Plasma cutting system and cutting management method

Abstract

The application belongs to the technical field of image recognition, and specifically provides a plasma cutting system and a cutting management method. The system each comprises a robot cutting device, and a cutting gun is arranged on the robot cutting device in cooperation. The system further comprises a mounting frame, the robot cutting device is fixedly arranged on the lower surface of the mounting frame, a visual sensor module is fixedly connected to one side of the robot cutting device, the robot cutting device is connected with the visual sensor module, the visual sensor module captures image information of a cutting path of a cutting target in real time, identifies an obstacle of a cutting track, and inputs an identification result to the robot cutting device, the robot cutting device controls a cutting action of itself according to the identification result, the visual sensor module further detects a workpiece after cutting, and outputs prompt information when the workpiece is unqualified. The method realizes rapid detection of an obstacle in a cutting track and a size deviation after cutting is completed.

Description

Technical Field

[0001] This invention relates to the field of image recognition technology, specifically to a plasma cutting system and a cutting control method. Background Technology

[0002] With the development of lightweighting in automobiles, integrated die casting of the vehicle body will lead a new revolution in vehicle body manufacturing. In particular, the integrated die casting of the front and rear floor panels is being gradually implemented in the automotive manufacturing industry. For the removal of the sprue in integrated die castings of the front and rear floor panels, domestic die casting manufacturers typically use robotic plasma cutting to achieve complete sprue removal.

[0003] In actual production, there are two main problems with robotic plasma cutting of die-cast parts. First, if obstacles are present in the robotic plasma cutting path and are not identified and cleared in advance, the cutting torch may be damaged if the cutting shank hits the obstacle. Second, after the robotic plasma cuts the shank, the cut workpiece may have dimensional deviations. Human inspection may result in missed or false detections. If these deviations are not detected in a timely and effective manner, it often leads to batch scrapping of workpieces with out-of-tolerance dimensions. Summary of the Invention

[0004] To address the problems associated with robotic plasma cutting, this invention provides a plasma cutting system and a cutting control method.

[0005] In a first aspect, the present invention provides a plasma cutting system, including a robotic cutting device with a cutting gun mounted on it; it also includes a mounting frame, the robotic cutting device being fixedly mounted on the lower surface of the mounting frame, a vision sensor module being mounted on one side of the robotic cutting device and fixedly connected to the mounting frame, and the robotic cutting device being communicatively connected to the vision sensor module.

[0006] The vision sensor module is used to capture image information of the cutting path of the cutting target in real time, identify obstacles in the cutting trajectory, and input the identification results to the robot cutting equipment. The robot cutting equipment controls its own cutting action according to the identification results. The vision sensor module also detects the cut workpiece and outputs prompt information when the workpiece is found to be unqualified.

[0007] The vision sensor module continuously monitors the surrounding environment during the cutting process, promptly detecting and identifying obstacles that may affect cutting safety. This effectively prevents collisions between the cutting torch and obstacles, reducing the risk of equipment damage and accidents, and improving workplace safety. The system extends beyond obstacle avoidance during cutting; it also performs quality inspection on the workpiece after cutting. If quality issues or non-compliance with specifications are detected, the vision sensor module immediately outputs alerts, allowing operators to take timely corrective or reprocessing measures, ensuring consistent and stable product quality.

[0008] The plasma cutting system provided by this invention integrates a vision sensor module, which realizes intelligent, automated and efficient cutting operations, significantly improves cutting accuracy, efficiency and safety, reduces production costs and labor intensity, and injects new vitality into the development of modern manufacturing.

[0009] As a preferred embodiment of the present invention, the robotic cutting device is equipped with a PLC controller, which is connected to the vision sensor module via the PROFINET communication protocol. The vision sensor module inputs the detection results to the PLC controller.

[0010] As a preferred embodiment of the technical solution of the present invention, the visual sensor module includes an image acquisition unit, a first processing unit, and a second processing unit;

[0011] The image acquisition unit captures image information of the cutting path of the cutting target in real time and labels the captured image information of the cutting path as an identification image and transmits it to the first processing unit; it is also used to acquire image information of the workpiece after cutting, label the image information of the workpiece after cutting as a size verification image and transmit it to the second processing unit.

