Intelligent robot spraying method based on spraying system

Abstract

The invention discloses an intelligent robot spraying method based on a spraying system. Firstly, three-dimensional contour data of a workpiece are obtained through three-dimensional scanning equipment, a model is constructed, and key information is marked. And then according to the model, the workpiece material, the spraying requirement and the environment condition, a spraying path is planned through an intelligent algorithm, coating parameters are set, a robot executes spraying in sequence, and real-time monitoring is conducted through a high-precision position and a visual sensor in the process. After spraying is completed, the quality is detected through detection equipment, and when the requirements are not met, the computer system adjusts parameters according to deviation and directs the robot to conduct supplementary spraying or re-spraying. Whole-course data are recorded, and technological parameters and a path algorithm are optimized through statistical analysis. According to the method, high-precision, intelligent and efficient spraying is achieved, the method adapts to various workpiece and process requirements, the spraying quality and the production efficiency are improved, good adaptability and expandability are achieved, and intelligent development of industrial spraying is powerfully promoted.

Description

[0001] Technology ZH

[0002] The present invention relates to the field related to robot spraying, and in particular to an intelligent robot spraying method based on a spraying system. Background Art

[0003] In modern industrial production, such as automobile manufacturing, furniture production, aerospace and many other fields, surface spraying of products is a key process. Traditional spraying methods often rely on manual operation, which is not only inefficient, but also difficult to ensure the consistency and stability of spraying quality. With the development of automation technology and robotics technology, intelligent robot spraying systems have gradually been widely used. However, the existing robotic spraying methods still have many shortcomings in spraying path planning, paint flow control, dynamic adjustment of spraying parameters, etc., and cannot fully meet the high-precision and high-quality spraying requirements of complex-shaped workpieces. In addition, they have poor adaptability when facing workpieces of different materials and different process requirements. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent robot spraying method based on a spraying system to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: an intelligent robot spraying method based on a spraying system, comprising the following steps:

[0006] Step 1: Artifact information acquisition and model building

[0007] Use 3D scanning equipment to perform all-round scanning on the workpiece to be sprayed to obtain the 3D contour data of the workpiece; transmit the above data to the computer control system, process and analyze the data through special data processing software, and build an accurate 3D model of the workpiece. At the same time, mark the areas that need to be sprayed, the spraying requirements of different areas, and the key data of the material information of the workpiece in the model;

[0008] Step 2: Spraying path planning

[0009] According to the three-dimensional model of the workpiece constructed in step 1, the computer control system uses an intelligent path planning algorithm to generate the robot's spray path. The algorithm comprehensively considers the shape, size, spray area and kinematic characteristics of the workpiece to ensure that the spray path is shortest, collision-free and can evenly cover the spray area. When planning the path, the workpiece surface is divided into multiple tiny spray units, and the optimal spray gun movement speed is calculated for each unit, so that the paint can be evenly attached to surfaces of different shapes;

[0010] Step 3: Paint parameter setting and mixing

[0011] According to the material of the workpiece, spraying requirements and environmental conditions, the computer control system automatically determines the appropriate paint type, paint viscosity, and paint flow parameters; during the paint mixing process, the flow sensor and viscosity sensor equipment connected to the paint supply system monitor the paint flow and viscosity in real time, and automatically adjust according to the set value to ensure that the mixing ratio and physical properties of the paint always meet the spraying requirements;

[0012] Step 4: Robot spraying operation execution

[0013] The intelligent robot starts the spraying operation according to the planned spraying path and the set paint parameters; during the spraying process, the robot arm carries the spray gun and moves accurately along the predetermined path, and the spray gun sprays the paint evenly on the surface of the workpiece according to the set paint flow and pressure; the robot is equipped with a high-precision position sensor and a visual sensor. The position sensor feeds back the robot's position information in real time to ensure its movement accuracy; the visual sensor monitors the spraying surface in real time to detect the thickness and uniformity of the coating and whether there are any defects such as missed spraying and over-spraying;

[0014] Step 5: Spraying quality inspection and feedback adjustment

[0015] After spraying an area or the entire workpiece, the non-contact thickness measuring instrument and gloss meter testing equipment are used to conduct a comprehensive test on the spraying quality; the test data is transmitted back to the computer control system for comparison and analysis with the preset quality standards; if the test results do not meet the requirements, the computer control system automatically adjusts the spraying parameters according to the deviation, and directs the robot to re-spray or re-spray the unqualified areas until the quality standards are met;

[0016] Step 6: Data recording and process optimization

[0017] During the entire spraying process, the computer control system records all spraying data in detail, including workpiece information, spraying path, coating parameters, and test results; through statistical analysis of a large amount of spraying data, it explores the patterns and potential problems, continuously optimizes the spraying process parameters and path planning algorithms, and improves the overall performance of the intelligent robot spraying system and the stability of the spraying quality to adapt to different types of workpieces and changing production needs.

[0018] Preferably, the specific implementation steps of the workpiece information acquisition and model building in step 1 are as follows:

[0019] Step 11, 3D scanning preparation: Before scanning the workpiece, the 3D scanning equipment needs to be calibrated; place the workpiece to be sprayed on the scanning workbench, and select the appropriate placement position and posture so that the scanning equipment can fully obtain the 3D contour data of the workpiece;

[0020] Step 12, 3D scanning data acquisition: according to the size, shape, complexity and required accuracy of the workpiece, set the scanning parameters of the 3D scanning equipment, including scanning resolution, scanning speed and scanning range;

[0021] Step 13, data transmission: After the scanning is completed, the raw data collected by the 3D scanning device is transmitted to the computer control system through wired or wireless transmission; during the transmission process, the integrity and accuracy of the data must be ensured, and a data check and error retransmission mechanism is used to prevent data loss or transmission errors;

[0022] Step 14: Data preprocessing:

[0023] Point cloud registration: After receiving the scanned data, the computer control system first performs data preprocessing operations; preprocessing includes data format conversion, noise removal, and data simplification steps;

[0024] Step 15, 3D model construction: The scanned data after preprocessing exists in the form of point cloud. Since the scanned data is obtained from multiple perspectives, the position of the point cloud data in space is relatively independent. The point cloud data is unified into a common coordinate system. Specifically, feature-based registration and iterative closest point algorithm are used for point cloud registration. Feature-based registration first extracts representative geometric features from the point cloud, and then calculates the initial transformation matrix by matching the correspondence between these features in different point clouds, and preliminarily aligns the point clouds.

