ABS (acrylonitrile butadiene styrene) plastic plate production and manufacturing workstation with cooperation of multiple robots

Abstract

The invention discloses a multi-robot cooperative ABS (acrylonitrile butadiene styrene) plastic plate production and manufacturing workstation, which comprises an extrusion molding auxiliary robot for assisting in feeding and taking and placing a mold; the plastic extruding machine is used for heating and melting the ABS plastic particles and then extruding the ABS plastic particles into an ABS plastic plate through a mold; the grinding robot is used for carrying a grinding tool and carrying out surface grinding and unhairing treatment on the surface of the ABS plastic plate; the detection robot is used for carrying out defect detection on the ABS plastic plate to obtain a defect detection result and storing the ABS plastic plate in different areas according to the defect detection result; and the material conveying module is arranged among the extrusion molding auxiliary robot, the extrusion molding machine, the polishing robot and the visual inspection robot and is used for conveying materials. The production efficiency and the production quality of the ABS plastic plates are improved, whether the produced ABS plastic plates are qualified or not is automatically detected, and the ABS plastic plates are stored in different areas.

Description

Technical Field

[0001] The invention relates to the technical field of production and manufacturing, and in particular to a multi-robot collaborative ABS plastic plate production and manufacturing workstation. Background Art

[0002] ABS plastic sheets have excellent mechanical properties, high stability, easy processing and excellent chemical properties, so they are widely used in the fields of home appliances, auto parts, office equipment, mechanical equipment and mold models. At present, there are technical problems such as low production efficiency and low production quality in the production process of ABS plastic sheets, and it is impossible to automatically detect whether the produced ABS plastic sheets are qualified. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems in the above-mentioned technology to a certain extent. To this end, the purpose of the present invention is to provide a multi-robot collaborative ABS plastic board production and manufacturing workstation, which improves the production efficiency and production quality of ABS plastic boards, automatically detects whether the produced ABS plastic boards are qualified, and stores them in different areas.

[0004] To achieve the above-mentioned purpose, the embodiment of the present invention proposes a multi-robot collaborative ABS plastic plate production and manufacturing workstation, comprising:

[0005] Extrusion auxiliary robot, used to assist in loading and removing molds;

[0006] Extruder, used to heat and melt ABS plastic particles and extrude them into ABS plastic sheets through a mold;

[0007] The grinding robot is used to carry grinding tools to grind and remove the burrs on the surface of ABS plastic plates;

[0008] The detection robot is used to perform defect detection on the ABS plastic plate, obtain defect detection results, and store the ABS plastic plate in different areas according to the defect detection results;

[0009] The material conveying module is arranged between the extrusion auxiliary robot, the extruder, the polishing robot and the visual inspection robot, and is used for material conveying.

[0010] According to some embodiments of the present invention, it further includes: a cooling device, which is arranged in a first preset range of the extruder; the cooling device is a cooling water tank or an air cooling device.

[0011] According to some embodiments of the present invention, the further comprising: a dust suction device, arranged in a second preset range of the grinding machine, for sucking away dust generated by grinding.

[0012] According to some embodiments of the present invention, the detection robot comprises:

[0013] The first detection module is used to detect the shape, size and surface defects of the ABS plastic plate to obtain a first detection result;

[0014] The second detection module is used to detect the thickness of the ABS plastic plate to obtain a second detection result;

[0015] The determination module is used to determine the defect detection result according to the first detection result and the second detection result.

[0016] According to some embodiments of the present invention, the material conveying module includes a conveyor belt or track and a positioning device; wherein the positioning device is used to determine the position information of the ABS plastic plate in each process.

[0017] According to some embodiments of the present invention, the first detection module includes:

[0018] The first acquisition module is used to acquire the surface image of the ABS plastic plate;

[0019] The stretching module is used to perform image grayscale processing on the surface image to obtain a grayscale image; perform linear expansion on the grayscale image based on a linear function, and stretch the grayscale of the pixel points in the grayscale image to obtain a stretched image;

[0020] Noise reduction module for:

[0021] Acquiring an image resolution of the stretched image, performing image decomposition on the stretched image according to the image resolution, and determining multi-order sub-images;

[0022] The texture features of each order of sub-images are obtained respectively, the corresponding denoising weights are determined according to the texture features, and the corresponding sub-images are denoised according to the denoising weights to obtain denoised images;

[0023] Edge detection module for:

[0024] Perform edge detection on the denoised image based on the Canny operator edge detection algorithm, generate edge lines, and obtain first detection information of shape and size based on the edge lines;

[0025] A comparison module is used to obtain the gray value of each pixel in the denoised image and compare it with a preset gray value, and to regard the pixel whose gray value is not within the preset gray value range as a defective pixel to obtain second detection information;

[0026] A first detection result is determined according to the first detection information and the second detection information.

