A CMM automatic coordinate measuring machine and method thereof

Abstract

The present application discloses a CMM automatic coordinate measuring machine and a method thereof, which uses artificial intelligence technology based on deep learning to perform data analysis on the material performance description of the measurement target object and the temperature data of the measurement environment, respectively extracts the time series feature pattern of the measurement environment temperature and the semantic feature representation of the material performance of the measurement target object, and then, through the interactive fusion of the features of the two, fully understands the geometric deformation state of the target object, and intelligently generates the corresponding temperature compensation coefficient estimation value, thereby realizing the compensation correction of the geometric parameters of the measurement target object. In this way, time and labor costs can be significantly saved, and more efficient temperature compensation can be achieved, thereby effectively improving the measurement accuracy and efficiency of the CMM automatic coordinate measuring machine.

Description

Technical Field

[0001] The present application relates to the field of intelligent calibration, and more specifically, to a CMM automatic coordinate measuring machine and a method thereof. Background Art

[0002] As the manufacturing industry develops towards precision and intelligence, the coordinate measuring machine (CMM) is an indispensable quality control tool in industrial production, and its measurement accuracy directly affects the quality level of the product. CMM can obtain three-dimensional coordinate data of the workpiece surface by contact or non-contact method, and is widely used in the fields of automobile, aviation, machinery manufacturing, etc. For example, Chinese patent CN111089551B discloses a coordinate measuring machine, which includes a moving mechanism and a probe, and the probe and the moving mechanism cooperate with each other to measure the geometric parameters of the measured target.

[0003] However, in actual applications, the measurement results of CMM may be affected by the ambient temperature, resulting in deviations in the measurement results. Changes in ambient temperature can cause thermal expansion or contraction of the target material, which in turn causes changes in geometric dimensions. Therefore, in order to improve the measurement accuracy of CMM, temperature compensation of the measurement results is required.

[0004] Traditional temperature compensation methods mainly reduce temperature changes by controlling the measurement environment, or pre-calibrate the measurement parameters used by CMM automatic coordinate measuring machines at different ambient temperatures. However, these methods have certain limitations. For example, the cost of ambient temperature control is high, and it is difficult to achieve precise control of all workpieces and measurement conditions. The pre-calibration method consumes a lot of time and manpower costs, and its compensation effect may not be ideal when facing complex and changing production environments.

[0005] Therefore, an optimized CMM automatic coordinate measuring machine and method thereof are expected. Summary of the invention

[0006] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a CMM automatic coordinate measuring machine and method thereof, which uses artificial intelligence technology based on deep learning to perform data analysis on the material performance description of the measurement target object and the temperature data of the measurement environment, respectively extracts the time series feature pattern of the measurement environment temperature and the semantic feature representation of the material performance of the measurement target object, and then, through the interactive feature fusion of the two, fully understands the geometric deformation state of the target object, and intelligently generates the corresponding temperature compensation coefficient estimate, thereby realizing the compensation correction of the geometric parameters of the measurement target object. In this way, time and labor costs can be significantly saved, and more efficient temperature compensation can be achieved, thereby effectively improving the measurement accuracy and efficiency of the CMM automatic coordinate measuring machine.

[0007] According to one aspect of the present application, a CMM automatic coordinate measuring machine is provided, comprising:

[0008] Probe;

[0009] A moving mechanism for driving the probe to move;

[0010] A data acquisition and processing system, wherein the data acquisition and processing system is used to calculate geometric parameters of a measured target object based on data collected by the probe;

[0011] A temperature compensation system, wherein the temperature compensation system is used to compensate and correct the geometric parameters of the measured target object based on the temperature characteristics of the measurement environment and the material properties of the measured target object to obtain optimized geometric parameter measurement values;

[0012] Wherein, the temperature compensation system comprises:

[0013] A temperature compensation information acquisition module, used to acquire a material property description of the measurement target object and a time queue of the measurement temperature collected by the temperature sensor;

[0014] A temperature measurement time series feature extraction module, used to extract a temperature measurement time series implicit feature vector from the temperature measurement time queue;

[0015] A material property description semantic coding module, used for semantically coding the material property description of the measurement target object to obtain a material property semantic coding vector of the measurement target object;

