Laser control system and control method based on machine learning

Abstract

An embodiment of the present application provides a laser control system and a control method based on machine learning, belonging to the field of machine learning, to improve the control flexibility of the laser control system and the processing / cutting efficiency. The method includes: an electronic device determines a model of a cut part in a component to be cut according to a model of the component to be cut and a model of a target component after the component to be cut is cut; wherein, the target component and the cut part are combined into the component to be cut; the electronic device divides the model of the cut part into multiple sub-models and determines the spatial information of the multiple sub-models; the electronic device processes the spatial information of the multiple sub-models through a deep neural network model to obtain a cutting light path output by the deep neural network model; wherein, the cutting light path is configured such that if the laser control system performs laser cutting on the component to be cut according to the cutting light path, the target component can be obtained.

Description

Technical Field

[0001] This application relates to the technical field of machine learning, and particularly to a laser control system and a control method based on machine learning. Background Art

[0002] The laser control system is an important part of laser processing equipment. It realizes the precise control and processing of the laser by integrating key technologies such as computer-aided design technology, numerical control technology, and computer manufacturing.

[0003] Specifically, computer-aided design technology is the first core technology in the laser control system. It reads the processing model through computer modeling or from an external file, performs graphic recognition, editing, and optimization processing, then typesets on the processing material, and outputs the model to be processed. Computer-aided manufacturing technology is the second core technology in the laser control system. It outputs the instructions for laser processing according to the process requirements related to the material, including laser power, pulse frequency, focus control, light path trajectory, and auxiliary gas. With the help of algorithms, it optimizes the processing path, sets the drilling position, and completes the processing plan for complex graphics with the optimal solution. Numerical control technology is the third core technology in the laser control system. It outputs the operation instructions for external devices such as motors, motion axes, and laser heads, controls the devices to implement specific processing details, and cooperates with the laser output instructions to complete the specific laser processing operations. It plays a key role in the processing of multi-axis platforms. Hardware design technology is the fourth core technology in the laser control system. It develops corresponding embedded software and hardware circuits according to the requirements of the laser industry, mainly focusing on safety and stability, such as equipment fault alarm, equipment safety alarm, cooling system monitoring, and software encryption. That is to say, the background technology of the laser control system mainly includes computer-aided design technology, computer-aided manufacturing technology, numerical control technology, and hardware design technology. The development and application of these technologies enable laser processing equipment to achieve high-precision, high-efficiency, and stable processing operations.

[0004] Currently, the light path of laser processing is mainly set manually, that is, manually set the corresponding light path according to the specific situation of the component to be cut, and the laser control system controls the laser processing / cutting according to the light path. However, this method is not flexible enough and the efficiency is relatively low. Summary of the Invention

[0005] The laser control system and control method based on machine learning provided by the embodiments of this application are used to improve the control flexibility of the laser control system and the processing / cutting efficiency.

[0006] To achieve the above object, the embodiments of this application adopt the following technical solutions:

[0007] In a first aspect, a laser control method based on machine learning is provided and applied to an electronic device. The method includes: The electronic device determines the model of the cut part in the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut; wherein, the target component and the cut part are combined into the component to be cut; The electronic device divides the model of the cut part into multiple sub-models and determines the spatial information of the multiple sub-models; The electronic device processes the spatial information of the multiple sub-models through a deep neural network model to obtain the cutting light path output by the deep neural network model; wherein, the cutting light path is configured such that if the laser control system performs laser cutting on the component to be cut according to the cutting light path, the target component can be obtained.

[0008] Optionally, the electronic device divides the model of the cut part into multiple sub-models and determines the spatial information of the multiple sub-models, including: The electronic device divides the model of the cut part into multiple sub-models in a three-dimensional model space according to the preset spatial equal division lines in the three-dimensional model space, where the spatial equal division lines are used to equally divide the three-dimensional model space into multiple cubic sub-spaces; The electronic device determines the spatial information of the multiple sub-models according to the respective volume sizes of the multiple sub-models.

[0009] Optionally, the coordinates of the three-dimensional model space are represented by the coordinates of the x-axis, the coordinates of the y-axis, and the coordinates of the z-axis. The spatial equal division lines include K1 dividing lines, K2 dividing lines, and K3 dividing lines. K1, K2, and K3 are all integers greater than 1. The K1 dividing lines are perpendicular to the plane formed by the x-axis and the y-axis, the K2 dividing lines are perpendicular to the plane formed by the x-axis and the z-axis, and the K3 dividing lines are perpendicular to the plane formed by the y-axis and the z-axis; The electronic device divides the model of the cut part into multiple sub-models in the three-dimensional model space according to the preset spatial equal division lines in the three-dimensional model space, including: The electronic device overlaps the model of the cut part with the cutting planes formed by any two mutually perpendicular dividing lines among the K1 dividing lines, the K2 dividing lines, and the K3 dividing lines, and divides the model of the cut part along the cutting planes, and a total of M sub-models are obtained by division. M is an integer greater than 2.

[0010] Optionally, the electronic device determines the spatial information of multiple sub-models according to the volume sizes of the multiple sub-models, including: for the i-th sub-model among the M sub-models, where i is any integer from 1 to M, the electronic device determines the spatial coordinate points within the i-th sub-model according to the volume size of the i-th sub-model, so as to obtain the i-th set of spatial coordinates, and a total of M sets of spatial coordinates are obtained; wherein, the number of spatial coordinate points within the i-th sub-model is negatively correlated with the volume size of the i-th sub-model; the volume size of the i-th sub-model refers to the volume size of the i-th sub-model in the three-dimensional model space, the spatial coordinate points within the i-th sub-model refer to the coordinates of the coordinate points within the i-th sub-model in the three-dimensional model space, the spatial coordinate points within the i-th sub-model can be used to generate and represent the cutting optical path, the i-th set of spatial coordinates is the spatial information of the i-th sub-model, and the i-th set of spatial coordinates includes the coordinates of the spatial coordinate points within the i-th sub-model respectively.

[0011] Optionally, for the i-th sub-model among the M sub-models, where i is any integer from 1 to M, the electronic device determines the spatial coordinate points within the i-th sub-model according to the volume size of the i-th sub-model and obtains the i-th set of spatial coordinates, including: the electronic device determines the volume size interval where the volume size of the i-th sub-model is located from different volume size intervals; the electronic device randomly generates the number of spatial coordinate points corresponding to the volume size interval within the i-th sub-model according to the number of spatial coordinate points corresponding to the volume size interval, so as to obtain the spatial coordinate points within the i-th sub-model; wherein, in the order of increasing volume, the number of spatial coordinate points corresponding to different volume size intervals gradually decreases; the electronic device generates the i-th set of spatial coordinates according to the spatial coordinate points within the i-th sub-model, and a total of M sets of spatial coordinates are obtained.

[0012] Optionally, the electronic device processes the spatial information of multiple sub-models through a deep neural network model to obtain the cutting optical path output by the deep neural network model, including: the electronic device interleaves the M sets of spatial coordinates to obtain an interleaved sequence of spatial coordinate sets; wherein, among the M sub-models, the positions of the spatial coordinate sets corresponding to at least two sub-models with volume sizes in the same volume size interval are concentrated in the interleaved sequence of spatial coordinate sets, and the positions of the spatial coordinate sets corresponding to at least two sub-models with volume sizes in different volume size intervals are discrete in the interleaved sequence of spatial coordinate sets; the electronic device inputs the M sets of spatial coordinates into the deep neural network model in the order of the interleaved sequence of spatial coordinate sets to obtain the cutting optical path output by the deep neural network model.