[0012] The first processing unit is used to compare the image information labeled as the recognition image with the pre-stored images of obstacles that have been labeled, perform obstacle recognition and positioning processing on the image information with obstacles found in the comparison, and input the processing result into the PLC controller.

[0013] The second processing unit is used to compare the image information labeled as a size verification image with the stored standard size workpiece image. If the comparison is consistent, it outputs a size qualified reminder message; if the comparison is inconsistent, it outputs a size unqualified reminder message.

[0014] By subdividing the vision sensor module into an image acquisition unit, a first processing unit, and a second processing unit, a refined image information processing workflow is achieved. The image acquisition unit is responsible for capturing the cutting path and workpiece image in real time and accurately, providing a high-quality data source for subsequent processing. The first and second processing units focus on obstacle recognition and size verification, respectively, significantly improving the efficiency and accuracy of image processing through parallel processing.

[0015] As a preferred embodiment of the technical solution of the present invention, the first processing unit performs contrast enhancement processing on the acquired image, adopts a gray-level difference method based on edge image pixels to extract the image contour, clears the pixels wrapped by the image edge while retaining the image contour, and realizes the extraction of obstacle edge features in the image; uses Gabor wavelet filter to extract obstacle features, completes the positioning of obstacle shape and position, and inputs the processing result to PLC controller.

[0016] The first processing unit first enhances the contrast of the acquired image, effectively improving image quality and making the boundary between obstacles and the background clearer. Then, it extracts the image contour using a gray-level difference method based on edge image pixels, clearing pixels inside the image edges and retaining only the contour information. This processing method greatly highlights the edge features of obstacles, providing a more accurate data foundation for subsequent feature extraction and recognition. Further obstacle feature extraction is performed using a Gabor wavelet filter, accurately capturing the shape and position information of obstacles, thus significantly improving the accuracy of obstacle recognition. The gray-level difference method based on edge image pixels and the Gabor wavelet filter have high computational efficiency in image processing. The first processing unit uses these methods to process the image, enabling obstacle feature extraction and recognition to be completed in a short time, ensuring real-time obstacle recognition. This is particularly important for cutting operations that require rapid response and dynamic adjustment, preventing cutting deviations or safety accidents caused by recognition delays.

[0017] As a preferred embodiment of the technical solution of the present invention, the first processing unit performs digital signal transformation by filtering the texture of the acquired image through a Gabor wavelet filter; decomposes the transformed digital signal into a set of wavelet functions, and improves the resolution of the image in the spatiotemporal domain and frequency domain by using a transformation factor and a dilation factor; during filtering, the output of the Gabor wavelet filter is adjusted by averaging the virtual and real filter masks of all convolutional pixels in the image.

[0018] As a preferred embodiment of the technical solution of the present invention, the second processing unit determines the image edge based on the grayscale changes on both sides of the image edge, connects the edges, and determines the image size of the cut workpiece by measuring the distance based on the position coordinates of the pixel points, and compares it with the image size of the standard workpiece; wherein, the standard workpiece image and the cut workpiece image have the same number of pixels.

[0019] The second processing unit accurately identifies image edges and determines their positions based on grayscale changes on both sides of the edges. This method effectively reduces the impact of edge blurring or noise on the measurement results. Subsequently, by connecting the edges and using the pixel coordinates for distance measurement, accurate measurement of the dimensions of the cut workpiece image is achieved. Since the standard workpiece image and the cut workpiece image have consistent pixel counts, this pixel-level comparison method further improves the accuracy of the dimension measurement, ensuring the accuracy and reliability of the measurement results.

[0020] Through precise dimensional measurement and comparison, the second processing unit can promptly detect deviations between the cut workpiece and the standard dimensions. Once a defective product is detected, the system will output a prompt message to alert the operator for corrective action. This real-time quality control mechanism helps ensure consistent product quality, reduce defect rates, and improve customer satisfaction.

[0021] As a preferred embodiment of the technical solution of the present invention, the system also includes a host computer, which is connected to the robot cutting equipment and vision sensor module via a wireless network to realize the functions of remotely monitoring the cutting process, adjusting cutting parameters, and receiving real-time feedback data.

