Segmentation method and device in machine tool track fitting, storage medium and computer equipment

Abstract

The invention discloses a segmentation method and device in machine tool track fitting, a storage medium and computer equipment, and the method comprises the steps: obtaining a plurality of groups of machine tool historical track points, selecting any group of machine tool historical track points, and according to the distance between adjacent machine tool historical track points in the selected group of machine tool historical track points, carrying out segmentation on the selected group of machine tool historical track points; the fitting types of the historical track points of the selected set of machine tools are determined according to the angle formed by the historical track points of the machine tools and the angle formed by the historical track points of the machine tools, and the fitting types comprise segmental fitting and point fitting; generating a machine tool track point training data set to train a machine tool track point classification model based on each group of machine tool historical track points and corresponding fitting types, the machine tool track point classification model being established based on a neural network; and based on the trained machine tool track point classification model, classifying the fitting types of each group of target fitting machine tool track points, and fitting each group of target fitting machine tool track points based on the classified fitting types to obtain a machine tool fitting track. And the online interpolation efficiency of the numerical control system is improved.

Description

Technical Field

[0001] The present application relates to the technical field of numerical control machining, and in particular to a segmentation method and device, a storage medium, and a computer device in machine tool trajectory fitting. Background Art

[0002] In the CNC machining process, spline curves are usually used to fit G01 tiny line segments to achieve trajectory compression and improve trajectory continuity. However, the complexity of spline curve fitting calculation is proportional to the number of tiny line segments to be fitted. In order to reduce the amount of single fitting calculation, the entire tool trajectory is usually divided into several intervals for segmented fitting. The artificial segmentation process will lead to low efficiency in overall trajectory calculation, which in turn will lead to a longer time for the overall fitting process, especially when there are a large number of G01 tiny line segment data, which is not conducive to the online interpolation work of the CNC system. Summary of the invention

[0003] In view of this, the present application provides a segmentation method and device, storage medium, and computer equipment in machine tool trajectory fitting, obtains multiple groups of machine tool historical trajectory points, selects any group of machine tool historical trajectory points, and determines the fitting type of the selected group of machine tool historical trajectory points according to the distance between adjacent machine tool historical trajectory points in the selected group of machine tool historical trajectory points and the angle formed between each machine tool historical trajectory point. The fitting type includes segment fitting and point fitting; based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network; based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified, and each group of target fitting machine tool trajectory points is fitted based on the classified fitting type to obtain a machine tool fitting trajectory. The online interpolation efficiency of the CNC system is improved.

[0004] According to one aspect of the present application, a segmentation method for machine tool trajectory fitting is provided, the method comprising:

[0005] Acquire multiple groups of machine tool historical trajectory points, wherein each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point, and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship;

[0006] Select any group of machine tool historical trajectory points, and determine the fitting type of the selected group of machine tool historical trajectory points according to the distances between adjacent machine tool historical trajectory points and the angles formed between the machine tool historical trajectory points in the selected group of machine tool historical trajectory points, wherein the fitting type includes segment fitting and point fitting;

[0007] Based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network;

[0008] Based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified, and each group of target fitting machine tool trajectory points is fitted based on the classified fitting type to obtain the machine tool fitting trajectory.

[0009] Optionally, determining the fitting type of the selected group of machine tool historical trajectory points according to the distances between adjacent machine tool historical trajectory points and the angles formed between the machine tool historical trajectory points in the selected group of machine tool historical trajectory points includes:

[0010] In the selected group of machine tool history trajectory points, determining a first connection line between the first machine tool history trajectory point and the second machine tool history trajectory point, and a second connection line between the second machine tool history trajectory point and the third machine tool history trajectory point;

[0011] If the distance between adjacent machine tool historical trajectory points meets the preset distance threshold range, and the angle formed by the first connecting line and the second connecting line meets the preset angle threshold range, then the fitting type of the selected group of machine tool historical trajectory points is segmented fitting;

[0012] If the distance between adjacent machine tool historical trajectory points does not meet the preset distance threshold range, and / or the angle formed by the first connecting line and the second connecting line does not meet the preset angle threshold range, the fitting type of the selected group of machine tool historical trajectory points is point fitting.

[0013] Optionally, fitting each group of target fitting machine tool trajectory points based on the classified fitting types to obtain the machine tool fitting trajectory includes:

[0014] For any group of target fitting machine tool trajectory points, when the fitting type classified by the group of target fitting machine tool trajectory points is segment fitting, the group of target fitting machine tool trajectory points is output as trajectory segments for machine tool trajectory fitting;

[0015] When the fitting type classified from the group of target fitting machine tool trajectory points is point fitting, the group of target fitting machine tool trajectory points are output as trajectory points for performing machine tool trajectory fitting to obtain a machine tool fitting trajectory.

[0016] Optionally, before classifying the fitting type of each group of target fitting machine tool trajectory points based on the trained machine tool trajectory point classification model, the method further includes:

[0017] A target fitting machine tool trajectory point sequence is obtained, and the target fitting machine tool trajectory point sequence is divided according to a preset moving step length to obtain multiple groups of target fitting machine tool trajectory points, wherein each group of target fitting machine tool trajectory points respectively includes a first target fitting machine tool trajectory point, a second target fitting machine tool trajectory point and a third target fitting machine tool trajectory point.