[0013] Optionally, the electronic device determines the spatial information of multiple sub-models according to the volume size of each sub-model, including: for the i-th sub-model among the M sub-models, where i is any integer from 1 to M, the electronic device copies the i-th sub-model into J copies according to the volume size of the i-th sub-model to obtain J sub-models #i, and the value of J is negatively correlated with the volume size of the i-th sub-model; the electronic device determines the spatial coordinate points within each of the J sub-models #i to obtain J spatial coordinate sets #i, and a total of N spatial coordinate sets for N sub-models are obtained, where N is a positive integer greater than M; among them, the volume size of each of the J sub-models #i refers to the volume of the sub-model #i in the three-dimensional model space, the spatial coordinate points within each of the J sub-models #i refer to the coordinates of the coordinate points within the sub-model #i in the three-dimensional model space, the spatial coordinate points within each of the J sub-models #i can be used to generate and represent the cutting optical path, and the J spatial coordinate sets #i are the spatial information of the i-th sub-model, and the J spatial coordinate sets #i include the coordinates of the spatial coordinate points within each of the J sub-models #i respectively.

[0014] Optionally, the electronic device copies the i-th sub-model into J copies according to the volume size of the i-th sub-model to obtain J sub-models #i, including: the electronic device determines the volume size interval where the volume size of the i-th sub-model is located from different volume size intervals; the electronic device copies the i-th sub-model into J copies according to the value of J corresponding to the volume size interval to obtain J sub-models #i; among them, in the order of increasing volume, the values of J corresponding to different volume size intervals gradually decrease; correspondingly, the electronic device determines the spatial coordinate points within each of the J sub-models #i to obtain J spatial coordinate sets #i, and a total of N spatial coordinate sets for N sub-models are obtained, including: the electronic device randomly generates a preset number of spatial coordinate points within each of the J sub-models #i, the spatial coordinate points within each of the J sub-models #i; the electronic device generates J spatial coordinate sets #i one-to-one according to the spatial coordinate points within each of the J sub-models #i, and a total of N spatial coordinate sets are obtained.

[0015] Optionally, the electronic device processes the spatial information of multiple sub-models through a deep neural network model to obtain the cutting optical path output by the deep neural network model, including: the electronic device interleaves the N spatial coordinate sets to obtain an interleaved sequence of spatial coordinate sets; among them, the positions of the J spatial coordinate sets #i corresponding to the J sub-models #i are concentrated in the interleaved sequence of spatial coordinate sets; the electronic device inputs the N spatial coordinate sets into the deep neural network model in the order of the interleaved sequence of spatial coordinate sets to obtain the cutting optical path output by the deep neural network model.

[0016] Optionally, the electronic device determines the model of the part to be cut in the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut, including: in response to a user input operation, the electronic device generates / obtains the model of the component to be cut and the model of the target component; the electronic device determines the different part between the model of the component to be cut and the model of the target component in the three-dimensional model space as the model of the part to be cut.

[0017] In a second aspect, a laser control system based on machine learning is provided. The system includes an electronic device, and the system is configured to: the electronic device determines the model of the part to be cut in the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut; wherein, the target component and the part to be cut are combined into the component to be cut; the electronic device divides the model of the part to be cut into multiple sub-models and determines the spatial information of the multiple sub-models; the electronic device processes the spatial information of the multiple sub-models through a deep neural network model to obtain the cutting optical path output by the deep neural network model; wherein, the cutting optical path is configured such that if the laser control system performs laser cutting on the component to be cut according to the cutting optical path, the target component can be obtained.

[0018] Optionally, the electronic device divides the model of the part to be cut into multiple sub-models and determines the spatial information of the multiple sub-models, including: the electronic device divides the model of the part to be cut into multiple sub-models in the three-dimensional model space according to the preset spatial equal division lines in the three-dimensional model space, where the spatial equal division lines are used to equally divide the three-dimensional model space into multiple cubic sub-spaces; the electronic device determines the spatial information of the multiple sub-models according to the respective volume sizes of the multiple sub-models.

[0019] Optionally, the coordinates of the three-dimensional model space are represented by the coordinates of the x-axis, the coordinates of the y-axis, and the coordinates of the z-axis. The spatial equal division lines include K1 dividing lines, K2 dividing lines, and K3 dividing lines. K1, K2, and K3 are all integers greater than 1. The K1 dividing lines are perpendicular to the plane formed by the x-axis and the y-axis, the K2 dividing lines are perpendicular to the plane formed by the x-axis and the z-axis, and the K3 dividing lines are perpendicular to the plane formed by the y-axis and the z-axis; the electronic device divides the model of the part to be cut into multiple sub-models in the three-dimensional model space according to the preset spatial equal division lines in the three-dimensional model space, including: the electronic device overlaps the model of the part to be cut with the cutting planes formed by any two mutually perpendicular dividing lines among the K1 dividing lines, the K2 dividing lines, and the K3 dividing lines, and divides the model of the part to be cut along the cutting planes, and a total of M sub-models are obtained by division, where M is an integer greater than 2.

[0020] Optionally, the electronic device determines the spatial information of multiple sub-models according to the volume size of each sub-model, including: for the i-th sub-model among the M sub-models, where i is any integer from 1 to M, the electronic device determines the spatial coordinate points within the i-th sub-model according to the volume size of the i-th sub-model, so as to obtain the i-th set of spatial coordinates, and a total of M sets of spatial coordinates are obtained; among them, the number of spatial coordinate points within the i-th sub-model is negatively correlated with the volume size of the i-th sub-model; the volume size of the i-th sub-model refers to the volume size of the i-th sub-model in the three-dimensional model space, the spatial coordinate points within the i-th sub-model refer to the coordinates of the coordinate points within the i-th sub-model in the three-dimensional model space, the spatial coordinate points within the i-th sub-model can be used to generate and represent the cutting optical path, the i-th set of spatial coordinates is the spatial information of the i-th sub-model, and the i-th set of spatial coordinates includes the coordinates of each spatial coordinate point within the i-th sub-model.

[0021] Optionally, for the i-th sub-model among the M sub-models, where i is any integer from 1 to M, the electronic device determines the spatial coordinate points within the i-th sub-model according to the volume size of the i-th sub-model and obtains the i-th set of spatial coordinates, including: the electronic device determines the volume size interval in which the volume size of the i-th sub-model is located from different volume size intervals; the electronic device randomly generates the number of spatial coordinate points corresponding to the volume size interval within the i-th sub-model according to the number of spatial coordinate points corresponding to the volume size interval, so as to obtain the spatial coordinate points within the i-th sub-model; among them, in the order of increasing volume, the number of spatial coordinate points corresponding to different volume size intervals gradually decreases; the electronic device generates the i-th set of spatial coordinates according to the spatial coordinate points within the i-th sub-model, and a total of M sets of spatial coordinates are obtained.

