A calibration system and method for centering a point laser module

Abstract

This invention discloses a calibration system and method for the pointing center of a point laser module, relating to the field of optical technology. The invention includes a basic positioning unit, a sensing feedback unit, a motion control unit, and an auxiliary positioning device. The basic positioning unit includes a six-degree-of-freedom module stage and a high-precision displacement platform; the sensing feedback unit includes a multi-axis attitude sensor array and a contact probe group; the motion control unit includes a multi-axis motion controller and a servo drive module; and the auxiliary positioning device includes a reference calibration component and an environmental compensation module. This invention establishes an initial coordinate system reference by extending a contact probe to contact a standard spherical target, monitors the environment in real time, synchronously inputs three-dimensional attitude and environmental parameters into an LSTM prediction model, generates real-time compensation quantities, constructs a measurement optical path, and uses moiré fringes for verification to achieve sub-pixel-level center locking, thus improving calibration accuracy.

Description

Technical Field

[0001] This invention belongs to the field of optical technology, and in particular relates to a calibration system and method for the pointing center of a point laser module. Background Technology

[0002] In fields such as precision optical measurement, industrial robot positioning, and aerospace, the pointing center calibration accuracy of point laser modules directly determines the overall system performance. Current technologies primarily rely on manual intervention and static calibration methods, which present the following technical bottlenecks:

[0003] (1) Calibration accuracy is limited

[0004] Traditional methods employ monocular vision or mechanical probe contact calibration, which are limited by image resolution (typically >1μm) or mechanical backlash error, making it difficult to achieve sub-micron level calibration accuracy. Especially under dynamic operating conditions, factors such as environmental temperature drift and mechanical vibration further reduce calibration repeatability.

[0005] (2) Insufficient real-time performance and adaptability

[0006] Existing systems mostly use fixed calibration parameters, which cannot be dynamically adjusted according to environmental changes (temperature, vibration, load disturbance). When the module undergoes slight deformation due to thermal expansion or mechanical stress, the calibration process needs to be restarted frequently, which seriously affects the continuity of operation.

[0007] (3) Reliability defects

[0008] Conventional sensor layouts (such as orthogonal triaxial accelerometers) are susceptible to single-point failure and are easily interfered with in complex electromagnetic environments, leading to the accumulation of attitude measurement errors.

[0009] (4) Lack of data management

[0010] Historical calibration parameters are mostly stored in linear databases, lacking correlation with operating conditions and anomaly recovery mechanisms. When calibration data is abnormal, the system must rely entirely on manual intervention for recalibration, leading to extended downtime. Summary of the Invention

[0011] The purpose of this invention is to provide a calibration system and method for the center pointing of a point laser module. By extending a contact probe to contact a standard spherical target, an initial coordinate system reference is established, the environment is monitored in real time, a measurement optical path is constructed, and sub-pixel-level center locking is achieved by verifying with moiré fringes. This solves the problems of insufficient calibration accuracy, insufficient real-time performance and adaptability, and lack of reliability in existing systems.

[0012] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution:

[0013] This invention is a calibration system for the centering of a point laser module, comprising a basic positioning unit, a sensing feedback unit, a motion control unit, and an auxiliary positioning device;

[0014] The basic positioning unit includes a six-degree-of-freedom modular stage and a high-precision displacement platform; the six-degree-of-freedom modular stage is a combination mechanism of slide rail and rotary table, used to realize X, Y, Z three-axis translation and rotation around the axis; the high-precision displacement platform adopts a nano-level electro-ceramic actuator and has a built-in grating ruler closed-loop feedback system.

[0015] The sensing feedback unit includes a multi-axis attitude sensor array and a contact probe group; the multi-axis attitude sensor array is used to monitor the attitude angle of the six-degree-of-freedom module stage in real time, and to construct a spatial position measurement network and establish a three-dimensional coordinate system mapping; the contact probe group is used to realize physical reference point contact detection through a tungsten carbide probe array.

[0016] The motion control unit includes a multi-axis motion controller and a servo drive module; the multi-axis motion controller is a real-time control module using FPGA and ARM architecture, and integrates a PID parameter self-tuning algorithm to dynamically adjust the servo response characteristics;

[0017] The auxiliary positioning device includes a reference calibration component and an environmental compensation module; the reference calibration component includes a grating plate with moiré fringes and an infrared through-beam sensor; the grating plate is used to provide a sub-micron level positioning reference; the infrared through-beam sensor constitutes a zero-point calibration system; the motion compensation module includes an active vibration isolation platform and a constant-temperature liquid cooling system; the active vibration isolation platform is used to eliminate ground vibration interference; the constant-temperature liquid cooling system is used to maintain the thermal stability of the mechanical structure.