[0027] According to some embodiments of the present invention, the second detection result includes:

[0028] The second acquisition module is used to send a first laser beam to the upper surface of the ABS plastic plate and acquire a first time after the first laser beam is reflected by the upper surface;

[0029] A third acquisition module is used to send a second laser beam to the lower surface of the ABS plastic plate and acquire a second time after the second laser beam is reflected by the upper surface;

[0030] A calculation module, used for calculating the thickness of the ABS plastic plate according to the first time and the second time, and obtaining a second detection result;

[0031]

[0032] Wherein, d is the thickness of the ABS plastic plate; T1 is the first time; T2 is the second time; λ is the refractive index of the ABS plastic plate; and c is the speed of light.

[0033] According to some embodiments of the present invention, the present invention further includes: a monitoring module, configured to:

[0034] Acquire a scene image of the area where the workstation is located;

[0035] Based on the preset trained scene object recognition model, each recognition frame is marked in the scene image, and the scene object corresponding to each recognition frame is determined;

[0036] Identify the status information of scene objects and generate monitoring information.

[0037] According to some embodiments of the present invention, the monitoring module identifies state information of a scene object and generates monitoring information, including:

[0038] Determine the type of scene objects, including equipment scene objects and material scene objects;

[0039] Obtain a grayscale image and a reference grayscale image of the device scene object; determine the texture feature value of each pixel in the grayscale image; compare the texture feature value of each pixel in the grayscale image with the texture feature value of the corresponding pixel in the reference grayscale image to obtain a first comparison result; obtain an HSV space image and a reference HSV space image of the device scene object; for each pixel in the HSV space image, determine the color feature value of the pixel according to the H, S, and V components of the pixel; compare the color feature value of the pixel in the HSV space image with the color feature value of the corresponding pixel in the reference HSV space image to obtain a second comparison result; determine the status information of the device scene object according to the first comparison result and the second comparison result;

[0040] Get the video clip of the material scene object. The motion trajectory of the particles in a video clip W×H×T is expressed as:

[0041] {(x(t),y(t))∣x∈[1,W],y∈[1,H],t∈[1,T]}

[0042] Where W is the width of the video clip; H is the height of the video clip; T represents the number of consecutive frames of the video clip, and the vector (x(t), y(t)) represents the position of the particle (x, y) at time t;

[0043] Cluster the motion trajectories of the particles to obtain the region patch(m) that matches the motion characteristics, where m is the number of the region;

[0044] determining state information of the material scene object according to the video clip of the material scene object;

[0045] Generate monitoring information based on the status information of the equipment scene object and the status information of the material scene object.

[0046] According to some embodiments of the present invention, determining state information of a material scene object according to a video clip of the material scene object includes:

[0047] Determine the conveying volume and conveying speed of the material scene object;

[0048] numel(m)=U{(x p (t),y p (t))∣(x p ,y p )∈patch(m),t∈[1,T]}

[0049]

[0050] Where numel(m) is the conveying volume of the material scene object; U is the number of elements in the area; (x p ,y p ) is the position of the particle in the region; (x p (t), y p (t)) is the particle (x p ,y p ) at time t; (x p (1), y p (1)) is the particle (x p ,y p ) is the position at the initial moment; v is the conveying speed of the material scene object;

[0051] According to the conveying volume and conveying speed of the material scene object, the state information of the material scene object is determined.

[0052] The present invention proposes a multi-robot collaborative ABS plastic board production and manufacturing workstation, which improves the production efficiency and production quality of the ABS plastic boards, automatically detects whether the produced ABS plastic boards are qualified, and stores them in different areas.