[0016] A temperature compensation coefficient estimation module, used to generate a temperature compensation coefficient estimation value based on the interactive fusion characteristics between the measured temperature time series implicit feature vector and the measured target object material property semantic coding vector;

[0017] A parameter modulation module, configured to modulate the geometric parameters of the measured target object based on the estimated value of the temperature compensation coefficient to obtain the optimized geometric parameter measurement value;

[0018] Wherein, the temperature compensation coefficient estimation module includes:

[0019] A feature interaction fusion unit, used for performing significant interaction fusion on the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector to obtain a measured temperature-material property significant interaction fusion representation vector;

[0020] The decoding estimation unit is used to input the measured temperature-material property significant interaction fusion representation vector into a decoder-based temperature compensation coefficient estimation module to obtain the temperature compensation coefficient estimation value.

[0021] According to another aspect of the present application, a CMM automatic coordinate measurement method is provided, which includes:

[0022] Obtaining a material property description of the measurement target object and a time queue of measurement temperatures collected by a temperature sensor;

[0023] Extracting a measured temperature time series implicit feature vector from the measured temperature time queue;

[0024] Semantically encoding the material property description of the measurement target object to obtain a semantic encoding vector of the material property of the measurement target object;

[0025] Generate a temperature compensation coefficient estimate based on the interactive fusion features between the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector;

[0026] Modulating the geometric parameters of the measured target object based on the estimated value of the temperature compensation coefficient to obtain an optimized geometric parameter measurement value;

[0027] The method of generating an estimated value of a temperature compensation coefficient based on the interactive fusion features between the implicit feature vector of the measured temperature time series and the semantic coding vector of the material performance of the measured target object includes:

[0028] Performing significant interactive fusion on the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector to obtain a measured temperature-material property significant interactive fusion representation vector;

[0029] The measured temperature-material property significant interaction fusion representation vector is input into a decoder-based temperature compensation coefficient estimation module to obtain the temperature compensation coefficient estimation value.

[0030] Compared with the prior art, the CMM automatic coordinate measuring machine and method provided by the present application use artificial intelligence technology based on deep learning to analyze the material performance description of the measurement target object and the temperature data of the measurement environment, respectively extract the time series feature pattern of the measurement environment temperature and the semantic feature representation of the material performance of the measurement target object, and then, through the interactive fusion of the features of the two, fully understand the geometric deformation state of the target object, and intelligently generate the corresponding temperature compensation coefficient estimation value, so as to achieve compensation correction of the geometric parameters of the measurement target object. In this way, time and labor costs can be significantly saved, and more efficient temperature compensation can be achieved, thereby effectively improving the measurement accuracy and efficiency of the CMM automatic coordinate measuring machine. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0032] Figure 1 is a block diagram of a CMM automatic coordinate measuring machine according to an embodiment of the present application;

[0033] Figure 2 A schematic diagram of data flow of a CMM automatic coordinate measuring machine according to an embodiment of the present application;

[0034] Figure 3 A block diagram of a temperature compensation system in a CMM automatic coordinate measuring machine according to an embodiment of the present application;

[0035] Figure 4 Flow chart of a CMM automatic coordinate measurement method according to an embodiment of the present application. DETAILED DESCRIPTION

[0036] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0037] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0038] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0039] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0040] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0041] Traditional temperature compensation methods mainly reduce temperature changes by controlling the measurement environment, or pre-calibrate the measurement parameters used by the CMM automatic coordinate measuring machine at different ambient temperatures. However, these methods have certain limitations. For example, the cost of ambient temperature control is high, and it is difficult to achieve precise control of all workpieces and measurement conditions. The pre-calibration method consumes more time and manpower costs, and its compensation effect may not be ideal when facing a complex and changeable production environment. Therefore, an optimized CMM automatic coordinate measuring machine and method thereof are expected.

[0042] In the technical solution of the present application, a CMM automatic coordinate measuring machine is proposed. Figure 1 4 is a block diagram of a CMM automatic coordinate measuring machine according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of a CMM automatic coordinate measuring machine according to an embodiment of the present application. Figure 1 and Figure 2 As shown, the CMM automatic coordinate measuring machine according to the embodiment of the present application includes: a probe 310; a moving mechanism 320 for driving the probe to move; a data acquisition and processing system 330, wherein the data acquisition and processing system is used to calculate the geometric parameters of the target object to be measured based on the data collected by the probe; a temperature compensation system 340, wherein the temperature compensation system is used to compensate and correct the geometric parameters of the target object to be measured based on the temperature characteristics of the measurement environment and the material properties of the target object to be measured, so as to obtain optimized geometric parameter measurement values.