[0018] As a preferred technical solution, the bottom of the six-degree-of-freedom module stage is rigidly connected to a high-precision displacement platform via an air-bearing vibration isolation interface; a multi-axis attitude sensor array is integrated on the surface of the six-degree-of-freedom module stage; the axis attitude sensor array includes a MEMS gyroscope and an accelerometer; the measurement extension arm of the six-degree-of-freedom module stage is equipped with a contact probe group; the tip of the contact probe group is in physical contact with the reference calibration component.

[0019] As a preferred technical solution, the high-precision displacement platform has a three-layer structure; the upper layer of the high-precision displacement platform is an air-bearing rotary platform for adjusting pitch and yaw angles; the middle layer of the high-precision displacement platform is a cross roller guide for fine-tuning in the X and Y axes; the bottom layer of the high-precision displacement platform is a piezoelectric ceramic nano-displacement stage for fine-tuning in the Z axis; the high-precision displacement platform is electrically connected to a multi-axis motion controller via a fiber optic encoder.

[0020] As a preferred technical solution, the multi-axis motion controller is electrically connected to the motion control unit via a CAN bus; the servo drive module is installed on the high-precision displacement platform; the reference calibration component also includes a laser interferometer reflector and a calibration ball target; the optical path of the laser interferometer reflector is parallel to the axis of the displacement platform; and the standard ball target is coaxially aligned with the contact probe.

[0021] This invention relates to a method for calibrating the center of a point laser module, comprising the following steps:

[0022] Step S1: The contact probe extends and contacts the standard ball target to establish an initial coordinate system reference;

[0023] Step S2: After the six-degree-of-freedom module stage is coarsely positioned by the cross roller guide, the piezoelectric ceramic nano displacement stage completes the fine adjustment of the Z-axis, and the air-bearing rotary platform provides pitch and yaw angle compensation.

[0024] Step S3: The multi-axis attitude sensor array acquires the three-dimensional attitude of the module;

[0025] Step S4: The environmental compensation module monitors environmental parameters in real time and compensates for changes in air refractive index through gradient field modeling;

[0026] Step S5: Simultaneously input 3D pose and environmental parameters into the LSTM prediction model to generate real-time compensation quantities;

[0027] Step S6: Construct the measurement optical path and verify sub-pixel-level center locking through moiré fringe verification;

[0028] Step S7: The calibration data is stored in a self-healing database, and the historical optimal parameters are automatically recalled in case of abnormal operating conditions;

[0029] Step S8: Input the historical best parameters into the motion control unit for calibration.

[0030] In step S1, the contact probe extends and contacts the standard spherical target to establish the initial coordinate system reference. The specific process is as follows:

[0031] Step S11: The contact probe extends and the probe tip contacts the surface of the standard ball target with constant pressure. The constant pressure is usually in the range of 0.5N-2N, and the trigger stroke error is controlled within ±5μm.

[0032] Step S12: The contact probe has a built-in micro-force sensor that monitors the contact force in real time. When the force reaches a preset threshold, a signal is triggered to lock the current position coordinates of the probe.

[0033] Step S13: Synchronously acquire the three-dimensional coordinates (X1, Y1, Z1) of the probe tip in the robotic arm coordinate system, the known geometric center coordinates (X0, Y0, Z0) of the standard spherical target in the global coordinate system, and the environmental temperature parameters acquired by the environmental compensation module. The environmental temperature parameters are used for thermal expansion compensation.

[0034] Step S14: Solve for the coordinate transformation matrix using the least squares method. The specific formula is as follows:

[0035] (X1,Y1,Z1) T =R×(X0,Y0,Z0) T +T;

[0036] In the formula, R is the rotation matrix, which is calculated by fitting the contact point between the probe and the target ball; T is the translation vector, which is directly obtained from the displacement of the probe robotic arm.

[0037] Step S15: The probe makes multi-point contact (at least four non-coplanar points) along the surface of the standard spherical target, and the positioning error of single-point contact is eliminated by fitting the spherical equation; the specific formula is as follows:

[0038] (xa) 2 +(yb) 2 +(zc) 2 =r 2 ;

[0039] In the formula, (a,b,c) are the coordinates of the sphere's center, and r is the radius of the standard sphere;

[0040] Step S16: Correct the coordinate system reference based on the ambient temperature parameters collected by the environmental compensation module; the formula for correcting the coordinate system reference is as follows:

[0041] ΔL=α×L×ΔT+β×ΔT 2 ;

[0042] In the formula, α is the coefficient of thermal expansion of the material, and β is the nonlinear compensation coefficient;

[0043] Step S17: When the error of the coordinate transformation matrix calibrated for three consecutive times is less than the threshold (the threshold is 5μm), it is determined that the initial coordinate system has been established. The stored reference parameters include the transformation matrix R / T, environmental compensation coefficient, timestamp, and temperature record.