[0053] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0054] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0056] Figure 1 The present invention is a block diagram of a multi-robot collaborative ABS plastic board production and manufacturing workstation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0058] like Figure 1 As shown, the embodiment of the present invention proposes a multi-robot collaborative ABS plastic plate production and manufacturing workstation, including:

[0059] Extrusion auxiliary robot, used to assist in loading and removing molds;

[0060] Extruder, used to heat and melt ABS plastic particles and extrude them into ABS plastic sheets through a mold;

[0061] The grinding robot is used to carry grinding tools to grind and remove the burrs on the surface of ABS plastic plates;

[0062] The detection robot is used to perform defect detection on the ABS plastic plate, obtain defect detection results, and store the ABS plastic plate in different areas according to the defect detection results;

[0063] The material conveying module is arranged between the extrusion auxiliary robot, the extruder, the polishing robot and the visual inspection robot, and is used for material conveying.

[0064] Working principle of the above technical solution: The extrusion auxiliary robot is mainly responsible for assisting in loading and picking up and placing the mold. Its precise operating ability ensures the accuracy and efficiency of the loading process, and can quickly and accurately pick up and place the mold to prepare for the extrusion process. Through automated operation, manual intervention is reduced, and production efficiency and safety are improved. The extrusion auxiliary robot uses a six-axis robot with moderate load and high precision.

[0065] The extruder is the core equipment for producing ABS plastic sheets. It heats and melts ABS plastic particles, and then extrude them through a mold to form ABS plastic sheets.

[0066] The grinding robot equipped with grinding tools is responsible for grinding and deburring the surface of ABS plastic plates. This ensures that the surface of the plastic plate is flat and smooth, meeting the quality requirements of subsequent processing or use. Automated grinding improves production efficiency while ensuring the consistency and stability of grinding quality. Grinding tools include sanding belts, grinding wheels, etc. The grinding robot selects a robot with high load and flexibility, which can adapt to the grinding needs of ABS plastic plates of different shapes and sizes, and is equipped with a force control system to ensure uniform grinding force and avoid over-grinding or under-grinding.

[0067] The inspection robot detects defects on ABS plastic plates, such as cracks, bubbles, impurities, etc. Based on the defect detection results, the robot stores the ABS plastic plates in different areas for subsequent processing. The inspection robot is equipped with a high-definition camera and image processing algorithm, which can capture and analyze image information on the surface of the plastic plate in real time. Automated inspection improves the accuracy and efficiency of inspection and reduces the cost and error of manual inspection. The storage area includes qualified areas and unqualified areas.

[0068] The material conveying module is set between the extrusion auxiliary robot, extruder, polishing robot and inspection robot, and is responsible for material conveying. It ensures the smooth connection between various production links and improves the automation level of the entire production line. Automated material conveying reduces the labor intensity of manual handling, improves production efficiency, and ensures the accurate and timely delivery of materials on the production line.

[0069] Beneficial effects of the above technical solution: The multi-robot collaborative ABS plastic board manufacturing workstation significantly improves production efficiency and product quality through a highly automated production method. The close collaboration and precise control between the various components ensure the stability and reliability of the entire production process.

[0070] In one embodiment, a central control system is also provided to uniformly manage and coordinate the various robots and equipment. The parameters and processes of each process are set by programming to realize the automatic operation of the entire workstation. The data interaction system enables data interaction between various robots and equipment, such as the extruder transmits data such as extrusion speed and temperature to the polishing robot and the detection robot so that they can make corresponding adjustments according to the actual situation. The monitoring system is equipped with monitoring cameras and sensors to monitor the operating status of the workstation in real time. Once an abnormal situation occurs, the system immediately issues an alarm and shuts down, and records relevant data so that maintenance personnel can troubleshoot.

[0071] According to some embodiments of the present invention, it further includes: a cooling device, which is arranged in a first preset range of the extruder; the cooling device is a cooling water tank or an air cooling device.

[0072] The working principle of the above technical solution is: the cooling device is arranged in the first preset range of the extruder, which is to ensure that the thermoplastic ABS plastic plate just extruded from the extruder can be cooled quickly and evenly. The specific position of the cooling device needs to be determined according to the layout of the production line, the model of the extruder and the requirements of cooling efficiency.

[0073] The beneficial effect of the above technical solution is that in the multi-robot collaborative ABS plastic plate production and manufacturing workstation, the setting of the cooling device facilitates improving production efficiency and product quality.

[0074] According to some embodiments of the present invention, the further comprising: a dust suction device, arranged in a second preset range of the grinding machine, for sucking away dust generated by grinding.

