for angle locking. The purpose of adjusting the first optical adjustment element 431 is to compensate for any angle disturbances that may be introduced during the adjustment of the second optical adjustment element 432, so that the angle returns to the locked state. When the light spot in the screen of the first detection unit 310 returns to the reference zero position (i.e., the angle offset recovers to no greater than the first preset threshold), it indicates that this round of compensation operation is complete.
[0057] The above-described process of "adjusting the second optical adjustment element 432 → confirming the angle state through the first detection unit 310 → adjusting the first optical adjustment element 431 to return the image of the first detection unit 310 to the reference zero position" constitutes a complete operation. Its physical essence is: introducing a positional change of known direction and magnitude through the second optical adjustment element 432, and then compensating for any potential angular disturbance introduced by this change back to the zero position through the first optical adjustment element 431, thus restoring the angle-locked state. In this operation, the second detection unit 330 is only used to initially read the displacement of the light spot along the X-axis, and as a reading tool after each operation when rotating the mechanical axis 180° to verify the positional offset; it is not used as a real-time feedback basis during the adjustment process.
[0058] After each round of the above operations, the operator rotates the mechanical axis 180° around its own axis, and then reads the current coordinates of the light spot after rotation through the second detection unit 330. The current coordinates are compared with the reference coordinates recorded before rotation, and the position offset ΔX' in the coordinate axis direction is recalculated in the same way as in S52. This ΔX' is the residual position offset in the X-axis direction after this round of iteration. The operator compares the absolute value of ΔX' with the second preset threshold (preferably 10 pixels in this embodiment): if |ΔX'| is still greater than 10 pixels, the above operation process is repeated until |ΔX'| is no greater than 10 pixels. During this process, the operator always uses the image of the first detection unit 310 as the real-time judgment basis when adjusting the first optical adjustment element 431; that is, "the zero position of the first detection unit is the reference". This reference is both the result of the angle domain locking in S51 and the target state that needs to be returned to each time the first optical adjustment element 431 is adjusted in the S53 iteration process, ensuring that the angle domain is completely restored to the locked state after each round of "second adjustment + first adjustment" operation. After the position offset converges to within the second preset threshold through multiple iterations, the position convergence in the X-axis direction is completed. Through this division of labor and cooperation method of "using the first detection unit as the real-time basis during adjustment and the second detection unit as the reading basis during verification", the operator can efficiently converge the position deviation while ensuring that the angle lock is not shaken, thus realizing the true decoupling of the angle domain and the position domain at the operational level.
[0059] In this embodiment, operators do not need to rely on complex mathematical models or precise calculations. They only need to follow the cyclical steps of "first deflection, then zeroing—rotation verification—reading and judgment" to complete the debugging, which significantly reduces the debugging threshold and the dependence on the operator's experience level.
[0060] S54: Adjust the position offset in the other coordinate axis direction according to the method of S53 until the position offset in the coordinate axis direction is not greater than the second preset threshold. After completing the X-axis position offset adjustment in S53, the operator adjusts the position offset of the other coordinate axis direction—the Y-axis direction—in the same way. Specifically, the operator uses the current Y-axis position offset ΔY as the adjustment target and repeats the same iterative operation as in S53: first, adjust the second optical adjustment element 432 in the group, observe the light spot move a certain displacement along the Y-axis direction through the display screen of the first detection unit 310, keep the second optical adjustment element 432 stationary, then adjust the first optical adjustment element 431 to make the light spot return to the reference position, then rotate the mechanical axis 180° again, read the coordinate difference before and after the rotation through the second detection unit 330 to calculate the current residual position offset ΔY' in the Y-axis direction, compare |ΔY'| with the second preset threshold (10 pixels), and if it is still greater than the threshold, repeat the above iterative process until it is not greater than the threshold.
[0061] Throughout the entire adjustment process along the Y-axis, the already adjusted X-axis results remain unaffected. Since the X-axis and Y-axis coordinates on the photosensitive surface of the second detection unit 330 are orthogonal and read independently, when the operator adjusts the second optical adjustment element 432 to shift the light spot along the Y-axis, the X-axis coordinate value remains essentially unchanged. Furthermore, the action of the first optical adjustment element 431 to return the light spot to its reference position is also performed on a two-dimensional plane. By observing the real-time coordinates of the light spot on the display screen, the operator only needs to focus on whether the Y-axis coordinate value has returned to the reference position; the X-axis coordinate value is already qualified and therefore requires no further adjustment. This characteristic of independent iteration and non-interference in each coordinate axis direction allows the positional deviations in both directions to converge to their respective acceptable ranges, avoiding the recurring oscillation problem common in two-dimensional coupled adjustment where "one axis is adjusted, but the other axis deviates."
[0062] During the Y-axis position offset adjustment process, similar to S53, the operator must always ensure that the angle offset does not exceed the first preset threshold. In each iteration, the operator must confirm that the angle deviation displayed on the first detection unit 310 screen does not exceed the limit. If the angle offset exceeds the first preset threshold due to the adjustment action of the first optical adjustment element 431, the Y-axis position adjustment is immediately paused. The first optical adjustment element 431 is then reverted to restore the angle offset to the acceptable range before continuing the Y-axis iterative convergence. When the Y-axis position offset before and after a 180° rotation of the mechanical axis does not exceed the second preset threshold (10 pixels), step S54 is completed. At this point, the deviations in the X-axis, Y-axis, and angle offset dimensions of the current group have all independently converged to within their respective preset thresholds, laying the foundation for the qualification judgment in S55.
[0063] It is worth noting that operators can choose to debug the X-axis first and then the Y-axis, or debug the Y-axis first and then the X-axis, depending on actual needs. Both orders can achieve the same convergence effect. The description of the order of X-axis first and then Y-axis in this embodiment is only for ease of understanding and does not constitute a limitation on the debugging order.