[0022] By introducing a host computer and establishing a wireless network connection with the robotic cutting equipment and vision sensor module, operators can monitor the cutting process in real time from a location far from the production site. This remote monitoring capability makes production management more convenient, no longer limited by geographical location, and improves the flexibility and efficiency of production management.

[0023] Through remote monitoring and real-time feedback mechanisms, the host computer can promptly detect safety hazards and abnormalities in the production process, such as equipment malfunctions and operational errors. Once a problem is detected, operators can take immediate action to address it, thereby preventing accidents and improving production safety.

[0024] As a preferred embodiment of the present invention, a lens protective cover is provided at the lens of the image acquisition unit of the vision sensor module. This prevents cutting dust from damaging the lens of the vision sensor module and causing misjudgments.

[0025] Secondly, the present invention provides a cutting control method based on the plasma cutting system described in the first aspect, comprising the following steps:

[0026] Signal interaction between the robot cutting equipment and the vision sensor module is established using the PROFINET communication protocol;

[0027] During the cutting process, the vision sensor module captures image information of the cutting path of the cutting target in real time, identifies obstacles in the cutting trajectory, and inputs the identification results to the robot cutting equipment;

[0028] The robotic cutting device controls its own cutting action based on the recognition result;

[0029] After cutting, the vision sensor module inspects the cut workpiece and outputs a prompt message when the workpiece is found to be unqualified.

[0030] Through the precise detection of the vision sensor module and the intelligent processing of the PLC controller, timely identification of cutting size deviations can be achieved, ensuring the quality and consistency of the cut workpieces, and further improving production efficiency and product qualification rate.

[0031] As a preferred embodiment of the technical solution of the present invention, during the cutting process, the visual sensor module captures image information of the cutting path of the cutting target in real time, identifies obstacles in the cutting trajectory, and inputs the identification results to the robotic cutting equipment, including the following steps:

[0032] The image acquisition unit captures the image information of the cutting path of the cutting target in real time and annotates the captured image information of the cutting path as a recognition image and transmits it to the first processing unit;

[0033] The first processing unit compares the image information labeled as the recognition image with the pre-stored images of obstacles that have been labeled, performs obstacle identification and location processing on the image information with obstacles found in the comparison, and inputs the processing result into the PLC controller.

[0034] As can be seen from the above technical solutions, this invention has the following advantages: By integrating a vision sensor module, the system can capture precise image information of the cutting target in real time and identify obstacles along the cutting path. This function not only reduces the need for manual intervention but also greatly improves cutting accuracy and avoids cutting deviations caused by human error. Simultaneously, automatic obstacle identification significantly improves the efficiency of the cutting operation and shortens the production cycle. Through the application of this invention in the field of robotic plasma cutting of automotive die-casting parts, efficient, rapid, accurate, and stable detection of obstacles within the cutting trajectory and dimensional deviations after cutting can be achieved, making it convenient, safe, and reliable to use. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a schematic block diagram of the system provided in the embodiments of the present invention.

[0037] Figure 2 This is a schematic flowchart of a method according to an embodiment of the present invention. Detailed Implementation

[0038] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0039] like Figure 1 As shown, this embodiment of the invention provides a plasma cutting system, including a robotic cutting device with a cutting gun mounted on it; it also includes a mounting frame, with the robotic cutting device fixedly mounted on the lower surface of the mounting frame, a vision sensor module mounted on one side of the robotic cutting device, and the vision sensor module being fixedly connected to the mounting frame, and the robotic cutting device being communicatively connected to the vision sensor module;

[0040] The vision sensor module is used to capture image information of the cutting path of the cutting target in real time, identify obstacles in the cutting trajectory, and input the identification results to the robot cutting equipment. The robot cutting equipment controls its own cutting action according to the identification results. The vision sensor module also detects the cut workpiece and outputs prompt information when the workpiece is found to be unqualified.