[0075] The working principle of the above technical solution is that the dust suction device is set within the second preset range of the grinding machine, which is determined according to the location and diffusion range of dust generated during the grinding process. By accurately setting the position of the dust suction device, it can be ensured that it can effectively remove the dust generated during the grinding process, thereby keeping the production environment clean.

[0076] The beneficial effects of the above technical solution are as follows: by reasonably setting up and using the dust collection device, the cleanliness of the production environment can be maintained, the health of the workers can be guaranteed, and environmental protection requirements can be met.

[0077] According to some embodiments of the present invention, the detection robot comprises:

[0078] The first detection module is used to detect the shape, size and surface defects of the ABS plastic plate to obtain a first detection result;

[0079] The second detection module is used to detect the thickness of the ABS plastic plate to obtain a second detection result;

[0080] The determination module is used to determine the defect detection result according to the first detection result and the second detection result.

[0081] The working principle of the above technical solution is as follows: the first detection module is responsible for detecting the shape, size and surface defects of the ABS plastic plate, and can identify whether the plastic plate has shape deformation, size deviation, surface scratches, cracks, bubbles and other defects. The second detection module generates a second detection result, which records the thickness distribution of the plastic plate in detail.

[0082] The beneficial effects of the above technical solution are as follows: through the coordinated work of the first detection module and the second detection module, the key parameters such as the shape, size, surface defects and thickness of the ABS plastic plate can be fully and accurately detected. The determination module generates reliable defect detection results, which is convenient for ensuring the stability and reliability of the entire production process.

[0083] According to some embodiments of the present invention, the material conveying module includes a conveyor belt or track and a positioning device; wherein the positioning device is used to determine the position information of the ABS plastic plate in each process.

[0084] The working principle and beneficial effects of the above technical solution: Through the coordinated work of the conveyor belt or track and the positioning device, it ensures the accurate and efficient transmission of the ABS plastic plate between various processes. This not only improves the continuity and stability of the production line, but also provides strong support for the realization of automated and intelligent production.

[0085] According to some embodiments of the present invention, the method further includes: a determination module, configured to determine whether defect information is contained according to the first detection result, and to generate prompt information when it is determined that the first detection result contains defect information;

[0086] The determining module comprises:

[0087] Building blocks for:

[0088] Obtain P sample inspection images and sample information that have been determined to contain defect information, wherein the sample information contains the inspection results of the shape, size, and surface defects of the sample ABS plastic plate, and whether the sample ABS plastic plate contains defect information; the inspection results of the P sample inspection images are formed into a sample inspection result matrix X, wherein the sample inspection result matrix X has P rows and 3 columns, and at the same time, the defect result vector Y corresponding to the P sample ABS plastic plates is formed into whether the defect information is contained, wherein the defect result vector Y has P values; when the sample ABS plastic plate contains defect information, the defect inspection result corresponding to the sample ABS plastic plate is 1, and when the sample ABS plastic plate does not contain defect information, the defect inspection result corresponding to the sample ABS plastic plate is 0;

[0089] Construct a defect learning function for the sample detection result matrix X and the defect result vector Y;

[0090]

[0091] Among them, f(X) is the constructed defect learning function, Y i is the i-th value of the defect result vector Y, X i,j is the value of the i-th row and j-th column of the sample detection result matrix X, λ0 is the detection constant term to be solved, and λ j is the weight coefficient of the detection item to be solved, i = 1, 2, 3...P, j = 1, 2, 3;

[0092] Determine the detection constant term λ0 and weight coefficient λ j ;

[0093]

[0094] in, Take the partial derivative of the defect learning function f(X) with respect to the detection constant term λ0, is the defect learning function f(X) for the weight coefficient λ j Take partial derivatives to form a system of 4 equations, which contains the detection constant term λ0 and 3 weight coefficients λ j There are 4 unknown quantities in total. By solving the above equations, we can get the detection constant term λ0 and 3 weight coefficients λ j ;

[0095] Determining whether defect information is contained according to the first detection result;

[0096]

[0097] Among them, P is the result determination value, K j is the detection result of the jth detection item in the first detection result;

[0098] The result determination value is compared with a preset threshold value, and when it is determined that the result determination value is less than the preset threshold value, a prompt message is generated.