[0018] Optionally, after dividing the target fitting machine tool trajectory point sequence according to a preset moving step length to obtain multiple groups of target fitting machine tool trajectory points, the method further includes:

[0019] If any group of target fitting machine tool trajectory points includes only one target fitting machine tool trajectory point, the target fitting machine tool trajectory point is output as a trajectory point;

[0020] If any group of target fitting machine tool trajectory points includes only two target fitting machine tool trajectory points, when the distance between the two target fitting machine tool trajectory points meets the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory segments; when the distance between the two target fitting machine tool trajectory points does not meet the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory points respectively.

[0021] Optionally, each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point within a preset spherical radius area from the first machine tool historical trajectory point, and a third machine tool historical trajectory point within a preset spherical radius area from the second machine tool historical trajectory point.

[0022] Optionally, the machine tool trajectory point training data set includes a preset number of groups of machine tool historical trajectory points.

[0023] According to another aspect of the present application, a segmentation device for machine tool trajectory fitting is provided, the device comprising:

[0024] A historical trajectory point acquisition module is used to acquire multiple groups of machine tool historical trajectory points, wherein each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship;

[0025] A historical trajectory point classification module is used to select any group of machine tool historical trajectory points, and determine the fitting type of the selected group of machine tool historical trajectory points according to the distance between adjacent machine tool historical trajectory points in the selected group of machine tool historical trajectory points and the angle formed between each machine tool historical trajectory point, wherein the fitting type includes segment fitting and point fitting;

[0026] A classification model training module, used for generating a machine tool trajectory point training data set based on each group of machine tool historical trajectory points and their corresponding fitting types, and training a machine tool trajectory point classification model based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network;

[0027] The machine tool trajectory fitting module is used to classify the fitting type of each group of target fitting machine tool trajectory points based on the trained machine tool trajectory point classification model, and fit each group of target fitting machine tool trajectory points based on the classified fitting type to obtain the machine tool fitting trajectory.

[0028] Optionally, the historical trajectory point classification module is further used to:

[0029] In the selected group of machine tool history trajectory points, determining a first connection line between the first machine tool history trajectory point and the second machine tool history trajectory point, and a second connection line between the second machine tool history trajectory point and the third machine tool history trajectory point;

[0030] If the distance between adjacent machine tool historical trajectory points meets the preset distance threshold range, and the angle formed by the first connecting line and the second connecting line meets the preset angle threshold range, then the fitting type of the selected group of machine tool historical trajectory points is segmented fitting;

[0031] If the distance between adjacent machine tool historical trajectory points does not meet the preset distance threshold range, and / or the angle formed by the first connecting line and the second connecting line does not meet the preset angle threshold range, the fitting type of the selected group of machine tool historical trajectory points is point fitting.

[0032] Optionally, the machine tool trajectory fitting module is further used for:

[0033] For any group of target fitting machine tool trajectory points, when the fitting type classified by the group of target fitting machine tool trajectory points is segment fitting, the group of target fitting machine tool trajectory points is output as trajectory segments for machine tool trajectory fitting;

[0034] When the fitting type classified from the group of target fitting machine tool trajectory points is point fitting, the group of target fitting machine tool trajectory points are output as trajectory points for performing machine tool trajectory fitting to obtain a machine tool fitting trajectory.

[0035] Optionally, the machine tool trajectory fitting module is further used for:

[0036] A target fitting machine tool trajectory point sequence is obtained, and the target fitting machine tool trajectory point sequence is divided according to a preset moving step length to obtain multiple groups of target fitting machine tool trajectory points, wherein each group of target fitting machine tool trajectory points respectively includes a first target fitting machine tool trajectory point, a second target fitting machine tool trajectory point and a third target fitting machine tool trajectory point.

[0037] Optionally, the machine tool trajectory fitting module is further used for:

[0038] If any group of target fitting machine tool trajectory points includes only one target fitting machine tool trajectory point, the target fitting machine tool trajectory point is output as a trajectory point;

[0039] If any group of target fitting machine tool trajectory points includes only two target fitting machine tool trajectory points, when the distance between the two target fitting machine tool trajectory points meets the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory segments; when the distance between the two target fitting machine tool trajectory points does not meet the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory points respectively.

[0040] Optionally, each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point within a preset spherical radius area from the first machine tool historical trajectory point, and a third machine tool historical trajectory point within a preset spherical radius area from the second machine tool historical trajectory point.

[0041] Optionally, the machine tool trajectory point training data set includes a preset number of groups of machine tool historical trajectory points.

[0042] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the segmentation method in the machine tool trajectory fitting is implemented.

[0043] According to another aspect of the present application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the above-mentioned segmentation method in machine tool trajectory fitting when executing the program.

