Shape and position integrated measurement method and system based on multi-sensor fusion

Abstract

The invention relates to a shape and position integrated measurement method and system based on multi-sensor fusion, and relates to the technical field of optical measurement. The method comprises the following steps: calibrating internal parameters and external parameters of a sensor group; determining the measurement uncertainty corresponding to the sensors in the sensor group; performing parameter optimization on the sensor group in combination with the measurement uncertainty corresponding to the sensor; performing tight coupling on the measurement uncertainty corresponding to the sensor to obtain tight coupling uncertainty corresponding to the sensor and total measurement uncertainty corresponding to the sensor group; and combining the total measurement uncertainty to construct an integrated measurement model, wherein the integrated measurement model is used for simultaneously measuring the shape and the position of the polyhedral common optical element. The efficient multi-sensor measurement system is designed for solving the problem that the deflection measurement cannot effectively measure the surface shape and the pose at the same time. The precision of system calibration and workpiece positioning is effectively enhanced from the aspect of hardware, and then the surface shape and pose measurement capability of deflection measurement is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of optical measurement technology, and particularly relates to a shape-position integrated measurement method based on multi-sensor fusion. BACKGROUND

[0002] With the increasing requirements of major equipment and advanced instruments on photoelectric performance and compactness of system structure, the demand for high-precision and reliable integrated optical systems is increasingly urgent in the fields of aerospace, astronomy, industry, etc. The traditional system integration process of manufacturing single optical elements first and then assembling the system is complex, and there are problems such as multiple assembly degrees of freedom, great difficulty in system integration, complex transmission error chain, poor stability, etc. In recent years, the co-body freeform surface has attracted widespread attention due to its greatly simplified assembly complexity and guaranteed functionality and stability [Beier, M. et al. Development, fabrication, and testing of an anamorphic imaging snap-together freeform telescope. Applied Optics 54, 3530-3542 (2015)]. However, due to the complex surface shape, strict precision requirements and high dynamic range of the co-body freeform surface, as well as the high requirements for the relative poses between functional surfaces and the difficulty in establishing the reference, many challenges are brought to the measurement. Therefore, it is urgent to develop a measurement method capable of simultaneously measuring the shape-position error of multiple functional surfaces.

[0003] Currently, there is no mature method for the form and position integrated measurement of conjunctive freeform surfaces [Lyu, H. et al. Measurement and Characterization of Position and Posture of Conjunctive Multifreeform Surfaces Using Multisensors. IEEE Transactions on Instrumentation and Measurement 72, 1003214 (2023)]. In theory, Coordinate Measuring Machine (CMM) has the corresponding measurement ability, but its measurement accuracy is only micron level, and contact measurement may damage the element. Therefore, some researchers use CMM for rough measurement, and after the pose meets the requirements, they use phase measuring deflectometry or profilometer to fine the surface [Blalock, T. et al. The Manufacturing of a Multi-surface Monolithic Telescope with Freeform Surfaces. The European Physical Journal Conferences 215, 06004 (2019)], but this method is low in efficiency. Some other researchers improve the measurement efficiency by replacing the contact probe with a non-contact sensor. In recent years, computer-generated hologram (CGH) based interferometric measurement has attracted widespread attention due to its high measurement accuracy [Chen, S. et al. Corrective machining of monolithic multiple freeform mirrors based on holographic null test. Optics and Lasers in Engineering 170, 107789 (2023)]. This method realizes the form and position accuracy detection of optical elements by machining alignment holography, main holography and reference holography areas on CGH. Its cost is high, and it requires the design and manufacture of workpiece reference and CGH to be synchronized, which increases the complexity. Moreover, due to the limited measurement range of interferometric measurement in a single interferogram, it is difficult to measure elements with significant pose differences between functional surfaces.Some researchers guarantee the reliability of measurement by designing and processing global references during processing [Kiontke, S. R. Monolithic freeform element. Proc. SPIE 9575, 95750G (2015)]. However, the processing of these references is very complex with the measurement process. Some researchers use chromatic confocal sensors (CCS) combined with high-precision multi-axis motion mechanisms for scanning measurement [Jiao, S. et al. Non-contact method of thickness measurement for thin-walled rotary shell parts based on chromatic confocal sensor. Measurement 224, 113794 (2024)], such methods are high in cost and low in flexibility. In summary, the measurement of monolithic optical elements is still a complex and costly challenge.