[0025] Surface reconstruction: After the point cloud registration is completed, the surface reconstruction operation is performed to convert the point cloud data into a continuous surface model. Specifically, the Delaunay triangulation algorithm is used to construct a triangular mesh that meets the properties of a specific empty circumscribed circle, and the point cloud data is connected into triangular patches to form an approximate surface model.

[0026] Model post-processing and marking: The constructed 3D surface model needs to be smoothed and defects repaired. The specific model smoothing adopts Laplace smoothing or mean curvature flow smoothing method to make the surface smoother and more natural, and reduce the jagged edges or uneven areas caused by data sampling or reconstruction process; some minor defects in the model, including isolated faces and self-intersection problems, are processed through automatic repair algorithms or manual editing tools to build an accurate 3D model of the workpiece; then, according to the spraying process requirements of the workpiece, mark the areas that need to be sprayed, the spraying requirements of different areas and the key data of the workpiece material information in the 3D model.

[0027] Preferably, the specific steps of the spraying path planning in step 2 are as follows:

[0028] Step 21: Initial division based on workpiece shape, size and spraying area

[0029] First, the computer control system reads the workpiece three-dimensional model data constructed in step 1, including the overall shape description and size parameters of the workpiece; determines the area to be sprayed according to the spraying process requirements; and divides the determined spraying area into multiple larger sub-areas according to certain rules;

[0030] Step 22: Spraying unit segmentation and spray gun accessibility analysis

[0031] For each larger sub-area, it is further divided into tiny spraying units. A grid-based division method is used to divide the plane area into uniform rectangular grids, and the curved surface area is adaptively divided into triangular grids or other curved surface grid units according to its curvature change; the size of the grid unit depends on the spraying range and accuracy requirements of the spray gun. The effective spraying diameter of the spray gun is set to D. In order to ensure uniform coverage, the side length or diagonal length of each grid unit can be set to d=kD, where k is a coefficient less than 1;

[0032] For each tiny spray unit, the computer control system combines the robot's kinematic model to perform a spray gun reachability analysis. The robot's kinematic model describes the relationship between the robot's joint angles and the spray gun's spatial position and posture. Through the kinematic forward solution, the range of spatial positions that the spray gun can reach under different joint angle combinations is calculated. For each spray unit, determine whether there is at least one robot joint angle configuration that allows the spray gun to reach the center of the unit and cover the entire unit with a suitable posture. If there is an unreachable spray unit, the robot's workspace or the placement of the workpiece needs to be adjusted or replanned;

[0033] Step 23: Path connection and optimization based on robot kinematic characteristics

[0034] After all reachable spraying units are determined, the movement path of the spray gun between these units is planned. Specifically, a graph search-based algorithm is used, in which each spraying unit is regarded as a node in the graph. The connecting edges between the nodes represent the possible paths for the spray gun to move from one unit to another. The path search algorithm is used to find the shortest path sequence from the starting spraying unit to all other units in the constructed graph. The evaluation function of the path search algorithm is f(n) = g(n) + h(n), where g(n) represents the actual cost from the starting node to the current node n, and h(n) represents the heuristic estimated cost from the current node n to the target node. By continuously expanding the nodes and selecting the node with the smallest f(n) value for exploration, an approximate shortest spraying path sequence is finally obtained.

[0035] Step 24: Calculate the movement speed of the spray gun

[0036] According to the flow characteristics of the paint, the spray pattern of the spray gun and the area of ​​the spray unit, the moving speed of the spray gun on each unit is calculated; let the flow rate of the paint be Q, the spray width of the spray gun be ω, and the area of ​​the spray unit be A. In order to ensure uniform coating thickness, the moving speed v of the spray gun is given by the formula Calculated, where t is the target coating thickness.

[0037] Preferably, the specific contents of the coating parameter setting and mixing in step 3 are as follows:

[0038] Determine the coating type: The computer control system first reads the workpiece information to obtain the workpiece material information marked in the model building step. Different materials have different requirements for the adhesion, corrosion resistance, and wear resistance of the coating, so the appropriate coating can be selected;

[0039] Paint viscosity setting: The computer control system determines the initial paint viscosity range based on the shape, size and spraying process of the workpiece; then determines the type of diluent required for a specific paint and the approximate dilution ratio range based on the information in the paint database; during the paint mixing process, the viscosity sensor monitors the viscosity of the paint in real time, compares the actual viscosity value with the set value, and automatically controls the diluent addition system to increase the amount of diluent added if the actual viscosity is higher than the set value; conversely, if the actual viscosity is lower than the set value, the amount of diluent added is reduced or a thickener is added until the viscosity reaches the set target range;

[0040] Paint flow setting: Calculate the paint flow based on the spraying area of ​​the workpiece, coating thickness requirements, and solid content of the paint. Assume that the total spraying area of ​​the workpiece is B, the required coating thickness is T, the solid content of the paint is C, the density of the paint is ρ, and the spraying time is t. The calculation formula for the paint flow Q is: During the spraying process, the flow sensor monitors the actual flow of the paint in real time;

[0041] Paint mixing process control: After the computer control system determines the paint type, viscosity, flow rate and other parameters, it sends instructions to the paint mixing system; the paint mixing system first selects the corresponding paint raw material barrel and diluent barrel according to the instructions, and starts the delivery pump to deliver the paint and diluent to the mixing tank according to the set ratio; during the mixing process, the flow sensor and viscosity sensor continue to work, and feed back real-time monitoring data to the computer control system, and the computer control system makes dynamic adjustments based on the feedback data.

