Punching cutting path generation method and system combined with graphic optimization

Abstract

The invention provides a method and a system for generating a punching cutting path in combination with graphic optimization, and relates to the technical field of path planning, and the method comprises the following steps: obtaining basic information of a punching cutting piece; the punching cutting sample image is matched, graph semantic segmentation is carried out, and punching cutting sample feature information is obtained and comprises geometric feature information and texture feature information; a punching cutting path is planned, and an initial punching cutting path is generated; performing graph optimization by minimizing the total path length, minimizing the repeated path length and minimizing the path crossing number to generate an optimized punching cutting path; pre-punching cutting is conducted, pre-punching cutting feedback data are collected, path feedback adjustment is conducted, and a target punching cutting path is obtained; and synchronizing to a punching cutting machine to guide a punching cutting path. The method solves the technical problems that a simple linear or geometric method is mostly adopted in traditional punching cutting path planning, effective optimization of the path is lacked, the metal cutting efficiency is low, and the metal cutting quality is poor.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and particularly to a punching and cutting path generation method and system combined with graphic optimization. Background Art

[0002] The punching and cutting technology is widely used in many fields, such as metal processing, electronic manufacturing, and the automotive industry. However, in the traditional punching and cutting process, path planning often relies on manual experience. The operator sets the cutting path according to experience. This method is not universal and is easily affected by human factors. Especially when dealing with workpieces with complex shapes and high-precision requirements, it is easy to lead to inefficient and inaccurate metal cutting paths. Moreover, although modern computer vision and deep learning technologies have made remarkable progress in image analysis, the existing punching and cutting path generation methods do not fully utilize these technologies. The application of image semantic segmentation and feature extraction methods in path planning is still limited, lacking in-depth analysis of complex punching and cutting paths, and unable to achieve accurate metal cutting path generation and optimization. Summary of the Invention

[0003] The present application provides a punching and cutting path generation method and system combined with graphic optimization, aiming to solve the technical problem that traditional punching and cutting path planning mostly uses simple linear or geometric methods, lacking effective optimization of the path, resulting in low metal cutting efficiency and poor quality.

[0004] In the first aspect disclosed in the present application, a punching and cutting path generation method combined with graphic optimization is provided. The method includes: obtaining basic information of the punching and cutting part, where the basic information of the punching and cutting part includes the model information and positioning information of the punching and cutting part; matching a punching and cutting sample image according to the model information of the punching and cutting part, and performing graphic semantic segmentation on the punching and cutting sample image to obtain punching and cutting sample feature information, where the punching and cutting sample feature information includes geometric feature information and texture feature information; based on the positioning information of the punching and cutting part, combining the punching and cutting sample feature information to perform punching and cutting path planning to generate an initial punching and cutting path; performing graphic optimization on the initial punching and cutting path to minimize the total path length, minimize the repeated path length, and minimize the number of path intersections to generate an optimized punching and cutting path; using the optimized punching and cutting path to perform pre-punching and cutting, and collecting pre-punching and cutting feedback data, and performing path feedback adjustment according to the pre-punching and cutting feedback data to obtain a target punching and cutting path; synchronizing the target punching and cutting path to a punching and cutting machine for punching and cutting path guidance.

[0005] The second aspect disclosed in this application provides a punching and cutting path generation system combined with graphic optimization. The system is used for the punching and cutting path generation method combined with graphic optimization as described above. The system includes: a basic information acquisition module for acquiring the basic information of the punching and cutting part, where the basic information of the punching and cutting part includes the model information and positioning information of the punching and cutting part; a graphic semantic segmentation module for matching the punching and cutting sample image according to the model information of the punching and cutting part and performing graphic semantic segmentation on the punching and cutting sample image to obtain the punching and cutting sample feature information, where the punching and cutting sample feature information includes geometric feature information and texture feature information; a path planning module for planning the punching and cutting path based on the positioning information of the punching and cutting part and combining the punching and cutting sample feature information to generate an initial punching and cutting path; a graphic optimization module for performing graphic optimization on the initial punching and cutting path to minimize the total path length, minimize the repeated path length, and minimize the number of path crossings, and generate an optimized punching and cutting path; a path feedback adjustment module for using the optimized punching and cutting path to perform pre-punching and cutting, collecting pre-punching and cutting feedback data, and performing path feedback adjustment according to the pre-punching and cutting feedback data to obtain a target punching and cutting path; a path guiding module for synchronizing the target punching and cutting path to a punching and cutting machine for punching and cutting path guidance.

[0006] One or more technical solutions provided in this application have at least the following beneficial effects: By acquiring the basic information of the punching and cutting part, including model information and positioning information, and combining the geometric and texture features of the sample image for cutting path planning, the accuracy of the path is ensured, and an initial path that meets the requirements can be generated for different cutting tasks; through graphic optimization, including minimizing the total path length, reducing repeated paths, and avoiding path crossings, the efficiency of the cutting process can be greatly improved, unnecessary path travel can be reduced, cutting time can be reduced, and mechanical wear can be reduced; by using the optimized path for pre-punching and cutting and collecting feedback data, problems existing in the path, such as pauses, crossings, or repetitions, can be identified and adjusted to ensure that the finally generated target cutting path is more accurate and efficient; by synchronizing the target path to the cutting machine for path guidance, the punching and cutting machine can be directly guided to operate according to the optimized path, thereby improving the overall production efficiency of metal cutting, reducing mistakes and waste in the metal cutting process, and ensuring the quality and production stability of metal cutting.