[0022] Optionally, the electronic device processes the spatial information of multiple sub-models through a deep neural network model to obtain the cutting optical path output by the deep neural network model, including: the electronic device interleaves the M sets of spatial coordinates to obtain an interleaved sequence of spatial coordinate sets; among them, among the M sub-models, the positions of the spatial coordinate sets corresponding to at least two sub-models with volume sizes in the same volume size interval are concentrated in the interleaved sequence of spatial coordinate sets, and the positions of the spatial coordinate sets corresponding to at least two sub-models with volume sizes in different volume size intervals are discrete in the interleaved sequence of spatial coordinate sets; the electronic device inputs the M sets of spatial coordinates into the deep neural network model in the order of the interleaved sequence of spatial coordinate sets to obtain the cutting optical path output by the deep neural network model.

[0023] Optionally, the electronic device determines the spatial information of multiple sub-models according to the volume size of each sub-model, including: for the i-th sub-model among the M sub-models, where i is any integer from 1 to M, the electronic device copies the i-th sub-model into J copies according to the volume size of the i-th sub-model to obtain J sub-models #i, and the value of J is negatively correlated with the volume size of the i-th sub-model; the electronic device determines the spatial coordinate points within each of the J sub-models #i to obtain J spatial coordinate sets #i, and a total of N spatial coordinate sets of N sub-models are obtained, where N is a positive integer greater than M; among them, the volume size of each of the J sub-models #i refers to the volume of the sub-model #i in the three-dimensional model space, the spatial coordinate points within each of the J sub-models #i refer to the coordinates of the coordinate points within the sub-model #i in the three-dimensional model space, the spatial coordinate points within each of the J sub-models #i can be used to generate and represent the cutting optical path, and the J spatial coordinate sets #i are the spatial information of the i-th sub-model, and the J spatial coordinate sets #i include the coordinates of the spatial coordinate points within each of the J sub-models #i respectively.

[0024] Optionally, the electronic device copies the i-th sub-model into J copies according to the volume size of the i-th sub-model to obtain J sub-models #i, including: the electronic device determines the volume size interval where the volume size of the i-th sub-model is located from different volume size intervals; the electronic device copies the i-th sub-model into J copies according to the value of J corresponding to the volume size interval to obtain J sub-models #i; among them, in the order of increasing volume, the values of J corresponding to different volume size intervals gradually decrease; correspondingly, the electronic device determines the spatial coordinate points within each of the J sub-models #i to obtain J spatial coordinate sets #i, and a total of N spatial coordinate sets of N sub-models are obtained, including: the electronic device randomly generates a preset number of spatial coordinate points within each of the J sub-models #i, the spatial coordinate points within each of the J sub-models #i; the electronic device generates J spatial coordinate sets #i one-to-one according to the spatial coordinate points within each of the J sub-models #i, and a total of N spatial coordinate sets are obtained.

[0025] Optionally, the electronic device processes the spatial information of multiple sub-models through a deep neural network model to obtain the cutting optical path output by the deep neural network model, including: the electronic device interleaves the N spatial coordinate sets to obtain an interleaved sequence of spatial coordinate sets; among them, the positions of the J spatial coordinate sets #i corresponding to the J sub-models #i are concentrated in the interleaved sequence of spatial coordinate sets; the electronic device inputs the N spatial coordinate sets into the deep neural network model in the order of the interleaved sequence of spatial coordinate sets to obtain the cutting optical path output by the deep neural network model.

[0026] Optionally, the electronic device determines the model of the cut part in the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut, including: in response to a user input operation, the electronic device generates / obtains the model of the component to be cut and the model of the target component; the electronic device determines the different part between the model of the component to be cut and the model of the target component in the three-dimensional model space as the model of the cut part.

[0027] In summary, through three-dimensional modeling, the electronic device can determine the model of the cut part in the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut, and divide the model of the cut part into multiple sub-models. In this way, the electronic device can obtain the cutting light path for the component to be cut output by the deep neural network model through the spatial information of the multiple sub-models. In addition, since the model of the cut part is not input into the deep neural network model as a whole for processing, but is processed in terms of multiple sub-models, the performance requirements for the deep neural network model can be reduced, and the robustness of the deep neural network model can be improved. Brief Description of the Drawings

[0028] Figure 1 is a schematic flowchart of a laser control method based on machine learning provided by an embodiment of the present application;

[0029] Figure 2 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed Embodiments

[0030] To facilitate understanding of the technical solutions provided by the embodiments of the present application, a brief introduction to the related technologies of the present application is first given. The brief introduction is as follows:

[0031] Machine learning is an important technical approach to realizing artificial intelligence. Machine learning can be divided into supervised learning, unsupervised learning, and reinforcement learning.

[0032] Supervised learning is based on the collected sample values and sample labels. Using machine learning algorithms, it learns the mapping relationship from sample values to sample labels and represents the learned mapping relationship with a machine learning model. The process of training a machine learning model is the process of learning this mapping relationship. For example, in signal detection, the received signal with noise is the sample, and the corresponding true constellation point of the signal is the label. Machine learning expects to learn the mapping relationship between the sample and the label through training, that is, to enable the machine learning model to learn a signal detector. During training, the model parameters are optimized by calculating the error between the predicted value of the model and the true label. Once the mapping relationship is learned, the learned mapping can be used to predict the label of each new sample. The mapping relationship learned by supervised learning can include linear mapping and non-linear mapping. According to the type of label, the learning tasks can be divided into classification tasks and regression tasks.

[0033] Unsupervised learning only relies on the collected sample values and uses algorithms to discover the internal patterns of the samples by itself. In unsupervised learning, there is a type of algorithm that uses the samples themselves as the supervision signal, that is, the model learns the mapping relationship from samples to samples, which is called self-supervised learning. During training, the model parameters are optimized by calculating the error between the predicted value of the model and the samples themselves. Self-supervised learning can be used in applications such as signal compression and decompression recovery. Common algorithms include autoencoders and adversarial generative networks, etc.

[0034] Reinforcement learning is different from supervised learning. It is a type of algorithm that learns strategies to solve problems by interacting with the environment. Different from supervised and unsupervised learning, there is no clear "correct" action label data in reinforcement learning problems. The algorithm needs to interact with the environment to obtain the reward signal feedback by the environment, and then adjust the decision-making actions to obtain a larger numerical value of the reward signal. For example, in downlink power control, the reinforcement learning model adjusts the downlink transmission power of each user according to the total system throughput rate feedback by the wireless network, and then expects to obtain a higher system throughput rate. The goal of reinforcement learning is also to learn the mapping relationship between the environmental state and the optimal decision-making action. However, because the label of the "correct action" cannot be obtained in advance, the network cannot be optimized by calculating the error between the action and the "correct action". The training of reinforcement learning is achieved through iterative interaction with the environment.

[0035] Deep neural network (DNN) is a specific implementation form of machine learning. According to the universal approximation theorem, a neural network can theoretically approximate any continuous function, enabling the neural network to have the ability to learn any mapping. Traditional communication systems need to rely on rich expert knowledge to design communication modules, while deep learning communication systems based on DNN can automatically discover the hidden pattern structures from a large amount of data sets, establish the mapping relationship between data, and obtain better performance than traditional modeling methods.