[0042] Preferably, the robot spraying operation in step 4 is specifically performed in the following steps:

[0043] Step 41, the computer control system sends the planned spray path data to the controller of the intelligent robot in a specific instruction format. The data includes a series of target position coordinates of the spray gun in space and corresponding posture information, as well as the movement speed and acceleration requirements between each path point;

[0044] Step 42, after receiving the command, the robot controller calculates the control signal required by each joint motor according to its own kinematic model and dynamic model, wherein the kinematic model is used to convert the target position and posture in space into the joint angle value, and the dynamic model is used to calculate the joint torque or force required to achieve the desired movement according to the mass, inertia, and friction factor of the robot;

[0045] Step 43, the robot arm carrying the spray gun starts to move, and the position sensor, specifically including the joint encoder and the laser tracker, monitors the actual position information of the robot in real time; the joint encoder is installed on each joint axis of the robot, and indirectly obtains the position information of the end effector by measuring the rotation angle and speed of the joint; the laser tracker directly measures the absolute position coordinates of the spray gun in space; the actual position information is compared with the preset path position information, and the position deviation Δp=p is calculated actual -p target , where p actual is the actual position vector, p target is the target position vector;

[0046] Step 44: According to the position deviation, the robot controller uses the PID control algorithm to calculate the corrected joint control signal. The output of the PID control algorithm is: Where K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient. Through the proportional, integral and differential coefficients, the path tracking performance of the robot is optimized so that the robot can respond to path changes quickly and accurately during movement and maintain the stability of movement.

[0047] Step 45, the computer control system sends a control signal to the control system of the spray gun according to the set paint flow and pressure parameters, and when the robot moves along the spraying path, the spray gun sprays the paint evenly on the surface of the workpiece according to the set paint flow and pressure;

[0048] Step 46, visual sensors are installed on the surface of the spray gun and the surface of the robot arm, and the visual sensors monitor the spraying surface in real time; the captured image data is transmitted to the computer control system in real time for processing and analysis, and first image preprocessing is performed, including image grayscale, filtering and noise reduction, and image enhancement operations; then, the image processing algorithm is used to detect the thickness and uniformity of the coating and whether there are any defects such as missed spraying and over-spraying; when the visual sensor detects that there are defects in the coating, the defect information is fed back to the computer control system, and the computer control system takes corresponding measures according to the type and severity of the defect.

[0049] Preferably, the specific steps of the spraying quality detection and feedback adjustment in step 6 are as follows:

[0050] Step 51, when a region or the entire workpiece is sprayed, the intelligent robot moves the workpiece to a designated inspection region, or the inspection device moves to the location of the workpiece, and then the non-contact thickness measuring instrument starts to work, and the measuring instrument measures the thickness of the workpiece surface point by point or region by region according to a preset scanning path to obtain a large number of thickness data points; then the gloss meter is used to detect the glossiness of the coating surface, that is, the gloss meter measures at multiple locations on the workpiece surface to obtain glossiness value distribution data, and transmits the above data back to the computer control system through wired or wireless communication;

[0051] Step 52, after receiving the thickness data points and the gloss value distribution data, the computer control system first reads the preset quality standard data for the workpiece from the database, and then compares and analyzes the test data with the preset standard point by point and area by area, and calculates the thickness deviation of each test point or area. If the average thickness deviation of a certain area exceeds the allowable range and the standard deviation is large, it means that not only the average thickness of the area does not meet the requirements, but also the thickness uniformity is poor;

[0052] Step 53. Based on the deviation analysis results, the computer control system determines the spraying parameters that need to be adjusted; the computer control system sends the adjusted spraying parameters to the robot and the paint supply system; then the robot is directed to move to the unqualified area for re-spraying or re-spraying. During the re-spraying process, the robot operates according to the new parameters and the optimized path, and the visual sensor continues to monitor the re-spraying area in real time to ensure the re-spraying effect; after the re-spraying or re-spraying is completed, quality inspection and data comparison analysis are performed again, and the above steps are repeated until the spraying quality of the entire workpiece reaches the preset quality standard.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] High-precision spraying: By building an accurate workpiece model and intelligent path planning algorithm through 3D scanning, high-precision spray path planning can be achieved for workpieces with complex shapes, ensuring that the paint adheres evenly and accurately to the workpiece surface, improving the consistency and stability of the coating quality.

[0055] Intelligent parameter control: Automatically set and adjust coating parameters according to workpiece material, spraying requirements and environmental conditions, and monitor and feedback adjustments in real time during the spraying process, effectively avoiding paint waste and quality fluctuations caused by human factors, and improving production efficiency and coating utilization.

[0056] Real-time quality detection and correction: Equipped with visual sensors and other detection equipment to monitor the spraying quality in real time, it can detect spraying defects in time and automatically perform re-spraying or re-spraying, which greatly reduces the scrap rate and improves the reliability of product quality.