[0004] Phase Measuring Deflectometry (PMD) is a powerful measurement technique, which is particularly suitable for the measurement of complex optical elements due to its high precision, flexibility and dynamic range. However, although PMD is highly sensitive to gradients, it is less sensitive to the position of the measured surface [Burke, J. et al. Deflectometry for specular surfaces: an overview. Advanced Optical Technologies 12, 1237687 (2023)]. Therefore, traditional PMD often obtains the mid-frequency topography of the measured object by eliminating the additional translation and tilt in the measurement results. However, the large pose measurement error of PMD greatly limits its applicability in simultaneous measurement of surface shape and position. SUMMARY

[0005] The present application relates to a shape and position integrated measurement system and method for multi-surface monolithic freeform optical elements, which can improve measurement accuracy and reliability according to measurement requirements. The technical solution is as follows: On the one hand, a shape and position integrated measurement method based on multi-sensor fusion is provided, characterized in that the method is applied to a computer device in a shape and position integrated measurement system based on multi-sensor fusion; The shape and position integrated measurement system based on multi-sensor fusion includes a screen, a camera, a sensor group, a turntable, a standard part and a computer device, and the sensor group includes a plurality of chromatic confocal sensors; The screen, camera, chromatic confocal sensor, turntable, and standard components are communicatively connected to the computer equipment. The computer device is used to acquire sensing data from the sensor array, and the method includes: The internal and external parameters of the sensor group are calibrated; Determine the measurement uncertainty of each sensor within the sensor group; Based on the measurement uncertainty corresponding to the sensor, the parameters of the sensor group are optimized. By tightly coupling the measurement uncertainty corresponding to the sensor, we obtain the tightly coupled uncertainty corresponding to the sensor and the total measurement uncertainty corresponding to the sensor group. An integrated measurement model is constructed based on the total measurement uncertainty. This integrated measurement model is used to simultaneously measure the shape and position of multi-faceted optical elements.

[0006] In an optional embodiment, before calibrating the internal and external parameters of the sensor group, the following steps are included: Determine the hardware parameters of the camera; Based on the camera's hardware parameters, the measurement range of the chromatic confocal sensor, the position of the chromatic confocal sensor, and the position of the turntable are determined.

[0007] In an optional embodiment, the calibration of the internal and external parameters of the sensor group includes: A reference calibration system is generated using the aforementioned standard components; The internal and external parameters of the sensor group are calibrated using the reference calibration system.

[0008] In an optional embodiment, determining the measurement uncertainty corresponding to a sensor within the sensor group includes: A virtual measurement system is constructed based on pre-calibrated parameters; Based on the virtual measurement system, the measurement uncertainty of each sensor within the sensor group is determined.

[0009] In an optional embodiment, the method further includes: Based on Bayesian inference, the configuration parameters of the sensor group are optimized.

[0010] In an optional embodiment, the step of tightly coupling the measurement uncertainty corresponding to the sensor to obtain a tightly coupled uncertainty corresponding to the sensor includes: Construct a tightly coupled virtual measurement system based on pre-calibrated parameters; Based on the tightly coupled virtual measurement system, the tightly coupled uncertainty corresponding to the sensor is determined.

[0011] In an optional embodiment, determining the total measurement uncertainty corresponding to the sensor group based on the tight coupling uncertainty includes: By combining simulation methods, the total measurement uncertainty corresponding to the sensor group is determined based on the tight coupling uncertainty.

[0012] In an optional embodiment, constructing an integrated measurement model based on the total measurement uncertainty includes: Based on the total measurement uncertainty, the integrated measurement model is constructed using the integrated workpiece positioning method and the integrated form and position measurement estimator.

[0013] In an optional embodiment, the integrated measurement estimator is a mathematical model-based estimator.

[0014] On the other hand, a shape and position integrated measurement system based on multi-sensor fusion is provided. The system includes computer equipment including a screen, camera, sensor group, turntable, standard parts and computer equipment. The sensor group includes multiple dispersive confocal sensors. The screen, camera, chromatic confocal sensor, turntable, and standard components are communicatively connected to the computer equipment. The computer device is used to perform any of the above-mentioned multi-sensor fusion-based shape and position measurement methods.