[0099] The working principle of the above technical solution is: collect P sample inspection images that have been determined to contain defect information and the corresponding sample information. This information includes the inspection results of the shape, size, and surface defects of the sample ABS plastic plate, and whether the sample ABS plastic plate contains defect information determined based on these inspection results. The inspection results of the P sample inspection images form a sample inspection result matrix X, which has P rows and 3 columns, where each row represents the three inspection results of a sample (shape, size, surface defects). At the same time, the defect information corresponding to the P sample ABS plastic plates is formed into a defect result vector Y, which has P values, a value of 1 indicates that it contains defect information, and a value of 0 indicates that it does not contain defect information. Use the logistic regression model to construct a defect learning function f(X), which can predict whether the sample contains defect information based on the sample inspection result matrix X. For the detection constant term λ0 and the weight coefficient λ j Solve, for the new first test result, use the trained model to calculate the result determination value P. Compare the result determination value with the preset threshold. If the result determination value is less than the preset threshold, it is determined that the ABS plastic plate corresponding to the first test result contains defect information, and generate prompt information. For example, the preset threshold is 0.5.

[0100] The beneficial effect of the above technical solution is that the determination module can accurately judge whether the ABS plastic plate contains defect information by constructing a defect learning function and using sample data for training. This method not only improves the accuracy of detection, but also reduces the cost and time of manual intervention.

[0101] According to some embodiments of the present invention, the first detection module includes:

[0102] The first acquisition module is used to acquire the surface image of the ABS plastic plate;

[0103] The stretching module is used to perform image grayscale processing on the surface image to obtain a grayscale image; perform linear expansion on the grayscale image based on a linear function, and stretch the grayscale of the pixel points in the grayscale image to obtain a stretched image;

[0104] Noise reduction module for:

[0105] Acquiring an image resolution of the stretched image, performing image decomposition on the stretched image according to the image resolution, and determining multi-order sub-images;

[0106] The texture features of each order of sub-images are obtained respectively, the corresponding denoising weights are determined according to the texture features, and the corresponding sub-images are denoised according to the denoising weights to obtain denoised images;

[0107] Edge detection module for:

[0108] Perform edge detection on the denoised image based on the Canny operator edge detection algorithm, generate edge lines, and obtain first detection information of shape and size based on the edge lines;

[0109] A comparison module is used to obtain the gray value of each pixel in the denoised image and compare it with a preset gray value, and to regard the pixel whose gray value is not within the preset gray value range as a defective pixel to obtain second detection information;

[0110] A first detection result is determined according to the first detection information and the second detection information.

[0111] The working principle of the above technical solution is as follows: the first acquisition module is used to acquire the surface image of the ABS plastic plate; the stretching module uses a linear function to stretch the pixel value of the grayscale image to enhance the image contrast and make the details in the image clearer. The denoising module decomposes the image into multiple sub-images of different orders according to the resolution of the image so as to process the texture features of different scales. For each sub-image, its texture features are extracted, and the preset texture feature-denoising weight data table is queried based on these features to determine the corresponding denoising weight. Then, the sub-image is denoised using these weights to obtain a clearer denoised image. The Canny operator is used to perform edge detection on the denoised image to generate edge lines. These edge lines can accurately reflect the shape and size information of the ABS plastic plate. Based on the edge lines, the shape and size features of the ABS plastic plate are extracted to generate the first detection information. The comparison module obtains the grayscale value of each pixel in the denoised image and compares it with the preset grayscale value. The pixel whose grayscale value is not within the preset grayscale value range is regarded as a defective pixel to obtain the second detection information. The shape, size and surface defects of the ABS plastic plate are comprehensively evaluated by combining the first detection information and the second detection information, and finally the first detection result is determined.

[0112] Beneficial effects of the above technical solution: The first detection module realizes accurate detection of the shape, size and surface defects of ABS plastic plates by integrating multiple image processing modules. This method not only improves the accuracy and efficiency of detection, but also provides reliable data support for subsequent automated processing.

[0113] According to some embodiments of the present invention, the second detection result includes:

[0114] The second acquisition module is used to send a first laser beam to the upper surface of the ABS plastic plate and acquire a first time after the first laser beam is reflected by the upper surface;

[0115] A third acquisition module is used to send a second laser beam to the lower surface of the ABS plastic plate and acquire a second time after the second laser beam is reflected by the upper surface;

[0116] A calculation module, used for calculating the thickness of the ABS plastic plate according to the first time and the second time, and obtaining a second detection result;

[0117]

[0118] Wherein, d is the thickness of the ABS plastic plate; T1 is the first time; T2 is the second time; λ is the refractive index of the ABS plastic plate; and c is the speed of light.