[0043] In particular, the probe 310. The probe is responsible for collecting data points at a specified position. According to different measurement requirements, the probe can be contact or non-contact.

[0044] In particular, the moving mechanism 320 is used to drive the probe to move. The moving mechanism is responsible for ensuring that the probe can accurately reach the predetermined measurement position and maintain stable movement during the entire measurement process. It is worth mentioning that the moving mechanism 320 and the probe 310 involved in the present application can adopt the structure disclosed in Chinese patent CN111089551B, and of course other equivalent structural configurations can also be adopted, which is not limited to this application.

[0045] In particular, the data acquisition and processing system 330, wherein the data acquisition and processing system is used to calculate the geometric parameters of the target object to be measured based on the data collected by the probe. In one example, after receiving the data collected by the probe, various geometric parameters such as length, width, height, diameter, roundness, flatness, straightness, etc. are calculated by the data acquisition and processing system. These parameters reflect the actual size and shape characteristics of the workpiece. It should be understood that calculating the geometric parameters of the target object to be measured based on the data collected by the probe is a prior art of the CNN automatic coordinate measuring machine. For example, it can use the principles disclosed in Chinese patent CN111089551B to calculate the geometric parameters of the target object. Of course, it can also use other principles to calculate the geometric parameters, which is not limited to this application.

[0046] In particular, the temperature compensation system 340 is used to compensate and correct the geometric parameters of the measured target object based on the temperature characteristics of the measurement environment and the material properties of the measured target object to obtain an optimized geometric parameter measurement value. In a specific example of the present application, Figure 3As shown, the temperature compensation system 340 includes: a temperature compensation information acquisition module 341, which is used to obtain the material property description of the measurement target object and the time queue of the measurement temperature collected by the temperature sensor; a measurement temperature time series feature extraction module 342, which is used to extract the measurement temperature time series implicit feature vector from the measurement temperature time queue; a material property description semantic encoding module 343, which is used to semantically encode the material property description of the measurement target object to obtain the measurement target object material property semantic encoding vector; a temperature compensation coefficient estimation module 344, which is used to generate a temperature compensation coefficient estimation value based on the interactive fusion characteristics between the measurement temperature time series implicit feature vector and the measurement target object material property semantic encoding vector; a parameter modulation module 345, which is used to modulate the geometric parameters of the measured target object based on the temperature compensation coefficient estimation value to obtain the optimized geometric parameter measurement value.

[0047] Specifically, the temperature compensation information acquisition module 341 is used to obtain the material property description of the measurement target object and the time queue of the measurement temperature collected by the temperature sensor. It should be understood that different materials have different physical properties such as thermal expansion coefficient and elastic modulus, which cause the materials to exhibit different thermal expansion or contraction behaviors under different temperature conditions. Therefore, obtaining the material property description of the measurement target object is the basis for evaluating the impact of temperature changes on its geometric shape. At the same time, since the geometric deformation of the measurement target object accumulates over time, in the technical solution of the present application, by continuously collecting temperature data to form a time queue of the measurement temperature, it is helpful to more accurately understand the dynamic impact of temperature changes on the geometry of the object, thereby achieving more accurate temperature compensation.

[0048] Specifically, the measured temperature time series feature extraction module 342 is used to extract the measured temperature time series implicit feature vector from the measured temperature time series. In order to fully capture the inherent laws and trends of temperature changes over time, the present application further performs time series encoding on the measured temperature time series to learn the time series dynamic change pattern of the measured temperature and generate the measured temperature time series implicit feature vector. In a specific example of the present application, a measured temperature time series encoder based on an LSTM model is used to encode the measured temperature time series. Due to its advantages in processing time series data, the LSTM model can effectively capture the long-term dependency of temperature changes and effectively perform time series modeling on the dynamic characteristics of temperature changes, thereby providing more accurate data support for subsequent temperature compensation.