[0007] The above description is only an overview of the technical solutions of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the description. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the specific implementation manners of this application are specifically given below. Brief Description of the Drawings

[0008] Figure 1 This is a schematic flow chart of the punching and cutting path generation method combined with graphic optimization provided by the embodiments of the present application.

[0009] Figure 2 This is a schematic structural diagram of the punching and cutting path generation system combined with graphic optimization provided by the embodiments of the present application.

[0010] Explanation of reference numerals: The basic information acquisition module 10, the graphic semantic segmentation module 20, the path planning module 30, the graphic optimization module 40, the path feedback adjustment module 50, and the path guidance module 60. Specific embodiments

[0011] By providing a punching and cutting path generation method and system combined with graphic optimization, the embodiments of the present application solve the technical problems that traditional punching and cutting path planning mostly uses simple linear or geometric methods, lacks effective optimization of the path, and results in low metal cutting efficiency and poor quality.

[0012] After introducing the basic principle of the present application, the following will specifically introduce various non-limiting embodiments of the present application in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0013] Embodiment 1, as Figure 1 shown, the embodiments of the present application provide a punching and cutting path generation method combined with graphic optimization, and the method includes: Obtain the basic information of the punching and cutting part, where the basic information of the punching and cutting part includes the model information of the punching and cutting part and the positioning information of the punching and cutting part.

[0014] Obtain the basic information of the punching and cutting part, which is used for subsequent path planning and optimization. Among them, the model information of the punching and cutting part refers to the model or identifier of the punching and cutting part, including information such as the size, material, and structure of the cutting part. By obtaining this information, the correct punching and cutting sample image and processing method can be selected; the positioning information of the punching and cutting part refers to the position data of the punching and cutting part on the production equipment, such as the coordinates or assembly position of the cutting part on the workbench. These positioning information helps to accurately plan the cutting path and ensure that the path planning matches the actual layout of the cutting part.

[0015] According to the model information of the punching and cutting part, match the punching and cutting sample image, and perform graphic semantic segmentation on the punching and cutting sample image to obtain the punching and cutting sample feature information, where the punching and cutting sample feature information includes geometric feature information and texture feature information.

[0016] Match the sample images corresponding to the punching and cutting part model information. These sample images have been labeled and contain key information such as punching and cutting. Perform graphic semantic segmentation on the sample images. Semantic segmentation is a technique in computer vision. Through this technique, each pixel in the image can be labeled as a specific category, such as hole positions, cutting paths, etc. The goal is to extract meaningful regions from the image for further feature analysis.

[0017] After graphic semantic segmentation, extract feature information from the segmented regions. Among them, geometric feature information includes aperture size, hole position, shape of the cutting profile, and intersection contact angle, etc.; texture feature information includes surface roughness, crack size, number of surface defects, etc. Texture features help to evaluate the quality and effect of punching and cutting, and affect subsequent path optimization and adjustment.

[0018] Based on the positioning information of the punching and cutting parts, combine the punching and cutting sample feature information to plan the punching and cutting path and generate the initial punching and cutting path.

[0019] Based on the positioning information of the punching and cutting parts, first determine the positions of the punching points. Each punching point represents the specific position where the punching machine needs to perform punching. These points will form a set for subsequent path planning; and determine the starting point and ending point of the cutting path. The starting point is the starting position of the punching and cutting machine, and the ending point is the completion position of the cutting. Through this information, form the two-end point set of the cutting path to provide basic data for subsequent path planning. Use the punching point set, cutting path starting point set, ending point set, and punching and cutting sample feature information to plan the initial cutting path. The specific path planning process will be detailed in the subsequent steps. This cutting path covers all punching points and has a reasonable route sequence.

[0020] Optimize the initial punching and cutting path in terms of minimizing the total path length, minimizing the repeated path length, and minimizing the number of path intersections to generate the optimized punching and cutting path.

[0021] The total path length refers to the total movement distance of the punching cutter from the starting point to the ending point. In path optimization, by adjusting the path direction, the movement distance of the machine is minimized as much as possible to save cutting time. This goal is achieved through optimization algorithms, such as approximation algorithms for the traveling salesman problem. During the cutting process, there may be some paths that are repeatedly traversed due to improper planning. Repeated paths not only waste time but also increase the burden on the equipment. Therefore, during optimization, it is necessary to eliminate or reduce the length of repeated paths to ensure that the path to each punching point and cutting position is passed through only once. Path intersection means that the cutting machine will encounter the intersection points of other paths during movement, which may cause the cutting machine to pause, turn, or require additional positioning time. The optimization process needs to minimize path intersections by adjusting the path direction so that there are no intersections or conflicts between paths.