[0064] S55: When the mechanical axis rotates before and after rotation, the angular offset detected by the first detection unit 310 is not greater than the first preset threshold and the position offset of each coordinate axis direction detected by the second detection unit 330 is not greater than the second preset threshold, the current group debugging is completed; After the operator completes the angle domain locking according to S51, the initial measurement of the position offset according to S52, and the iterative convergence of the position offset in the X-axis and Y-axis directions according to S53 and S54 respectively, a comprehensive judgment needs to be made on the debugging results of the current group. Specifically, the operator rotates the mechanical axis around its own axis (preferably 180°), and after the rotation is completed, simultaneously reads the detection data of the first detection unit 310 and the second detection unit 330: the first detection unit 310 is used to verify whether the angle offset is still not greater than the first preset threshold (i.e., 2 pixels), and the second detection unit 330 is used to verify whether the position offsets in the X-axis and Y-axis directions (i.e., the absolute values of ΔX and ΔY) are both not greater than the second preset threshold (i.e., 10 pixels). Only when all three judgment conditions are met simultaneously, i.e., angle offset ≤ 2 pixels and |ΔX| ≤ 10 pixels and |ΔY| ≤ 10 pixels, can the current group be considered to have passed the debugging. If any dimension fails to meet the corresponding threshold requirement, the corresponding debugging step must be returned to re-converge—if the angle is not qualified, return to S51; if a certain axial position is not qualified, return to S53 or S54, until all three dimensions meet the threshold conditions.
[0065] It is particularly important to note that the judgments in the above three dimensions are based on an "AND" logic relationship, not an "OR" logic relationship. Failure in any dimension constitutes a sufficient condition for the current group's debugging to be incomplete. This multi-dimensional joint judgment method ensures that the mechanical axis corresponding to the current group simultaneously meets the design requirements in both the angle and position dimensions, avoiding a "false pass" situation where one dimension is passable while another exceeds the tolerance. When all three dimensions meet the threshold requirements, the current group's debugging is complete. The operator then locks and fixes all the pose adjustment mechanisms (such as pitch adjustment screws, yaw adjustment screws, and translation adjustment mechanisms) of the first optical adjustment element 431 and the second optical adjustment element 432 in the current group, ensuring that the pose of the reflectors in this group does not change during subsequent debugging. At this point, the mechanical axis error corresponding to the current group has independently converged to the acceptable range in both the angle and position domains, and the angle and position are decoupled and do not interfere with each other. The debugging results of this group are completely locked. Once locked, the current group does not participate in the establishment of the debugging benchmark for any subsequent group of reflectors. The debugging of subsequent groups will restart with the global reference optical axis established in S2 as the sole benchmark. This ensures that the error of each group of reflectors is independent of other groups, thus breaking the physical chain of error accumulation and transmission in the debugging process. In actual operation, to enhance the reliability of the judgment results, the operator can rotate the mechanical axis completely one full turn (i.e., 360°) after the qualified judgment and observe whether the movement trajectory of the light spot on the display screen of the first detection unit 310 and the second detection unit 330 is always within the qualified range during the entire rotation. If there is an out-of-tolerance error during the full 360° stroke, it indicates that there may be non-debugging factors such as the shaking of the mechanical axis system itself or the loose installation of the reflector. These factors need to be investigated and eliminated before re-verification.
[0066] S6: Lock the pose of all optical adjustment elements in the current group, and then use the global reference optical axis as the only reference to independently debug the next group in the manner of S5 until all groups are debugged; during the debugging of any group, the state of the outgoing beam corresponding to that group does not participate in the reference establishment of the next group.
[0067] Specifically, for the first optical adjustment element 431 (i.e., Kuder mirror I) and the second optical adjustment element 432 (i.e., Kuder mirror II) in the current group, the operator locks all the attitude adjustment mechanisms on their mounting bases, including the pitch adjustment locking screw, the yaw adjustment locking screw, and the translation adjustment locking mechanism, to ensure that the spatial position and angular attitude of the two mirrors in this group will not change in any subsequent debugging process. After locking, the debugging error of this group of mirrors has been fixed within its respective qualified range, and the direction of the emitted beam of this group of mirrors is also fixed.
[0068] Subsequently, the operator shifts the focus of debugging to the next set of reflectors. Before starting the debugging of the next set, the operator must understand that the starting reference for the debugging of the next set is not the actual state of the emitted beam after the debugging of the current set, but the global reference optical axis established in S2. In specific operation, the operator does not use the optical axis of the emitted beam of the current set as the debugging reference for the next set of reflectors, but realigns the incident light path of the next set of reflectors with the global reference optical axis; that is, the adjustment target of the first optical adjustment element 431 of the next set is to ensure that the angular deviation of the emitted beam of the corresponding set during the rotation of the mechanical axis is not greater than the first preset threshold. The reference for this angular deviation is the global reference optical axis, not the optical axis of the emitted beam of the current set. Since the emitted beam of the current set may have a small residual error (although it is qualified), if it is used as the reference, these residual errors will be inherited by the next set. However, by directly using the global reference optical axis as the reference, the debugging of the next set can be completely independent of the residual error of the current set, thus cutting off the chain of error accumulation and transmission along the optical path from the root.
[0069] After locking the current group and confirming that the next group uses the global reference optical axis as the reference, the operator independently debugs the next group according to the complete steps S51 to S55: first, rotate the mechanical axis corresponding to the next group, observe the angle deviation through the first detection unit 310, and adjust the first optical adjustment element 431 of the group to complete the angle domain locking; then, obtain the position coordinates and calculate the position offset through the second detection unit 330; then, while maintaining the angle domain locking, iteratively converge the position offsets in the X and Y axes to within the second preset threshold by alternately adjusting the second optical adjustment element 432 and the first optical adjustment element 431 of the group; finally, comprehensively determine that all three dimensions are qualified. During the debugging of each group, the state of the emitted beam corresponding to the group is only used to determine whether the group itself is qualified. After the group is debugged, the actual optical axis state of its emitted beam, regardless of its residual error, is not transmitted to the next group. That is, the starting reference of the next group is always the global reference optical axis, not the actual optical axis of the emitted beam of the previous group.