[0119] The working principle of the above technical solution is: once the laser beam touches the upper surface of the plastic plate, it will be quickly reflected back. At this time, the second acquisition module will accurately capture and record the first time spent in this reflection process. The laser beam will also be reflected after touching the lower surface of the plastic plate. The third acquisition module will be responsible for recording the second time spent in this reflection process. The calculation module calculates the thickness of the ABS plastic plate based on the first time and the second time.

[0120] Beneficial effects of the above technical solution: The second detection result provides reliable quantitative information for the thickness of the ABS plastic plate by combining laser ranging technology and precise time measurement, thereby improving the accuracy of thickness detection.

[0121] According to some embodiments of the present invention, the present invention further includes: a monitoring module, configured to:

[0122] Acquire a scene image of the area where the workstation is located;

[0123] Based on the preset trained scene object recognition model, each recognition frame is marked in the scene image, and the scene object corresponding to each recognition frame is determined;

[0124] Identify the status information of scene objects and generate monitoring information.

[0125] The working principle of the above technical solution is as follows: The monitoring module first captures the real-time scene image of the area where the workstation is located. The captured scene image is deeply analyzed using a preset and fully trained scene object recognition model. Each recognition frame is accurately marked in the image, and each recognition frame corresponds to a specific object in the image. Based on the recognition results of the model, a specific scene object label is assigned to each recognition frame. Through image analysis and machine learning algorithms, the objects in each recognition frame are deeply analyzed to obtain their status information. The status information includes a variety of information such as the location, size, shape, color, and activity status of the object.

[0126] The beneficial effects of the above technical solution are: by real-time monitoring of key objects in the workstation, such as equipment, personnel, etc., the operation safety of the workstation is ensured. By monitoring each object and its status on the production line, the refined management of the production process is achieved and the production efficiency is improved. When an abnormal state or behavior is identified, the monitoring module can quickly issue an early warning to provide strong support for timely measures.

[0127] According to some embodiments of the present invention, the monitoring module identifies state information of a scene object and generates monitoring information, including:

[0128] Determine the type of scene objects, including equipment scene objects and material scene objects;

[0129] Obtain a grayscale image and a reference grayscale image of the device scene object; determine the texture feature value of each pixel in the grayscale image; compare the texture feature value of each pixel in the grayscale image with the texture feature value of the corresponding pixel in the reference grayscale image to obtain a first comparison result; obtain an HSV space image and a reference HSV space image of the device scene object; for each pixel in the HSV space image, determine the color feature value of the pixel according to the H, S, and V components of the pixel; compare the color feature value of the pixel in the HSV space image with the color feature value of the corresponding pixel in the reference HSV space image to obtain a second comparison result; determine the status information of the device scene object according to the first comparison result and the second comparison result;

[0130] Get the video clip of the material scene object. The motion trajectory of the particles in a video clip W×H×T is expressed as:

[0131] {(x(t),y(t))∣x∈[1,W],y∈[1,H],t∈[1,T]}

[0132] Where W is the width of the video clip; H is the height of the video clip; T represents the number of consecutive frames of the video clip, and the vector (x(t), y(t)) represents the position of the particle (x, y) at time t;

[0133] Cluster the motion trajectories of the particles to obtain the region patch(m) that matches the motion characteristics, where m is the number of the region;

[0134] determining state information of the material scene object according to the video clip of the material scene object;

[0135] Generate monitoring information based on the status information of the equipment scene object and the status information of the material scene object.

[0136] The working principle of the above technical solution is as follows: the monitoring module determines the type of scene objects, including equipment scene objects and material scene objects; the equipment scene objects include extrusion auxiliary robots, extruders, polishing robots, inspection robots, etc. The material scene objects are the materials transported by the material transport module in the extrusion auxiliary robots, extruders, polishing robots and visual inspection robots.