[0044] By means of the above technical scheme, the present application provides a segmentation method and device, storage medium, and computer equipment in machine tool trajectory fitting, obtains multiple groups of machine tool historical trajectory points, selects any group of machine tool historical trajectory points, and determines the fitting type of the selected group of machine tool historical trajectory points according to the distance between adjacent machine tool historical trajectory points in the selected group of machine tool historical trajectory points and the angle formed between each machine tool historical trajectory point. The fitting type includes segment fitting and point fitting; based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network; based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified, and each group of target fitting machine tool trajectory points is fitted based on the classified fitting type to obtain a machine tool fitting trajectory. The online interpolation efficiency of the CNC system is improved.

[0045] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0047] Figure 1 A schematic flow chart of a segmentation method in machine tool trajectory fitting provided in an embodiment of the present application is shown;

[0048] Figure 2 A schematic diagram of a machine tool trajectory point classification model training provided by an embodiment of the present application is shown;

[0049] Figure 3 A schematic flow chart of another segmentation method in machine tool trajectory fitting provided by an embodiment of the present application is shown;

[0050] Figure 4 A schematic diagram showing the classification of fitting types of a machine tool historical trajectory point group provided by an embodiment of the present application is shown;

[0051] Figure 5 A schematic flow chart of another segmentation method in machine tool trajectory fitting provided in an embodiment of the present application is shown;

[0052] Figure 6 A schematic structural diagram of a segmentation device in machine tool trajectory fitting provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0053] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0054] In this embodiment, a segmentation method for machine tool trajectory fitting is provided, such as Figure 1 As shown, the method includes:

[0055] Step 101 , obtaining multiple groups of machine tool historical trajectory points, wherein each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship.

[0056] Global trajectory smooth fitting refers to fitting discrete trajectory points into smooth continuous curves or paths through mathematical models or algorithms. It is very important in CNC machining or robot motion control, and can improve the motion accuracy and stability of the machine. Machine learning can learn general rules from limited observation data and use these rules to predict unknown data. In the field of CNC technology, machine tool trajectories are usually fitted through G code programs. G code programs contain a large number of G01 tiny line segments. In order to reduce the amount of calculation for a single fitting, the entire tool rule is usually compressed into several segments for fitting separately.

[0057] In the above embodiment of the present application, multiple groups of machine tool historical trajectory points are obtained, wherein each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point, and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship. The preset three-dimensional spatial relationship is, for example, the spatial three-dimensional coordinates corresponding to any trajectory point in the machine tool trajectory.

[0058] Step 102, select any group of machine tool historical trajectory points, and determine the fitting type of the selected group of machine tool historical trajectory points based on the distance between adjacent machine tool historical trajectory points in the selected group and the angle formed by each machine tool historical trajectory point, wherein the fitting type includes segment fitting and point fitting.

[0059] Next, select any group of machine tool history trajectory points, and determine the fitting type of the selected group of machine tool history trajectory points based on the distance between adjacent machine tool history trajectory points in the selected group and the angle between each machine tool history trajectory point. The fitting types include segment fitting and point fitting, in preparation for subsequent model training.

[0060] Step 103, based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network.

[0061] Next, based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set. The machine tool trajectory point classification model is established based on a neural network. The model training process can be regarded as a classification task of points in three-dimensional space. A fully connected neural network (DNN) is selected to implement it. The three coordinates of each point (the first machine tool historical trajectory point, the second machine tool historical trajectory point, and the third machine tool historical trajectory point in a group of machine tool historical trajectory points) are directly expanded as input features. The hidden layer can be set to a layer of 32 neurons, and the output layer is a fully connected layer with 2 neurons. At the same time, the cross entropy loss function and the adam optimizer are used to train the model. Figure 2As shown, each group of machine tool historical trajectory points is input into the machine tool trajectory point classification model, and the output is a fitting type 0 or 1, 1 represents segment fitting, and 0 represents point fitting.

[0062] Step 104 , based on the trained machine tool trajectory point classification model, classify the fitting type of each group of target fitting machine tool trajectory points, and fit each group of target fitting machine tool trajectory points based on the classified fitting type to obtain the machine tool fitting trajectory.

[0063] Finally, based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified, and each group of target fitting machine tool trajectory points is fitted based on the classified fitting type to obtain the machine tool fitting trajectory.

[0064] By applying the technical solution of this embodiment, multiple groups of machine tool historical trajectory points are obtained, and any group of machine tool historical trajectory points is selected. According to the distance between adjacent machine tool historical trajectory points in the selected group of machine tool historical trajectory points and the angle formed between each machine tool historical trajectory point, the fitting type of the selected group of machine tool historical trajectory points is determined, and the fitting type includes segment fitting and point fitting; based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network; based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified, and each group of target fitting machine tool trajectory points is fitted based on the classified fitting type to obtain a machine tool fitting trajectory. The online interpolation efficiency of the CNC system is improved.

[0065] Further, as a refinement and expansion of the specific implementation of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, another segmentation method in machine tool trajectory fitting is provided, such as Figure 3 As shown, the method includes:

[0066] Step 201 , obtaining multiple groups of machine tool historical trajectory points, wherein each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship.