[0070] The operator sequentially completes the debugging of all groups one by one according to the above method until each group of reflectors corresponding to all mechanical axes converges independently to their respective qualified range. Taking a typical Couder optical path containing four Couder reflectors and two mechanical axes as an example, the operator first completes the debugging and locking of the first group (Couder reflector I and Couder reflector II) corresponding to mechanical axis I according to S51 to S55. Then, using the global reference optical axis as the sole reference, the operator completes the debugging and locking of the second group (Couder reflector III and Couder reflector IV) corresponding to mechanical axis II according to the same S51 to S55. At this point, all debugging is completed. Compared with the step-by-step transmission mode of "Couder I error → Couder II reference offset → Couder III error accumulation → Couder IV error further amplification" in the traditional sequential debugging method, the debugging process of this invention allows the error of each group to converge independently and not to be superimposed. The residual error of the first group is locked within its own group and is not transmitted to the second group. The debugging accuracy of the second group depends only on its own adjustment accuracy and the accuracy of the global reference optical axis, and is unrelated to the debugging results of the first group. This fully independent debugging architecture between groups ensures that the total error of the final output beam is equal to the envelope of the errors of each group rather than the algebraic sum of the errors of each group. This fundamentally overcomes the inherent defect of error accumulation and amplification along the optical path in existing technologies. At the same time, it allows operators to independently complete the debugging of each group without tracing the source of error, greatly reducing rework rate and debugging time.
[0071] Furthermore, in S6, the starting reference for the next group of debugging is the global reference optical axis established in S2, rather than the actual optical axis of the emitted beam after the current group's debugging is completed; after the current group's debugging is completed, its residual error is not propagated to the next group, and the angle deviation and position deviation of the next group are independent of the residual error of the current group.
[0072] In the debugging method of this invention, the global reference optical axis and the actual optical axis of each group of outgoing beams do not coincide in physical space. Since each group of reflectors has residual errors within an acceptable range, the actual optical axis of the outgoing beam of each group after debugging will have a slight deviation relative to the global reference optical axis. If the actual optical axis of the previous group of outgoing beams is used as the debugging reference for the next group, all residual errors of the previous group in the angular and positional domains will be inherited and superimposed by the next group as the "correct reference." Therefore, this invention explicitly defines the debugging reference for the next group as the global reference optical axis established in S2, rather than the actual optical axis of the outgoing beam after the current group has been debugged. Specifically, when the operator debugs the next group of reflectors, the reference zero point for the light spot trajectory radius observed by the first detection unit 310 and the light spot position offset measured by the second detection unit 330 is the ideal outgoing state corresponding to the global reference optical axis—that is, the spatial position and direction the outgoing beam should reach when all reflectors in the Kuder optical path 442 are in the ideal position, rather than the position of the light spot in the detection screen after the current group has been debugged.
[0073] The residual error of the current group after adjustment is not propagated to the next group. The essence of this technique is that the goal of adjusting the next group is to bring its corresponding emitted beam within a acceptable range relative to the global reference optical axis, rather than to make it identical to the emitted beam of the previous group. For example, if the previous group has a residual angular deviation of +1 pixel in the angular domain and a residual positional deviation of +5 pixels in the X-axis direction in the positional domain, and if the emitted beam of the previous group is used as the reference, the total error after the next group's adjustment is at least the algebraic sum of the residual errors of the previous group and the next group's own residual error (i.e., total angular error ≥ 1 + δ1 pixels, total positional error ≥ 5 + δ2 pixels). However, when using the global reference optical axis as the reference, the total angular error after the next group's adjustment is only δ1 pixels, and the total positional error is only δ2 pixels. δ1 and δ2 are the residual errors of the next group when it is adjusted to the acceptable range, and are completely independent of the +1 and +5 pixels of the previous group. Therefore, the adjustment accuracy of each group of reflectors depends only on the adjustment accuracy of that group itself and the adjustment quality of the operator, and is unrelated to the adjustment results of any previous group of reflectors. Since the residual error of each group is limited to its own preset threshold, the total error of the final emitted beam is equal to the envelope of the group with the largest error, rather than the algebraic sum of the errors of each group, which fundamentally avoids the amplification and accumulation of multi-level errors along the optical path.
[0074] At the operational level, at the start of each debugging session, operators use the preset reference coordinates on the photosensitive surface of the second detection unit 330 and the preset reference trajectory on the photosensitive surface of the first detection unit as references. These preset reference coordinates and trajectories are derived from the global reference optical axis, rather than the actual coordinates of the light spot on the photosensitive surface after the previous debugging session. Taking the position domain as an example, regardless of where the light spot is located on the display screen of the second detection unit 330 after the previous debugging session, the reference position referred to in "returning the beam to the reference position" in the next debugging session is always the same position; that is, the preset reference coordinates corresponding to the global reference optical axis. This debugging mode of "each group independently zeroing and each group restarting" eliminates the need for operators to record or trace the residual errors of the previous set of reflectors. The debugging process of each group starts from a unified global reference, and the debugging logic between groups is completely decoupled. Once all groups have been debugged, the output beam accuracy of the Kuder 442 optical path is determined by the envelope of the independent convergence accuracy of each group, rather than by the stepwise accumulation of errors from each group. This reliably guarantees the acceptance criteria of an output beam angle deviation better than 0.01 mrad and a composite position deviation better than 35 μm. Furthermore, if the final inspection reveals that the output beam accuracy is unacceptable, because the errors of each group are independent and do not propagate to each other, operators can quickly locate the group with the out-of-tolerance error based on the test data of each group. Only that group needs to be re-adjusted, eliminating the need to disassemble all the reflectors and readjust them step by step from scratch. This significantly shortens the rework cycle and reduces debugging costs.