[0015] The beneficial effects of the technical solution provided by this invention include at least the following: To address the limitation of simultaneous effective measurement of surface shape and pose in deflection measurement, a highly efficient multi-sensor measurement system was designed. Hardware improvements significantly enhance the accuracy of system calibration and workpiece positioning, thereby improving the measurement capabilities of deflection measurement for both surface shape and pose. Furthermore, the system effectively avoids the numerical deviations inherent in traditional numerical optimization methods in surface shape reconstruction, enhancing the determinism of the measurement. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0017] Figure 1 This illustration shows a structural block diagram of a shape and position integrated measurement system based on multi-sensor fusion, provided in an exemplary embodiment of this application.

[0018] Figure 2The illustration shows a flowchart of a geometrical measurement method based on multi-sensor fusion provided in an exemplary embodiment of this application.

[0019] Figure 3 The illustration shows a flowchart of another geometric-position integrated measurement method based on multi-sensor fusion provided in an exemplary embodiment of this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0021] Figure 1 This illustration shows a structural block diagram of a shape-position integrated measurement system based on multi-sensor fusion, provided in an exemplary embodiment of this application. The system includes a screen 110, a camera 120, a sensor group 130, a turntable 140, standard components 150, and a computer device 160. The sensor group 130 includes multiple dispersive confocal sensors 131. The screen, camera, dispersive confocal sensors, turntable, and standard components are communicatively connected to the computer device. The computer device is used to acquire sensing data from the sensor group.

[0022] exist Figure 1 Under the system architecture, Figure 2 This illustration shows a flowchart of a form-position integrated measurement method based on multi-sensor fusion, provided in an exemplary embodiment of this application, illustrating the application of this method... Figure 1 Taking the system shown as an example, the method includes: Step 201: Calibrate the internal and external parameters of the sensor group.

[0023] This process involves calibrating the sensors within the sensor array. Optionally, prior to calibrating the internal and external parameters of the sensors, there are corresponding equipment setup and calibration steps.

[0024] Step 202: Determine the measurement uncertainty of each sensor within the sensor group.

[0025] In this embodiment, a virtual measurement system is constructed based on pre-calibrated parameters, the error sources of the system are analyzed, an error propagation chain is established, and the measurement uncertainty of the sensor itself is modeled based on the uncertainty of the parameters.

[0026] Step 203: Optimize the parameters of the sensor group based on the measurement uncertainty of the sensor.

[0027] Optionally, this parameter optimization process is used to further improve the parameter accuracy of the pre-calibration system.

[0028] Step 204: Tightly couple the measurement uncertainty corresponding to the sensor to obtain the tightly coupled uncertainty corresponding to the sensor.

[0029] This process involves determining the tightness of coupling for the sensor by considering system factors.

[0030] Step 205: Determine the total measurement uncertainty corresponding to the sensor group based on the tight coupling uncertainty.

[0031] Step 206: Construct an integrated measurement model by combining the total measurement uncertainty.

[0032] In this embodiment, the integrated model is used to simultaneously measure the shape and position of a multi-faceted optical element.

[0033] Figure 3 This illustration shows a flowchart of another form-position integrated measurement method based on multi-sensor fusion provided in an exemplary embodiment of this application, demonstrating the application of this method... Figure 1 Taking the computer device within the system shown as an example, the method includes: Step 301: Determine the camera's hardware parameters.

[0034] Step 302: Based on the camera's hardware parameters, determine the measurement range of the chromatic confocal sensor, the position of the chromatic confocal sensor, and the position of the turntable.

[0035] In this embodiment, using optical simulation software within a computer device, the camera's hardware parameters are first set, including the camera target surface size and lens focal length. Then, the position of the test object is determined based on edge ray tracing, and the position of the screen is further determined by tracing reflected rays. Figure 1 As shown. Finally, the positions of the CCS and the turntable can be set according to the measurement range of the CCS.

[0036] Step 303: Generate a reference calibration system using standard parts.