[0036] The idea of DNN comes from the neuron structure of brain tissue. Each neuron performs a weighted sum operation on its input values and generates an output by passing the result of the weighted sum through a non-linear function. For example, assume the input of a neuron is , the weights corresponding to the input are , the bias of the weighted sum is , and the form of the non-linear function can be diverse. For example, it can be the maximum function of . Based on this, the execution effect of a neuron can be . The weights of each neuron are the so-called model parameters of DNN. The model parameters can be optimized through the training process, enabling DNN to have the ability to extract data features and express mapping relationships. DNN generally uses supervised learning or unsupervised learning strategies to optimize the model parameters.

[0037] DNN generally has a multi-layer structure. Each layer of DNN can contain multiple neurons. After the input layer of DNN processes the received values through neurons, it passes them to the intermediate hidden layers. DNN generally has more than one hidden layer, and the hidden layer often directly affects the ability to extract information and fit functions. Increasing the number of hidden layers of DNN or expanding the width of each layer can improve the function fitting ability of DNN. Then, after the hidden layers of DNN process the received values through neurons, they pass the calculation results to the final output layer to generate the final output of DNN.

[0038] According to the construction method of the network, DNN can be divided into feed forward neural network (FNN), convolutional neural networks (CNN), and recurrent neural network (RNN).

[0039] The characteristic of the FNN network is that neurons between adjacent layers are completely connected in pairs, which usually requires a large amount of storage space and leads to a high computational complexity.

[0040] CNN is a neural network specifically designed to process data with a similar grid structure. For example, time series data (discrete sampling on the time axis) and image data (two-dimensional discrete sampling) can both be considered data with a similar grid structure. CNN does not perform operations using all the input information at once, but instead uses a fixed-size window to intercept part of the information for convolution operations, which greatly reduces the computational amount of model parameters. In addition, according to the different types of information intercepted by the window (such as people and objects in the same picture being different types of information), each window can use different convolution kernels for operations, enabling CNN to better extract the features of the input data.

[0041] The RNN is a type of DNN network that utilizes feedback time series information. Its input includes the new input value at the current moment and its own output value at the previous moment. The RNN is suitable for obtaining sequence features that are relevant in time, and is particularly applicable to applications such as speech recognition and channel coding and decoding.

[0042] Aspects, embodiments, or features of the present application will be presented in the context of a system that may include multiple devices, components, modules, etc. It should be understood and appreciated that each system may include additional devices, components, modules, etc., and / or may not include all of the devices, components, modules, etc. discussed in conjunction with the figures. Additionally, combinations of these solutions may be used.

[0043] Furthermore, in the embodiments of the present application, words such as "exemplary" and "for example" are used to provide examples, illustrations, or explanations. Any embodiment or design solution described as "exemplary" in the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of the word "exemplary" is intended to present concepts in a specific manner.

[0044] In the embodiments of the present application, "information", "signal", "message", "channel", and "signaling" may sometimes be used interchangeably. It should be noted that when the differences are not emphasized, their intended meanings are consistent. "Of", "corresponding", and "corresponding to" may sometimes be used interchangeably. It should be noted that when the differences are not emphasized, their intended meanings are consistent. Additionally, " / " mentioned in the present application may be used to represent an "or" relationship.

[0045] The network architectures and service scenarios described in the embodiments of the present application are for the purpose of more clearly illustrating the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those of ordinary skill in the art will understand that as the network architecture evolves and new service scenarios emerge, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0046] Exemplarily, Figure 1 The flowchart of the laser control method based on machine learning provided by the embodiments of the present application. This method can be executed by an electronic device, and will be specifically introduced below.

[0047] As Figure 1 shown, the process of this communication method is as follows:

[0048] S101. The electronic device determines the model of the cut part in the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut.

[0049] The target component and the cut part are combined into the component to be cut. In response to the user's input operation, the electronic device generates / obtains the model of the component to be cut and the model of the target component. That is, the user can perform 3D modeling through the electronic device to generate the model of the component to be cut and the model of the target component. Of course, the user can also directly import the established models of the component to be cut and the target component into the electronic device. The electronic device can determine the different part between the model of the component to be cut and the model of the target component in the 3D model space as the model of the cut part. That is, subtracting the model of the target component from the model of the component to be cut gives the model of the cut part.

[0050] S102. The electronic device divides the model of the cut part into multiple sub-models and determines the spatial information of the multiple sub-models.

[0051] The electronic device can divide the model of the cut part into multiple sub-models in the 3D model space according to the preset spatial equal division lines in the 3D model space, where the spatial equal division lines are used to equally divide the 3D model space into multiple cubic sub-spaces. For example, the coordinates of the 3D model space are represented by the coordinates of the x-axis, the coordinates of the y-axis, and the coordinates of the z-axis. The spatial equal division lines include K1 dividing lines, K2 dividing lines, and K3 dividing lines. K1, K2, and K3 are all integers greater than 1, and the specific values are not limited as long as the model of the cut part can be reasonably divided into multiple sub-models, such as the number being between 10 and 20. The K1 dividing lines are perpendicular to the plane formed by the x-axis and the y-axis, the K2 dividing lines are perpendicular to the plane formed by the x-axis and the z-axis, and the K3 dividing lines are perpendicular to the plane formed by the y-axis and the z-axis. In this way, the spatial equal division lines used to equally divide the 3D model space into multiple cubic sub-spaces can be vividly understood as each block in a Rubik's Cube. On this basis, the electronic device can overlap the model of the cut part with the cutting planes formed by any two mutually perpendicular dividing lines among the K1 dividing lines, the K2 dividing lines, and the K3 dividing lines, and divide the model of the cut part along the cutting planes, and a total of M sub-models are obtained, where M is an integer greater than 2.

[0052] The electronic device determines the spatial information of the multiple sub-models according to the volume sizes of the multiple sub-models respectively, and there are two specific ways to achieve this.

[0053] Method 1:

[0054] For the $i$-th sub-model among $M$ sub-models, where $i$ is any integer ranging from 1 to $M$, the electronic device can determine the spatial coordinate points within the $i$-th sub-model according to the volume size of the $i$-th sub-model, so as to obtain the $i$-th set of spatial coordinates, and a total of $M$ sets of spatial coordinates are obtained. Among them, the number of spatial coordinate points within the $i$-th sub-model is negatively correlated with the volume size of the $i$-th sub-model. That is, the smaller the $i$-th sub-model, the higher the requirement for cutting accuracy may be. Therefore, more spatial coordinate points need to be set to facilitate the subsequent deep neural network model to generate a more refined cutting optical path for the $i$-th sub-model. The volume size of the $i$-th sub-model refers to the volume size of the $i$-th sub-model in the three-dimensional model space. The spatial coordinate points within the $i$-th sub-model refer to the coordinates of the coordinate points within the $i$-th sub-model in the three-dimensional model space. The spatial coordinate points within the $i$-th sub-model can be used to generate and represent the cutting optical path. The $i$-th set of spatial coordinates is the spatial information of the $i$-th sub-model, and the $i$-th set of spatial coordinates includes the coordinates of each of the spatial coordinate points within the $i$-th sub-model.