[0075] Furthermore, during the iteration of S53, the angle offset detected by the first detection unit 310 is monitored in real time. If the angle offset exceeds the first preset threshold, the position adjustment is paused and the first optical adjustment element 431 is called back. Position adjustment continues only after the angle offset recovers to no greater than the first preset threshold.
[0076] In the debugging method of this invention, the angle domain locking completed in S51 is not a one-time solution. During the subsequent position iterations in S53 and S54, the operator's repeated adjustments to the second optical adjustment element 432 and the first optical adjustment element 431 may theoretically still cause minor disturbances to the accuracy of the locked angle domain. This is because although the first optical adjustment element 431 is mainly responsible for adjusting the beam deflection angle, its adjustment action may couple in additional position changes; similarly, although the second optical adjustment element 432 is mainly responsible for adjusting the beam spatial position, its adjustment action may also couple in minor angle changes. To address this bidirectional coupling disturbance, this step explicitly stipulates that throughout the entire position iteration process in S53, the operator must continuously monitor the angle offset detected by the first detection unit 310 in real time to ensure that it is never greater than the first preset threshold.
[0077] Specifically, in each round of position iteration, after the operator completes "adjusting the second optical adjustment element 432 to shift the light spot by a certain displacement along a certain coordinate axis" but before "adjusting the first optical adjustment element 431 to return the light spot to the reference position," the operator needs to read the current angle offset through the display screen of the first detection unit 310. After completing the zeroing adjustment of the first optical adjustment element 431, the operator reads the display screen of the first detection unit 310 again to confirm whether the angle offset is still within the acceptable range. More preferably, the operator can read the detection data of the first detection unit 310 three times at the beginning, middle, and end of each round of iteration to capture angle overshoot or angle drift that may occur during the adjustment process. If the angular offset exceeds the first preset threshold (i.e., 2 pixels) during any real-time monitoring, the operator immediately pauses the ongoing position adjustment operation and stops adjusting the second optical adjustment element 432. Instead, the operator prioritizes reverting the first optical adjustment element 431—that is, while keeping the current pose of the second optical adjustment element 432 unchanged, the operator fine-tunes the pitch and yaw angles of the first optical adjustment element 431 to restore the radius of the light spot trajectory displayed on the first detection unit 310 screen to no greater than the first preset threshold. After the angular offset is confirmed to have returned to the acceptable range, the operator resumes the previously paused position adjustment process, continuing subsequent operations such as returning the light spot to the reference position or verifying the rotation of the mechanical axis.
[0078] To ensure the smooth execution of the "pause-callback-resume" process, the operator must maintain constant observation of the display screen of the first detection unit 310 during position iteration. The display screens of the first detection unit 310 and the second detection unit 330 can be set side-by-side on the same display device or displayed simultaneously via split-screen, allowing the operator to monitor the real-time status of both the angle and position domains during the adjustment process. If an angle offset exceeds the limit in a certain iteration, the spot offset adjustment action already performed (i.e., the position change introduced by the second optical adjustment element 432) is retained after the first optical adjustment element 431 is called back. The callback action only corrects the angle domain and does not change the pose of the second optical adjustment element 432. After the angle domain returns to acceptable levels, the operator continues to perform the operation of returning the spot to the reference position that was not completed in this round, and then rotates the mechanical axis to verify the position offset after this iteration. Through the above-mentioned "real-time interlock verification during position iteration" mechanism, even if the adjustment action of the second optical adjustment element 432 inevitably causes a slight disturbance to the angle domain, the disturbance will be captured by the first detection unit 310 and eliminated in time, so that the angle domain remains locked throughout the entire position convergence process and will not gradually drift out of the qualified range due to the cumulative effect of the position adjustment action.
[0079] In extreme cases, if the angle offset exceeds the limit significantly in a certain iteration, and simply reverting the first optical adjustment element 431 cannot restore the angle offset to the acceptable range—for example, if the adjustment range is too large, requiring a significant revert to restore angle accuracy—the operator may choose to "re-execute the iteration." That is, after reverting the first optical adjustment element 431 to restore the angle to the acceptable range, the operator will restart the current position adjustment operation from "adjusting the second optical adjustment element 432 to offset the light spot by a certain displacement," and appropriately reduce the step size of the light spot offset to avoid triggering the angle over-limit again. This adaptive adjustment of the iteration step size is a conventional choice for those skilled in the art and does not affect the implementation of the core interlocking mechanism of this step. Through the closed-loop control described above, which involves "real-time monitoring - pausing position adjustment - calling back the first optical adjustment element - resuming position adjustment," the angle domain and the position domain remain independent and do not interfere with each other throughout the entire adjustment process. Even though the physical adjustment actions of the two optical adjustment elements are coupled in terms of mechanism, the interlocking verification mechanism in this step achieves complete decoupling at the functional level, ensuring that the angle adjustment results and the position adjustment results can converge independently and do not affect each other. This also avoids the repeated oscillation problem of "adjusting the position to deviate from the angle, adjusting the angle to deviate from the position" in traditional methods, further consolidating the effectiveness and reliability of the S51 to S54 domain-specific adjustment strategy.
[0080] Furthermore, the reference element 420 is a planar reflective element with a scale mark 421, which is located in the central region of the planar reflective element.