[0037] Step 304: In conjunction with the reference calibration system, calibrate the internal and external parameters of the sensor group.

[0038] Optionally, the internal and external parameters of the sensor are calibrated. The camera's internal parameters can be obtained through calibration by Zhang Zhengyou. The geometric pose of the system is calibrated using standard components, obtained by rotating a standard ball driven by a turntable. The observations are performed by the deflection system and the dispersive confocal system, respectively, and unified into a unified coordinate system using the standard component coordinate system as an intermediate bridge.

[0039] Step 305: Construct a virtual measurement system based on the pre-calibrated parameters.

[0040] Step 306: Based on the virtual measurement system, determine the measurement uncertainty of the sensors within the sensor group.

[0041] Optionally, the uncertainty in the deflection measurement mainly originates from camera photon noise, and the intensity noise follows a Poisson distribution, as shown in Equation 1 below: Formula 1: ; In the formula, σ I 2 The variance representing the intensity of the noise. K Represents gain, I dark Indicates dark current. σ d 2 The variance of all noise readouts and amplified noise. σ q 2 =1 / 12 (DN 2 ) represents quantization noise.

[0042] After determining the noise distribution of the error sources, this embodiment of the application also establishes the error propagation chain of the PMD. The intensity noise of the camera is transmitted to the phase through encoding and decoding, and further transmitted to the correspondence between the camera and the screen through ray tracing, thereby directly affecting the measurement gradient. Therefore, the reprojection error is used to measure the measurement uncertainty of the PMD.

[0043] Step 307: Optimize the configuration parameters of the sensor group based on Bayesian inference.

[0044] In the embodiment of this application, the system configuration parameters are optimized by combining Formula 2 and applying the Bayesian inference principle: Formula 2: ; In the formula, s represents the system configuration parameters. O F and O L These represent observations from the PMD system and the CCS system, respectively. P ( s | O F , O L ), P ( O F , O L | s )and P ( s ) represent the posterior probability, likelihood probability, and prior probability, respectively.P ( O F , O L The constant denoted by ) represents the normalization constant, which is omitted for simplicity. Furthermore, directly calculating the posterior probability is very difficult, and prior knowledge is often unavailable before system calibration. Therefore, the solver described above can be transformed into solving the problem of maximizing the likelihood probability as shown in Equation 3: Formula 3: ; Furthermore, when the measured noise distribution follows a Gaussian distribution, maximizing the likelihood can be transformed into a graphical optimization problem as shown in Equation 4: Formula 4: ; In the formula, and The information matrix, i.e., the inverse of the covariance matrix, is used to construct the weights for optimization.

[0045] Step 308: Construct a tightly coupled virtual measurement system based on the pre-calibrated parameters.

[0046] Step 309: Based on the tightly coupled virtual measurement system, determine the tightly coupled uncertainty corresponding to the sensor.

[0047] In this embodiment, considering both the uncertainty of the sensor itself and its pose parameters, similar to a system optimization process based on Bayesian inference, the overall measurement uncertainty of the system is determined through Monte Carlo simulation, i.e., the information matrix is ​​updated. .

[0048] Step 310: Using simulation methods, determine the total measurement uncertainty corresponding to the sensor group based on the tight coupling uncertainty.

[0049] Step 311: Combining the total measurement uncertainty, an integrated measurement model is constructed based on the integrated workpiece positioning method and the integrated form and position measurement estimator.

[0050] In this embodiment, the workpiece positioning and system configuration parameter optimization are placed within a unified optimization framework, and the calibration module is used as the regularization term of the positioning solver. This shortens the error propagation chain and improves the convergence accuracy. The process is shown in Equation 5 below: Formula 5: ; In the formula, m Represents workpiece positioning parameters. λ (·) indicates a regular expression injected by the calibration module. At this point, the location of the device under test and the optimal system configuration parameters can be obtained simultaneously.

[0051] In this embodiment, the observation noise distribution of the tightly coupled system is verified, a shape and position estimator is constructed based on the system's observation data, and the calibration and sensor uncertainties are marginalized and integrated into the solver, providing physical constraints for the measurement framework and improving measurement fidelity. This process is shown in Equation 6 below: Formula 6: ; In the formula, w These represent the mathematical model coefficients of the measured form and position data. Substituting them into the mathematical model will simultaneously yield the form and position data of the measured part. That is, in the embodiments of this application, the integrated measurement estimator is an estimator based on a mathematical model. Furthermore, the integrated measurement model is used for simultaneous form and position measurement of multifaceted optical elements.