[0044] As a preferred technical solution, in step S3, the multi-axis attitude sensor array (including a three-axis accelerometer and a gyroscope) is installed at the four vertices of the six-degree-of-freedom module platform surface, forming a spatially redundant measurement network. The calibration matrices of the four multi-axis attitude sensor arrays are uniformly transformed to the platform coordinate system to ensure consistent attitude calculation references. Kalman filtering is used to fuse the acceleration and angular velocity data from the four sensors to eliminate single-point measurement noise. The three-dimensional attitude angles of the platform are calculated using the least squares method, including pitch, yaw, and roll; satisfying the formula:

[0045]

[0046] In the formula, S i T represents the actual measured value from the sensor. i (θ) represents the theoretical model value;

[0047] Real-time monitoring of data deviations from each sensor eliminates abnormal nodes, such as failed sensors caused by temperature drift; then, by correlating the sensor spacing with the kinematic model, measurement errors caused by mechanical deformation are compensated.

[0048] The geometry of the four sensor nodes is parameterized to establish an ideal regular tetrahedron mathematical model (side length L and vertex coordinates can be analytically expressed), which serves as the reference shape for the mechanical structure. When mechanical deformation occurs, the distance {L} between each node is measured in real time. ij} Calculate the deviation between the actual geometric solid and the ideal regular tetrahedron. Quantify the degree of deformation.

[0049] Derive the transfer equation of the effect of mechanical deformation on sensor measurements:

[0050]

[0051] In the formula, K is the stiffness matrix (obtained through finite element analysis or experimental calibration). To address process noise, joint suppression of deformation error and sensor noise is achieved;

[0052] A three-dimensional temperature field ΔL is established using temperature sensors arranged in a ring. thermal =α×f V (TT n ); α is the thermal expansion data of the material, and T0 is the reference temperature.

[0053] As a preferred technical solution, the training process of the LSTM prediction model in step S5 is as follows:

[0054] Step S51: Collect sensor data and environmental data, align them according to timestamps, and construct a multi-dimensional time series input matrix;

[0055] Step S52: Use a time series generator to divide the input sequence, and output the compensation amount Δθ for the next time step;

[0056] Step S53: Normalize the multi-source data;

[0057] Step S54: Construct the LSTM model architecture; where the loss function is weighted MSE, and the pose parameter weights are set to 3 times the environmental parameters; regularization is achieved by setting weight_decay = 0.01 in the Adam optimizer to implement L2 regularization;

[0058] Step S55: Convert the trained model to ONNX format and deploy it to the embedded AI acceleration module to achieve a real-time inference latency of <5ms.

[0059] When deploying the LSTM prediction model, the trained model is converted to ONNX format and deployed to an embedded AI acceleration module (such as NVIDIA Jetson) to achieve a real-time inference latency of <5ms. The prediction compensation is sent to the six-DOF platform controller via the EtherCAT bus.

[0060] By measuring the phase difference of the laser beam between the target mirror and the reference mirror, and combining this with the periodic characteristics of the moiré fringes, sub-pixel-level displacement detection can be achieved. When two coherent beams are superimposed, a tiny displacement will cause a phase shift in the interference fringes. By analyzing the deformation and positional changes of the moiré fringes, the sub-micron-level center shift can be calculated.

[0061] As a preferred technical solution, the specific process for constructing the measurement optical path and verifying it through moiré fringes in step S6 is as follows:

[0062] Step S61: Configure the interferometer parameters, including configuring the reference mirror and target mirror parameters;

[0063] Step S62: Insert a high-density diffraction grating (3000 lines / mm) into the laser path to generate a periodic moiré fringe substrate;

[0064] Step S63: Monitor the optical path offset in real time using a four-quadrant detector, and drive the piezoelectric ceramic fine-tuning mirror to complete beam collimation (accuracy ≤ 0.1 μrad);

[0065] Step S64: The piezoelectric ceramic modulator finely adjusts the phase of the reference light at a frequency of 10 kHz to generate dynamic moiré fringes;

[0066] Step S65: The global shutter camera captures a stripe image at a rate of 2000fps and transmits it to the image processing unit via optical fiber;

[0067] Step S66: The image processing unit performs a two-dimensional Fourier transform on the image to extract the phase information of the fundamental frequency component; the specific formula is as follows:

[0068]

[0069] In the formula, F(u,v) represents the frequency domain signal, and φ represents the phase distribution FF1B.

[0070] Step S67: The center of the stripe is fitted to subpixel using cubic spline interpolation, with a positioning accuracy of 1 / 50 pixel (equivalent to 0.02 μm);

[0071] Step S68: Input the center offset Δd into the PID controller to generate a high-precision displacement platform drive signal, which drives the servo drive module to complete the position correction; the calculation formula for generating the high-precision displacement platform drive signal is as follows:

[0072]

[0073] Step S69: When the standard deviation of the offset for 5 consecutive iterations is less than 0.1 μm, the calibration is considered complete.

[0074] As a preferred technical solution, in step S7, the self-healing database architecture includes a real-time layer, an optimization layer, and a disaster recovery layer; the real-time layer uses a time-series database to store the calibration raw data updated at the second level; the optimization layer is based on a time-series clustering algorithm, which stores the historical optimal parameters according to working conditions in a relational database and establishes a working condition feature vector index; the disaster recovery layer includes distributive redundant storage of parameters through blockchain technology.