[0137] The grayscale image is a real-time image of the device scene object, and the reference grayscale image is a grayscale image of the device scene object in a standard state. The method for determining the texture eigenvalue of each pixel in the grayscale image includes a grayscale co-occurrence matrix, a local binary pattern, and a Gabor filter. The texture eigenvalue of each pixel in the grayscale image is calculated, and compared with the texture eigenvalue of the corresponding pixel in the reference grayscale image to obtain a first comparison result. This step helps to detect subtle changes on the surface of the device, such as scratches, wear, etc. The HSV space image of the device scene object and its reference HSV space image are obtained. HSV stands for hue, saturation, and value, respectively. For each pixel in the HSV space image, its color eigenvalue is determined according to the H, S, and V components, and compared with the color eigenvalue of the corresponding pixel in the reference image to obtain a second comparison result. This step can detect color changes and reflect changes in the state of the device, such as oxidation, pollution, etc. Combining the first comparison result and the second comparison result, the monitoring module can determine the specific state information of the device scene object.

[0138] Get a video clip of the material scene object, which has a specific width W, height H, and time length T. Perform cluster analysis on the motion trajectories of all particles in the video clip to obtain the region patch(m) that matches the motion characteristics, where m is the number of the region. This step helps to identify the flow pattern, stacking state, etc. of the material. Based on the clustering results and the overall analysis of the video clip, the monitoring module can determine the status information of the material scene object. The monitoring module integrates the status information of the equipment scene object and the status information of the material scene object, reflecting the status of the equipment and materials in the scene.

[0139] The beneficial effects of the above technical solution are as follows: through sophisticated image and video analysis technology, the equipment scene objects and material scene objects are monitored separately, achieving comprehensive monitoring of the scene object status and ensuring the accuracy and practicality of the monitoring information.

[0140] According to some embodiments of the present invention, determining state information of a material scene object according to a video clip of the material scene object includes:

[0141] Determine the conveying volume and conveying speed of the material scene object;

[0142] numel(m)=U{(xp (t),y p (t))∣(x p ,y p )∈patch(m),t∈[1,T]}

[0143]

[0144] Where numel(m) is the conveying volume of the material scene object; U is the number of elements in the area; (x p ,y p ) is the position of the particle in the region; (x p (t), y p (t)) is the particle (x p ,y p ) at time t; (x p (1), y p (1)) is the particle (x p ,y p ) is the position at the initial moment; v is the conveying speed of the material scene object;

[0145] According to the conveying volume and conveying speed of the material scene object, the state information of the material scene object is determined.

[0146] The working principle and beneficial effects of the above technical solution are: accurately determining the conveying volume and conveying speed of the material scene object, and then accurately determining the status information of the material scene object.

[0147] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A multi-robot collaborative ABS plastic board production and manufacturing workstation, characterized in that: include: Extrusion auxiliary robot, used to assist in loading and removing molds; Extruder, used to heat and melt ABS plastic particles and extrude them into ABS plastic sheets through a mold; The grinding robot is used to carry grinding tools to grind and remove the burrs on the surface of ABS plastic plates; The detection robot is used to perform defect detection on the ABS plastic plate, obtain defect detection results, and store the ABS plastic plate in different areas according to the defect detection results; The material conveying module is arranged between the extrusion auxiliary robot, the extruder, the polishing robot and the visual inspection robot, and is used for material conveying.

2. The multi-robot collaborative ABS plastic board manufacturing workstation as claimed in claim 1, characterized in that: Also includes: The cooling device is arranged in the first preset range of the extruder; the cooling device is a cooling water tank or an air cooling device.

3. The multi-robot collaborative ABS plastic board manufacturing workstation as claimed in claim 1, characterized in that: Also includes: The dust suction device is arranged in the second preset range of the grinding machine and is used for sucking away the dust generated by grinding.

4. The multi-robot collaborative ABS plastic board manufacturing workstation as claimed in claim 1, characterized in that: The detection robot comprises: The first detection module is used to detect the shape, size and surface defects of the ABS plastic plate to obtain a first detection result; The second detection module is used to detect the thickness of the ABS plastic plate to obtain a second detection result; The determination module is used to determine the defect detection result according to the first detection result and the second detection result.

5. The multi-robot collaborative ABS plastic board manufacturing workstation as claimed in claim 1, characterized in that: The material conveying module includes a conveyor belt or track and a positioning device; wherein the positioning device is used to determine the position information of the ABS plastic plate in each process.