[0055] For example, the electronic device can determine the volume size interval in which the volume size of the $i$-th sub-model is located from different volume size intervals. The electronic device can randomly generate the number of spatial coordinate points corresponding to the volume size interval within the $i$-th sub-model according to the number of spatial coordinate points corresponding to the volume size interval, so as to obtain the spatial coordinate points within the $i$-th sub-model. Among them, in the order of increasing volume, the number of spatial coordinate points corresponding to different volume size intervals gradually decreases, that is, the above-mentioned negative correlation is realized. The specific values of the number of spatial coordinate points corresponding to different volume size intervals are not limited and can be set according to the actual situation. For example, the number of spatial coordinate points corresponding to volume size interval 1 is 2000, volume size interval 1 < volume size interval 2, the number of spatial coordinate points corresponding to volume size interval 2 is 1500, volume size interval 2 < volume size interval 3, and the number of spatial coordinate points corresponding to volume size interval 3 is 1000. In this way, the electronic device can generate the $i$-th set of spatial coordinates according to the spatial coordinate points within the $i$-th sub-model, and a total of $M$ sets of spatial coordinates are obtained.

[0056] Method 2:

[0057] For the i-th sub-model among the M sub-models, where i is any integer ranging from 1 to M, the electronic device can copy the i-th sub-model into J copies according to the volume size of the i-th sub-model, obtaining J sub-models #i. The value of J is negatively correlated with the volume size of the i-th sub-model. The electronic device can determine the spatial coordinate points within each sub-model #i among the J sub-models #i, obtaining J spatial coordinate sets #i, and a total of N spatial coordinate sets for the N sub-models are obtained. N is a positive integer greater than M. The volume size of each sub-model #i among the J sub-models #i refers to the volume size of this sub-model #i in the three-dimensional model space. The spatial coordinate points within each sub-model #i among the J sub-models #i refer to the coordinates of the coordinate points within this sub-model #i in the three-dimensional model space. The spatial coordinate points within each sub-model #i among the J sub-models #i can be used to generate and represent the cutting optical path. The J spatial coordinate sets #i are the spatial information of the i-th sub-model, and the J spatial coordinate sets #i include the coordinates of the spatial coordinate points within each of the J sub-models #i respectively.

[0058] For example, the electronic device can determine the volume size interval in which the volume size of the i-th sub-model is located from different volume size intervals. The electronic device can copy the i-th sub-model into J copies according to the value of J corresponding to the volume size interval, obtaining J sub-models #i. Among them, in the order of increasing volume, the values of J corresponding to different volume size intervals gradually decrease. Correspondingly, the electronic device can randomly generate a preset number of spatial coordinate points within each sub-model #i among the J sub-models #i (where random generation can ensure the same number, and the coordinate positions are different as a whole). The electronic device generates J spatial coordinate sets #i one by one according to the spatial coordinate points within each sub-model #i among the J sub-models #i, and a total of N spatial coordinate sets are obtained. For example, J = 3 corresponding to volume size interval 1, volume size interval 1 < volume size interval 2, J = 2 corresponding to volume size interval 2, volume size interval 2 < volume size interval 3, and J = 1 corresponding to volume size interval 3.

[0059] It can be seen that the purposes of Method 1 and Method 2 are the same, that is, to set more spatial coordinate points to facilitate the subsequent deep neural network model to generate a more refined cutting optical path for the i-th sub-model. However, the implementation means of Method 1 and Method 2 are different. Method 1 directly increases the number of spatial coordinate points within one sub-model, while the other method is to increase the number of the same sub-model, and the number of spatial coordinate points within each sub-model is the same.

[0060] S103. The electronic device processes the spatial information of multiple sub-models through the deep neural network model to obtain the cutting optical path output by the deep neural network model.

[0061] Among them, the cutting optical path is configured such that if the laser control system performs laser cutting on the component to be cut according to the cutting optical path, the target component can be obtained.

[0062] For the above-mentioned method 1:

[0063] The electronic device can interleave the M sets of spatial coordinates to obtain a sequence of interleaved sets of spatial coordinates. Among them, in the M sub-models, the sets of spatial coordinates corresponding to at least two sub-models with volume sizes in the same volume size interval are concentrated in the positions in the sequence of interleaved sets of spatial coordinates, and the sets of spatial coordinates corresponding to at least two sub-models with volume sizes in different volume size intervals are discrete in the positions in the sequence of interleaved sets of spatial coordinates. For example, for sub-models 1, 2, 3, 4, 5, and 6, sub-model 1 and sub-model 4 are in the same volume size interval, sub-model 2 and sub-model 5 are in the same volume size interval, and sub-model 3 and sub-model 6 are in the same volume size interval. The set of spatial coordinates of sub-model 1 is spatial coordinate set #1, the set of spatial coordinates of sub-model 2 is spatial coordinate set #2, the set of spatial coordinates of sub-model 3 is spatial coordinate set #3, the set of spatial coordinates of sub-model 4 is spatial coordinate set #4, the set of spatial coordinates of sub-model 5 is spatial coordinate set #5, and the set of spatial coordinates of sub-model 6 is spatial coordinate set #6. In the order of the number of spatial coordinate points from more to less (which can also be understood as the energy level from high to low, and here the method of interleaving in wireless communication is borrowed, that is, it is reused in the field of this application), the sequence of interleaved sets of spatial coordinates is {spatial coordinate set #1, spatial coordinate set #4, spatial coordinate set #2, spatial coordinate set #5, spatial coordinate set #3, spatial coordinate set #6}. The electronic device inputs the M sets of spatial coordinates in the order in the sequence of interleaved sets of spatial coordinates into the deep neural network model to obtain the cutting optical path output by the deep neural network model. The cutting optical path can be generated by the above-mentioned spatial coordinates, that is, it contains multiple spatial coordinates, and the cutting optical path can be one or more.

[0064] For the above-mentioned method 2:

[0065] The electronic device can interleave N sets of spatial coordinates to obtain a sequence of interleaved sets of spatial coordinates. Among them, the positions of the J sets of spatial coordinates #i corresponding to the J sub-models #i in the sequence of interleaved sets of spatial coordinates are concentrated. For example, sub-model 1 is replicated into 2, resulting in the 1st sub-model 1 and the 2nd sub-model 1, sub-model 2 is replicated into 3, obtaining the 1st sub-model 2, the 2nd sub-model 2, and sub-model 3 is not replicated. The set of spatial coordinates of sub-model 1 is the set of spatial coordinates #11 and the set of spatial coordinates #12, the set of spatial coordinates of sub-model 2 is the set of spatial coordinates #21, the set of spatial coordinates #22, and the set of spatial coordinates #23, and the set of spatial coordinates of sub-model 3 is the set of spatial coordinates #3. In the order from the most to the least number of spatial coordinate points (which can also be understood as from the highest to the lowest energy level. Here, the interleaving method in wireless communication is borrowed, that is, it is reused in the field of this application), the sequence of interleaved sets of spatial coordinates is {set of spatial coordinates #11, set of spatial coordinates #12, set of spatial coordinates #21, set of spatial coordinates #22, set of spatial coordinates #23, set of spatial coordinates #3}. The electronic device can input the N sets of spatial coordinates in the order of the sequence of interleaved sets of spatial coordinates into the deep neural network model to obtain the cutting optical path output by the deep neural network model.