[0067] Step 202, select any group of machine tool historical trajectory points, and determine a first connection line between the first machine tool historical trajectory point and the second machine tool historical trajectory point, and a second connection line between the second machine tool historical trajectory point and the third machine tool historical trajectory point in the selected group of machine tool historical trajectory points.

[0068] Step 203: If the distance between adjacent machine tool historical trajectory points meets a preset distance threshold range, and the angle formed by the first connecting line and the second connecting line meets a preset angle threshold range, then the fitting type of the selected group of machine tool historical trajectory points is segmented fitting.

[0069] Step 204: if the distance between adjacent machine tool historical trajectory points does not meet the preset distance threshold range, and / or the angle formed by the first connecting line and the second connecting line does not meet the preset angle threshold range, then the fitting type of the selected group of machine tool historical trajectory points is point fitting.

[0070] In the above embodiment of the present application, multiple groups of machine tool historical trajectory points are obtained, and each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point, and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship, in preparation for subsequent model training. Specifically, the multiple groups of machine tool historical trajectory points obtained are shown in Table 1, and each group of machine tool historical trajectory points respectively represents the coordinates of a trajectory point on the x-axis, y-axis, and z-axis:

[0071] Table 1

[0072]

[0073]

[0074] Specifically, every three machine tool history trajectory points of the G01 micro segment are processed. First, a point P is randomly generated. 1 (the first machine tool history track point), in P 1 A point P is randomly generated in the spherical area corresponding to the preset spherical radius of the sphere center. 2 (The second machine tool history track point), followed by P 2 As the center of the sphere, a point P3 (the third machine tool historical trajectory point) is randomly generated in the spherical area corresponding to the preset spherical radius.

[0075] Select any group of machine tool history trajectory points, and determine the first connection line between the first machine tool history trajectory point and the second machine tool history trajectory point, and the second connection line between the second machine tool history trajectory point and the third machine tool history trajectory point in the selected group of machine tool history trajectory points. To this end, P can be obtained by calculation. 1 P 2 The length of the line segment between 2 P 3 The length of the line segment between 1 P 2 P 3The angle formed is used as a data of the training set. If the distance between adjacent machine tool historical trajectory points meets the preset distance threshold range, and the angle formed by the first line and the second line meets the preset angle threshold range, then the fitting type of the selected group of machine tool historical trajectory points is segment fitting. If the distance between adjacent machine tool historical trajectory points does not meet the preset distance threshold range, and / or the angle formed by the first line and the second line does not meet the preset angle threshold range, then the fitting type of the selected group of machine tool historical trajectory points is point fitting.

[0076] Specifically, by comparing with the threshold, there are eight classification situations corresponding to the three points, such as Figure 4 As shown in the figure, the corresponding labels (fitting types) are finally generated. There are two types: 0 and 1. LY means that the distance between the two points meets the threshold condition, and LN does not meet it; AY means that the angle formed by the three points meets the threshold condition, and AN does not meet it. Label 0 means that the output needs to be P 2 The output is separated by breakpoints, and label 1 represents the output as a whole. 1 To C 8 Represents 8 situations respectively.

[0077] Step 205, based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network, and the fitting types include segment fitting and point fitting.

[0078] Next, based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set. The machine tool trajectory point classification model is established based on a neural network, and the fitting types include segment fitting and point fitting.

[0079] Step 206 , classifying the fitting type of each group of target fitting machine tool trajectory points based on the trained machine tool trajectory point classification model.

[0080] Step 207 : for any group of target fitting machine tool trajectory points, when the fitting type classified by the group of target fitting machine tool trajectory points is segment fitting, the group of target fitting machine tool trajectory points is output as trajectory segments for machine tool trajectory fitting.

[0081] Step 208: When the fitting type classified from the group of target fitting machine tool trajectory points is point fitting, the group of target fitting machine tool trajectory points are output as trajectory points to perform machine tool trajectory fitting to obtain a machine tool fitting trajectory.

[0082] Next, the trained neural network model is deployed to the FPGA. At the same time, combined with the parallel computing characteristics of the neural network, it is particularly suitable for FPGA chips to perform calculations, and has potential advantages, that is, through the intelligent segmentation method, the above-mentioned embodiment of the present application can be run alone on the FPGA chip of the numerical control system hardware. Specifically, deploying the machine learning model on the FPGA involves converting the optimized and quantized model from a high-performance computing platform (such as a CPU or GPU) into a form suitable for the FPGA architecture to achieve more efficient computing performance and lower power consumption, including selecting an appropriate machine learning model, training and quantization, and then generating hardware description language (HDL) code through a model conversion tool, and finally achieving real-time reasoning and acceleration on the FPGA. Based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified. That is, the G01 data points (each group of target fitting machine tool trajectory points) are passed into the neural network every three, and the data is further processed according to the predicted corresponding labels (fitting types): the three points meet all threshold conditions, and are all output as spline fitting type value points, and the three points do not meet any threshold conditions, and are output in segments for subsequent fitting operations. The threshold conditions include distance threshold conditions and angle threshold conditions. That is, for any group of target fitting machine tool trajectory points, when the fitting type classified by the group target fitting machine tool trajectory points is segment fitting, the group target fitting machine tool trajectory points are output as trajectory segments for machine tool trajectory fitting. When the fitting type classified by the group target fitting machine tool trajectory points is point fitting, the group target fitting machine tool trajectory points are output as trajectory points for machine tool trajectory fitting to obtain a machine tool fitting trajectory.