[0057] Process optimization and adaptability: By recording and analyzing a large amount of spraying data, the spraying process and path planning algorithm are continuously optimized, so that the intelligent robot spraying system can quickly adapt to different types of workpieces and changing production needs, thereby improving the flexibility and competitiveness of the entire production system. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] See also Figure 1 The present invention provides a technical solution: an intelligent robot spraying method based on a spraying system, comprising the following steps:

[0061] Step 1: Artifact information acquisition and model building

[0062] Use 3D scanning equipment to perform all-round scanning on the workpiece to be sprayed to obtain the 3D contour data of the workpiece; transmit the above data to the computer control system, process and analyze the data through special data processing software, and build an accurate 3D model of the workpiece. At the same time, mark the areas that need to be sprayed, the spraying requirements of different areas, and the key data of the material information of the workpiece in the model;

[0063] The specific implementation steps for artifact information acquisition and model construction are as follows:

[0064] Step 11, 3D scanning preparation: Before scanning the workpiece, the 3D scanning equipment needs to be calibrated. For example, for laser scanners, the accuracy of the laser beam emission angle, intensity and scanning range must be ensured; for scanning equipment based on structured light, the pattern clarity, brightness uniformity of the structured light projector and the focal length, aperture and other parameters of the camera need to be calibrated to ensure high-quality scanning data. The calibration process is usually carried out according to the calibration tools and standard operating procedures provided by the equipment manufacturer. It may be necessary to use a specific calibration plate or a standard object of known geometric shape. By scanning these standard objects and comparing the difference between the actual measured value and the standard value, the various parameters of the equipment are adjusted until the specified accuracy requirements are met; Place the workpiece to be sprayed on the scanning workbench, select the appropriate placement position and posture, so that the scanning equipment can fully obtain the 3D contour data of the workpiece. For some irregularly shaped or special structured workpieces, it may be necessary to design a special fixture or support device to ensure that the workpiece remains stable and does not move or deform during the scanning process. At the same time, set positioning marks or reference points on the workbench. These marks will help determine the coordinate system and orientation of the workpiece in the subsequent data processing process. For example, you can stick high-precision reflective stickers or place positioning blocks with special shapes at the four corners or specific locations of the workbench. These marks can be clearly identified in the scan data, thus providing a benchmark for data registration and model construction.

[0065] Step 12, 3D scanning data acquisition: According to the size, shape, complexity and required accuracy of the workpiece, set the scanning parameters of the 3D scanning device, including scanning resolution, scanning speed and scanning range. For smaller workpieces with higher accuracy requirements, such as precision mechanical parts or electronic components, a higher scanning resolution should be set, which may reach sub-millimeter or even micron level, and the scanning speed should be appropriately reduced to ensure data accuracy; for large workpieces, such as car bodies or large mechanical equipment shells, the resolution can be appropriately reduced and the scanning speed can be increased to improve scanning efficiency while ensuring that the overall shape and key features can be accurately captured. The scanning range needs to be adjusted according to the actual size of the workpiece to ensure that the entire workpiece can be included in the scanning field of view; at the same time, since most 3D scanning devices cannot obtain all the information of the workpiece in one scan, especially for workpieces with complex shapes and internal structures, scanning from multiple perspectives is required. When performing multi-perspective scanning, the position and posture of the scanning device are adjusted in sequence according to the pre-planned scanning path and angle sequence to perform a full-scale scan of the workpiece. For example, for a car engine cylinder with complex curved surfaces and internal cavities, first scan it from the top perspective to obtain the contour data of the cylinder top plane, cylinder port and other parts; then move the scanning device to the side and scan the cylinder side curved surface, heat sink and other structures at different heights and angles; finally, for the internal cavity, special endoscopic scanning equipment can be used or auxiliary tools such as reflectors can be placed inside the cavity to scan from a suitable entrance position to ensure that the contour data of the internal structure can also be fully collected. After each perspective switch, ensure that there is a certain overlap between the scanning areas so that the scanning data from different perspectives can be accurately spliced ​​together in subsequent data processing;

[0066] Step 13, data transmission: After the scanning is completed, the raw data collected by the 3D scanning device is transmitted to the computer control system through wired or wireless transmission; during the transmission process, the integrity and accuracy of the data must be ensured, and a data check and error retransmission mechanism is used to prevent data loss or transmission errors;

[0067] Step 14: Data preprocessing:

[0068] Point cloud registration: After receiving the scanned data, the computer control system first performs data preprocessing operations; preprocessing includes data format conversion, noise removal, and data simplification steps;

[0069] Step 15, 3D model construction: The pre-processed scan data exists in the form of point cloud. Since the scan data is obtained from multiple perspectives, the position of the point cloud data in space is relatively independent. The point cloud data is unified into a common coordinate system. Specifically, feature-based registration and iterative closest point algorithm are used for point cloud registration; feature-based registration first extracts representative geometric features from the point cloud, and then calculates the initial transformation matrix by matching the correspondence between these features in different point clouds, and preliminarily aligns the point clouds; among them, the commonly used point cloud registration methods are feature-based registration and iterative closest point (ICP) algorithm. Feature-based registration first extracts representative geometric features from the point cloud, such as corners, edges, etc., and then calculates the initial transformation matrix by matching the correspondence between these features in different point clouds, and preliminarily aligns the point clouds. The ICP algorithm is an iterative optimization method. It first gives an initial estimated transformation matrix, then continuously searches for the nearest point pair between the two point clouds, calculates the distance error between the corresponding points, and updates the transformation matrix by minimizing this error. After multiple iterations, the two point clouds are accurately registered. For example, for a workpiece point cloud data scanned from three different perspectives, the feature-based registration method is first used to find the approximate correspondence between the three point clouds to obtain an initial transformation matrix, and then the ICP algorithm is applied for fine registration. After multiple iterations, the three point clouds are accurately fused together to form a complete workpiece point cloud model;

[0070] Surface reconstruction: After the point cloud registration is completed, the surface reconstruction operation is performed to convert the point cloud data into a continuous surface model. Specifically, the Delaunay triangulation algorithm is used to construct a triangular mesh that meets the properties of a specific empty circumscribed circle, and the point cloud data is connected into triangular patches to form an approximate surface model.