[0022] To achieve the above optimization goals, optimization algorithms such as the greedy algorithm, genetic algorithm, and simulated annealing are used. By continuously iterating to optimize the path, after optimization, the generated optimized punching and cutting path will minimize time, improve cutting efficiency, and reduce friction and wear of the equipment, thereby enhancing production efficiency and cutting quality.

[0023] Perform pre-punching and cutting using the optimized punching and cutting path, and collect pre-punching and cutting feedback data. Adjust the path based on the pre-punching and cutting feedback data to obtain the target punching and cutting path.

[0024] Transfer the optimized punching and cutting path to the actual punching cutter for pre-cutting operations. This is a trial operation stage, and the main purpose is to verify the effectiveness of path planning. During pre-punching and cutting, the punching cutter cuts the material according to this path and real-time collects data related to the cutting process to obtain pre-punching and cutting feedback data, including processing movement pause data, path intersection and repetition data, etc.

[0025] Based on the collected feedback data, adjust the path. Specifically, if the feedback data shows the existence of path intersections or repetitions, the path will be adjusted to ensure that the machine avoids these problems during the next cutting. If long pauses are recorded, the path needs to be re-planned to avoid excessive repositioning of the machine. The goal of path feedback adjustment is to improve path planning through the feedback data of the pre-punching and cutting process, making the final target cutting path more accurate and effective.

[0026] Synchronize the target punching and cutting path to the punching cutter for punching and cutting path guidance.

[0027] Synchronize the final target punching and cutting path to the punching and cutting machine. The machine automatically guides and performs the cutting operation according to the path data. The machine will execute according to each punching point and cutting path on the path to ensure that all punching and cutting actions meet the requirements, making the punching and cutting process more accurate and efficient.

[0028] Furthermore, for the graphic semantic segmentation of the punching and cutting sample image to obtain the punching and cutting sample feature information, the method includes: Perform graphic semantic segmentation on the punching and cutting sample image to obtain the punching and cutting identification area; extract geometric feature information from the punching and cutting identification area, where the geometric feature information includes aperture size features, aperture position features, cutting profile shape features, and junction contact angle features; extract texture feature information from the punching and cutting identification area, where the texture feature information includes surface roughness features, crack size features, and surface defect quantity features.

[0029] Graphic semantic segmentation is a computer vision technology used to label each pixel in an image as a different category. The goal is to identify the punching area, cutting path, and other relevant areas from the punching and cutting sample image. During the image semantic segmentation process, the punching area and cutting path are labeled as different areas to obtain the punching and cutting identification area, which will serve as the basis for subsequent feature extraction.

[0030] Extract geometric feature information for the punching and cutting identification area in the image to describe the geometric features of punching and cutting. Among them, the aperture size feature describes the diameter or other size features of the punching; the aperture position feature describes the position of the hole, including its coordinates relative to the cut piece or its position on the workpiece; the cutting profile shape feature describes the profile shape of the cutting, such as the linearity, curvature, angle, etc. of the cutting path; the junction contact angle feature describes the contact angle between the cutting path and other areas. For example, at the intersection of two cutting paths, the size of the contact angle will affect the cutting quality and efficiency.

[0031] Extract the texture features of the punched cutting identification area in the image to obtain the texture feature information for judging the cutting quality and surface flatness. Among them, the surface roughness feature describes the smoothness or roughness of the cutting surface, which is usually obtained by measuring the details of the surface texture, such as the fluctuation amplitude, wavelength, etc. A rough surface means poor cutting quality or improper cutting parameter settings; the crack size feature describes whether there are cracks on the surface and their size features. Cracks are usually related to factors such as material properties, cutting conditions, tool wear, etc. The appearance of cracks may affect the final cutting quality; the surface defect quantity feature describes the number of defects on the surface, including cracks, pores, scratches, etc. The number of surface defects usually reflects the instability during the cutting process or equipment problems.

[0032] Furthermore, the method for performing graphic semantic segmentation on the punched cutting sample image to obtain the punched cutting identification area includes: Based on the punched cutting detection record, obtain the historical punched cutting image set; identify and shape-identify the punched cutting areas in the historical punched cutting image set to obtain the sample punched cutting recognition result set; based on graphic semantic segmentation, construct an encoder and a decoder, and use the historical punched cutting image set and the sample punched cutting recognition result set for training to obtain a punched cutting area recognizer; use the punched cutting area recognizer to identify the punched cutting area of the punched cutting sample image to obtain the punched cutting identification area.

[0033] Obtain the punched cutting detection record, which contains the historical punched cutting image set of the previously executed punched cutting tasks. The historical punched cutting image set contains different cutting areas, including information such as punching positions and cutting paths. These images can be collected from previous cutting tasks or obtained through an automatic detection system. These image sets will be used as basic data for subsequent pattern recognition and training.