[0075] The self-healing database uses an LSTM network to monitor the calibration data stream in real time. When sensor noise > 3σ or environmental parameter mutation is detected, the self-healing process is triggered. Based on the weighted Euclidean distance, the current working condition is matched with the historical best parameter library, and the calibration parameter group with a matching degree > 90% is called first.

[0076] The present invention has the following beneficial effects:

[0077] (1) This invention establishes an initial coordinate system reference by extending a contact probe to contact a standard ball target, monitors the environment in real time, and synchronously inputs three-dimensional attitude and environmental parameters into the LSTM prediction model to generate real-time compensation quantities, constructs a measurement optical path, and uses moiré fringes to verify sub-pixel-level center locking, thereby improving calibration accuracy.

[0078] (2) The present invention establishes an initial coordinate system by contacting a standard ball target with a contact probe, eliminates single-point positioning error by using a 4-point contact method, improves accuracy to 0.3μm, uses PID algorithm to adjust contact pressure to avoid damage to the target ball surface, maintains the coordinate system of the robotic arm and the global coordinate system, supports real-time coordinate transformation, and improves positioning accuracy.

[0079] (3) This invention locks the deformation propagation path through a regular tetrahedral structure and rigid geometric relationship, reducing the dimension of the compensation model from 12 degrees of freedom in the traditional cubic layout to 6 degrees of freedom. The real-time feedback of the deformation ΔL keeps the compensation delay within 1ms, which is better than the 10ms level of the traditional temperature compensation scheme, thus improving the system's environmental adaptability.

[0080] (4) This invention stores calibration data in a self-healing database and automatically calls the historical optimal parameters under abnormal working conditions. The historical optimal parameters are then input into the motion control unit for calibration, which makes it convenient for users to directly call the optimal parameters under the next abnormal situation, thus improving calibration efficiency.

[0081] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0082] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0083] Figure 1 This is a flowchart of a method for calibrating the center of a point laser module according to the present invention;

[0084] Figure 2 This is a schematic diagram of a calibration system for pointing a point laser module to the center, according to the present invention. Detailed Implementation

[0085] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0086] Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0087] To make the purpose, technical solution, and advantages of this application clearer, the following will be described in conjunction with the appendix. Figure 1 The embodiments of this application will be described in further detail.

[0088] Before introducing the embodiments of this application, the role of the calibration technology is explained as follows:

[0089] (1) Calibrate the pointing center accuracy

[0090] Ensure that the actual emission direction of the laser beam strictly coincides with the theoretical optical axis to eliminate pointing deviation caused by assembly errors; achieve sub-pixel level positioning by accurately matching the geometric contour center with the reference calibration component (such as the center of a circular hole).

[0091] (2) Ensure system performance stability

[0092] It compensates for the effects of environmental disturbances such as temperature, humidity, and vibration on the quality of the laser beam, maintaining long-term operational stability; it verifies the fluctuation range of laser wavelength and energy output to meet industrial-grade application standards.

[0093] (3) Supports collaborative operation of multiple devices

[0094] In laser display systems, ensuring spatial synchronization of images from multiple projection modules avoids image misalignment; it also provides a unified spatial coordinate system benchmark for robot navigation and automated production lines.

[0095] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figure 1 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0096] Example 1

[0097] Please see Figure 1 As shown, the present invention is a calibration method for the center pointing of a point laser module, comprising the following steps:

[0098] Step S1: The contact probe extends and contacts the standard ball target to establish an initial coordinate system reference;

[0099] Step S2: After the six-degree-of-freedom module stage is coarsely positioned by the cross roller guide, the piezoelectric ceramic nano displacement stage completes the fine adjustment of the Z-axis, and the air-bearing rotary platform provides pitch and yaw angle compensation.

[0100] Step S3: The multi-axis attitude sensor array acquires the three-dimensional attitude of the module;

[0101] Step S4: The environmental compensation module monitors environmental parameters in real time and compensates for changes in air refractive index through gradient field modeling;

[0102] Step S5: Simultaneously input 3D pose and environmental parameters into the LSTM prediction model to generate real-time compensation quantities;

[0103] Step S6: Construct the measurement optical path and verify sub-pixel-level center locking through moiré fringe verification;

[0104] Step S7: The calibration data is stored in a self-healing database, and the historical optimal parameters are automatically recalled in case of abnormal operating conditions;

[0105] Step S8: Input the historical best parameters into the motion control unit for calibration.

[0106] In step S1, the contact probe extends and contacts the standard spherical target to establish the initial coordinate system reference. The specific process is as follows:

[0107] Step S11: The contact probe extends and the probe tip contacts the surface of the standard ball target with constant pressure. The constant pressure is usually in the range of 0.5N-2N, and the trigger stroke error is controlled within ±5μm.