[0071] Model post-processing and marking: The constructed 3D surface model needs to be smoothed and defects repaired. The specific model smoothing adopts Laplace smoothing or mean curvature flow smoothing method to make the surface smoother and more natural, and reduce the jagged edges or uneven areas caused by data sampling or reconstruction. Some minor defects in the model, including isolated patches and self-intersection problems, are processed through automatic repair algorithms or manual editing tools to build an accurate 3D model of the workpiece. Then, according to the spraying process requirements of the workpiece, mark the areas that need to be sprayed, the spraying requirements of different areas, and the key data of the workpiece material information in the 3D model.

[0072] Step 2: Spraying path planning

[0073] According to the three-dimensional model of the workpiece constructed in step 1, the computer control system uses an intelligent path planning algorithm to generate the robot's spray path. The algorithm comprehensively considers the shape, size, spray area and kinematic characteristics of the workpiece to ensure that the spray path is shortest, collision-free and can evenly cover the spray area. When planning the path, the workpiece surface is divided into multiple tiny spray units, and the optimal spray gun movement speed is calculated for each unit, so that the paint can be evenly attached to surfaces of different shapes;

[0074] The specific steps of spray path planning are as follows:

[0075] Step 21: Initial division based on workpiece shape, size and spraying area

[0076] First, the computer control system reads the three-dimensional model data of the workpiece constructed in step one, including the overall shape description and size parameters of the workpiece; the range of the area that needs to be sprayed is determined according to the spraying process requirements. For some complex workpieces, there may be some areas that do not need to be sprayed or require special treatment. For example, for a mechanical part with an internal cavity, the inside of the cavity may not need to be sprayed, so the external surface is identified and marked as the main spraying area; the determined spraying area is divided into multiple larger sub-areas according to certain rules. The basis for the division can be the geometric shape characteristics of the workpiece (such as plane areas, cylindrical areas, curved surface transition areas, etc.) or functional areas (such as the front, side, back, etc. of the product). The purpose of this step is to provide a macroscopic framework and hierarchy for subsequent more refined spray unit division and path planning;

[0077] Step 22: Spraying unit segmentation and spray gun accessibility analysis

[0078] For each larger sub-area, it is further divided into tiny spraying units. A grid-based division method is used to divide the plane area into uniform rectangular grids, and the curved surface area is adaptively divided into triangular grids or other curved surface grid units according to its curvature change; the size of the grid unit depends on the spraying range and accuracy requirements of the spray gun. The effective spraying diameter of the spray gun is set to D. In order to ensure uniform coverage, the side length or diagonal length of each grid unit can be set to d=kD, where k is a coefficient less than 1;

[0079] For each tiny spray unit, the computer control system combines the robot's kinematic model to perform a spray gun reachability analysis. The robot's kinematic model describes the relationship between the robot's joint angles and the spray gun's spatial position and posture. Through the kinematic forward solution, the range of spatial positions that the spray gun can reach under different joint angle combinations is calculated. For each spray unit, determine whether there is at least one robot joint angle configuration that allows the spray gun to reach the center of the unit and cover the entire unit with a suitable posture. If there is an unreachable spray unit, the robot's workspace or the placement of the workpiece needs to be adjusted or replanned;

[0080] Step 23: Path connection and optimization based on robot kinematic characteristics

[0081] After all reachable spraying units are determined, the movement path of the spray gun between these units is planned. Specifically, a graph search-based algorithm is used, in which each spraying unit is regarded as a node in the graph. The connecting edges between the nodes represent the possible paths for the spray gun to move from one unit to another. The path search algorithm is used to find the shortest path sequence from the starting spraying unit to all other units in the constructed graph. The evaluation function of the path search algorithm is f(n) = g(n) + h(n), where g(n) represents the actual cost from the starting node to the current node n, and h(n) represents the heuristic estimated cost from the current node n to the target node. By continuously expanding the nodes and selecting the node with the smallest f(n) value for exploration, an approximate shortest spraying path sequence is finally obtained.

[0082] Step 24: Calculate the movement speed of the spray gun

[0083] According to the flow characteristics of the paint, the spray pattern of the spray gun and the area of ​​the spray unit, the moving speed of the spray gun on each unit is calculated; let the flow rate of the paint be Q, the spray width of the spray gun be ω, and the area of ​​the spray unit be A. In order to ensure uniform coating thickness, the moving speed v of the spray gun is given by the formula Calculated, where t is the target coating thickness

[0084] Step 3: Paint parameter setting and mixing

[0085] According to the material of the workpiece, spraying requirements and environmental conditions, the computer control system automatically determines the appropriate paint type, paint viscosity, and paint flow parameters; during the paint mixing process, the flow sensor and viscosity sensor equipment connected to the paint supply system monitor the paint flow and viscosity in real time, and automatically adjust according to the set value to ensure that the mixing ratio and physical properties of the paint always meet the spraying requirements;

[0086] The specific contents of parameter setting and adjustment are as follows:

[0087] Determine the coating type: The computer control system first reads the workpiece information obtained and the workpiece material information marked in the model building step. Different materials have different requirements for the adhesion, corrosion resistance, and wear resistance of the coating, so the appropriate coating can be selected accordingly; for example, for metal workpieces, such as steel components, if a good anti-corrosion effect is required, epoxy primers containing anti-rust pigments such as zinc powder may be preferred; for plastic products, due to their low surface energy, it is necessary to select coatings with good affinity to the plastic surface, such as acrylic coatings or special plastic coatings, to ensure the adhesion of the coating;