[0083] By applying the technical solution of this embodiment, multiple groups of machine tool historical trajectory points are obtained, and any group of machine tool historical trajectory points is selected. In the selected group of machine tool historical trajectory points, the first connecting line between the first machine tool historical trajectory point and the second machine tool historical trajectory point, and the second connecting line between the second machine tool historical trajectory point and the third machine tool historical trajectory point are determined. If the distance between adjacent machine tool historical trajectory points meets the preset distance threshold range, and the angle formed by the first connecting line and the second connecting line meets the preset angle threshold range, the fitting type of the selected group of machine tool historical trajectory points is segment fitting. If the distance between adjacent machine tool historical trajectory points does not meet the preset distance threshold range, and / or the angle formed by the first connecting line and the second connecting line does not meet the preset angle threshold range, the fitting type of the selected group of machine tool historical trajectory points is point fitting. Based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set. Based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified. For any group of target fitting machine tool trajectory points, when the fitting type classified by the group target fitting machine tool trajectory points is segment fitting, the group target fitting machine tool trajectory points are output as trajectory segments for machine tool trajectory fitting. When the fitting type classified by the group target fitting machine tool trajectory points is point fitting, the group target fitting machine tool trajectory points are output as trajectory points for machine tool trajectory fitting. The online interpolation efficiency of the CNC system is improved.

[0084] Further, as a refinement and expansion of the specific implementation of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, another segmentation method in machine tool trajectory fitting is provided, such as Figure 5 As shown, the method includes:

[0085] Step 301, obtaining multiple groups of machine tool historical trajectory points, wherein each group of machine tool historical trajectory points respectively includes a first machine tool historical trajectory point, a second machine tool historical trajectory point and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship, and each group of machine tool historical trajectory points respectively includes a first machine tool historical trajectory point, a second machine tool historical trajectory point within a preset spherical radius area from the first machine tool historical trajectory point, and a third machine tool historical trajectory point within a preset spherical radius area from the second machine tool historical trajectory point.

[0086] In the above embodiment of the present application, multiple groups of machine tool historical trajectory points are obtained, and each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship. Specifically, the first machine tool historical trajectory point can be obtained first, and then the second machine tool historical trajectory point can be obtained based on a preset spherical radius area from the first machine tool historical trajectory point, and based on the second machine tool historical trajectory point, the third machine tool historical trajectory point can be obtained within a preset spherical radius area from the second machine tool historical trajectory point, to prepare for subsequent model training.

[0087] Step 302, select any group of machine tool historical trajectory points, and determine the fitting type of the selected group of machine tool historical trajectory points based on the distance between adjacent machine tool historical trajectory points in the selected group and the angle formed by each machine tool historical trajectory point, wherein the fitting type includes segment fitting and point fitting.

[0088] Next, any group of machine tool historical trajectory points is selected, and the fitting type of the selected group of machine tool historical trajectory points is determined according to the distance between adjacent machine tool historical trajectory points in the selected group and the angle formed by each machine tool historical trajectory point. The fitting types include segment fitting and point fitting. By dividing the original large number of trajectory points into segment output or point output, the amount of fitting data can be reduced in the process of machine tool trajectory fitting, thereby improving the trajectory fitting efficiency.

[0089] Step 303, based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network, and the machine tool trajectory point training data set includes a preset number of groups of machine tool historical trajectory points.

[0090] Next, based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set. The machine tool trajectory point classification model is established based on a neural network. At the same time, in order to improve the accuracy of model training and expand the data coverage in the data set, the data set can be set to include one thousand or more groups of machine tool historical trajectory points.

[0091] Step 304, obtaining a target fitting machine tool trajectory point sequence, dividing the target fitting machine tool trajectory point sequence according to a preset moving step length, and obtaining multiple groups of target fitting machine tool trajectory points, wherein each group of target fitting machine tool trajectory points respectively includes a first target fitting machine tool trajectory point, a second target fitting machine tool trajectory point, and a third target fitting machine tool trajectory point.

[0092] Next, start to classify the trajectory points that need to be fitted. Specifically, obtain the target fitting machine tool trajectory point sequence, divide the target fitting machine tool trajectory point sequence according to the preset moving step length, and obtain multiple groups of target fitting machine tool trajectory points. For example, the obtained target fitting machine tool trajectory point sequence includes point 1, point 2, point 3, point 4, point 5, point 6, point 7, and point 8. The preset moving step length can be set to 3, then the divided groups of target fitting machine tool trajectory points are "Group 1: Point 1, Point 2, Point 3", "Group 2: Point 4, Point 5, Point 6", and "Group 3: Point 7, Point 8".