[0081] In this embodiment, the reference element 420 is preferably a planar reflective element marked with graduation 421, commonly referred to as a flat mirror or collimating mirror. The main body of the flat mirror is a high-precision optical planar reflector, and its substrate material is preferably K9 optical glass or fused silica, which has excellent optical uniformity and thermal stability, ensuring that the reflective surface shape remains stable even when the ambient temperature changes. One surface of the flat mirror is processed into a high-precision optical plane, and its surface shape accuracy is preferably better than λ / 10 (λ is the working wavelength, usually taken as 632.8nm), to ensure that the wavefront quality of the reflected beam meets the detection requirements; this high-precision optical plane is also coated with a high-reflectivity film layer to increase the reflectivity to the working wavelength, so that the beam reflected back to the collimating light source 410 has sufficient intensity, making it easy for the operator to clearly observe the self-collimated image. The flat mirror is fixed to the incident end of the Couder optical path 442 by structural components (such as positioning brackets or two-dimensional adjustable mounting bases). The scale mark 421 is located in the central area of the planar reflective element, preferably in the form of a crosshair scale line 421 or a concentric ring mark, with a preferred line width of 0.1 mm. The two lines of the crosshair scale 421 are perpendicular to each other, corresponding to two orthogonal directions in space; the concentric ring mark forms multiple equally spaced concentric rings with the center point as the center, used to help determine the degree of deviation of the light spot from the center. The geometric center of the scale mark 421 represents the center point of the entire reference element 420, providing a visual alignment reference for subsequent reference alignment operations. When installing the reference element 420, the operator compares the scale mark 421 with the pre-marked optical axis center mark at the entrance of the Kud optical path 442, and adjusts the translational and rotational degrees of freedom of the structural components to precisely align the center of the scale mark 421 with the optical axis center of the Kud optical path 442. Simultaneously, in the subsequent self-collimation operation of step S2, the operator observes the beam image reflected from the reference element 420 through the collimating light source 410 (such as an internal focusing telescope). When the reflected image coincides with the reference line on the reticle of the collimating light source 410, and the center of the reflected image is simultaneously aligned with the center of the scale mark 421 on the plane mirror, precise self-collimation between the collimating light source 410 and the reference element 420 is achieved. The setting of the scale mark 421 transforms the "center alignment" operation from an abstract geometric concept into a concrete visual reference that can be observed and determined, significantly reducing the operational difficulty of establishing the reference and the dependence on the operator's experience.
[0082] In alternative embodiments, the 421 mark can also be a single diameter line combined with a center dot mark, or symmetrically distributed four quadrant marks, as long as they can provide a reference for the center position. These alternative forms are all equivalent embodiments of the present invention.
[0083] Furthermore, the collimating light source 410 is an internal focusing telescope. An internal focusing telescope is an optical instrument that integrates collimated beam emission and self-collimating observation functions. Internally, it typically includes a light source, a reticle, a focusing lens group, and an eyepiece system. During commissioning, after the operator turns on the light source of the internal focusing telescope, the light emitted by the light source passes through the reticle and focusing lens group and is emitted forward as a parallel beam. When this parallel beam illuminates the optical reflecting surface of the reference element 420 (i.e., the plane reflecting element marked with scale 421), it is reflected back along its original path and re-enters the internal focusing telescope. The returned reflected light is guided through the internal optical path system of the internal focusing telescope (including the focusing lens group and the semi-transparent, semi-reflective beam splitter) to the eyepiece system, allowing the operator to observe the reflected crosshair image or light spot image through the eyepiece.
[0084] The advantage of an internally focused telescope lies in its integrated transmission and reception functions. It can independently perform autocollimation without requiring additional observation equipment, making it convenient to use and suitable for various field environments. Furthermore, the focusing function of the internally focused telescope allows it to adapt to reference elements 420 at different distances. When the spatial layout of the Cood optical path 442 entrance changes, a clear image can be achieved simply by rotating the focusing handwheel, demonstrating strong adaptability to different field conditions.
[0085] In another alternative embodiment, the collimating light source 410 is a collimator. A collimator is also an optical instrument capable of producing a collimated beam. Its basic structure includes a light source, a reticle, and an objective lens (or collimating lens group) located at the focal point of the reticle. The light source illuminates the reticle located at the focal plane of the objective lens. The pattern on the reticle (such as crosshairs) is emitted outward as a collimated beam after passing through the objective lens, forming a collimated beam. Unlike internally focusing telescopes, collimators typically do not have built-in autocollimation observation capabilities. Therefore, when using a collimator as the collimating light source 410, an additional independent observation device (such as a theodolite with an autocollimating eyepiece or a dedicated autocollimating telescope) is required to observe the reflected image.
[0086] In practice, the operator fixes the collimator in a suitable position, causing it to emit a collimated beam that illuminates the reference element 420. The reference element 420 reflects the beam back to the collimator along its original path. The operator observes the position of the reflected reticle image relative to the reticle lines on the collimator itself using a self-collimating eyepiece or external observation device positioned on the side of the collimator's optical path. Based on this, the operator adjusts the collimator's orientation until the reflected image and the reticle lines are completely aligned, achieving self-collimation. The collimator, as a classic collimating light source 410 device, has advantages such as structural stability, high output beam parallelism, and good long-term reliability, making it suitable for debugging scenarios with high requirements for reference stability. Regardless of whether an internal focusing telescope or a collimator is used as the collimating light source 410, its core function is to emit a reference collimated beam with a known direction and establish a one-to-one correspondence between the direction of this beam and the normal of the reflecting surface of the reference element 420 through the self-collimation principle, providing a directional reference for subsequent Couder mirror debugging. Those skilled in the art can flexibly select one of the two collimating light sources 410 mentioned above according to the on-site equipment conditions, accuracy requirements and operating habits. Both can achieve the self-collimation reference establishment function required by the present invention.
[0087] Furthermore, in S53, after the beam is offset by a certain displacement along a certain coordinate axis, the second detection unit 330 reads the current coordinates and calculates the position offset in the coordinate axis direction.