[0041] By integrating a vision sensor module, this system can capture precise image information of the cutting target in real time. During the cutting process, the vision sensor module continuously monitors the surrounding environment, promptly detecting and identifying obstacles that may affect cutting safety. This effectively avoids collisions between the cutting torch and obstacles, reducing the risk of equipment damage and accidents, and improving on-site safety. Combining the vision sensor module with robotic cutting equipment achieves an automated and intelligent upgrade to the cutting operation. This not only reduces the labor intensity of workers but also makes the cutting process more flexible and efficient through the application of intelligent algorithms, providing strong support for the development of industrial automation and intelligent manufacturing.

[0042] The vision sensor module employs a 3D vision sensor to capture and process the three-dimensional spatial information of the cutting target in real time. Using a 3D vision sensor as an auxiliary tool significantly improves the accuracy and efficiency of plasma cutting. By acquiring the three-dimensional data of the cutting target in real time, the system can automatically identify obstacles, reduce material waste, and improve overall processing quality.

[0043] In some embodiments, the robotic cutting device is equipped with a PLC controller, which communicates with the vision sensor module via the PROFINET communication protocol. The vision sensor module inputs the detection results to the PLC controller.

[0044] The vision sensor module is installed directly above the workpiece, approximately 1.8 meters from its surface. The mounting plate features an adjustable threaded design with elongated holes. A lens protector is added to the lens to prevent cutting dust from damaging it and causing misjudgments. A PROFINET communication mechanism is designed between the PLC controller and the vision sensor module of the robotic plasma cutting equipment, establishing signal interaction between them and ensuring that the vision sensor module's output is effectively fed back to the PLC controller for further processing. The foreign object detection function of the vision sensor module can be used to detect obstacles along the cutting trajectory of die-cast parts. Before robotic plasma cutting, after the die-casting machine produces qualified die-cast parts with free of oil and water stains, the upper computer feeds back the information to the PLC controller of the robotic plasma cutting equipment. The robotic plasma cutting equipment then uses the standard qualified workpiece produced by the die-casting machine as a calibration object to collect and store its image information. The image of the standard workpiece and the subsequently acquired image must have the same pixel count for consistent pixel size during later dimensional inspection. In this embodiment of the invention, when inspecting the cut workpiece, in addition to dimensional detection, image recognition methods can be used to detect oil stains, water stains, foreign objects, etc., on the workpiece surface. Oil stains, water stains, and foreign objects are identified as defective parts. Simultaneously, the vision sensor module identifies defective parts and feeds them back to the PLC controller for processing. Upon receiving an NG signal from the vision sensor module, the PLC controller pauses robot operation and sends a feedback to the host computer. Upon receiving the alarm information, the host computer manually removes the obstacles before resuming cutting, effectively avoiding collisions between the cutting torch and obstacles. The PLC controller of the robotic plasma equipment can set a detection threshold range for the calibrated dimensions of the calibration object. All subsequent die-cast parts cut are compared and analyzed with the calibration object. If the dimensional deviation of the cut die-cast part exceeds the detection threshold range, the vision sensor module identifies and sends a defective signal back to the robot. Upon receiving the defective signal, the PLC controller of the robotic plasma equipment sends a feedback to the host computer, which determines that the cut workpiece is defective.

[0045] In some embodiments, the visual sensor module includes an image acquisition unit, a first processing unit, and a second processing unit;

[0046] The image acquisition unit captures image information of the cutting path of the cutting target in real time and labels the captured image information of the cutting path as an identification image and transmits it to the first processing unit; it is also used to acquire image information of the workpiece after cutting, label the image information of the workpiece after cutting as a size verification image and transmit it to the second processing unit.

[0047] The first processing unit is used to compare the image information labeled as the recognition image with the pre-stored images of obstacles that have been labeled, perform obstacle recognition and positioning processing on the image information with obstacles found in the comparison, and input the processing result into the PLC controller.

[0048] The second processing unit is used to compare the image information labeled as a size verification image with the stored standard size workpiece image. If the comparison is consistent, it outputs a size qualified reminder message; if the comparison is inconsistent, it outputs a size unqualified reminder message.