[0088] Paint viscosity setting: The computer control system determines the initial paint viscosity range based on the shape, size and spraying process of the workpiece; then determines the type of diluent required for a specific paint and the approximate dilution ratio range based on the information in the paint database; during the paint mixing process, the viscosity sensor monitors the viscosity of the paint in real time, compares the actual viscosity value with the set value, and automatically controls the diluent addition system to increase the amount of diluent added if the actual viscosity is higher than the set value; conversely, if the actual viscosity is lower than the set value, the amount of diluent added is reduced or a thickener is added until the viscosity reaches the set target range;

[0089] Paint flow setting: Calculate the paint flow based on the spraying area of ​​the workpiece, coating thickness requirements, and solid content of the paint. Assume that the total spraying area of ​​the workpiece is B, the required coating thickness is T, the solid content of the paint is C, the density of the paint is ρ, and the spraying time is t. The calculation formula for the paint flow Q is: During the spraying process, the flow sensor monitors the actual flow of the paint in real time;

[0090] Paint mixing process control: After the computer control system determines the paint type, viscosity, flow rate and other parameters, it sends instructions to the paint mixing system; the paint mixing system first selects the corresponding paint raw material barrel and diluent barrel according to the instructions, and starts the delivery pump to deliver the paint and diluent to the mixing tank in the set ratio; during the mixing process, the flow sensor and viscosity sensor continue to work, and feed back the real-time monitoring data to the computer control system. The computer control system makes dynamic adjustments based on the feedback data. After the mixing is completed, the paint is briefly stirred and allowed to stand in the mixing tank to ensure that the paint is evenly mixed and there are no defects such as bubbles inside. The stirring time and standing time can be pre-set in the computer control system according to the type and characteristics of the paint. For example, for some water-based paints that are prone to bubble generation, the stirring time may be controlled at 5-10 minutes and the standing time at 2-3 minutes to ensure that the bubbles are fully released and to avoid bubble defects during the spraying process that affect the coating quality;

[0091] Step 4: Robot spraying operation execution

[0092] The intelligent robot starts the spraying operation according to the planned spraying path and the set paint parameters; during the spraying process, the robot arm carries the spray gun and moves accurately along the predetermined path, and the spray gun sprays the paint evenly on the surface of the workpiece according to the set paint flow and pressure; the robot is equipped with a high-precision position sensor and a visual sensor. The position sensor feeds back the robot's position information in real time to ensure its movement accuracy; the visual sensor monitors the spraying surface in real time to detect the thickness and uniformity of the coating and whether there are any defects such as missed spraying and over-spraying;

[0093] The specific steps for robot spraying operation are as follows:

[0094] Step 41, the computer control system sends the planned spray path data to the controller of the intelligent robot in a specific instruction format. The data includes a series of target position coordinates of the spray gun in space and corresponding posture information, as well as the movement speed and acceleration requirements between each path point;

[0095] Step 42, after receiving the command, the robot controller calculates the control signal required by each joint motor according to its own kinematic model and dynamic model, wherein the kinematic model is used to convert the target position and posture in space into the joint angle value, and the dynamic model is used to calculate the joint torque or force required to achieve the desired movement according to the mass, inertia, and friction factor of the robot;

[0096] Step 43, the robot arm carrying the spray gun starts to move, and the position sensor, specifically including the joint encoder and the laser tracker, monitors the actual position information of the robot in real time; the joint encoder is installed on each joint axis of the robot, and indirectly obtains the position information of the end effector by measuring the rotation angle and speed of the joint; the laser tracker directly measures the absolute position coordinates of the spray gun in space; the actual position information is compared with the preset path position information, and the position deviation Δp=p is calculated actual -p target , where p actual is the actual position vector, p target is the target position vector;

[0097] Step 44: According to the position deviation, the robot controller uses the PID control algorithm to calculate the corrected joint control signal. The output of the PID control algorithm is: Where K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient. Through the proportional, integral and differential coefficients, the path tracking performance of the robot is optimized so that the robot can respond to path changes quickly and accurately during movement and maintain the stability of movement.

[0098] Step 45, the computer control system sends a control signal to the control system of the spray gun according to the set paint flow and pressure parameters, and when the robot moves along the spraying path, the spray gun sprays the paint evenly on the surface of the workpiece according to the set paint flow and pressure;

[0099] Step 46. Install visual sensors on the surface of the spray gun and the surface of the robot arm, and use the visual sensors to monitor the spray surface in real time; the captured image data is transmitted to the computer control system in real time for processing and analysis, and first image preprocessing is performed, including image grayscale, filtering and noise reduction, and image enhancement operations; then, the image processing algorithm is used to detect the thickness and uniformity of the coating, as well as the presence of missed spraying and over-spraying defects; when the visual sensor detects defects in the coating, the defect information is fed back to the computer control system, and the computer control system takes corresponding measures according to the type and severity of the defect. Step 5. Spraying quality inspection and feedback adjustment

[0100] After spraying an area or the entire workpiece, the non-contact thickness measuring instrument and gloss meter testing equipment are used to conduct a comprehensive test on the spraying quality; the test data is transmitted back to the computer control system for comparison and analysis with the preset quality standards; if the test results do not meet the requirements, the computer control system automatically adjusts the spraying parameters according to the deviation, and directs the robot to re-spray or re-spray the unqualified areas until the quality standards are met;

[0101] The specific steps for spraying quality inspection and feedback adjustment are as follows:

[0102] Step 51, when a region or the entire workpiece is sprayed, the intelligent robot moves the workpiece to a designated inspection region, or the inspection device moves to the location of the workpiece, and then the non-contact thickness measuring instrument starts to work, and the measuring instrument measures the thickness of the workpiece surface point by point or region by region according to a preset scanning path to obtain a large number of thickness data points; then the gloss meter is used to detect the glossiness of the coating surface, that is, the gloss meter measures at multiple locations on the workpiece surface to obtain glossiness value distribution data, and transmits the above data back to the computer control system through wired or wireless communication;