[0088] In this embodiment, the core of step S53 lies in gradually approximating and eliminating the positional offset in a certain coordinate axis direction through an iterative strategy of first offsetting and then zeroing. Specifically, after the operator completes the acquisition and calculation of the position coordinates before and after the mechanical axis rotation for the first time, the initial values of the positional offsets in the X-axis and Y-axis directions have been obtained. Taking one of the coordinate axes (e.g., the X-axis) as the current debugging target, the operator first adjusts the angle of the second optical adjustment element 432 (i.e., the Kuder mirror II) corresponding to the mechanical axis, and observes the movement direction and distance of the light spot in real time through the display screen of the second detection unit 330, so that the light spot moves a certain displacement along the X-axis direction. For example, if the currently measured X-axis positional offset is 30 pixels, the operator can first adjust the second optical adjustment element 432 to move the light spot about 20 pixels along the X-axis. Subsequently, keeping the second optical adjustment element 432 stationary, the operator adjusts the first optical adjustment element 431 (i.e., the Kuder mirror I) corresponding to the mechanical axis, so that the light spot returns to the reference position (e.g., the center area of the screen or the initial coordinate position) in the display screen of the first detection unit 310. After completing the above set of "adjustment of the second optical adjustment element 432 + adjustment of the first optical adjustment element 431", it is not necessary to directly determine whether the axis is qualified, but rather to quantitatively confirm the effect of this adjustment. To this end, the operator rotates the mechanical axis again (also rotating 180°), and again reads the current coordinates of the light spot after rotation through the second detection unit 330. This current coordinate is then compared with the reference coordinates recorded before rotation, and the current position offset of the coordinate axis is calculated in the same way. The calculated value is the residual position offset of the axis after this iteration. The operator compares this residual offset with a second preset threshold (preferably 10 pixels). If the residual offset is still greater than the second preset threshold, it indicates that this iteration has not yet adjusted the axis to a qualified state, and the above process of "adjusting the second optical adjustment element 432 to shift the light spot - adjusting the first optical adjustment element 431 to return the light spot to the reference position - rotating the mechanical axis - reading the current coordinates and calculating the offset" needs to be repeated until the calculated position offset is not greater than the second preset threshold. Through quantitative measurement and calculation after each iteration, operators can objectively and accurately grasp the current debugging progress, avoiding insufficient or over-adjustment caused by subjective judgment alone. At the same time, each iteration gradually reduces the residual position offset, which can converge to the qualified range after several iterations.
[0089] In this embodiment, the operation of "reading the current coordinates and calculating the position offset in the direction of the coordinate axis" needs to be performed in each iteration. It is completely consistent with the calculation method in step S5, that is, the offset is determined by the coordinate difference before and after the mechanical axis rotates. The only difference is that step S5 is the initial measurement, while this step is to re-measure and verify the residual offset after each adjustment to ensure that the results of the debugging process are controllable, quantifiable and traceable.
[0090] Furthermore, the first preset threshold is configured such that when the spot offset detected by the first detection unit 310 does not exceed 3 pixels, the corresponding outgoing beam angle deviation is not greater than 8.17 μrad. In this embodiment, the first preset threshold essentially includes two dimensions: a detection and judgment standard at the pixel level and a precision index at the physical level. As mentioned earlier, the first detection unit 310 consists of an imaging lens 320 with a focal length of 90mm and an image acquisition device with a pixel size of 2.4μm. According to the principle of geometrical optics imaging, when there is an angular deviation θ in the outgoing beam, the spot displacement y caused by this angular deviation on the photosensitive surface of the image acquisition device after imaging by the imaging lens 320 satisfies the following relationship: y=F·tanθ, where F is the focal length of the imaging lens 320 and θ is the angular deviation of the outgoing beam relative to the collimating optical axis 441. When the spot offset does not exceed 3 pixels, since the physical size of each pixel is 2.4 μm, the maximum displacement of the spot on the photosensitive surface is 3×2.4 μm=7.2 μm (i.e., 0.0072 mm). Substituting F=90 mm and y=0.0072 mm into the above formula, the corresponding angular deviation θ=arctan(y / F)=arctan(0.0072 / 90) can be calculated. 8.17×10 -6 rad, or 8.17 μrad. Therefore, configuring the first preset threshold as "spot offset not exceeding 3 pixels" and "outgoing beam angle deviation not exceeding 8.17 μrad" are physically equivalent, and a clear conversion relationship is established between the two through the optical imaging formula. In actual debugging operations, the operator directly observes and reads the pixel displacement of the spot through the display screen of the first detection unit 310. When the pixel displacement does not exceed 3 pixels, it means that the angle deviation of the current outgoing beam has been controlled within 8.17 μrad. This configuration method allows the operator to intuitively judge whether the angle deviation is qualified without having professional knowledge of optical calculation, thus reducing the technical threshold of debugging operations. It should be noted that in this embodiment, the acceptance standard for angle deviation is expressed in two ways: "2 pixels" and "3 pixels". The two are not contradictory: 2 pixels correspond to the strict acceptance threshold in actual operation, while 3 pixels correspond to the equivalent conversion boundary of the physical accuracy index of 8.17 μrad. In actual debugging, if 2 pixels are used as the acceptance threshold, the corresponding actual angle deviation is about 5.45 μrad, which is more stringent and more conducive to ensuring the final debugging quality.