[0049] The image acquisition unit is responsible for capturing the cutting path and workpiece images in real time and accurately, providing a high-quality data source for subsequent processing. The first and second processing units focus on obstacle recognition and size verification, respectively, significantly improving the efficiency and accuracy of image processing through parallel processing. The first processing unit compares the captured cutting path image with pre-stored obstacle images, quickly identifying and locating potential obstacles. This real-time and accurate obstacle recognition capability provides timely and effective obstacle avoidance guidance for the robotic cutting equipment, ensuring the safety and continuity of the cutting operation. The second processing unit automatically verifies whether the workpiece dimensions meet requirements by comparing the image of the cut workpiece with images of standard-sized workpieces. This function not only reduces the burden of manual inspection but also improves the accuracy and efficiency of size verification, ensuring the stability and consistency of product quality. The refined design of the vision sensor module enables the entire cutting system to achieve a higher level of intelligence. By integrating advanced image processing technology and intelligent algorithms, the system can autonomously complete multiple stages such as obstacle recognition, path planning, cutting operations, and size verification, realizing intelligent control throughout the entire process from raw materials to finished products. Because the vision sensor module adopts a modular design, it can be flexibly configured and expanded according to actual needs. For example, the parameters of the image acquisition unit can be adjusted according to different cutting targets and workpiece requirements, or new processing units can be added to expand the system's functionality. This flexibility and scalability enable the system to adapt to diverse production needs and to be continuously upgraded and optimized as technology advances.

[0050] In some embodiments, the first processing unit performs contrast enhancement processing on the acquired image, uses a gray-level difference method based on edge image pixels to extract the image contour, clears the pixels wrapped by the image edge while retaining the image contour, and realizes the extraction of obstacle edge features in the image; uses Gabor wavelet filter to extract obstacle features, completes the localization of obstacle shape and position, and inputs the processing result to PLC controller.

[0051] The first processing unit first enhances the contrast of the acquired image. Then, it extracts the image contour using a gray-level difference method based on edge image pixels, clearing pixels inside the image edges and retaining only the contour information. Gabor wavelet filters are then used for further obstacle feature extraction, accurately capturing the shape and position information of obstacles, thus significantly improving the accuracy of obstacle recognition. In practical applications, the cutting environment may be affected by various factors such as changes in lighting and shadow interference, leading to a decrease in image quality and affecting the obstacle recognition effect. By introducing contrast enhancement and edge contour extraction techniques, the first processing unit can effectively cope with these complex environments, ensuring accurate obstacle recognition under various conditions. This greatly enhances the system's adaptability to complex environments and improves its stability and reliability. By accurately identifying the shape and position of obstacles, the first processing unit provides reliable obstacle avoidance guidance for the robotic cutting equipment. The robotic cutting equipment can adjust its cutting path based on this information to avoid collisions with obstacles, thereby optimizing the overall efficiency and quality of the cutting operation. This path planning strategy based on real-time obstacle recognition makes the cutting operation more flexible and efficient.

[0052] By promptly identifying and avoiding obstacles, robotic cutting equipment can prevent unnecessary collisions and friction with its surroundings, thereby reducing wear and tear and the failure rate. This helps extend the equipment's lifespan and reduce repair and replacement costs due to equipment damage. Simultaneously, reducing noise and vibration caused by collisions also helps improve the working environment, increasing worker comfort and work efficiency.

[0053] In some embodiments, the first processing unit filters the texture of the acquired image using a Gabor wavelet filter to perform digital signal transformation, thereby improving the accuracy of obstacle recognition. The Gabor wavelet filter digital signal transformation method is as follows:

[0054]

[0055] In the formula, Represents digital signals; These represent the conversion factor and the expansion factor, respectively. This represents the basic wavelet.

[0056] The transformed digital signal is decomposed into a set of wavelet functions, and the image resolution is improved in the spatiotemporal and frequency domains by using transformation and dilation factors; the two-dimensional function of the Gabor wavelet filter is:

[0057]

[0058] In the formula, Indicates Gaussian distribution in Shaft and Axial variants, Represents the frequency of the sine curve. This indicates the direction of the sine curve.

[0059] During filtering, the output of the Gabor wavelet filter is adjusted by averaging the virtual and real filter masks of all convolutional pixels in the image. The calculation formula is as follows:

[0060]

[0061] In the formula, Indicates the magnitude of filter output adjustment; This represents the result of convolving the image region with the Gabor wavelet filter mask; It is the result of convolving the image region using Gabor wavelet filter virtual masking code.