[0093] Step 305 , based on the trained machine tool trajectory point classification model, classify the fitting type of each group of target fitting machine tool trajectory points, and fit each group of target fitting machine tool trajectory points based on the classified fitting type to obtain the machine tool fitting trajectory.

[0094] Based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified, and each group of target fitting machine tool trajectory points is fitted based on the classified fitting type to obtain the machine tool fitting trajectory. For example, after group 1 is classified as segmented fitting using the machine tool trajectory point classification model, point 1, point 2, and point 3 are output as segments for fitting. After group 2 is classified as point fitting using the machine tool trajectory point classification model, point 4, point 5, and point 6 are directly output for fitting.

[0095] Step 306: if any group of target fitting machine tool trajectory points includes only one target fitting machine tool trajectory point, the target fitting machine tool trajectory point is output as a trajectory point.

[0096] Step 307: if any group of target fitting machine tool trajectory points includes only two target fitting machine tool trajectory points, when the distance between the two target fitting machine tool trajectory points meets the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory segments; when the distance between the two target fitting machine tool trajectory points does not meet the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory points respectively.

[0097] In particular, if any group of target fitting machine tool trajectory points includes only one target fitting machine tool trajectory point, the target fitting machine tool trajectory point is output as a trajectory point. For example, if the target fitting machine tool trajectory point sequence is point 1, point 2, point 3, point 4, point 5, point 6, and point 7, then the divided group 3 only has point 7, and point 7 is directly output for fitting.

[0098] If any group of target fitting machine tool trajectory points includes only two target fitting machine tool trajectory points, for example, group 3 is divided into point 7 and point 8, then when the distance between the two target fitting machine tool trajectory points (point 7 and point 8) meets the preset distance threshold range, the two target fitting machine tool trajectory points (point 7 and point 8) are output as trajectory segments; when the distance between the two target fitting machine tool trajectory points (point 7 and point 8) does not meet the preset distance threshold range, the two target fitting machine tool trajectory points (point 7 and point 8) are respectively output as trajectory points for fitting, which can improve the efficiency of trajectory fitting.

[0099] By applying the technical solution of this embodiment, multiple groups of machine tool historical trajectory points are obtained, and any group of machine tool historical trajectory points is selected. According to the distance between adjacent machine tool historical trajectory points in the selected group of machine tool historical trajectory points and the angle formed between each machine tool historical trajectory point, the fitting type of the selected group of machine tool historical trajectory points is determined, and a machine tool trajectory point training data set is generated based on each group of machine tool historical trajectory points and their corresponding fitting types, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set, and a target fitting machine tool trajectory point sequence is obtained, and the target fitting machine tool trajectory point sequence is divided according to a preset moving step length to obtain multiple groups of target fitting machine tool trajectory points. Based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified, and each group of target fitting machine tool trajectory points is fitted based on the classified fitting type to obtain a machine tool fitting trajectory. If any group of target fitting machine tool trajectory points includes only one target fitting machine tool trajectory point, the target fitting machine tool trajectory point is output as a trajectory point. If any group of target fitting machine tool trajectory points includes only two target fitting machine tool trajectory points, when the distance between the two target fitting machine tool trajectory points meets the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory segments, and when the distance between the two target fitting machine tool trajectory points does not meet the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory points respectively. The online interpolation efficiency of the CNC system is improved.

[0100] Further, as Figure 1 The specific implementation of the method, the embodiment of the present application provides a segmentation device for machine tool trajectory fitting, such as Figure 6 As shown, the device comprises:

[0101] A historical trajectory point acquisition module 401 is used to acquire multiple groups of machine tool historical trajectory points, wherein each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship;

[0102] The historical trajectory point classification module 402 is used to select any group of machine tool historical trajectory points, and determine the fitting type of the selected group of machine tool historical trajectory points according to the distance between adjacent machine tool historical trajectory points and the angle formed by each machine tool historical trajectory point in the selected group of machine tool historical trajectory points, wherein the fitting type includes segment fitting and point fitting;

[0103] A classification model training module 403 is used to generate a machine tool trajectory point training data set based on each group of machine tool historical trajectory points and their corresponding fitting types, and train a machine tool trajectory point classification model based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network;

[0104] The machine tool trajectory fitting module 404 is used to classify the fitting type of each group of target fitting machine tool trajectory points based on the trained machine tool trajectory point classification model, and fit each group of target fitting machine tool trajectory points based on the classified fitting type to obtain the machine tool fitting trajectory.

[0105] Optionally, the historical trajectory point classification module 402 is further used for:

[0106] In the selected group of machine tool history trajectory points, determining a first connection line between the first machine tool history trajectory point and the second machine tool history trajectory point, and a second connection line between the second machine tool history trajectory point and the third machine tool history trajectory point;

[0107] If the distance between adjacent machine tool historical trajectory points meets the preset distance threshold range, and the angle formed by the first connecting line and the second connecting line meets the preset angle threshold range, then the fitting type of the selected group of machine tool historical trajectory points is segmented fitting;

[0108] If the distance between adjacent machine tool historical trajectory points does not meet the preset distance threshold range, and / or the angle formed by the first connecting line and the second connecting line does not meet the preset angle threshold range, the fitting type of the selected group of machine tool historical trajectory points is point fitting.