[0034] Analyze each image in the historical image set to identify the punched cutting areas therein. For each punched cutting area, identify it according to the shape features, such as circular, elliptical, etc. The identified punched cutting areas and their shape information will form the sample punched cutting recognition result set. These recognition results will be used as labeled data for subsequent model training.

[0035] The task of semantic segmentation usually adopts convolutional neural networks, especially network structures such as U-Net and FCN, which can effectively perform image segmentation. An encoder and a decoder are constructed based on graphic semantic segmentation. Among them, the encoder is used to extract feature information in the image. The encoder converts the input image into a higher-level feature representation through multiple convolutional layers, and extracts the deep semantic information of the image. The decoder is responsible for restoring the features extracted by the encoder into an image, accurately performing pixel-level classification, and labeling each region in the image (such as the punching region, cutting path, etc.). The cooperation of the encoder and the decoder makes the segmentation effect of the image more accurate, and can better identify and label the punching and cutting regions.

[0036] Use the historical punching and cutting image set and the sample punching and cutting recognition result set for training. The sample recognition result set provides the annotation data of the image. The model learns through these data and gradually adjusts its internal parameters. The training process requires multiple rounds of iteration. Through continuous forward propagation and backward propagation, the loss function of the model is optimized, so as to improve the accuracy of the model in identifying the punching and cutting regions. Through the above training, a punching and cutting region recognizer is finally obtained. This recognizer can accurately identify the punching region, cutting path and other relevant regions based on the input punching and cutting sample image.

[0037] Input the punching and cutting sample image into the punching and cutting region recognizer. According to the previous training results, the recognizer processes the input image and identifies the target regions such as the punching region and the cutting path. During the image recognition process, the recognizer assigns a label to each pixel in the image to identify whether it belongs to the punching region, the cutting path region or other regions. In this way, the punching and cutting identification regions of the entire image will be extracted to form the punching and cutting identification regions. These identification regions provide important data for subsequent path planning and optimization.

[0038] Furthermore, based on the positioning information of the punching and cutting part, combined with the feature information of the punching and cutting sample, perform punching and cutting path planning to generate an initial punching and cutting path. The method includes: Based on the positioning information of the punching and cutting part, extract the punching point set, the cutting path start point set, and the cutting path end point set; randomly extract the first punching point, the first cutting path start point, and the corresponding first cutting path end point; based on the first punching point, the first cutting path start point, and the first cutting path end point, collect the first punching and cutting record data, where the first punching and cutting record data includes multiple groups of punching and cutting paths and multiple groups of punching and cutting images; based on the geometric feature information and texture feature information of the punching and cutting sample feature information, screen the multiple groups of punching and cutting images to obtain the punching and cutting image screening result; extract the punching and cutting path corresponding to the punching and cutting image screening result as the initial punching and cutting path.

[0039] The positioning information of the punched and cut parts includes the position data of the punched and cut parts in the equipment. This positioning information includes the accurate coordinates of each punching point and cutting point on the workpiece. According to the positioning information, all punching points are extracted from the punched and cut parts. Each punching point represents the position where the punching operation needs to be performed in the cutting path. According to the positioning information, the set of starting points of the cutting path and the set of ending points of the cutting path are extracted. The starting point of the cutting path refers to the starting position of the cutting path, and the ending point of the cutting path refers to the ending position of the cutting path. Each cutting path needs to clarify the starting point and the ending point to ensure the completion of cutting and the accuracy of the position.

[0040] Randomly select a punching point from the set of extracted punching points as the analysis object, which is the first punching point. Similarly, randomly select a starting point from the set of starting points of the cutting path and match it with the corresponding ending point. The starting point and the ending point together define a complete cutting path. The purpose of this process is to introduce some randomness, avoid the rigidity of path planning, and increase the diversity and flexibility of the path.

[0041] According to the first punching point, the first starting point of the cutting path, and the first ending point of the cutting path, analyze their corresponding relationships, and obtain the first punched and cut record data obtained from actual or simulated cutting operations according to the corresponding relationships. Among them, each punched and cut operation will generate a specific cutting path. By executing different paths multiple times, multiple punched and cut path data are collected. These path data include information such as the path length and cutting order of each cut. In addition to the path data, multiple groups of punched and cut images are also collected to record the effect after each cut. These images can show the punched area, the cutting path, and the cutting quality of the material.

[0042] For each group of punched and cut images, perform screening according to the geometric features and texture features of the images, calculate the difference between each group of images and the predetermined standard, and select the qualified images according to the matching degree of these features. After screening, the screening result of the punched and cut images is obtained, which contains the qualified images. This screening result will provide accurate basic data for subsequent path extraction and optimization.

[0043] Extract the corresponding punched and cut path from each qualified image. These paths represent the movement trajectory of the cutting machine during the punched and cut operation, including the direction, order, and possible intersection points of the cutting path. The extracted punched and cut path is used as the initial punched and cut path. This initial path is generated based on the screened image data and reflects the actual cutting situation at different punching points and cutting paths.