[0108] Step S12: The contact probe has a built-in micro-force sensor that monitors the contact force in real time. When the force reaches a preset threshold, a signal is triggered to lock the current position coordinates of the probe.

[0109] Step S13: Synchronously acquire the three-dimensional coordinates (X1, Y1, Z1) of the probe tip in the robotic arm coordinate system, the known geometric center coordinates (X0, Y0, Z0) of the standard spherical target in the global coordinate system, and the environmental temperature parameters acquired by the environmental compensation module. The environmental temperature parameters are used for thermal expansion compensation.

[0110] Step S14: Solve for the coordinate transformation matrix using the least squares method. The specific formula is as follows:

[0111] (X1,Y1,Z1) T =R×(X0,Y0,Z0) T +T;

[0112] In the formula, R is the rotation matrix, which is calculated by fitting the contact point between the probe and the target ball; T is the translation vector, which is directly obtained from the displacement of the probe robotic arm.

[0113] Step S15: The probe makes multi-point contact (at least four non-coplanar points) along the surface of the standard spherical target, and the positioning error of single-point contact is eliminated by fitting the spherical equation; the specific formula is as follows:

[0114] (xa) 2 +(yb) 2 +(zc) 2 =r 2 ;

[0115] In the formula, (a,b,c) are the coordinates of the sphere's center, and r is the radius of the standard sphere;

[0116] Step S16: Correct the coordinate system reference based on the ambient temperature parameters collected by the environmental compensation module; the formula for correcting the coordinate system reference is as follows:

[0117] ΔL=α×L×ΔT+β×ΔT 2 ;

[0118] In the formula, α is the coefficient of thermal expansion of the material, and β is the nonlinear compensation coefficient;

[0119] Step S17: When the error of the coordinate transformation matrix calibrated for three consecutive times is less than the threshold (the threshold is 5μm), it is determined that the initial coordinate system has been established. The stored reference parameters include the transformation matrix R / T, environmental compensation coefficient, timestamp, and temperature record.

[0120] In step S3, the multi-axis attitude sensor array (including a three-axis accelerometer and a gyroscope) is installed at the four vertices of the six-degree-of-freedom module platform surface, forming a spatially redundant measurement network. The calibration matrices of the four multi-axis attitude sensor arrays are uniformly transformed to the platform coordinate system to ensure consistent attitude calculation references. Kalman filtering is used to fuse the acceleration and angular velocity data from the four sensors to eliminate single-point measurement noise. The three-dimensional attitude angles of the platform, including pitch, yaw, and roll, are calculated using the least squares method, satisfying the formula:

[0121]

[0122] In the formula, S i T represents the actual measured value from the sensor. i (θ) represents the theoretical model value;

[0123] Real-time monitoring of data deviations from each sensor eliminates abnormal nodes, such as failed sensors caused by temperature drift; then, by correlating the sensor spacing with the kinematic model, measurement errors caused by mechanical deformation are compensated.

[0124] The geometry of the four sensor nodes is parameterized to establish an ideal regular tetrahedron mathematical model (side length L and vertex coordinates can be analytically expressed), which serves as the reference shape for the mechanical structure. When mechanical deformation occurs, the distance {L} between each node is measured in real time. ij} Calculate the deviation between the actual geometric solid and the ideal regular tetrahedron. Quantify the degree of deformation.

[0125] Derive the transfer equation of the effect of mechanical deformation on sensor measurements:

[0126]

[0127] In the formula, K is the stiffness matrix (obtained through finite element analysis or experimental calibration). To address process noise, joint suppression of deformation error and sensor noise is achieved;

[0128] A three-dimensional temperature field ΔL is established using temperature sensors arranged in a ring. thermal =α×f V (TT n ); α is the thermal expansion data of the material, and T0 is the reference temperature.

[0129] In step S5, the training process of the LSTM prediction model is as follows:

[0130] Step S51: Collect sensor data and environmental data, align them according to timestamps, and construct a multi-dimensional time series input matrix;

[0131] Step S52: Use a time series generator to divide the input sequence, and output the compensation amount Δθ for the next time step;

[0132] Step S53: Normalize the multi-source data;

[0133] Step S54: Construct the LSTM model architecture; where the loss function is weighted MSE, and the pose parameter weights are set to 3 times the environmental parameters; regularization is achieved by setting weight_decay = 0.01 in the Adam optimizer to implement L2 regularization;

[0134] Step S55: Convert the trained model to ONNX format and deploy it to the embedded AI acceleration module to achieve a real-time inference latency of <5ms.

[0135] When deploying the LSTM prediction model, the trained model is converted to ONNX format and deployed to an embedded AI acceleration module (such as NVIDIA Jetson) to achieve a real-time inference latency of <5ms. The prediction compensation is sent to the six-DOF platform controller via the EtherCAT bus.