[0049] Specifically, the material property description semantic encoding module 343 is used to semantically encode the material property description of the measurement target object to obtain the material property semantic encoding vector of the measurement target object. Since the material property description of the measurement target object is text data, such as "aluminum, thermal expansion coefficient is 23×10^ -6 / ℃". Therefore, in order to convert the unstructured text information into a digital format that can be processed by the machine learning model, it is necessary to further semantically encode the material property description of the measured target object to map it to a high-dimensional semantic feature space and capture the key information in the material description, thereby obtaining a semantic encoding vector of the material property of the measured target object. In a specific example of the present application, the BERT model is used to semantically encode the material property description of the measured target object, so as to utilize the bidirectional semantic encoding capability of the BERT model and fully understand the contextual semantic relationship in the material property description to obtain the semantic encoding vector of the material property of the measured target object.

[0050] Specifically, the temperature compensation coefficient estimation module 344 is used to generate an estimated value of the temperature compensation coefficient based on the interactive fusion characteristics between the implicit feature vector of the measured temperature series and the semantic coding vector of the material properties of the measured target object. In order to comprehensively utilize the complementary information of temperature changes and material properties descriptions, the present application further interactively fuses the implicit feature vector of the measured temperature series and the semantic coding vector of the material properties of the measured target object to more comprehensively and accurately describe the geometric deformation state of the measured target object. It is worth mentioning that, considering that traditional feature fusion methods usually adopt simple feature splicing or weighted summation methods, they are often unable to fully explore the correlation and interaction between the two. In this regard, the present application adopts a feature saliency interactive fusion method based on cross-modal implicit joint features, which uses the implicit joint features between temperature and material properties to guide the interaction and fusion between the two, and can effectively identify and strengthen the key connection between the two, thereby achieving a deeper level of feature fusion.

[0051] In an embodiment of the present application, a temperature compensation coefficient estimate is generated based on the interactive fusion features between the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector, including: first, inputting the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector into a cross-modal implicit joint feature capture network to obtain a measured temperature-material property cross-modal implicit joint feature vector; and performing feature activation based on a Sigmoid function on the measured temperature-material property cross-modal implicit joint feature vector to obtain the conditional feature vector. That is, a shared implicit joint feature is extracted from the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector through a cross-modal implicit joint feature capture network, and the extracted cross-modal implicit joint feature is nonlinearly transformed using a Sigmoid function to emphasize key information and suppress irrelevant noise, so as to obtain the measured temperature-material performance cross-modal implicit joint feature vector as a conditional feature vector; then, based on the semantic contribution of the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector relative to the conditional feature vector, feature modulation is performed on the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector to obtain a modulated measured temperature time series implicit feature vector and a modulated measured target object material performance semantic coding vector; that is, the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector are separated into The correlation between the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector is evaluated to determine the importance of the two to the geometric deformation state of the target object, and the original measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector are weighted modulated to effectively emphasize the key information in the two and suppress irrelevant information; further, the measured temperature time series implicit feature vector after modulation is used as the query vector, the measured target object material performance semantic coding vector after modulation is used as the key vector and the conditional feature vector is used as the value vector, and the measured temperature time series implicit feature vector after modulation, the measured target object material performance semantic coding vector after modulation and the conditional feature vector are input into the feature-to-feature significant guided interaction module based on the transformer structure. Here, the Transformer architecture is used to guide the interaction between features, so as to realize the significant guided fusion of cross-modal features, so as to generate the measured temperature-material performance significant interaction fusion representation vector. Finally, the measured temperature-material performance significant interaction fusion representation vector is input into the temperature compensation coefficient estimation module based on the decoder to obtain the estimated value of the temperature compensation coefficient. That is, the measured temperature-material property significant interaction fusion representation vector is input into the decoder-based temperature compensation coefficient estimation module for decoding and estimation.The decoder uses its internal multi-layer perception network structure to perform nonlinear transformation on the measured temperature-material property significant interaction fusion representation vector, which can effectively learn the interaction between temperature changes and material properties, thereby accurately predicting the estimated value of the temperature compensation coefficient corresponding to the geometric deformation of the target object under the current temperature conditions.