[0044] Furthermore, based on the geometric feature information and texture feature information of the punching and cutting sample feature information, screening the multiple groups of punching and cutting images to obtain a punching and cutting image screening result, the method comprising: Extracting multiple groups of control geometric feature information and multiple groups of control texture feature information of the multiple groups of punching and cutting images; calculating multiple groups of geometric feature deviations between the multiple groups of control geometric feature information and the geometric feature information; calculating multiple groups of texture feature deviations between the multiple groups of control texture feature information and the texture feature information; and performing image screening on the multiple groups of punching and cutting images based on the multiple groups of geometric feature deviations and the multiple groups of texture feature deviations to obtain the punching and cutting image screening result.

[0045] Using the punching and cutting area recognizer constructed in the foregoing steps to perform punching and cutting area recognition on multiple groups of punching and cutting images, and performing geometric feature extraction and texture feature extraction according to the recognition results to obtain multiple groups of control geometric feature information and multiple groups of control texture feature information. The specific feature extraction process has been described in detail in the foregoing steps and will not be elaborated here. The control geometric feature information refers to the geometric features extracted from the control image and is used to compare with the geometric features in the current image; the control texture feature information refers to the texture features extracted from the control image and serves as the comparison basis for the texture features of the current image.

[0046] Compare the geometric features of the current image with the geometric features extracted from the control image, and calculate the deviation between them. The deviation can be a numerical difference, an angular difference, or a shape similarity, etc. Common calculation methods include Euclidean distance, cosine similarity, etc. Calculate the difference between each pair of geometric features. For example, the distance between two punching positions, the angular difference between two cutting paths, etc. Compare the geometric features in multiple control images with the geometric features of the current image to obtain multiple groups of geometric feature deviations. Through these deviations, the gap between the current image and multiple historical sample images can be understood.

[0047] Similar to the geometric features, calculate the deviation between the texture features of the current image and the texture features of the control image. The texture features can include surface roughness, the size of cracks, the number of defects, etc. The deviation calculation is obtained based on the numerical differences of these texture features to obtain multiple groups of texture feature deviations. These data can help determine whether there are obvious quality differences in the image, such as surface cracks, uneven roughness, etc.

[0048] Based on the calculated deviation results, multiple groups of punching and cutting images are screened. Through comparison, those images with smaller deviations in geometric features and texture features are selected, as they are considered to be more matched with the target cutting path. If the deviations of some images are too large, they will be excluded because they may represent cutting results with substandard quality. After screening, the screening results of the punching and cutting images are finally obtained. These images meet the set standards and can be used for subsequent path extraction and optimization.

[0049] Furthermore, based on the multiple groups of geometric feature deviations and the multiple groups of texture feature deviations, the image screening of the multiple groups of punching and cutting images is performed to obtain the screening results of the punching and cutting images. The method includes: Obtain the geometric feature deviation threshold and the texture feature deviation threshold; eliminate the punching and cutting images in the multiple groups of geometric feature deviations that do not meet the geometric feature deviation threshold to obtain the first image screening result; eliminate the punching and cutting images in the multiple groups of texture feature deviations that do not meet the texture feature deviation threshold to obtain the second image screening result; integrate the first image screening result and the second image screening result to obtain the screening results of the punching and cutting images.

[0050] The geometric feature deviation threshold refers to the maximum allowable difference during the comparison of geometric features. For example, the maximum allowable distance between two punching positions, the maximum allowable deviation of the cutting path angle, etc. If the difference between the two is greater than the set threshold, then this image will be considered unqualified. This threshold is set according to the requirements of specific applications. For example, for larger or more complex punching and cutting tasks, the threshold can be relatively loose, while for tasks with higher precision requirements, the threshold may need to be more stringent.

[0051] The texture feature deviation threshold refers to the maximum allowable surface quality difference. For example, the maximum allowable deviation of surface roughness, the maximum allowable difference in crack size, etc. If the deviation of the texture feature is greater than this threshold, the image will be considered to have unqualified cutting quality and should be eliminated. This threshold is set based on the surface quality standard of the workpiece. For tasks with high surface quality requirements, the threshold should be set lower.

[0052] For each group of punching and cutting images, compare the calculated geometric feature deviation with the geometric feature deviation threshold. If the deviation is greater than the set threshold, it indicates that the geometric features of this image do not meet the requirements, and then it will be eliminated from the screening results. Retain those images with deviations within the threshold to obtain the first image screening result. The geometric features of this group of images meet the requirements and can continue to be used for subsequent texture feature screening.

[0053] For each set of punched and cut images, compare the calculated texture feature deviation with the texture feature deviation threshold. If the texture feature deviation is greater than the set threshold, it indicates that the texture quality of the image does not meet the requirements, and then remove it from the screening results. Retain those images whose texture features meet the standards to obtain the second image screening result. The texture features of these images meet the standards and can be used for subsequent extraction of the punched and cut path.

[0054] Merge the first image screening result and the second image screening result to ensure that the finally screened images meet the requirements of both geometric features and texture features. Through the integrated screening result, finally obtain the punched and cut images that meet all requirements. These images can be used as the basic data for subsequent extraction and optimization of the cutting path.