[0066] In summary, through 3D modeling, the electronic device can determine the model of the cut part in the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut, and divide the model of the cut part into multiple sub-models. In this way, the electronic device can obtain the cutting optical path for the component to be cut output by the deep neural network model through the spatial information of multiple sub-models. In addition, since the model of the cut part is not input into the deep neural network model as a whole for processing, but is processed in units of multiple sub-models, the performance requirements for the deep neural network model can be reduced, and the robustness of the deep neural network model can be improved.

[0067] The communication method provided in the embodiments of the present application has been described in detail above. The following will describe in detail the system for executing the method provided in the embodiments of the present application.

[0068] A laser control system based on machine learning, the system includes an electronic device, and the system is configured to: the electronic device determines the model of the cut part in the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut; wherein, the target component and the cut part are combined into the component to be cut; the electronic device divides the model of the cut part into multiple sub-models and determines the spatial information of the multiple sub-models; the electronic device processes the spatial information of the multiple sub-models through a deep neural network model to obtain the cutting optical path output by the deep neural network model; wherein, the cutting optical path is configured to: if the laser control system performs laser cutting on the component to be cut according to the cutting optical path, the target component can be obtained.

[0069] Optionally, the electronic device divides the model of the cut part into multiple sub-models and determines the spatial information of the multiple sub-models, including: the electronic device divides the model of the cut part into multiple sub-models in the three-dimensional model space according to the preset spatial equal division lines in the three-dimensional model space, wherein the spatial equal division lines are used to equally divide the three-dimensional model space into multiple cubic sub-spaces; the electronic device determines the spatial information of the multiple sub-models according to the respective volume sizes of the multiple sub-models.

[0070] Optionally, the coordinates of the three-dimensional model space are represented by the coordinates of the x-axis, the coordinates of the y-axis, and the coordinates of the z-axis. The spatial equal division lines include K1 dividing lines, K2 dividing lines, and K3 dividing lines. K1, K2, and K3 are all integers greater than 1. The K1 dividing lines are perpendicular to the plane formed by the x-axis and the y-axis. The K2 dividing lines are perpendicular to the plane formed by the x-axis and the z-axis. The K3 dividing lines are perpendicular to the plane formed by the y-axis and the z-axis; the electronic device divides the model of the cut part into multiple sub-models in the three-dimensional model space according to the preset spatial equal division lines in the three-dimensional model space, including: the electronic device overlaps the model of the cut part with the cutting planes formed by any two mutually perpendicular dividing lines among the K1 dividing lines, the K2 dividing lines, and the K3 dividing lines, and divides the model of the cut part along the cutting planes, and a total of M sub-models are obtained by division, where M is an integer greater than 2.

[0071] Optionally, the electronic device determines the spatial information of multiple sub-models according to the volume size of each of the multiple sub-models, including: for the i-th sub-model among the M sub-models, where i is any integer from 1 to M, the electronic device determines the spatial coordinate points within the i-th sub-model according to the volume size of the i-th sub-model, so as to obtain the i-th set of spatial coordinates, and a total of M sets of spatial coordinates are obtained; wherein, the number of spatial coordinate points within the i-th sub-model is negatively correlated with the volume size of the i-th sub-model; the volume size of the i-th sub-model refers to the volume size of the i-th sub-model in the three-dimensional model space, the spatial coordinate points within the i-th sub-model refer to the coordinates of the coordinate points within the i-th sub-model in the three-dimensional model space, the spatial coordinate points within the i-th sub-model can be used to generate and represent the cutting optical path, the i-th set of spatial coordinates is the spatial information of the i-th sub-model, and the i-th set of spatial coordinates includes the coordinates of each of the spatial coordinate points within the i-th sub-model.

[0072] Optionally, for the i-th sub-model among the M sub-models, where i is any integer from 1 to M, the electronic device determines the spatial coordinate points within the i-th sub-model according to the volume size of the i-th sub-model and obtains the i-th set of spatial coordinates, including: the electronic device determines the volume size interval where the volume size of the i-th sub-model is located from different volume size intervals; the electronic device randomly generates, within the i-th sub-model, the number of spatial coordinate points corresponding to the volume size interval according to the number of spatial coordinate points corresponding to the volume size interval, so as to obtain the spatial coordinate points within the i-th sub-model; wherein, in the order of increasing volume, the number of spatial coordinate points corresponding to different volume size intervals gradually decreases; the electronic device generates the i-th set of spatial coordinates according to the spatial coordinate points within the i-th sub-model, and a total of M sets of spatial coordinates are obtained.

[0073] Optionally, the electronic device processes the spatial information of multiple sub-models through a deep neural network model to obtain the cutting optical path output by the deep neural network model, including: the electronic device interleaves the M sets of spatial coordinates to obtain an interleaved sequence of sets of spatial coordinates; wherein, among the M sub-models, the positions of the sets of spatial coordinates corresponding to at least two sub-models with volume sizes in the same volume size interval are concentrated in the interleaved sequence of sets of spatial coordinates, and the positions of the sets of spatial coordinates corresponding to at least two sub-models with volume sizes in different volume size intervals are discrete in the interleaved sequence of sets of spatial coordinates; the electronic device inputs the M sets of spatial coordinates into the deep neural network model in the order of the interleaved sequence of sets of spatial coordinates to obtain the cutting optical path output by the deep neural network model.

[0074] Optionally, the electronic device determines the spatial information of multiple sub-models according to the volume size of each sub-model, including: for the i-th sub-model among the M sub-models, where i is any integer from 1 to M, the electronic device copies the i-th sub-model into J copies according to the volume size of the i-th sub-model to obtain J sub-models #i, and the value of J is negatively correlated with the volume size of the i-th sub-model; the electronic device determines the spatial coordinate points in each of the J sub-models #i to obtain J sets of spatial coordinates #i, and a total of N sets of spatial coordinates of N sub-models are obtained, where N is a positive integer greater than M; among them, the volume size of each of the J sub-models #i refers to the volume size of the sub-model #i in the three-dimensional model space, the spatial coordinate points in each of the J sub-models #i refer to the coordinates of the coordinate points in the sub-model #i in the three-dimensional model space, the spatial coordinate points in each of the J sub-models #i can be used to generate and represent the cutting optical path, and the J sets of spatial coordinates #i are the spatial information of the i-th sub-model, and the J sets of spatial coordinates #i include the coordinates of the spatial coordinate points in each of the J sub-models #i respectively.

[0075] Optionally, the electronic device copies the i-th sub-model into J copies according to the volume size of the i-th sub-model to obtain J sub-models #i, including: the electronic device determines the volume size interval where the volume size of the i-th sub-model is located from different volume size intervals; the electronic device copies the i-th sub-model into J copies according to the value of J corresponding to the volume size interval to obtain J sub-models #i; among them, in the order of increasing volume, the values of J corresponding to different volume size intervals gradually decrease; correspondingly, the electronic device determines the spatial coordinate points in each of the J sub-models #i to obtain J sets of spatial coordinates #i, and a total of N sets of spatial coordinates of N sub-models are obtained, including: the electronic device randomly generates a preset number of spatial coordinate points in each of the J sub-models #i, the spatial coordinate points in each of the J sub-models #i; the electronic device generates J sets of spatial coordinates #i one-to-one according to the spatial coordinate points in each of the J sub-models #i, and a total of N sets of spatial coordinates are obtained.