[0109] Optionally, the machine tool trajectory fitting module 404 is further used for:

[0110] For any group of target fitting machine tool trajectory points, when the fitting type classified by the group of target fitting machine tool trajectory points is segment fitting, the group of target fitting machine tool trajectory points is output as trajectory segments for machine tool trajectory fitting;

[0111] When the fitting type classified from the group of target fitting machine tool trajectory points is point fitting, the group of target fitting machine tool trajectory points are output as trajectory points for performing machine tool trajectory fitting to obtain a machine tool fitting trajectory.

[0112] Optionally, the machine tool trajectory fitting module 404 is further used for:

[0113] A target fitting machine tool trajectory point sequence is obtained, and the target fitting machine tool trajectory point sequence is divided according to a preset moving step length to obtain multiple groups of target fitting machine tool trajectory points, wherein each group of target fitting machine tool trajectory points respectively includes a first target fitting machine tool trajectory point, a second target fitting machine tool trajectory point and a third target fitting machine tool trajectory point.

[0114] Optionally, the machine tool trajectory fitting module 404 is further used for:

[0115] If any group of target fitting machine tool trajectory points includes only one target fitting machine tool trajectory point, the target fitting machine tool trajectory point is output as a trajectory point;

[0116] If any group of target fitting machine tool trajectory points includes only two target fitting machine tool trajectory points, when the distance between the two target fitting machine tool trajectory points meets the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory segments; when the distance between the two target fitting machine tool trajectory points does not meet the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory points respectively.

[0117] Optionally, each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point within a preset spherical radius area from the first machine tool historical trajectory point, and a third machine tool historical trajectory point within a preset spherical radius area from the second machine tool historical trajectory point.

[0118] Optionally, the machine tool trajectory point training data set includes a preset number of groups of machine tool historical trajectory points.

[0119] It should be noted that for other corresponding descriptions of the functional units involved in the segmentation device in the machine tool trajectory fitting provided in the embodiment of the present application, reference can be made to Figure 1 , Figure 3 and Figure 5 The corresponding description in the method will not be repeated here.

[0120] Based on the above Figure 1 , Figure 3 and Figure 5 The method shown in the embodiment of the present application accordingly provides a storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned Figure 1 , Figure 3 and Figure 5 The segmentation method in machine tool trajectory fitting is shown.

[0121] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0122] Based on the above Figure 1 , Figure 3 and Figure 5 The method shown, and Figure 6 In order to achieve the above-mentioned purpose, the embodiment of the present application further provides a computer device, which can be a personal computer, a server, a network device, etc. The computer device includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to achieve the above-mentioned Figure 1 , Figure 3 and Figure 5 The segmentation method in machine tool trajectory fitting is shown.

[0123] Optionally, the computer device may further include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a WI-FI module, etc. The user interface may include a display, an input unit such as a keyboard, etc., and the optional user interface may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a Bluetooth interface, a WI-FI interface), etc.

[0124] Those skilled in the art will appreciate that the computer device structure provided in this embodiment does not limit the computer device, and may include more or fewer components, or a combination of certain components, or different component arrangements.

[0125] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages and saves the hardware and software resources of the computer device, and supports the operation of information processing programs and other software and / or programs. The network communication module is used to realize communication between the components inside the storage medium, and communication with other hardware and software in the physical device.

[0126] Through the description of the above implementation methods, the technicians in this field can clearly understand that the present application can be implemented by means of software plus the necessary general hardware platform, or by hardware to obtain multiple groups of machine tool historical trajectory points, select any group of machine tool historical trajectory points, and determine the fitting type of the selected group of machine tool historical trajectory points according to the distance between adjacent machine tool historical trajectory points in the selected group of machine tool historical trajectory points and the angle formed between each machine tool historical trajectory point. The fitting type includes segment fitting and point fitting; based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network; based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified, and each group of target fitting machine tool trajectory points is fitted based on the classified fitting type to obtain a machine tool fitting trajectory. The online interpolation efficiency of the CNC system is improved.

[0127] Those skilled in the art will appreciate that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily necessary for implementing the present application. Those skilled in the art will appreciate that the modules in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the description of the implementation scenario, or can be changed accordingly and located in one or more devices different from the present implementation scenario. The modules of the above-mentioned implementation scenario can be combined into one module, or can be further split into multiple submodules.

[0128] The above serial numbers of this application are only for description and do not represent the advantages and disadvantages of the implementation scenarios. The above disclosure is only a few specific implementation scenarios of this application, but this application is not limited to them, and any changes that can be thought of by technicians in this field should fall within the scope of protection of this application.