[0052] Among them, the process of inputting the measured temperature time series implicit feature vector and the measured target object material property semantic coding vector into the cross-modal implicit joint feature capture network to obtain the measured temperature-material property cross-modal implicit joint feature vector includes: fusing the measured temperature time series implicit feature vector and the measured target object material property semantic coding vector according to the position points and passing them through a neural network layer based on the tanh function to obtain the measured temperature-material property cross-modal implicit joint feature vector.

[0053] More specifically, based on the semantic contribution of the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector relative to the conditional feature vector, the process of feature modulating the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector to obtain the modulated measured temperature time series implicit feature vector and the modulated measured target object material performance semantic coding vector includes: calculating the first semantic contribution of the measured temperature time series implicit feature vector relative to the conditional feature vector; calculating the second semantic contribution of the measured target object material performance semantic coding vector to the conditional feature vector; normalizing the first semantic contribution and the second semantic contribution, and using the normalized first semantic contribution and the second semantic contribution to weighted modulate the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector to obtain the modulated measured temperature time series implicit feature vector and the modulated measured target object material performance semantic coding vector. Among them, calculating the first semantic contribution of the measured temperature time series implicit feature vector relative to the conditional feature vector includes: calculating the dot division vector between the measured temperature time series implicit feature vector and the conditional feature vector, and calculating the base-two logarithmic value of the absolute value of each eigenvalue in the dot division vector to obtain a semantic correlation representation vector; calculating the dot product vector between the semantic correlation representation vector and the measured temperature time series implicit feature vector, and calculating an exponential function value with base e and the sum of the eigenvalues ​​of the dot product vector as an exponent to obtain the first semantic contribution.

[0054] More specifically, the process of inputting the modulated measured temperature time series implicit feature vector, the modulated measured target object material property semantic coding vector and the conditional feature vector into the feature-to-feature significant guided interaction module based on the converter structure to obtain the measured temperature-material property significant interaction fusion representation vector includes: multiplying the modulated measured temperature time series implicit feature vector by the transposed vector of the modulated measured target object material property semantic coding vector and then dividing it by the square root of the length of the measured target object material property semantic coding vector to obtain the measured temperature-material property query attention score matrix; passing the measured temperature-material property query attention score matrix through a softmax function and then multiplying it by the conditional feature vector to obtain the measured temperature-material property significant interaction fusion representation vector.

[0055] In summary, in the above embodiment, the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector are significantly interactively fused to obtain the measured temperature-material property significant interactive fusion representation vector, including: processing the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector with the following feature interactive fusion formula to obtain the measured temperature-material property significant interactive fusion representation vector, wherein the feature interactive fusion formula is:

[0056]

[0057]

[0058] in, represents the implicit eigenvector of the measured temperature time series, represents the semantic coding vector of the material performance of the measured target object, Indicates adding by position point. represents the implicit joint eigenvector of measured temperature-material properties across modalities, and Represent the weight matrix and bias parameters of the neural network layer, respectively. represents the hyperbolic tangent function, represents the sigmoid activation function, represents the conditional eigenvector, , and Respectively represent the first The characteristic value of the measured target object material property semantic encoding vector eigenvalue and the conditional eigenvector eigenvalues, is the absolute value symbol, represents the logarithmic function with base 2, represents the exponential function with base e, represents the length of the measured temperature time series implicit feature vector, and the measured temperature time series implicit feature vector, the measured target object material property semantic encoding vector and the conditional feature vector have the same length, and Represent the first semantic contribution and the second semantic contribution respectively, and They represent the normalized first semantic contribution and second semantic contribution respectively. and They represent the implicit feature vector of the measured temperature time series after modulation and the semantic encoding vector of the material performance of the measured target object after modulation, represents the transpose of a vector, represents the matrix multiplication operation, is the normalized exponential function, A vector representing the significant fusion of the measured temperature-material property interactions is shown.