[0103] Step 52, after receiving the thickness data points and the gloss value distribution data, the computer control system first reads the preset quality standard data for the workpiece from the database, and then compares and analyzes the test data with the preset standard point by point and area by area, and calculates the thickness deviation of each test point or area. If the average thickness deviation of a certain area exceeds the allowable range and the standard deviation is large, it means that not only the average thickness of the area does not meet the requirements, but also the thickness uniformity is poor;

[0104] Step 53. Based on the deviation analysis results, the computer control system determines the spraying parameters that need to be adjusted; the computer control system sends the adjusted spraying parameters to the robot and the paint supply system; then the robot is directed to move to the unqualified area for re-spraying or re-spraying. During the re-spraying process, the robot operates according to the new parameters and the optimized path, and the visual sensor continues to monitor the re-spraying area in real time to ensure the re-spraying effect; after the re-spraying or re-spraying is completed, quality inspection and data comparison analysis are performed again, and the above steps are repeated until the spraying quality of the entire workpiece reaches the preset quality standard.

[0105] Step 6: Data recording and process optimization

[0106] During the entire spraying process, the computer control system records all spraying data in detail, including workpiece information, spraying path, coating parameters, and test results; through statistical analysis of a large amount of spraying data, it explores the patterns and potential problems, continuously optimizes the spraying process parameters and path planning algorithms, and improves the overall performance of the intelligent robot spraying system and the stability of the spraying quality to adapt to different types of workpieces and changing production needs.

[0107] The present invention uses a three-dimensional scanning device to perform an all-round scan of the workpiece, and uses a special data processing software to build an accurate three-dimensional model, which can capture the shape, size and other information of the workpiece in great detail. The spray path planning based on this model comprehensively considers the complex geometric features of the workpiece, the kinematic characteristics of the robot and other factors, so that the spray gun can move on the surface of the workpiece with the best posture and path. For example, when processing workpieces with complex curved surfaces and fine structures, such as automobile engine cylinders or aerospace parts, the intelligent path planning algorithm can ensure that the paint is evenly and accurately covered to every detail, effectively avoiding the quality problems such as uneven coating thickness, missed spraying, and multiple spraying caused by the difficulty in controlling the angle and path of the spray gun by manual operation, greatly improving the consistency and stability of the coating quality, and meeting the requirements of high-precision spraying; the high-precision position sensor and visual sensor equipped in the spraying process play a key role. The position sensor feeds back the position information of the robot in real time, ensuring the high precision of the robot's movement, so that the spray gun can run strictly according to the predetermined path, and the error is controlled within a very small range. The visual sensor monitors the spraying surface in real time, and can detect the thickness, uniformity and presence of defects of the coating in time. Once a problem is found, such as the coating thickness in a certain area deviates from the set standard, the computer control system can automatically adjust the spraying parameters such as paint flow, spray gun speed, and the distance between the spray gun and the workpiece according to the deviation, and command the robot to accurately spray or re-spray the unqualified area. This real-time monitoring and feedback adjustment mechanism effectively ensures the uniformity and high standards of the coating quality on the entire workpiece surface, significantly reduces the scrap rate, and improves the overall quality reliability of the product.

[0108] According to the material of the workpiece, the spraying requirements and the environmental conditions, the computer control system of the present invention can automatically determine the appropriate parameters such as the type of coating, the viscosity of the coating, and the flow rate of the coating. This automated process avoids the subjectivity and uncertainty caused by the setting of parameters by manual experience in the traditional spraying process. For example, when facing workpieces of different materials, such as metal, plastic, composite materials, etc., the system can automatically match the most suitable coating formula according to the material characteristics to ensure that the coating has good adhesion and compatibility with the surface of the workpiece. At the same time, during the coating preparation process, the flow rate and viscosity of the coating are monitored in real time by connecting the flow sensor and viscosity sensor of the coating supply system, and automatically adjusted according to the set value, ensuring that the mixing ratio and physical properties of the coating always meet the spraying requirements. This intelligent parameter setting and preparation mechanism not only improves production efficiency, but also reduces the waste of coating and reduces production costs; the intelligent robot operates according to the planned spraying path and can complete the spraying task in an efficient manner. Since the path planning has fully considered the shortest path, collision-free and other factors, the robot does not need to frequently perform unnecessary start and stop and path adjustment during the spraying process, thereby greatly shortening the spraying time of a single workpiece. For example, when batch-producing the same type of workpieces, compared with traditional manual spraying or robot spraying without optimized paths, the method of the present invention can significantly improve production efficiency and meet the needs of large-scale production. In addition, the robot's automated operation can also achieve 24-hour uninterrupted operation, further improving production capacity and efficiency, and winning an advantage for enterprises in the fierce market competition.

[0109] To sum up, the intelligent robot spraying method based on the spraying system of the present invention has outstanding advantages in improving spraying quality, improving production efficiency, enhancing system adaptability, etc., and brings a new solution to spraying operations in modern industrial production. It has broad application prospects and significant economic and social benefits.