[0055] Furthermore, the method for performing path feedback adjustment according to the pre-punched and cut feedback data to obtain the target punched and cut path includes: Extract the processing motion pause data, path intersection and repetition data according to the pre-punched and cut feedback data; based on the processing motion pause data, the path intersection and repetition data, perform marking of processing path defects, and perform path feedback adjustment according to the results of the processing path defect marking.

[0056] Processing motion pause refers to the pause or suspension state of the punching cutting machine when executing the cutting path. The pause may occur for multiple reasons, such as path adjustment, equipment error, too long interval between punching points, too slow cutting speed, etc. Extract the time and position of all pause events from the feedback data, as well as the duration of each pause, the position where it occurs, and the reason for the pause, etc.

[0057] Path intersection occurs when different parts of the cutting path meet, which may lead to incorrect movement or low efficiency of the cutting machine. Path intersection is usually due to unreasonable path planning or ineffective allocation of multiple cutting paths. Extract the data of path intersection, indicate at which positions and under what circumstances path intersection occurs, as well as the specific coordinates of the intersection point. These data are helpful for subsequent optimization of path planning.

[0058] Path repetition means that when the cutting machine is performing cutting, some paths are walked repeatedly, which may be because the path planning is inaccurate or redundant paths are not removed. Extract the information of the repeated paths from the feedback data, point out which paths are walked repeatedly, as well as the length and position of the repeated paths. These data are helpful for identifying and removing unnecessary repeated cutting.

[0059] By analyzing the machining motion pauses, path intersections, and duplicate data, the defective areas in the path are marked, including pause points, intersection points, and duplicate paths. The marked results will help identify the non-optimal parts during the cutting process and provide a basis for subsequent path adjustment. Based on the results of the machining path defect marking, path adjustment is carried out, including removing unnecessary pauses, avoiding path intersections, eliminating duplicate paths, etc. The purpose of the path feedback adjustment is to improve the cutting efficiency, reduce errors and unnecessary motions, and finally generate a more efficient and accurate cutting path.

[0060] In summary, the punching and cutting path generation method combined with graphic optimization provided by the embodiments of the present application has the following technical effects: By obtaining the basic information of the punching and cutting part, including model information and positioning information, and combining the geometric and texture features of the sample image for cutting path planning, the accuracy of the path is ensured, and an initial path that meets the requirements can be generated for different cutting tasks; through graphic optimization, including minimizing the total path length, reducing duplicate paths, and avoiding path intersections, the efficiency of the cutting process can be greatly improved, unnecessary path travel can be reduced, the cutting time can be shortened, and mechanical wear can be reduced; by using the optimized path for pre-punching and cutting and collecting feedback data, problems existing in the path, such as pauses, intersections, or duplicates, can be identified and adjusted to ensure that the finally generated target cutting path is more accurate and efficient; by synchronizing the target path to the cutting machine for path guidance, the punching and cutting machine can be directly guided to operate according to the optimized path, thereby improving the overall production efficiency of metal cutting, reducing mistakes and waste during the metal cutting process, and ensuring the quality and production stability of metal cutting.

[0061] Embodiment 2, based on the same inventive concept as the punching and cutting path generation method combined with graphic optimization in the foregoing embodiment, as Figure 2 shown, the embodiments of the present application provide a punching and cutting path generation system combined with graphic optimization, and the system includes: A basic information acquisition module 10, configured to acquire basic information of a punching and cutting part, where the basic information of the punching and cutting part includes model information of the punching and cutting part and positioning information of the punching and cutting part.

[0062] A graphic semantic segmentation module 20, configured to match a punching and cutting sample image according to the model information of the punching and cutting part, and perform graphic semantic segmentation on the punching and cutting sample image to obtain punching and cutting sample feature information, where the punching and cutting sample feature information includes geometric feature information and texture feature information.

[0063] A path planning module 30, configured to perform punching and cutting path planning based on the positioning information of the punching and cutting part, and combine the punching and cutting sample feature information to generate an initial punching and cutting path.

[0064] The graphic optimization module 40 is used to optimize the initial punching and cutting path to minimize the total path length, minimize the repeated path length, and minimize the number of path intersections, so as to generate an optimized punching and cutting path.

[0065] The path feedback adjustment module 50 is used to perform pre-punching and cutting using the optimized punching and cutting path, collect pre-punching and cutting feedback data, and perform path feedback adjustment according to the pre-punching and cutting feedback data to obtain a target punching and cutting path.

[0066] The path guidance module 60 is used to synchronize the target punching and cutting path to a punching and cutting machine for punching and cutting path guidance.

[0067] Furthermore, the graphic semantic segmentation module 20 includes: A semantic segmentation unit for performing graphic semantic segmentation on the punching and cutting sample image to obtain a punching and cutting identification area; a geometric feature extraction unit for extracting geometric features from the punching and cutting identification area to obtain the geometric feature information, where the geometric feature information includes aperture size features, aperture position features, cutting profile shape features, and junction contact angle features; a texture feature extraction unit for extracting texture features from the punching and cutting identification area to obtain the texture feature information, where the texture feature information includes surface roughness features, crack size features, and surface defect quantity features.