[0091] The second preset threshold is configured such that when the spot offset of each coordinate axis direction detected by the second detection unit 330 does not exceed 10 pixels, the corresponding outgoing beam synthesis position deviation is not greater than 34.6 μm; In this embodiment, the second detection unit 330 is an image acquisition device without an imaging lens 320. The position of the light spot on its photosensitive surface directly reflects the actual translational position of the emitted light beam in space. The second detection unit 330 and the first detection unit 310 use the same image acquisition device with the same pixel size of 2.45 μm. When the light spot offset in each coordinate axis direction does not exceed 10 pixels, the maximum positional deviation in a single coordinate axis direction is 10 × 2.45 μm = 24.5 μm (i.e., 0.0245 mm), where the maximum positional deviation in the X-axis direction is denoted as... The maximum positional deviation in the Y-axis direction is denoted as The actual position deviation of the emitted beam in space is the resultant of the position deviations in two orthogonal directions, the X-axis and the Y-axis. According to the formula for calculating the distance between two points in a two-dimensional plane, the resultant position deviation... satisfy: .when and When the maximum value of 24.5 μm is taken, the synthesis position deviation This is approximately 34.6 μm. Therefore, configuring the second preset threshold as "the spot offset in each coordinate axis direction does not exceed 10 pixels" is physically equivalent to "the composite position deviation of the emitted beam does not exceed 34.6 μm". In actual debugging operations, the operator reads the pixel displacement in the X and Y axes through the display screen of the second detection unit 330. When the pixel displacement in both directions does not exceed 10 pixels, it means that the composite position deviation of the emitted beam has been controlled within 34.6 μm. By clearly defining the pixel threshold with specific physical accuracy indicators (8.17 μrad and 34.6 μm), even if the pixel size of the image acquisition device changes due to model changes in the future, those skilled in the art can deduce the corresponding pixel threshold from the physical accuracy indicators through simple conversion. The physical accuracy defined in the dependent claim can still provide substantial definitional support for the first and second preset thresholds in the embodiments, thereby preventing competitors from circumventing the pixel threshold limitation by changing cameras with different pixel sizes in patent infringement determination.
[0092] Further, S52 includes the following steps: S521: The second detection unit 330 obtains the first position coordinates of the light spot of the emitted light beam on the photosensitive surface of the second detection unit 330 before the mechanical shaft rotates; S522: Rotate the mechanical shaft 180° around its axis, and obtain the second position coordinates of the light spot on the photosensitive surface after the mechanical shaft is rotated by the second detection unit 330; S523: Calculate the difference between the second position coordinate and the first position coordinate in the X-axis direction, and use it as the position offset in the X-axis direction; calculate the difference between the second position coordinate and the first position coordinate in the Y-axis direction, and use it as the position offset in the Y-axis direction.
[0093] In this embodiment, the second detection unit 330 is an image acquisition device (preferably an industrial camera) without an imaging lens 320. Its photosensitive surface directly receives the emitted light beam and forms a light spot. Before the mechanical axis rotates, the operator first observes the position of the current light spot in the image through the display device connected to the second detection unit 330. The image acquisition device outputs the position coordinates (in pixels) of each pixel on its photosensitive surface along with the corresponding light spot image data to the display device or data processing terminal. The operator or data processing terminal extracts the center position of the current light spot using an image processing algorithm (such as the centroid method or grayscale centroid method) and reads the coordinate value of the center position with reference to the preset coordinate system of the photosensitive surface, which is recorded as the first position coordinate (X1, Y1). This first position coordinate represents the initial incident position of the emitted light beam on the photosensitive surface of the second detection unit 330 when the mechanical axis is at the initial zero position. To ensure the accuracy of subsequent calculation results, it is preferable to acquire multiple frames of images and calculate the average value of the center coordinates of the light spot when reading the coordinates to eliminate the influence of image sensor noise and random jitter on the coordinate reading. If irregular spot shape or uneven energy distribution is found during the reading process, the operator also needs to check whether there is stray light interference or optical element contamination in the Kude optical path 442. Coordinate reading can only be performed after the spot quality meets the requirements.
[0094] In this embodiment, after the operator completes the reading of the first position coordinates, the mechanical axis is rotated 180° around its own axis. Rotating 180° is the preferred method in the debugging of the Kude optical path 442. Its advantage is that when the mechanical axis rotates 180°, the spot of the emitted beam on the photosensitive surface of the second detection unit 330 will move to a position symmetrical to the initial position with respect to the mechanical axis wobble error, making the positional deviation most fully and significantly exposed, which is beneficial for subsequent accurate measurement and correction of the deviation. After the mechanical axis has rotated to its position and stabilized, the operator observes the spot position again through the display screen of the second detection unit 330, and extracts the center coordinates of the spot using the same image processing algorithm as in step S51, denoted as the second position coordinates (X2, Y2). These coordinates represent the incident position of the emitted beam on the photosensitive surface of the second detection unit 330 after the mechanical axis has rotated 180°. It should be noted that during the rotation of the mechanical shaft, the operator should keep the position of the second detection unit 330 and its entire debugging fixture fixed and should not touch or move any part of the debugging fixture, so as to ensure that the first position coordinate and the second position coordinate are collected under the same reference system, thereby ensuring the authenticity and validity of the subsequent calculation results.
[0095] In this embodiment, the data processing terminal or operator calculates the position offset ΔX in the X-axis direction (X1, Y1) and the position offset ΔY in the Y-axis direction (X2, Y2) based on the first position coordinates (X1, Y1) and the second position coordinates (X2, Y2) obtained in steps S521 and S522, as follows: ΔX = X2 - X1, ΔY = Y2 - Y1. The sign of ΔX indicates the offset direction of the light spot in the X-axis direction (positive for positive offset, negative for negative offset), and the sign of ΔY indicates the offset direction of the light spot in the Y-axis direction. The absolute value of ΔX indicates the degree of position offset in the X-axis direction, and the absolute value of ΔY indicates the degree of position offset in the Y-axis direction; both are measured in pixels. Subsequent adjustments in the X-axis direction are based on whether the absolute value of ΔX is greater than a second preset threshold. Through iterative adjustment, the absolute value of ΔX is gradually reduced and eventually converges to within the second preset threshold. Similarly, adjustments in the Y-axis direction are based on the absolute value of ΔY. By decomposing the position offset into independent components in two orthogonal directions, the X and Y axes, subsequent steps can iteratively adjust each direction separately without interference, avoiding the complex problem of "adjusting one axis while changing the other" in two-dimensional coupled adjustment. This is one of the important technical means of achieving inter-axis decoupling in this invention. In this embodiment, the second preset threshold is preferably 10 pixels. When the absolute value of ΔX is not greater than 10 pixels and the absolute value of ΔY is not greater than 10 pixels, the position offset of the mechanical axis is considered to meet the qualification requirements.