[0076] Optionally, the electronic device processes the spatial information of multiple sub-models through a deep neural network model to obtain the cutting optical path output by the deep neural network model, including: the electronic device interleaves the N sets of spatial coordinates to obtain an interleaved sequence of spatial coordinate sets; among them, the positions of the J sets of spatial coordinates #i corresponding to the J sub-models #i are concentrated in the interleaved sequence of spatial coordinate sets; the electronic device inputs the N sets of spatial coordinates into the deep neural network model in the order of the interleaved sequence of spatial coordinate sets to obtain the cutting optical path output by the deep neural network model.

[0077] Optionally, the electronic device determines the model of the cut part in the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut, including: in response to a user input operation, the electronic device generates / obtains the model of the component to be cut and the model of the target component; the electronic device determines the different part between the model of the component to be cut and the model of the target component in the three-dimensional model space as the model of the cut part.

[0078] Figure 2 Structural schematic of the electronic device provided in the embodiments of the present application Figure 2 . Exemplarily, the electronic device may be a terminal, or a chip (system) or other components or assemblies that can be set in a terminal. As Figure 2 shown, the electronic device 200 may include a processor 201. Optionally, the electronic device 200 may further include a memory 202 and / or a transceiver 203. Among them, the processor 201 is coupled to the memory 202 and the transceiver 203, such as being connected through a communication bus. In addition, the electronic device 200 may also be a chip, such as including a processor 201. At this time, the transceiver may be an input / output interface of the chip.

[0079] The following combines Figure 2 to specifically introduce each component of the electronic device 200:

[0080] Among them, the processor 201 is the control center of the electronic device 200, which may be a single processor or a collective term for multiple processing elements. For example, the processor 201 is one or more central processing units (CPUs), or may be an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application, such as: one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs).

[0081] Optionally, the processor 201 may execute various functions of the electronic device 200 by running or executing software programs stored in the memory 202 and calling data stored in the memory 202, such as executing the above method.

[0082] In a specific implementation, as an embodiment, the processor 201 may include one or more CPUs, such as Figure 2 the CPU0 and CPU1 shown in

[0083] In a specific implementation, as an example, the electronic device 200 may also include multiple processors. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here can refer to one or more devices, circuits, and / or processing cores for processing data (such as computer programs or instructions).

[0084] Among them, the memory 202 is used to store the software program for executing the solution of this application and is controlled by the processor 201 for execution. The specific implementation manner can refer to the above method embodiment and will not be elaborated here.

[0085] Optionally, the memory 202 can be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or can also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 202 can be integrated with the processor 201 or exist independently and is coupled to the processor 201 through the interface circuit ( Figure 2 not shown in the figure), and this application embodiment does not make specific limitations on this.

[0086] The transceiver 203 is used for communication with other electronic devices. For example, when the electronic device 200 is a terminal, the transceiver 203 can be used for communication with a network device or with another terminal device. Another example is that when the electronic device 200 is a network device, the transceiver 203 can be used for communication with a terminal or with another network device.

[0087] Optionally, the transceiver 203 can include a receiver and a transmitter ( Figure 2 not shown separately). Among them, the receiver is used to implement the receiving function, and the transmitter is used to implement the sending function.

[0088] Optionally, the transceiver 203 can be integrated with the processor 201 or exist independently and is coupled to the processor 201 through the interface circuit (Figure 2 is not shown in the figure) and is coupled to the processor 201. The embodiments of the present application do not make specific limitations in this regard.

[0089] It can be understood that Figure 2 the structure of the electronic device 200 shown in the figure does not constitute a limitation on the electronic device. The actual electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0090] In addition, the technical effects of the electronic device 200 can refer to the technical effects of the method described in the above method embodiments, and will not be elaborated here.

[0091] It should be understood that the processor in the embodiments of the present application may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0092] It should also be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0093] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer program or instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer program or instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, or magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0094] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this document generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood with reference to the context.

[0095] In the present application, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or a similar expression refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0096] It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0097] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0098] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0099] In several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of devices or units can be electrical, mechanical, or other forms.

[0100] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0101] In addition, the functional units in each embodiment of this application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0102] When the above-mentioned functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0103] As described above, the above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A laser control method based on machine learning, characterized in that: Applied to electronic equipment, the method comprises: The electronic device determines the model of the cut portion of the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut; wherein the target component and the cut portion are combined into the component to be cut; The electronic device divides the model of the cut portion into a plurality of sub-models, and determines spatial information of the plurality of sub-models; The electronic device processes the spatial information of the multiple sub-models through a deep neural network model to obtain a cutting optical path output by the deep neural network model; wherein the cutting optical path is configured such that if a laser control system performs laser cutting on the component to be cut according to the cutting optical path, the target component can be obtained; The electronic device divides the model of the cut portion into a plurality of sub-models, and determines spatial information of the plurality of sub-models, including: The electronic device divides the model of the cut portion into the plurality of sub-models in the three-dimensional model space according to a space dividing line preset in the three-dimensional model space, wherein the space dividing line is used to divide the three-dimensional model space into a plurality of cubic sub-spaces; The electronic device determines the spatial information of the multiple sub-models according to the volume sizes of the multiple sub-models; The coordinates of the three-dimensional model space are represented by the coordinates of the x-axis, the coordinates of the y-axis, and the coordinates of the z-axis, the space bisectors include K1 bisectors, K2 bisectors, and K3 bisectors, K1, K2, and K3 are all integers greater than 1, the K1 bisectors are perpendicular to the plane formed by the x-axis and the y-axis, the K2 bisectors are perpendicular to the plane formed by the x-axis and the z-axis, and the K3 bisectors are perpendicular to the plane formed by the y-axis and the z-axis; the electronic device, in the three-dimensional model space, divides the model of the cut portion into the multiple sub-models according to the space bisectors preset in the three-dimensional model space, including: The electronic device overlaps the cutting plane formed by any two mutually perpendicular cutting lines among the K1 cutting lines, the K2 cutting lines and the K3 cutting lines according to the model of the cut part, and cuts the model of the cut part along the cutting plane to obtain M sub-models in total, where M is an integer greater than 2; The electronic device determines the spatial information of the multiple sub-models according to the respective volume sizes of the multiple sub-models, including: For an i-th submodel of the M submodels, i is any integer from 1 to M, the electronic device determines a spatial coordinate point in the i-th submodel according to the volume of the i-th submodel to obtain an i-th spatial coordinate set, and obtains a total of M spatial coordinate sets; Among them, the number of spatial coordinate points in the i-th sub-model is negatively correlated with the volume size of the i-th sub-model; the volume size of the i-th sub-model refers to the volume size of the i-th sub-model in the three-dimensional model space, the spatial coordinate points in the i-th sub-model refer to the coordinates of the coordinate points in the i-th sub-model in the three-dimensional model space, the spatial coordinate points in the i-th sub-model can be used to generate and represent the cutting light path, the i-th spatial coordinate set is the spatial information of the i-th sub-model, and the i-th spatial coordinate set includes the coordinates of the respective spatial coordinate points in the i-th sub-model.