[0068] Furthermore, the graphic semantic segmentation module 20 includes: A historical image acquisition unit for acquiring a set of historical punching and cutting images based on punching and cutting detection records; a shape identification unit for identifying and shape-identifying the punching and cutting areas in the set of historical punching and cutting images to obtain a set of sample punching and cutting identification results; a training unit for constructing an encoder and a decoder based on graphic semantic segmentation and training using the set of historical punching and cutting images and the set of sample punching and cutting identification results to obtain a punching and cutting area identifier; an area identification unit for using the punching and cutting area identifier to identify the punching and cutting area of the punching and cutting sample image to obtain the punching and cutting identification area.

[0069] Furthermore, the path planning module 30 includes: The point extraction unit is used to extract a set of punching points, a set of starting points of the cutting path, and a set of ending points of the cutting path based on the positioning information of the punching and cutting parts; the first punching point extraction unit is used to randomly extract the first punching point, the first starting point of the cutting path, and the corresponding first ending point of the cutting path; the cutting record data extraction unit is used to collect the first punching and cutting record data based on the first punching point, the first starting point of the cutting path, and the first ending point of the cutting path, wherein the first punching and cutting record data includes multiple sets of punching and cutting paths and multiple sets of punching and cutting images; the screening unit is used to screen the multiple sets of punching and cutting images based on the geometric feature information and texture feature information of the punching and cutting sample feature information to obtain the punching and cutting image screening result; the initial punching and cutting path acquisition unit is used to extract the punching and cutting path corresponding to the punching and cutting image screening result as the initial punching and cutting path.

[0070] Furthermore, the path planning module 30 includes: The control geometric feature information extraction unit is used to extract multiple sets of control geometric feature information and multiple sets of control texture feature information of the multiple sets of punching and cutting images; the geometric feature deviation calculation unit is used to calculate multiple sets of geometric feature deviations between the multiple sets of control geometric feature information and the geometric feature information; the texture feature deviation calculation unit is used to calculate multiple sets of texture feature deviations between the multiple sets of control texture feature information and the texture feature information; the image screening unit is used to perform image screening on the multiple sets of punching and cutting images based on the multiple sets of geometric feature deviations and the multiple sets of texture feature deviations to obtain the punching and cutting image screening result.

[0071] Furthermore, the path planning module 30 includes: The threshold acquisition unit is used to acquire the geometric feature deviation threshold and the texture feature deviation threshold; the first elimination unit is used to eliminate the punching and cutting images in the multiple sets of geometric feature deviations that do not meet the geometric feature deviation threshold to obtain the first image screening result; the second elimination unit is used to eliminate the punching and cutting images in the multiple sets of texture feature deviations that do not meet the texture feature deviation threshold to obtain the second image screening result; the screening result acquisition unit is used to integrate the first image screening result and the second image screening result to obtain the punching and cutting image screening result.

[0072] Furthermore, the path feedback adjustment module 50 includes: The data extraction unit is used to extract the processing motion pause data, path intersection and repetition data according to the pre-punching and cutting feedback data; the feedback adjustment unit is used to perform processing path defect marking based on the processing motion pause data and the path intersection and repetition data, and perform path feedback adjustment according to the processing path defect marking result.

[0073] Through the foregoing detailed description of the method for generating a punching cutting path optimized in combination with a graph, those skilled in the art can clearly know the system for generating a punching cutting path optimized in combination with a graph in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description in the method part.

[0074] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating a punching and cutting path optimized in combination with a graph, characterized in that The method includes: Obtaining basic information of the punching and cutting part, where the basic information of the punching and cutting part includes the model information of the punching and cutting part and the positioning information of the punching and cutting part; According to the model information of the punching and cutting part, matching the punching and cutting sample image, and performing graphic semantic segmentation on the punching and cutting sample image to obtain punching and cutting sample feature information, where the punching and cutting sample feature information includes geometric feature information and texture feature information; Based on the positioning information of the punching and cutting part, combining the punching and cutting sample feature information to perform punching and cutting path planning, and generating an initial punching and cutting path; Minimizing the total path length, minimizing the repeated path length, and minimizing the number of path intersections to optimize the initial punching and cutting path graphically to generate an optimized punching and cutting path; Using the optimized punching and cutting path to perform pre-punching and cutting, and collecting pre-punching and cutting feedback data, and performing path feedback adjustment according to the pre-punching and cutting feedback data to obtain a target punching and cutting path; Synchronizing the target punching and cutting path to a punching and cutting machine for punching and cutting path guidance.

2. The punching cutting path generation method combined with graphic optimization according to claim 1, wherein The method of performing graphic semantic segmentation on the punching and cutting sample image to obtain punching and cutting sample feature information includes: Performing graphic semantic segmentation on the punching and cutting sample image to obtain a punching and cutting identification area; Performing geometric feature extraction on the punching and cutting identification area to obtain the geometric feature information, where the geometric feature information includes aperture size feature, aperture position feature, cutting profile shape feature, and junction contact angle feature; Performing texture feature extraction on the punching and cutting identification area to obtain the texture feature information, where the texture feature information includes surface roughness feature, crack size feature, and surface defect quantity feature.