[0062] In some embodiments, the second processing unit determines the image edge based on the grayscale changes on both sides of the image edge, connects the edges, and determines the image size of the cut workpiece by measuring the distance based on the position coordinates of the pixels, and compares it with the image size of the standard workpiece; wherein the standard workpiece image and the cut workpiece image have the same number of pixels.

[0063] Traditional dimensional verification often relies on manual measurement or the use of complex measuring equipment, which is not only time-consuming and labor-intensive but also prone to human error. The second processing unit, however, automates and digitizes dimensional verification, greatly simplifying the process. Operators simply input the image of the cut workpiece into the system, which automatically performs dimensional measurement and comparison, significantly improving work efficiency.

[0064] The second processing unit integrates advanced image processing technologies and algorithms, enabling it to autonomously perform tasks such as image edge recognition, size measurement, and comparison. This intelligent processing method not only reduces the workload of manual labor but also improves the system's automation and intelligence levels, making the cutting system more adaptable to the demands of modern manufacturing for efficiency, precision, and intelligence.

[0065] The second processing unit records detailed image data and measurement results during the measurement and comparison process. This data can be stored and traced as production records. When it is necessary to investigate or analyze product quality issues, this data can be easily retrieved for review and verification, thereby improving the traceability and transparency of the production process.

[0066] In some embodiments, the system also includes a host computer that connects to the robotic cutting equipment and vision sensor module via a wireless network to remotely monitor the cutting process, adjust cutting parameters, and receive real-time feedback data.

[0067] By introducing a host computer and establishing a wireless network connection with the robotic cutting equipment and vision sensor module, the host computer allows operators to quickly adjust cutting parameters such as cutting speed, power, and angle according to actual needs while providing remote monitoring. This instant adjustment capability makes the cutting process more flexible, enabling it to quickly adapt to workpieces of different materials, shapes, and sizes, thereby improving production efficiency and product quality.

[0068] The host computer can receive real-time feedback data from the robotic cutting equipment and vision sensor modules, including cutting progress, obstacle recognition results, and workpiece size verification information. This real-time data provides operators with comprehensive production information, helping them to promptly identify, analyze, and resolve problems, thereby ensuring the smooth operation of the production process.

[0069] The host computer can receive real-time feedback data and record and save this data. These production records not only help with subsequent production analysis and optimization, but also provide strong support for tracing product quality issues. When it is necessary to trace the quality of a batch of products, relevant data can be easily retrieved for review and verification.

[0070] It should be noted that, in this embodiment of the invention, a lens protective cover is provided at the lens of the image acquisition unit of the vision sensor module. This is to prevent cutting dust from damaging the lens of the vision sensor module and causing misjudgment.

[0071] In some embodiments, the mounting frame is further provided with an automatic calibration device for periodically or as needed to calibrate the position of the vision sensor module, ensuring the continuous stability of cutting accuracy.

[0072] The addition of an automatic calibration device can automatically compensate for the positional shift of the vision sensor module caused by long-term use or environmental changes, ensuring that it maintains high-precision measurement and positioning capabilities throughout the entire working cycle, further improving the reliability and stability of cutting operations.

[0073] It should be further noted here that the automatic calibration device includes a displacement sensor, a fine-tuning mechanism, and a control unit;

[0074] A displacement sensor is installed between the mounting frame and the vision sensor module to accurately measure minute positional changes in the vision sensor module during actual operation, including horizontal offset and vertical height changes. The fine-tuning mechanism, directly connected to the vision sensor module, consists of a precision motor, transmission mechanism, and guiding elements. It automatically adjusts the position of the vision sensor module based on feedback signals from the displacement sensor, achieving precise calibration. The fine-tuning mechanism should have sufficient adjustment range and accuracy to meet calibration requirements in different scenarios. The control unit, as the core of the automatic calibration device, is responsible for receiving data input from the displacement sensor, processing and analyzing it, and then issuing control commands to the fine-tuning mechanism. The control unit should also have self-diagnostic capabilities, enabling it to monitor the operating status of the calibration device in real time and promptly issue alarms when abnormalities occur.