Claims

1. A segmentation method for machine tool trajectory fitting, characterized in that: The method comprises: Acquire multiple groups of machine tool historical trajectory points, wherein each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point, and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship; Select any group of machine tool historical trajectory points, and determine the fitting type of the selected group of machine tool historical trajectory points according to the distances between adjacent machine tool historical trajectory points and the angles formed between the machine tool historical trajectory points in the selected group of machine tool historical trajectory points, wherein the fitting type includes segment fitting and point fitting; Based on each group of machine tool historical trajectory points and their corresponding fitting types, a machine tool trajectory point training data set is generated, and a machine tool trajectory point classification model is trained based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network; Based on the trained machine tool trajectory point classification model, the fitting type of each group of target fitting machine tool trajectory points is classified, and each group of target fitting machine tool trajectory points is fitted based on the classified fitting type to obtain the machine tool fitting trajectory.

2. The method according to claim 1, characterized in that The method of determining the fitting type of the selected group of machine tool historical trajectory points according to the distances between adjacent machine tool historical trajectory points and the angles formed between the machine tool historical trajectory points in the selected group of machine tool historical trajectory points includes: In the selected group of machine tool history trajectory points, determining a first connection line between the first machine tool history trajectory point and the second machine tool history trajectory point, and a second connection line between the second machine tool history trajectory point and the third machine tool history trajectory point; If the distance between adjacent machine tool historical trajectory points meets the preset distance threshold range, and the angle formed by the first connecting line and the second connecting line meets the preset angle threshold range, then the fitting type of the selected group of machine tool historical trajectory points is segmented fitting; If the distance between adjacent machine tool historical trajectory points does not meet the preset distance threshold range, and / or the angle formed by the first connecting line and the second connecting line does not meet the preset angle threshold range, the fitting type of the selected group of machine tool historical trajectory points is point fitting.

3. The method according to claim 1, characterized in that The step of fitting each group of target fitting machine tool trajectory points based on the classified fitting types to obtain the machine tool fitting trajectory includes: For any group of target fitting machine tool trajectory points, when the fitting type classified by the group of target fitting machine tool trajectory points is segment fitting, the group of target fitting machine tool trajectory points is output as trajectory segments for machine tool trajectory fitting; When the fitting type classified from the group of target fitting machine tool trajectory points is point fitting, the group of target fitting machine tool trajectory points are output as trajectory points for performing machine tool trajectory fitting to obtain a machine tool fitting trajectory.

4. The method according to claim 1, characterized in that: Before classifying the fitting type of each group of target fitting machine tool trajectory points based on the trained machine tool trajectory point classification model, the method further includes: A target fitting machine tool trajectory point sequence is obtained, and the target fitting machine tool trajectory point sequence is divided according to a preset moving step length to obtain multiple groups of target fitting machine tool trajectory points, wherein each group of target fitting machine tool trajectory points respectively includes a first target fitting machine tool trajectory point, a second target fitting machine tool trajectory point and a third target fitting machine tool trajectory point.

5. The method according to claim 4, characterized in that After dividing the target fitting machine tool trajectory point sequence according to the preset moving step length to obtain multiple groups of target fitting machine tool trajectory points, the method further includes: If any group of target fitting machine tool trajectory points includes only one target fitting machine tool trajectory point, the target fitting machine tool trajectory point is output as a trajectory point; If any group of target fitting machine tool trajectory points includes only two target fitting machine tool trajectory points, when the distance between the two target fitting machine tool trajectory points meets the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory segments; when the distance between the two target fitting machine tool trajectory points does not meet the preset distance threshold range, the two target fitting machine tool trajectory points are output as trajectory points respectively.

6. The method according to any one of claims 1 to 6, characterized in that Each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point within a preset spherical radius area from the first machine tool historical trajectory point, and a third machine tool historical trajectory point within a preset spherical radius area from the second machine tool historical trajectory point.

7. The method according to claim 6, characterized in that The machine tool trajectory point training data set includes a preset number of groups of machine tool historical trajectory points.

8. A segmentation device for machine tool trajectory fitting, characterized in that: The device comprises: A historical trajectory point acquisition module is used to acquire multiple groups of machine tool historical trajectory points, wherein each group of machine tool historical trajectory points includes a first machine tool historical trajectory point, a second machine tool historical trajectory point and a third machine tool historical trajectory point that conform to a preset three-dimensional spatial relationship; A historical trajectory point classification module is used to select any group of machine tool historical trajectory points, and determine the fitting type of the selected group of machine tool historical trajectory points according to the distance between adjacent machine tool historical trajectory points in the selected group of machine tool historical trajectory points and the angle formed between each machine tool historical trajectory point, wherein the fitting type includes segment fitting and point fitting; A classification model training module, used for generating a machine tool trajectory point training data set based on each group of machine tool historical trajectory points and their corresponding fitting types, and training a machine tool trajectory point classification model based on the machine tool trajectory point training data set, wherein the machine tool trajectory point classification model is established based on a neural network; The machine tool trajectory fitting module is used to classify the fitting type of each group of target fitting machine tool trajectory points based on the trained machine tool trajectory point classification model, and fit each group of target fitting machine tool trajectory points based on the classified fitting type to obtain the machine tool fitting trajectory.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for machine tool trajectory fitting according to any one of claims 1 to 7 is implemented.

10. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the method for machine tool trajectory fitting according to any one of claims 1 to 7 is implemented.

Images