3. The punching cutting path generation method optimized in combination with graphics according to claim 2, wherein, The method of performing graphic semantic segmentation on the punching and cutting sample image to obtain a punching and cutting identification area includes: Based on the punching and cutting detection records, obtaining a historical punching and cutting image set; Identifying and shape-marking the punching and cutting areas in the historical punching and cutting image set to obtain a sample punching and cutting recognition result set; Based on graphic semantic segmentation, constructing an encoder and a decoder, and training using the historical punching and cutting image set and the sample punching and cutting recognition result set to obtain a punching and cutting area recognizer; Using the punching and cutting area recognizer to perform punching and cutting area recognition on the punching and cutting sample image to obtain the punching and cutting identification area.

4. The punching cutting path generation method combined with graphic optimization according to claim 1, wherein The method of performing punching and cutting path planning based on the positioning information of the punching and cutting part and combining the punching and cutting sample feature information to generate an initial punching and cutting path includes: Based on the positioning information of the punching and cutting part, extracting a set of punching points, a set of cutting path starting points, and a set of cutting path ending points; Randomly extracting a first punching point, a first cutting path starting point, and the corresponding first cutting path ending point; Based on the first punching point, the first cutting path starting point, and the first cutting path ending point, collecting first punching and cutting record data, where the first punching and cutting record data includes multiple sets of punching and cutting paths and multiple sets of punching and cutting images; Based on the geometric feature information and texture feature information of the punched cutting sample feature information, screen the multiple groups of punched cutting images to obtain the punched cutting image screening result; Extract the punched cutting path corresponding to the punched cutting image screening result as the initial punched cutting path.

5. The punching cutting path generation method combined with graphic optimization according to claim 4, wherein The method for screening the multiple groups of punched cutting images based on the geometric feature information and texture feature information of the punched cutting sample feature information to obtain the punched cutting image screening result includes: Extract multiple groups of control geometric feature information and multiple groups of control texture feature information of the multiple groups of punched cutting images; Calculate multiple groups of geometric feature deviations between the multiple groups of control geometric feature information and the geometric feature information; Calculate multiple groups of texture feature deviations between the multiple groups of control texture feature information and the texture feature information; Based on the multiple groups of geometric feature deviations and the multiple groups of texture feature deviations, perform image screening on the multiple groups of punched cutting images to obtain the punched cutting image screening result.

6. The punching cutting path generation method combined with graphic optimization according to claim 5, characterized in that The method for performing image screening on the multiple groups of punched cutting images based on the multiple groups of geometric feature deviations and the multiple groups of texture feature deviations to obtain the punched cutting image screening result includes: Obtain a geometric feature deviation threshold and a texture feature deviation threshold; Eliminate the punched cutting images in the multiple groups of geometric feature deviations that do not meet the geometric feature deviation threshold to obtain a first image screening result; Eliminate the punched cutting images in the multiple groups of texture feature deviations that do not meet the texture feature deviation threshold to obtain a second image screening result; Integrate the first image screening result and the second image screening result to obtain the punched cutting image screening result.

7. The punching cutting path generation method optimized in combination with a graph according to claim 1, characterized in that The method for performing path feedback adjustment according to the pre-punched cutting feedback data to obtain the target punched cutting path includes: Extract processing motion pause data, path intersection and repetition data according to the pre-punched cutting feedback data; Based on the processing motion pause data, the path intersection and repetition data, perform processing path defect marking, and perform path feedback adjustment according to the processing path defect marking result.

8. The punching and cutting path generation system optimized in combination with graphics, characterized in that, A system for implementing the punched cutting path generation method combined with graphic optimization according to any one of claims 1-7, the system includes: A basic information acquisition module for acquiring basic information of a punched cutting part, where the basic information of the punched cutting part includes punched cutting part model information and punched cutting part positioning information; A graphic semantic segmentation module for matching a punched cutting sample image according to the punched cutting part model information, and performing graphic semantic segmentation on the punched cutting sample image to obtain punched cutting sample feature information, where the punched cutting sample feature information includes geometric feature information and texture feature information; A path planning module for performing punched cutting path planning based on the punched cutting part positioning information and combining the punched cutting sample feature information to generate an initial punched cutting path; A graphic optimization module for performing graphic optimization on the initial punched cutting path to minimize the total path length, minimize the repeated path length, and minimize the number of path intersections, and generate an optimized punched cutting path; A path feedback adjustment module, which is used to perform pre-punching cutting using the optimized punching cutting path, collect pre-punching cutting feedback data, and perform path feedback adjustment according to the pre-punching cutting feedback data to obtain a target punching cutting path; A path guidance module, which is used to synchronize the target punching cutting path to a punching cutting machine for punching cutting path guidance.

Images