[0136] By measuring the phase difference of the laser beam between the target mirror and the reference mirror, and combining this with the periodic characteristics of the moiré fringes, sub-pixel-level displacement detection can be achieved. When two coherent beams are superimposed, a tiny displacement will cause a phase shift in the interference fringes. By analyzing the deformation and positional changes of the moiré fringes, the sub-micron-level center shift can be calculated.

[0137] In step S6, the specific process for constructing the measurement optical path and verifying it through moiré fringes is as follows:

[0138] Step S61: Configure the interferometer parameters, including configuring the reference mirror and target mirror parameters;

[0139] Step S62: Insert a high-density diffraction grating (3000 lines / mm) into the laser path to generate a periodic moiré fringe substrate;

[0140] Step S63: Monitor the optical path offset in real time using a four-quadrant detector, and drive the piezoelectric ceramic fine-tuning mirror to complete beam collimation (accuracy ≤ 0.1 μrad);

[0141] Step S64: The piezoelectric ceramic modulator finely adjusts the phase of the reference light at a frequency of 10 kHz to generate dynamic moiré fringes;

[0142] Step S65: The global shutter camera captures a stripe image at a rate of 2000fps and transmits it to the image processing unit via optical fiber;

[0143] Step S66: The image processing unit performs a two-dimensional Fourier transform on the image to extract the phase information of the fundamental frequency component; the specific formula is as follows:

[0144]

[0145] In the formula, F(u,v) represents the frequency domain signal, and φ represents the phase distribution FF1B.

[0146] Step S67: The center of the stripe is fitted to subpixel using cubic spline interpolation, with a positioning accuracy of 1 / 50 pixel (equivalent to 0.02 μm);

[0147] Step S68: Input the center offset Δd into the PID controller to generate a high-precision displacement platform drive signal, which drives the servo drive module to complete the position correction; the calculation formula for generating the high-precision displacement platform drive signal is as follows:

[0148]

[0149] Step S69: When the standard deviation of the offset for 5 consecutive iterations is less than 0.1 μm, the calibration is considered complete.

[0150] In step S7, the self-healing database architecture includes a real-time layer, an optimization layer, and a disaster recovery layer. The real-time layer uses a time-series database to store the calibration raw data that is updated in seconds. The optimization layer is based on a time-series clustering algorithm, which stores the historical optimal parameters according to working conditions in a relational database and establishes a working condition feature vector index. The disaster recovery layer includes distributive redundant storage of parameters through blockchain technology.

[0151] The self-healing database uses an LSTM network to monitor the calibration data stream in real time. When sensor noise > 3σ or environmental parameter mutation is detected, the self-healing process is triggered. Based on the weighted Euclidean distance, the current working condition is matched with the historical best parameter library, and the calibration parameter group with a matching degree > 90% is called first.

[0152] Example 2

[0153] See Figure 2As shown, the present invention is a calibration system for the center pointing of a point laser module, which can be used to execute the method content of Embodiment 1 of the present invention, including: a basic positioning unit, a sensing feedback unit, a motion control unit, and an auxiliary positioning device;

[0154] The basic positioning unit includes a six-degree-of-freedom modular stage and a high-precision displacement platform; the six-degree-of-freedom modular stage is a combination of slide rail and rotary table, used to realize X, Y, Z three-axis translation and rotation around the axis; the high-precision displacement platform adopts a nano-level electro-ceramic actuator and has a built-in grating ruler closed-loop feedback system.

[0155] The sensing feedback unit includes a multi-axis attitude sensor array and a contact probe group; the multi-axis attitude sensor array is used to monitor the attitude angle of the six-degree-of-freedom module stage in real time, and to construct a spatial position measurement network and establish a three-dimensional coordinate system mapping; the contact probe group is used to realize physical reference point contact detection through a tungsten carbide probe array.

[0156] The motion control unit includes a multi-axis motion controller and a servo drive module; the multi-axis motion controller is a real-time control module using FPGA and ARM architecture, and integrates a PID parameter self-tuning algorithm to dynamically adjust the servo response characteristics;

[0157] The auxiliary positioning device includes a reference calibration component and an environmental compensation module; the reference calibration component includes a grating plate with moiré fringes and an infrared through-beam sensor; the grating plate is used to provide a sub-micron level positioning reference; the infrared through-beam sensor constitutes a zero-point calibration system; the motion compensation module includes an active vibration isolation platform and a constant temperature liquid cooling system; the active vibration isolation platform is used to eliminate ground vibration interference; the constant temperature liquid cooling system is used to maintain the thermal stability of the mechanical structure.

[0158] The bottom of the six-degree-of-freedom module stage is rigidly connected to the high-precision displacement platform through an air-bearing vibration isolation interface; the surface of the six-degree-of-freedom module stage integrates a multi-axis attitude sensor array; the axis attitude sensor array includes a MEMS gyroscope and an accelerometer; the measurement extension arm of the six-degree-of-freedom module stage is equipped with a contact probe group; the tip of the contact probe group is in physical contact with the reference calibration component.