[0059] Preferably, in an example of the present application, the step of passing the measured temperature-material property significant interaction fusion representation vector through a decoder-based temperature compensation coefficient estimation module to obtain a temperature compensation coefficient estimation value comprises:

[0060] Calculate the absolute value sum of each eigenvalue of the measured temperature-material property significant interaction fusion representation vector to obtain a first measured temperature-material property significant interaction fusion spatial structure value, and calculate the square root of the sum of the squares of each eigenvalue of the measured temperature-material property significant interaction fusion representation vector to obtain a second measured temperature-material property significant interaction fusion spatial structure value;

[0061] Multiply each eigenvalue of the measured temperature-material property significant interaction fusion representation vector by the first measured temperature-material property significant interaction fusion spatial structure value and the second measured temperature-material property significant interaction fusion spatial structure value to obtain a first measured temperature-material property significant interaction fusion structure reference value and a second measured temperature-material property significant interaction fusion structure reference value corresponding to each eigenvalue;

[0062] Multiply each eigenvalue of the measured temperature-material property significant interaction fusion representation vector by the length of the measured temperature-material property significant interaction fusion representation vector and the square root of the length to obtain a first measured temperature-material property significant interaction fusion scale transformation value and a second measured temperature-material property significant interaction fusion scale transformation value corresponding to each eigenvalue;

[0063] Dividing the first measured temperature-material property significant interaction fusion structure reference value by the difference between the first measured temperature-material property significant interaction fusion space structure value and the first measured temperature-material property significant interaction fusion scale transformation value to obtain the first measured temperature-material property significant interaction fusion transformation adjustment value;

[0064] Dividing the second measured temperature-material property significant interactive fusion structure reference value by the difference between the second measured temperature-material property significant interactive fusion space structure value and the second measured temperature-material property significant interactive fusion scale transformation value to obtain a second measured temperature-material property significant interactive fusion transformation adjustment value;

[0065] Calculating a weighted sum of the first measured temperature-material property significant interaction fusion transformation adjustment value and the second measured temperature-material property significant interaction fusion transformation adjustment value to obtain each eigenvalue of the optimized measured temperature-material property significant interaction fusion representation vector; and

[0066] The optimized measured temperature-material property significant interaction fusion representation vector is passed through a decoder-based temperature compensation coefficient estimation module to obtain a temperature compensation coefficient estimation value.

[0067] Here, the measured temperature-material property significant interaction fusion representation vector is denoted as The optimization expression is:

[0068]

[0069]

[0070] in, is the vector representing the significant interaction between measured temperature and material properties, are the eigenvalues ​​of the vector representing the significant interaction between the measured temperature and material properties, is the length of the vector representing the measured temperature-material property significant interaction, represents the field of real numbers, is the first measured temperature-material property significant interaction integrating spatial structure value, The second measured temperature-material property significant interaction fusion spatial structure value, is the first measured temperature-material property significant interaction fusion transformation adjustment value, is the second measured temperature-material property significant interaction fusion transformation adjustment value, is a first measured temperature-material property significant interactive fusion transformation adjustment vector composed of each of the first measured temperature-material property significant interactive fusion transformation adjustment values, is a second measured temperature-material property significant interactive fusion transformation adjustment vector composed of the respective second measured temperature-material property significant interactive fusion transformation adjustment values, is the weight hyperparameter, is the dot product, It's a little bit more. is the optimized measured temperature-material property significant interaction fusion representation vector.

[0071] That is, the applicant of the present application took into account that the measured temperature time series implicit feature vector and the measured target object material property semantic coding vector respectively represent the measured temperature time series correlation feature and the measured target object material property text semantic coding feature, and when significantly guiding the interaction between their features with the joint implicit feature as conditional information, the difference in feature modal expression between the measured temperature time series implicit feature vector and the measured target object material property semantic coding vector will make the feature manifold of the joint implicit feature between the two have more significant heterogeneity relative to the measured temperature time series implicit feature vector and the measured target object material property semantic coding vector, thereby making the cross-modal fusion feature representation of the measured temperature-material property significant interaction fusion representation vector have a complex spatial structure, thereby expecting to improve its decoding regression convergence and generalization effects under complex spatial structures.

[0072] Therefore, in the preferred example, with respect to the spatial structure information of the feature set of the measured temperature-material property significant interaction fusion representation vector in the high-dimensional space, the scale-based frame transformation of each eigenvalue of the measured temperature-material property significant interaction fusion representation vector is performed by taking the class-norm spatial structured representation of the measured temperature-material property significant interaction fusion representation vector as a reference window, and the spatial structure-based frame attention weight adjustment of each eigenvalue of the measured temperature-material property significant interaction fusion representation vector is implemented to ensure the invariance of the spatial transformation (translation, scaling and rotation) of the measured temperature-material property significant interaction fusion representation vector under the feature space interaction, thereby improving the convergence and generalization effect of the decoding regression of the feature set of the measured temperature-material property significant interaction fusion representation vector under the complex spatial structure representation, and improving the accuracy of the temperature compensation coefficient estimation value obtained by passing it through the decoder-based temperature compensation coefficient estimation module.