[0075] The control unit has a built-in calibration program, including preset calibration procedures and parameter settings. Users can select a suitable calibration mode (such as periodic calibration, on-demand calibration, etc.) according to their actual needs, and set the calibration cycle and accuracy requirements. Specific workflow:

[0076] The user initiates the automatic calibration program via the host computer or control panel. Upon receiving the command, the control unit activates the displacement sensor and fine-tuning mechanism, placing them in standby mode. The displacement sensor begins operation, acquiring the current position information of the vision sensor module in real time and transmitting the data to the control unit. The control unit processes and analyzes the received data to determine if there is any positional shift in the vision sensor module. If a positional shift is detected, the control unit sends a control command to the fine-tuning mechanism according to preset calibration parameters and procedures. Upon receiving the command, the fine-tuning mechanism drives the vision sensor module to adjust to the correct position until the predetermined calibration accuracy is achieved. After calibration, the control unit feeds back the calibration results to the host computer or control panel and automatically saves the calibration record and parameter settings. Simultaneously, the control unit performs self-diagnosis on the calibration device to ensure that all components are functioning correctly.

[0077] An automatic calibration device ensures that the vision sensor module maintains high-precision measurement and positioning capabilities throughout the entire work cycle, thereby improving the accuracy and reliability of cutting operations. The automatic calibration device automates the calibration process without manual intervention, reducing labor intensity and human error. Regular or on-demand automatic calibration can promptly detect and correct positional deviations, avoiding cutting errors and downtime caused by positional shifts, thus improving production efficiency and equipment utilization. The automatic calibration device can adapt to changes in different working environments and cutting requirements, achieving more flexible and accurate calibration results by adjusting calibration parameters and procedures.

[0078] like Figure 2 As shown, an embodiment of the present invention provides a cutting control method based on the plasma cutting system described in the first aspect, comprising the following steps:

[0079] Step 1: Establish signal interaction between the robot cutting equipment and the vision sensor module via the PROFINET communication protocol;

[0080] Step 2: During the cutting process, the vision sensor module captures image information of the cutting path of the cutting target in real time, identifies obstacles in the cutting trajectory, and inputs the identification results to the robot cutting device; the robot cutting device controls its own cutting action according to the identification results;

[0081] Step 3: After cutting, the vision sensor module detects the cut workpiece and outputs a prompt message when the workpiece is found to be unqualified.

[0082] Through the precise detection of the vision sensor module and the intelligent processing of the PLC controller, timely identification of cutting size deviations can be achieved, ensuring the quality and consistency of the cut workpieces, and further improving production efficiency and product qualification rate.

[0083] In some embodiments, during the cutting process, the visual sensor module captures image information of the cutting path of the cutting target in real time, and the steps of identifying obstacles in the cutting trajectory and inputting the identification results to the robotic cutting equipment include:

[0084] The image acquisition unit captures the image information of the cutting path of the cutting target in real time and annotates the captured image information of the cutting path as a recognition image and transmits it to the first processing unit;

[0085] The first processing unit compares the image information labeled as the recognition image with the pre-stored images of obstacles that have been labeled, performs obstacle identification and location processing on the image information with obstacles found in the comparison, and inputs the processing result into the PLC controller.

[0086] Through the precise detection of the vision sensor module and the intelligent processing of the PLC controller, timely identification of cutting size deviations can be achieved, ensuring the quality and consistency of the cut workpieces, and further improving production efficiency and product qualification rate.

[0087] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0088] Although the present invention has been described in detail with reference to the accompanying drawings and preferred embodiments, the present invention is not limited thereto. Various equivalent modifications or substitutions can be made to the embodiments of the present invention by those skilled in the art without departing from the spirit and essence of the invention, and such modifications or substitutions should all be within the scope of the present invention. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should also be covered within the protection scope of the present invention.