[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A Kuder optical path debugging fixture, characterized in that, include: The carrier frame (100), and the beam splitting element (200), the first detection unit (310) and the second detection unit (330) disposed on the carrier frame (100); The beam splitter (200) is disposed on the outgoing beam path of the Couder optical path (442) and is used to split the outgoing beam into a first beam (211) and a second beam (212). The first detection unit (310) is disposed in the optical path of the first beam (211) and is used to detect the angle information of the emitted beam; The second detection unit (330) is disposed in the optical path of the second beam (212) and is used to detect the position information of the emitted beam; The first detection unit (310) includes an image acquisition device and an imaging lens (320) disposed in front of the image acquisition device, and the second detection unit (330) is an image acquisition device without an imaging lens (320).
2. The Kuder optical path debugging fixture according to claim 1, characterized in that, The support frame (100) includes a first arm (110) and a second arm (120) that are integrally formed. The first arm (110) and the second arm (120) are respectively provided with a first mounting part (111) and a second mounting part (121), which are used to install the first detection unit (310) and the second detection unit (330). A mounting frame (130) is provided at the connection between the first arm (110) and the second arm (120), and the mounting frame (130) is provided with a receiving groove (131) for accommodating the beam splitter (200).
3. The Kuder optical path debugging fixture according to claim 2, characterized in that, The second arm (120) is provided with a first type of fixing ear (122) on the outside, and the mounting frame (130) is provided with a second type of fixing ear (132) on both sides of the receiving groove (131). The beam splitter (200) is confined within the receiving groove (131) by a number of pressure blocks (133).
4. The Kuder optical path debugging fixture according to claim 1, characterized in that, The beam splitter (200) is configured as a beam splitter, and the first detection unit (310) and the second detection unit (330) are both configured as cameras.
5. A method for debugging a Kuder optical path, employing a Kuder optical path debugging fixture as described in any one of claims 1-4, characterized in that, include: S1: A reference element (420) is set at the incident end of the Kud optical path (442) so that the center of the reference element (420) is aligned with the center of the optical axis of the Kud optical path (442); S2: Using a collimated light source (410) to emit a collimated beam to the reference element (420), and adjusting the collimated light source (410) to align the center of its collimated optical axis (441) with the center of the reference element (420) to achieve self-collimation, so as to establish a global reference optical axis; S3: Set the Kude optical path debugging fixture at the output end of the Kude optical path (442); S4: Divide the multiple Couder mirrors in the Couder optical path (442) into several groups according to their corresponding mechanical axes, and each group includes at least a first optical adjustment element (431) and a second optical adjustment element (432). S5: Perform the following debugging steps on the current group: S51: Rotate the mechanical axis corresponding to the current group, observe the trajectory of the emitted beam during the rotation process through the first detection unit (310), adjust the first optical adjustment element (431) in the group, so that the angular offset of the beam during the rotation process is not greater than the first preset threshold, and complete the angular domain locking of the current group; S52: The position coordinates of the emitted beam before and after the mechanical shaft rotates are obtained by the second detection unit (330), and the position offset in each coordinate axis direction is calculated respectively; S53: While keeping the angle offset no greater than the first preset threshold, adjust the second optical adjustment element (432) in the group to make the beam offset along a certain coordinate axis direction, and then adjust the first optical adjustment element (431) to make the beam return to the reference position. Repeat the process until the position offset in the coordinate axis direction is no greater than the second preset threshold. S54: Adjust the position offset in the other coordinate axis direction according to the method of S53 until the position offset in the coordinate axis direction is not greater than the second preset threshold. S55: When the mechanical shaft rotates before and after, the angular offset detected by the first detection unit (310) is not greater than the first preset threshold and the position offset of each coordinate axis direction detected by the second detection unit (330) is not greater than the second preset threshold, the current group debugging is completed; S6: Lock the pose of all optical adjustment elements in the current group, and then use the global reference optical axis as the only reference to independently debug the next group in the manner of S5 until all groups are debugged; during the debugging of any group, the state of the outgoing beam corresponding to that group does not participate in the reference establishment of the next group.
6. The method for debugging a Kuder optical path according to claim 5, characterized in that, In S6, the starting reference for the next group of debugging is the global reference optical axis established in S2, rather than the actual optical axis of the emitted beam after the current group of debugging is completed. Once the current group has been debugged, its residual error will not be passed on to the next group. The angle deviation and position deviation of the next group are independent of the residual error of the current group.
7. The method for debugging a Kuder optical path according to claim 5, characterized in that, During the iteration of S53, the angle offset detected by the first detection unit (310) is monitored in real time. If the angle offset exceeds the first preset threshold, the position adjustment is paused and the first optical adjustment element (431) is called back. The position adjustment is continued after the angle offset recovers to no greater than the first preset threshold.
8. The method for debugging a Kuder optical path according to claim 5, characterized in that, The reference element (420) is a planar reflective element with a scale (421) mark, which is located in the central region of the planar reflective element.
9. A method for debugging a Kuder optical path according to claim 5, characterized in that, The first preset threshold is 2 pixels, and the second preset threshold is 10 pixels.
10. A method for debugging a Kuder optical path according to claim 5, characterized in that, S52 includes the following steps: S521: The first position coordinates of the light spot of the emitted beam on the photosensitive surface of the second detection unit (330) before the mechanical shaft rotates are obtained by the second detection unit (330); S522: Rotate the mechanical shaft 180° around its axis, and obtain the second position coordinates of the light spot on the photosensitive surface after the mechanical shaft is rotated by the second detection unit (330); S523: Calculate the difference between the second position coordinate and the first position coordinate in the X-axis direction, and use it as the position offset in the X-axis direction; calculate the difference between the second position coordinate and the first position coordinate in the Y-axis direction, and use it as the position offset in the Y-axis direction.