[0073] Specifically, the parameter modulation module 345 is used to modulate the geometric parameters of the measured target object based on the temperature compensation coefficient estimate to obtain the optimized geometric parameter measurement value. In an example of the present application, when the geometric parameters of the measured target object are modulated based on the temperature compensation coefficient estimate, the measurement error is compensated by addition operation to obtain the optimized geometric parameter measurement value.

[0074] As described above, the CMM automatic coordinate measuring machine 300 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a CMM automatic coordinate measurement algorithm. In a possible implementation, the CMM automatic coordinate measuring machine 300 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the CMM automatic coordinate measuring machine 300 can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the CMM automatic coordinate measuring machine 300 can also be one of the many hardware modules of the wireless terminal.

[0075] Alternatively, in another example, the CMM automatic coordinate measuring machine 300 and the wireless terminal may also be separate devices, and the CMM automatic coordinate measuring machine 300 may be connected to the wireless terminal through a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0076] Furthermore, a CMM automatic coordinate measurement method is also provided.

[0077] Figure 4 FIG. 1 is a flow chart of a CMM automatic coordinate measurement method according to an embodiment of the present application. Figure 4 As shown, the CMM automatic coordinate measurement method according to the embodiment of the present application includes the steps of: S1, acquiring the material property description of the measurement target object and the time queue of the measurement temperature collected by the temperature sensor; S2, extracting the measurement temperature time series implicit feature vector from the measurement temperature time queue; S3, semantically encoding the material property description of the measurement target object to obtain the measurement target object material property semantic encoding vector; S4, generating a temperature compensation coefficient estimate based on the interactive fusion characteristics between the measurement temperature time series implicit feature vector and the measurement target object material property semantic encoding vector; S5, modulating the geometric parameters of the measured target object based on the temperature compensation coefficient estimate to obtain optimized geometric parameter measurement values.

[0078] In summary, the CMM automatic coordinate measurement method according to the embodiment of the present application is explained, which uses artificial intelligence technology based on deep learning to perform data analysis on the material performance description of the measurement target object and the temperature data of the measurement environment, and respectively extracts the time series feature pattern of the measurement environment temperature and the semantic feature representation of the material performance of the measurement target object, and then, through the interactive fusion of the features of the two, the geometric deformation state of the target object is fully understood, and the corresponding temperature compensation coefficient estimation value is intelligently generated, so as to achieve compensation correction of the geometric parameters of the measurement target object. In this way, time and labor costs can be significantly saved, and more efficient temperature compensation can be achieved, thereby effectively improving the measurement accuracy and efficiency of the CMM automatic coordinate measuring machine.

[0079] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A CMM automatic coordinate measuring machine, characterized in that: include: Probe; A moving mechanism for driving the probe to move; A data acquisition and processing system, wherein the data acquisition and processing system is used to calculate geometric parameters of a measured target object based on data collected by the probe; A temperature compensation system, wherein the temperature compensation system is used to compensate and correct the geometric parameters of the measured target object based on the temperature characteristics of the measurement environment and the material properties of the measured target object to obtain optimized geometric parameter measurement values; Wherein, the temperature compensation system comprises: A temperature compensation information acquisition module, used to acquire a material property description of the measurement target object and a time queue of the measurement temperature collected by the temperature sensor; A temperature measurement time series feature extraction module, used to extract a temperature measurement time series implicit feature vector from the temperature measurement time queue; A material property description semantic coding module, used for semantically coding the material property description of the measurement target object to obtain a material property semantic coding vector of the measurement target object; A temperature compensation coefficient estimation module, used to generate a temperature compensation coefficient estimation value based on the interactive fusion characteristics between the measured temperature time series implicit feature vector and the measured target object material property semantic coding vector; A parameter modulation module, configured to modulate the geometric parameters of the measured target object based on the estimated value of the temperature compensation coefficient to obtain the optimized geometric parameter measurement value; Wherein, the temperature compensation coefficient estimation module includes: A feature interaction fusion unit, used for performing significant interaction fusion on the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector to obtain a measured temperature-material property significant interaction fusion representation vector; The decoding estimation unit is used to input the measured temperature-material property significant interaction fusion representation vector into a decoder-based temperature compensation coefficient estimation module to obtain the temperature compensation coefficient estimation value.