6. The multi-robot collaborative ABS plastic board manufacturing workstation as claimed in claim 4, characterized in that: The first detection module includes: A first acquisition module is used to acquire a surface image of an ABS plastic plate; The stretching module is used to perform image grayscale processing on the surface image to obtain a grayscale image; perform linear expansion on the grayscale image based on a linear function, and stretch the grayscale of the pixel points in the grayscale image to obtain a stretched image; Noise reduction module for: Acquiring an image resolution of the stretched image, performing image decomposition on the stretched image according to the image resolution, and determining multi-order sub-images; The texture features of each order of sub-images are obtained respectively, the corresponding denoising weights are determined according to the texture features, and the corresponding sub-images are denoised according to the denoising weights to obtain denoised images; Edge detection module for: Performing edge detection on the denoised image based on the Canny operator edge detection algorithm to generate edge lines, and obtaining first detection information of shape and size based on the edge lines; A comparison module is used to obtain the gray value of each pixel in the denoised image and compare it with a preset gray value, and to regard the pixel whose gray value is not within the preset gray value range as a defective pixel to obtain second detection information; A first detection result is determined according to the first detection information and the second detection information.

7. The multi-robot collaborative ABS plastic board manufacturing workstation as claimed in claim 4, characterized in that: The second detection result includes: The second acquisition module is used to send a first laser beam to the upper surface of the ABS plastic plate and acquire a first time after the first laser beam is reflected by the upper surface; A third acquisition module is used to send a second laser beam to the lower surface of the ABS plastic plate and acquire a second time after the second laser beam is reflected by the upper surface; A calculation module, used for calculating the thickness of the ABS plastic plate according to the first time and the second time, and obtaining a second detection result; Wherein, d is the thickness of the ABS plastic plate; T1 is the first time; T2 is the second time; λ is the refractive index of the ABS plastic plate; and c is the speed of light.

8. The multi-robot collaborative ABS plastic board manufacturing workstation as claimed in claim 1, characterized in that: Also includes: Monitoring module for: Acquire a scene image of the area where the workstation is located; Based on the preset trained scene object recognition model, each recognition frame is marked in the scene image, and the scene object corresponding to each recognition frame is determined; Identify the status information of scene objects and generate monitoring information.

9. The multi-robot collaborative ABS plastic board manufacturing workstation as claimed in claim 8, characterized in that: The monitoring module identifies the state information of the scene object and generates monitoring information, including: Determine the type of scene objects, including equipment scene objects and material scene objects; Obtain a grayscale image and a reference grayscale image of the device scene object; determine the texture feature value of each pixel in the grayscale image; compare the texture feature value of each pixel in the grayscale image with the texture feature value of the corresponding pixel in the reference grayscale image to obtain a first comparison result; obtain an HSV space image and a reference HSV space image of the device scene object; for each pixel in the HSV space image, determine the color feature value of the pixel according to the H, S, and V components of the pixel; compare the color feature value of the pixel in the HSV space image with the color feature value of the corresponding pixel in the reference HSV space image to obtain a second comparison result; determine the status information of the device scene object according to the first comparison result and the second comparison result; Get the video clip of the material scene object. The motion trajectory of the particles in a video clip W×H×T is expressed as: {(x(t),y(t))∣x∈[1,W],y∈[1,H],t∈[1,T]} Where W is the width of the video clip; H is the height of the video clip; T represents the number of consecutive frames of the video clip, and the vector (x(t), y(t)) represents the position of the particle (x, y) at time t; Cluster the motion trajectories of the particles to obtain the region patch(m) that matches the motion characteristics, where m is the number of the region; determining state information of the material scene object according to the video clip of the material scene object; Generate monitoring information based on the status information of the equipment scene object and the status information of the material scene object.

10. The multi-robot collaborative ABS plastic board manufacturing workstation as claimed in claim 9, characterized in that: The state information of the material scene object is determined according to the video clip of the material scene object, including: Determine the conveying volume and conveying speed of the material scene object; numel(m)=U{(x p (t),y p (t))∣(x p ,y p )∈patch(m),t∈[1,T]} Where numel(m) is the conveying volume of the material scene object; U is the number of elements in the area; (x p ,y p ) is the position of the particle in the region; (x p (t), y p (t)) is the particle (x p ,y p ) at time t; (x p (1), y p (1)) is the particle (x p ,y p ) is the position at the initial moment; v is the conveying speed of the material scene object; According to the conveying volume and conveying speed of the material scene object, the state information of the material scene object is determined.

Images