Claims

1. A plasma cutting system, comprising a robotic cutting device, wherein a cutting gun is configured on the robotic cutting device; characterized in that, It also includes a mounting frame, the robot cutting device is fixedly mounted on the lower surface of the mounting frame, a vision sensor module is provided on one side of the robot cutting device, and the vision sensor module is fixedly connected to the mounting frame, and the robot cutting device is communicatively connected to the vision sensor module; The vision sensor module is used to capture image information of the cutting path of the cutting target in real time, identify obstacles in the cutting trajectory, and input the identification results to the robot cutting device. The robot cutting device controls its own cutting action according to the identification results. The vision sensor module also inspects the cut workpiece and outputs a prompt message when the workpiece is found to be defective. The robotic cutting equipment is equipped with a PLC controller, which communicates with the vision sensor module via the PROFINET communication protocol. The vision sensor module inputs the detection results to the PLC controller. The visual sensor module includes an image acquisition unit, a first processing unit, and a second processing unit; The image acquisition unit captures image information of the cutting path of the cutting target in real time and labels the captured image information of the cutting path as an identification image and transmits it to the first processing unit; it is also used to acquire image information of the workpiece after cutting, label the image information of the workpiece after cutting as a size verification image and transmit it to the second processing unit. The first processing unit is used to compare the image information labeled as the recognition image with the pre-stored images of obstacles that have been labeled, perform obstacle recognition and positioning processing on the image information with obstacles found in the comparison, and input the processing result into the PLC controller. The second processing unit is used to compare the image information labeled as a size verification image with the stored standard size workpiece image. If the comparison is consistent, it outputs a size qualified reminder message; if the comparison is inconsistent, it outputs a size unqualified reminder message. The first processing unit performs contrast enhancement processing on the acquired image, and uses a gray-level difference method based on edge image pixels to extract the image contour. By clearing the pixels wrapped by the image edge and retaining the image contour, the edge features of obstacles in the image are extracted. The obstacle feature is extracted using a Gabor wavelet filter to complete the localization of the obstacle shape and position, and the processing results are input to the PLC controller. The first processing unit performs digital signal transformation by filtering the texture of the acquired image using a Gabor wavelet filter. The transformed digital signal is decomposed into a set of wavelet functions, and the resolution of the image is improved in the time, space and frequency domains by using transformation factors and dilation factors. During filtering, the output of the Gabor wavelet filter is adjusted by averaging the virtual and real filter masks of all convolutional pixels in the image.

2. The plasma cutting system according to claim 1, characterized in that, The second processing unit determines the image edge based on the grayscale changes on both sides of the image edge, connects the edges, and determines the image size of the cut workpiece by measuring the distance based on the position coordinates of the pixels. It then compares the image size with the standard workpiece image size, wherein the standard workpiece image and the cut workpiece image have the same number of pixels.

3. The plasma cutting system according to claim 2, characterized in that, The system also includes a host computer that connects to the robotic cutting equipment and vision sensor module via a wireless network, enabling remote monitoring of the cutting process, adjustment of cutting parameters, and receipt of real-time feedback data.

4. The plasma cutting system according to claim 3, characterized in that, The image acquisition unit of the vision sensor module is equipped with a lens cover.

5. A cutting control method based on the plasma cutting system according to any one of claims 1-4, characterized in that, Includes the following steps: Signal interaction between the robot cutting equipment and the vision sensor module is established using the PROFINET communication protocol; During the cutting process, the vision sensor module captures image information of the cutting path of the cutting target in real time, identifies obstacles in the cutting trajectory, and inputs the identification results to the robot cutting equipment; The robotic cutting device controls its own cutting action based on the recognition result; After cutting is completed, the vision sensor module inspects the cut workpiece and outputs a prompt message when the workpiece is found to be unqualified. During the cutting process, the vision sensor module captures image information of the cutting path of the target in real time, identifies obstacles in the cutting trajectory, and inputs the identification results to the robotic cutting equipment. The steps include: The image acquisition unit captures the image information of the cutting path of the cutting target in real time and annotates the captured image information of the cutting path as a recognition image and transmits it to the first processing unit; The first processing unit compares the image information labeled as the recognition image with the pre-stored images of obstacles that have been labeled, performs obstacle identification and location processing on the image information with obstacles found in the comparison, and inputs the processing result into the PLC controller.

Images