2. The CMM automatic coordinate measuring machine according to claim 1, characterized in that: The temperature measurement time series feature extraction module is used to: The time queue of the measured temperature is input into the measured temperature time series encoder based on the LSTM model to obtain the measured temperature time series implicit feature vector.

3. The CMM automatic coordinate measuring machine according to claim 2, characterized in that: The feature interactive fusion unit includes: A condition information construction subunit, used for extracting a cross-modal implicit joint feature between the measured temperature time series implicit feature vector and the measured target object material performance semantic encoding vector as a condition feature vector; A feature modulation subunit is used to perform feature modulation on the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector based on the semantic contribution of the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector relative to the conditional feature vector to obtain a modulated measured temperature time series implicit feature vector and a modulated measured target object material performance semantic coding vector; The cross-domain interactive fusion subunit is used to perform cross-domain interactive fusion encoding based on the attention mechanism on the implicit feature vector of the measured temperature time series after modulation, the semantic encoding vector of the material performance of the measured target object after modulation, and the conditional feature vector to obtain the measured temperature-material performance significant interactive fusion representation vector.

4. The CMM automatic coordinate measuring machine according to claim 3, characterized in that: The condition information construction subunit is used to: Inputting the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector into a cross-modal implicit joint feature capture network to obtain a measured temperature-material property cross-modal implicit joint feature vector; The measured temperature-material property cross-modal implicit joint feature vector is subjected to feature activation based on a Sigmoid function to obtain the conditional feature vector.

5. The CMM automatic coordinate measuring machine according to claim 4, characterized in that: The characteristic modulation subunit is used for: Calculating a first semantic contribution of the measured temperature time series implicit feature vector relative to the conditional feature vector; Calculate a second semantic contribution of the measured target object material property semantic encoding vector to the conditional feature vector; The first semantic contribution and the second semantic contribution are normalized, and the normalized first semantic contribution and the second semantic contribution are used to weighted modulate the measured temperature time series implicit feature vector and the measured target object material performance semantic coding vector to obtain the modulated measured temperature time series implicit feature vector and the modulated measured target object material performance semantic coding vector.

6. The CMM automatic coordinate measuring machine according to claim 5, characterized in that: The cross-domain interaction fusion subunit is used to: Using the implicit feature vector of the measured temperature time series after modulation as the query vector, the semantic encoding vector of the material performance of the measured target object after modulation as the key vector and the conditional feature vector as the value vector, the implicit feature vector of the measured temperature time series after modulation, the semantic encoding vector of the material performance of the measured target object after modulation and the conditional feature vector are input into a significant guided interaction module between features based on a converter structure to obtain the measured temperature-material performance significant interaction fusion representation vector.

7. A CMM automatic coordinate measuring method of a CMM automatic coordinate measuring machine according to claim 1, characterized in that: include: Obtaining a material property description of the measurement target object and a time queue of measurement temperatures collected by a temperature sensor; Extracting a measured temperature time series implicit feature vector from the measured temperature time queue; Semantically encoding the material property description of the measurement target object to obtain a semantic encoding vector of the material property of the measurement target object; Generate a temperature compensation coefficient estimate based on the interactive fusion features between the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector; Modulating the geometric parameters of the measured target object based on the estimated value of the temperature compensation coefficient to obtain an optimized geometric parameter measurement value; The method of generating an estimated value of a temperature compensation coefficient based on the interactive fusion features between the implicit feature vector of the measured temperature time series and the semantic coding vector of the material performance of the measured target object includes: Performing significant interactive fusion on the measured temperature time series implicit feature vector and the measured target object material property semantic encoding vector to obtain a measured temperature-material property significant interactive fusion representation vector; The measured temperature-material property significant interaction fusion representation vector is input into a decoder-based temperature compensation coefficient estimation module to obtain the temperature compensation coefficient estimation value.

Images