[0159] The high-precision displacement platform has a three-layer structure; the top layer is an air-bearing rotary platform for pitch and yaw angle adjustment; the middle layer is a cross roller guide for fine-tuning in the X and Y axes; and the bottom layer is a piezoelectric ceramic nano-displacement stage for fine-tuning in the Z axis. The high-precision displacement platform is electrically connected to a multi-axis motion controller via a fiber optic encoder.

[0160] The multi-axis motion controller is electrically connected to the motion control unit via a CAN bus; the servo drive module is installed on the high-precision displacement platform; the reference calibration components also include a laser interferometer reflector and a calibration ball target; the optical path of the laser interferometer reflector is parallel to the axis of the displacement platform; the standard ball target and the contact probe are coaxially aligned.

[0161] It is worth noting that the various units included in the above system embodiments are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0162] Furthermore, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware, and the corresponding program can be stored in a computer-readable storage medium.

[0163] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A calibration system for the centering of a point laser module, comprising a basic positioning unit, a sensing feedback unit, a motion control unit, and an auxiliary positioning device, characterized in that: The basic positioning unit includes a six-degree-of-freedom modular stage and a high-precision displacement platform; the six-degree-of-freedom modular stage is a combination mechanism of slide rail and rotary table, used to realize X, Y, Z three-axis translation and rotation around the axis; the high-precision displacement platform adopts a nano-level electro-ceramic actuator and has a built-in grating ruler closed-loop feedback system. The sensing feedback unit includes a multi-axis attitude sensor array and a contact probe group; the multi-axis attitude sensor array is used to monitor the attitude angle of the six-degree-of-freedom module stage in real time, and to construct a spatial position measurement network and establish a three-dimensional coordinate system mapping; the contact probe group is used to realize physical reference point contact detection through a tungsten carbide probe array. The motion control unit includes a multi-axis motion controller and a servo drive module; the multi-axis motion controller is a real-time control module using FPGA and ARM architecture, and integrates a PID parameter self-tuning algorithm to dynamically adjust the servo response characteristics; The auxiliary positioning device includes a reference calibration component and an environmental compensation module; the reference calibration component includes a grating plate with moiré fringes and an infrared beam sensor; The grating plate is used to provide a submicron-level positioning reference; the infrared through-beam sensor constitutes a zero-point calibration system; the motion compensation module includes an active vibration isolation platform and a constant-temperature liquid cooling system; the active vibration isolation platform is used to eliminate ground vibration interference; the constant-temperature liquid cooling system is used to maintain the thermal stability of the mechanical structure.

2. The calibration system for pointing the center of a point laser module according to claim 1, characterized in that, The bottom of the six-degree-of-freedom module stage is rigidly connected to the high-precision displacement platform through an air-bearing vibration isolation interface; the surface of the six-degree-of-freedom module stage integrates a multi-axis attitude sensor array; the axis attitude sensor array includes a MEMS gyroscope and an accelerometer; the measurement extension arm of the six-degree-of-freedom module stage is equipped with a contact probe group; the tip of the contact probe group is in physical contact with the reference calibration component.

3. The calibration system for pointing the center of a point laser module according to claim 1, characterized in that, The high-precision displacement platform has a three-layer structure; the upper layer is an air-bearing rotary platform for pitch and yaw angle adjustment; the middle layer is a cross roller guide for fine-tuning in the X and Y axes; and the bottom layer is a piezoelectric ceramic nano-displacement stage for fine-tuning in the Z axis. The high-precision displacement platform is electrically connected to a multi-axis motion controller via a fiber optic encoder.

4. The calibration system for pointing the center of a point laser module according to claim 1, characterized in that, The multi-axis motion controller is electrically connected to the motion control unit via a CAN bus; the servo drive module is installed on the high-precision displacement platform; the reference calibration assembly also includes a laser interferometer reflector and a calibration ball target; the optical path of the laser interferometer reflector is parallel to the axis of the displacement platform; the standard ball target is coaxially aligned with the contact probe.

5. A method for calibrating the center of a point laser module, characterized in that, Includes the following steps: Step S1: The contact probe extends and contacts the standard ball target to establish an initial coordinate system reference; Step S2: After the six-degree-of-freedom module stage is coarsely positioned by the cross roller guide, the piezoelectric ceramic nano displacement stage completes the fine adjustment of the Z-axis, and the air-bearing rotary platform provides pitch and yaw angle compensation. Step S3: The multi-axis attitude sensor array acquires the three-dimensional attitude of the module; Step S4: The environmental compensation module monitors environmental parameters in real time and compensates for changes in air refractive index through gradient field modeling; Step S5: Simultaneously input 3D pose and environmental parameters into the LSTM prediction model to generate real-time compensation quantities; Step S6: Construct the measurement optical path and verify sub-pixel-level center locking through moiré fringe verification; Step S7: The calibration data is stored in a self-healing database, and the historical optimal parameters are automatically recalled in case of abnormal operating conditions; Step S8: Input the historical best parameters into the motion control unit for calibration.