2. The method according to claim 1, characterized in that For the i-th sub-model among the M sub-models, i is any integer from 1 to M, the electronic device determines the spatial coordinate points in the i-th sub-model according to the volume of the i-th sub-model, and obtains the i-th spatial coordinate set, including; The electronic device determines the volume size interval in which the volume size of the i-th sub-model is located from different volume size intervals; The electronic device randomly generates, in the i-th sub-model, a number of spatial coordinate points equal to the number of spatial coordinate points corresponding to the volume size interval according to the number of spatial coordinate points corresponding to the volume size interval, so as to obtain spatial coordinate points in the i-th sub-model; wherein, in order from small to large volume, the number of spatial coordinate points corresponding to each of the different volume size intervals gradually decreases; The electronic device generates the i-th spatial coordinate set according to the spatial coordinate points in the i-th sub-model, and obtains the M spatial coordinate sets in total.

3. The method according to claim 2, characterized in that The electronic device processes the spatial information of the multiple sub-models through a deep neural network model to obtain a cutting optical path output by the deep neural network model, including: The electronic device interleaves the M spatial coordinate sets to obtain an interleaved spatial coordinate set sequence; wherein, among the M sub-models, the spatial coordinate sets corresponding to at least two sub-models whose volume sizes are in the same volume size interval are concentrated in the position of the interleaved spatial coordinate set sequence, and the spatial coordinate sets corresponding to at least two sub-models whose volume sizes are in different volume size intervals are discrete in the position of the interleaved spatial coordinate set sequence; The electronic device inputs the M spatial coordinate sets into the deep neural network model in the order of the interleaved spatial coordinate set sequence to obtain the cutting optical path output by the deep neural network model.

4. The method according to claim 1, characterized in that: The electronic device determines the spatial information of the multiple sub-models according to the respective volume sizes of the multiple sub-models, including: For the i-th sub-model among the M sub-models, i is any integer from 1 to M, the electronic device copies the i-th sub-model into J copies according to the volume of the i-th sub-model to obtain J sub-models #i, and the value of J is negatively correlated with the volume of the i-th sub-model; The electronic device determines the spatial coordinate points in each sub-model #i of the J sub-models #i to obtain J spatial coordinate sets #i, and obtains N spatial coordinate sets of the N sub-models in total, where N is a positive integer greater than M; Among them, the volume size of each sub-model #i among the J sub-models #i refers to the volume size of the sub-model #i in the three-dimensional model space, the spatial coordinate point in each sub-model #i among the J sub-models #i refers to the coordinate of the coordinate point in the sub-model #i in the three-dimensional model space, the spatial coordinate point in each sub-model #i among the J sub-models #i can be used to generate and represent the cutting light path, the J spatial coordinate sets #i are the spatial information of the i-th sub-model, and the J spatial coordinate sets #i include the coordinates of the respective spatial coordinate points in the J sub-models #i.

5. The method according to claim 4, characterized in that The electronic device copies the i-th sub-model into J copies according to the volume of the i-th sub-model to obtain J sub-models #i, including; The electronic device determines the volume size interval in which the volume size of the i-th sub-model is located from different volume size intervals; The electronic device copies the i-th sub-model into J copies according to the value of J corresponding to the volume size interval, to obtain J sub-models #i; wherein, in order from small to large volume, the values ​​of J corresponding to the different volume size intervals gradually decrease; Correspondingly, the electronic device determines the spatial coordinate points in each sub-model #i of the J sub-models #i, obtains J spatial coordinate sets #i, and obtains N spatial coordinate sets of N sub-models in total, including: The electronic device randomly generates a preset number of spatial coordinate points in each sub-model #i in the J sub-models #i, the spatial coordinate points in each sub-model #i in the J sub-models #i; The electronic device generates the J spatial coordinate sets #i in a one-to-one correspondence according to the spatial coordinate points in each sub-model #i in the J sub-models #i, and obtains the N spatial coordinate sets in total.

6. The method according to claim 5, characterized in that The electronic device processes the spatial information of the multiple sub-models through a deep neural network model to obtain a cutting optical path output by the deep neural network model, including: The electronic device interleaves the N spatial coordinate sets to obtain an interleaved spatial coordinate set sequence; wherein the positions of the J spatial coordinate sets #i corresponding to the J sub-models #i in the interleaved spatial coordinate set sequence are concentrated; The electronic device inputs the N spatial coordinate sets into the deep neural network model in the order of the interleaved spatial coordinate set sequence to obtain the cutting optical path output by the deep neural network model.

7. A laser control system based on machine learning, characterized in that: The system comprises an electronic device, wherein the system is configured to: The electronic device determines the model of the cut portion of the component to be cut according to the model of the component to be cut and the model of the target component after the component to be cut is cut; wherein the target component and the cut portion are combined into the component to be cut; The electronic device divides the model of the cut portion into a plurality of sub-models, and determines spatial information of the plurality of sub-models; The electronic device processes the spatial information of the multiple sub-models through a deep neural network model to obtain a cutting optical path output by the deep neural network model; wherein the cutting optical path is configured such that if a laser control system performs laser cutting on the component to be cut according to the cutting optical path, the target component can be obtained; The electronic device divides the model of the cut portion into a plurality of sub-models, and determines spatial information of the plurality of sub-models, including: The electronic device divides the model of the cut portion into the plurality of sub-models in the three-dimensional model space according to a space dividing line preset in the three-dimensional model space, wherein the space dividing line is used to divide the three-dimensional model space into a plurality of cubic sub-spaces; The electronic device determines the spatial information of the multiple sub-models according to the volume sizes of the multiple sub-models; The coordinates of the three-dimensional model space are represented by the coordinates of the x-axis, the coordinates of the y-axis, and the coordinates of the z-axis, the space bisectors include K1 bisectors, K2 bisectors, and K3 bisectors, K1, K2, and K3 are all integers greater than 1, the K1 bisectors are perpendicular to the plane formed by the x-axis and the y-axis, the K2 bisectors are perpendicular to the plane formed by the x-axis and the z-axis, and the K3 bisectors are perpendicular to the plane formed by the y-axis and the z-axis; the electronic device, in the three-dimensional model space, divides the model of the cut portion into the multiple sub-models according to the space bisectors preset in the three-dimensional model space, including: The electronic device overlaps the cutting plane formed by any two mutually perpendicular cutting lines among the K1 cutting lines, the K2 cutting lines and the K3 cutting lines according to the model of the cut part, and cuts the model of the cut part along the cutting plane to obtain M sub-models in total, where M is an integer greater than 2; The electronic device determines the spatial information of the multiple sub-models according to the respective volume sizes of the multiple sub-models, including: For an i-th submodel of the M submodels, i is any integer from 1 to M, the electronic device determines a spatial coordinate point in the i-th submodel according to the volume of the i-th submodel to obtain an i-th spatial coordinate set, and obtains a total of M spatial coordinate sets; Among them, the number of spatial coordinate points in the i-th sub-model is negatively correlated with the volume size of the i-th sub-model; the volume size of the i-th sub-model refers to the volume size of the i-th sub-model in the three-dimensional model space, the spatial coordinate points in the i-th sub-model refer to the coordinates of the coordinate points in the i-th sub-model in the three-dimensional model space, the spatial coordinate points in the i-th sub-model can be used to generate and represent the cutting light path, the i-th spatial coordinate set is the spatial information of the i-th sub-model, and the i-th spatial coordinate set includes the coordinates of the respective spatial coordinate points in the i-th sub-model.

Images