[0052] In summary, the method provided in this application addresses the problem that deflection measurement cannot effectively measure both surface shape and pose simultaneously, and designs a highly efficient multi-sensor measurement system. It effectively enhances the accuracy of system calibration and workpiece positioning from a hardware perspective, thereby improving the measurement capability of deflection measurement for both surface shape and pose. Furthermore, it effectively avoids the numerical deviations inherent in traditional numerical optimization in surface shape reconstruction, enhancing the determinism of the measurement.

[0053] The above are merely optional embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A geometric-position integrated measurement method based on multi-sensor fusion, characterized in that, The method is applied to computer equipment within a shape and position integrated measurement system based on multi-sensor fusion; The shape and position integrated measurement system based on multi-sensor fusion includes a screen, camera, sensor group, turntable, standard parts and computer equipment, and the sensor group includes multiple dispersive confocal sensors; The screen, camera, chromatic confocal sensor, turntable, and standard components are communicatively connected to the computer equipment. The computer device is used to acquire sensing data from the sensor array, and the method includes: The internal and external parameters of the sensor group are calibrated; Determine the measurement uncertainty of each sensor within the sensor group; Based on the measurement uncertainty corresponding to the sensor, the parameters of the sensor group are optimized. By tightly coupling the measurement uncertainty corresponding to the sensor, we obtain the tightly coupled uncertainty corresponding to the sensor and the total measurement uncertainty corresponding to the sensor group. An integrated measurement model is constructed based on the total measurement uncertainty. This integrated measurement model is used to simultaneously measure the shape and position of multi-faceted optical elements.

2. The method according to claim 1, characterized in that, Before calibrating the internal and external parameters of the sensor group, the following steps are included: Determine the hardware parameters of the camera; Based on the camera's hardware parameters, the measurement range of the chromatic confocal sensor, the position of the chromatic confocal sensor, and the position of the turntable are determined.

3. The method according to claim 1, characterized in that, The calibration of the internal and external parameters of the sensor group includes: A reference calibration system is generated using the aforementioned standard components; The internal and external parameters of the sensor group are calibrated using the reference calibration system.

4. The method according to claim 1, characterized in that, Determining the measurement uncertainty of the sensors within the sensor group includes: A virtual measurement system is constructed based on pre-calibrated parameters; Based on the virtual measurement system, the measurement uncertainty of each sensor within the sensor group is determined.

5. The method according to claim 4, characterized in that, The method further includes: Based on Bayesian inference, the configuration parameters of the sensor group are optimized.

6. The method according to claim 1, characterized in that, The step of tightly coupling the measurement uncertainty corresponding to the sensor to obtain the tightly coupled uncertainty corresponding to the sensor and the total measurement uncertainty corresponding to the sensor group includes: Construct a tightly coupled virtual measurement system based on pre-calibrated parameters; Based on the tightly coupled virtual measurement system, determine the tightly coupled uncertainty corresponding to the sensor; By combining simulation methods, the total measurement uncertainty corresponding to the sensor group is determined.

7. The method according to claim 1, characterized in that, The construction of an integrated measurement model based on the total measurement uncertainty includes: Based on the total measurement uncertainty, the integrated measurement model is constructed using the integrated workpiece positioning method and the integrated form and position measurement estimator.

8. The method according to claim 8, characterized in that, The integrated measurement estimator is an estimator based on a mathematical model.

9. A shape and position integrated measurement system based on multi-sensor fusion, characterized in that, The shape and position integrated measurement system based on multi-sensor fusion includes a screen, camera, sensor group, turntable, standard parts and computer equipment, and the sensor group includes multiple dispersive confocal sensors; The screen, camera, chromatic confocal sensor, turntable, and standard components are communicatively connected to the computer equipment. The computer device is used to execute the multi-sensor fusion-based shape and position measurement method as described in any one of claims 1 to 8.

Images