6. The method for calibrating the center of a point laser module according to claim 5, characterized in that, In step S1, the specific process for establishing the initial coordinate system reference by extending the contact probe to contact the standard spherical target is as follows: Step S11: The contact probe extends, and the probe tip contacts the surface of the standard ball target with constant pressure; Step S12: The contact probe has a built-in micro-force sensor that monitors the contact force in real time. When the force reaches a preset threshold, a signal is triggered to lock the current position coordinates of the probe. Step S13: Synchronously acquire the three-dimensional coordinates (X1, Y1, Z1) of the probe tip in the robotic arm coordinate system, the known geometric center coordinates (X0, Y0, Z0) of the standard spherical target in the global coordinate system, and the environmental temperature parameters acquired by the environmental compensation module. Step S14: Solve for the coordinate transformation matrix using the least squares method. The specific formula is as follows: (X1,Y1,Z1) T =R×(X0,Y0,Z0) T +T; In the formula, R is the rotation matrix and T is the translation vector; Step S15: The probe makes multi-point contact along the surface of the standard spherical target, and the positioning error of single-point contact is eliminated by fitting the spherical equation; Step S16: Correct the coordinate system reference based on the ambient temperature parameters collected by the environmental compensation module; Step S17: When the error of the coordinate transformation matrix calibrated for three consecutive times is less than the threshold, it is determined that the initial coordinate system has been established.

7. The method for calibrating the center of a point laser module according to claim 5, characterized in that, In step S3, the multi-axis attitude sensor array is installed at the four vertices of the six-degree-of-freedom module platform. The calibration matrices of the four multi-axis attitude sensor arrays are uniformly transformed to the platform coordinate system. Kalman filtering is used to fuse the acceleration and angular velocity data of the four sensors. The three-dimensional attitude angle of the platform is calculated by the least squares method. The data deviation of each sensor is monitored in real time, and abnormal nodes are eliminated. Then, the measurement error caused by mechanical deformation is compensated by associating the sensor spacing with the kinematic model.

8. The method for calibrating the center of a point laser module according to claim 5, characterized in that, In step S5, the training process of the LSTM prediction model is as follows: Step S51: Collect sensor data and environmental data, align them according to timestamps, and construct a multi-dimensional time series input matrix; Step S52: Use a time series generator to divide the input sequence, and output the compensation amount Δθ for the next time step; Step S53: Normalize the multi-source data; Step S54: Construct the LSTM model architecture; where the loss function is weighted MSE, and the pose parameter weights are set to 3 times the environmental parameters; regularization is achieved by setting weight_decay = 0.01 in the Adam optimizer to implement L2 regularization; Step S55: Convert the trained model to ONNX format and deploy it to the embedded AI acceleration module to achieve a real-time inference latency of <5ms.

9. The method for calibrating the center of a point laser module according to claim 5, characterized in that, In step S6, the specific process for constructing the measurement optical path and verifying it using moiré fringes is as follows: Step S61: Configure interferometer parameters; Step S62: Insert a high-density diffraction grating into the laser path to generate a periodic moiré fringe substrate; Step S63: Monitor the optical path offset in real time using a four-quadrant detector, and drive the piezoelectric ceramic fine-tuning mirror to complete beam collimation; Step S64: The piezoelectric ceramic modulator finely adjusts the phase of the reference light at a frequency of 10 kHz to generate dynamic moiré fringes; Step S65: The global shutter camera captures a stripe image at a rate of 2000fps and transmits it to the image processing unit via optical fiber; Step S66: The image processing unit performs a two-dimensional Fourier transform on the image to extract the phase information of the fundamental frequency component; Step S67: Subpixel fitting of the fringe center is performed using cubic spline interpolation; Step S68: Input the center offset Δd into the PID controller to generate a high-precision displacement platform drive signal, which drives the servo drive module to complete the position correction; Step S69: When the standard deviation of the offset is less than the threshold for 5 consecutive iterations, the calibration is considered complete.

10. The method for calibrating the center of a point laser module according to claim 5, characterized in that, In step S7, the self-healing database architecture includes a real-time layer, an optimization layer, and a disaster recovery layer. The real-time layer uses a time-series database to store calibration raw data updated at the second level. The optimization layer uses a time-series clustering algorithm to classify and store historical optimal parameters according to working conditions in a relational database and establish a working condition feature vector index. The disaster recovery layer includes distributive redundant storage of parameters through blockchain technology. The self-healing database uses an LSTM network to monitor the calibration data stream in real time. When sensor noise > 3σ or environmental parameter mutation is detected, the self-healing process is triggered. Based on the weighted Euclidean distance, the current working condition is matched with the historical best parameter library, and the calibration parameter group with a matching degree > 90% is called first.

Images