[0110] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent robot spraying method based on a spraying system, characterized in that: The following steps are involved: Step 1: Artifact information acquisition and model construction Use 3D scanning equipment to perform all-round scanning on the workpiece to be sprayed to obtain the 3D contour data of the workpiece; transmit the above data to the computer control system, process and analyze the data through special data processing software, and build an accurate 3D model of the workpiece. At the same time, mark the areas that need to be sprayed, the spraying requirements of different areas, and the key data of the material information of the workpiece in the model; Step 2: Spraying path planning According to the three-dimensional model of the workpiece constructed in step 1, the computer control system uses an intelligent path planning algorithm to generate the robot's spray path. The algorithm comprehensively considers the shape, size, spray area and kinematic characteristics of the workpiece to ensure that the spray path is shortest, collision-free and can evenly cover the spray area. When planning the path, the workpiece surface is divided into multiple tiny spray units, and the optimal spray gun movement speed is calculated for each unit, so that the paint can be evenly attached to surfaces of different shapes; Step 3: Paint parameter setting and mixing According to the material of the workpiece, spraying requirements and environmental conditions, the computer control system automatically determines the appropriate paint type, paint viscosity, and paint flow parameters; during the paint mixing process, the flow sensor and viscosity sensor equipment connected to the paint supply system monitor the paint flow and viscosity in real time, and automatically adjust according to the set value to ensure that the mixing ratio and physical properties of the paint always meet the spraying requirements; Step 4: Robot spraying operation execution The intelligent robot starts spraying according to the planned spraying path and set paint parameters; During the spraying process, the robot arm carries the spray gun and moves precisely along the predetermined path. At the same time, the spray gun sprays the paint evenly on the surface of the workpiece according to the set paint flow and pressure. The robot is equipped with high-precision position sensors and visual sensors. The position sensors provide real-time feedback on the robot's position information to ensure its movement accuracy. The visual sensor monitors the spraying surface in real time to detect the coating thickness, uniformity, and whether there are any defects such as missed spraying or over-spraying. Step 5: Spraying quality inspection and feedback adjustment After spraying an area or the entire workpiece, the non-contact thickness measuring instrument and gloss meter testing equipment are used to conduct a comprehensive test on the spraying quality; the test data is transmitted back to the computer control system for comparison and analysis with the preset quality standards; if the test results do not meet the requirements, the computer control system automatically adjusts the spraying parameters according to the deviation, and directs the robot to re-spray or re-spray the unqualified areas until the quality standards are met; Step 6: Data recording and process optimization During the entire spraying process, the computer control system records all spraying data in detail, including workpiece information, spraying path, coating parameters, and test results; through statistical analysis of a large amount of spraying data, it explores the patterns and potential problems, continuously optimizes the spraying process parameters and path planning algorithms, and improves the overall performance of the intelligent robot spraying system and the stability of the spraying quality to adapt to different types of workpieces and changing production needs.

2. The intelligent robot spraying method based on the spraying system according to claim 1 is characterized in that: The specific implementation steps of the artifact information acquisition and model construction in step 1 are as follows: Step 11, 3D scanning preparation: Before scanning the workpiece, the 3D scanning equipment needs to be calibrated; place the workpiece to be sprayed on the scanning workbench, and select the appropriate placement position and posture so that the scanning equipment can fully obtain the 3D contour data of the workpiece; Step 12, 3D scanning data acquisition: according to the size, shape, complexity and required accuracy of the workpiece, set the scanning parameters of the 3D scanning equipment, including scanning resolution, scanning speed and scanning range; Step 13, data transmission: after scanning is completed, the raw data collected by the 3D scanning device is transmitted to the computer control system via wired or wireless transmission; During the transmission process, the integrity and accuracy of the data must be ensured, and data verification and error retransmission mechanisms must be used to prevent data loss or transmission errors; Step 14: Data preprocessing: Point cloud registration: After receiving the scanned data, the computer control system first performs data preprocessing operations; preprocessing includes data format conversion, noise removal, and data simplification steps; Step 15, 3D model construction: The scanned data after preprocessing exists in the form of point cloud. Since the scanned data is obtained from multiple perspectives, the position of the point cloud data in space is relatively independent. The point cloud data is unified into a common coordinate system. Specifically, feature-based registration and iterative closest point algorithm are used for point cloud registration. Feature-based registration first extracts representative geometric features from the point cloud, and then calculates the initial transformation matrix by matching the correspondence between these features in different point clouds, and preliminarily aligns the point clouds. Surface reconstruction: After the point cloud registration is completed, the surface reconstruction operation is performed to convert the point cloud data into a continuous surface model. Specifically, the Delaunay triangulation algorithm is used to construct a triangular mesh that meets the properties of a specific empty circumscribed circle, and the point cloud data is connected into triangular patches to form an approximate surface model. Model post-processing and marking: The constructed 3D surface model needs to be smoothed and defects repaired. The specific model smoothing adopts Laplace smoothing or mean curvature flow smoothing method to make the surface smoother and more natural, and reduce the jagged edges or uneven areas caused by data sampling or reconstruction process; some minor defects in the model, including isolated faces and self-intersection problems, are processed through automatic repair algorithms or manual editing tools to build an accurate 3D model of the workpiece; then, according to the spraying process requirements of the workpiece, mark the areas that need to be sprayed, the spraying requirements of different areas and the key data of the workpiece material information in the 3D model.

3. The intelligent robot spraying method based on the spraying system according to claim 1 is characterized in that: The specific steps of the spraying path planning in step 2 are as follows: Step 21: Initial division based on workpiece shape, size and spraying area First, the computer control system reads the workpiece three-dimensional model data constructed in step 1, including the overall shape description and size parameters of the workpiece; determines the area to be sprayed according to the spraying process requirements; and divides the determined spraying area into multiple larger sub-areas according to certain rules; Step 22: Spraying unit segmentation and spray gun accessibility analysis For each larger sub-area, it is further divided into tiny spraying units. A grid-based division method is used to divide the plane area into uniform rectangular grids, and the curved surface area is adaptively divided into triangular grids or other curved surface grid units according to its curvature change; the size of the grid unit depends on the spraying range and accuracy requirements of the spray gun. The effective spraying diameter of the spray gun is set to D. In order to ensure uniform coverage, the side length or diagonal length of each grid unit can be set to d=kD, where k is a coefficient less than 1; For each tiny spray unit, the computer control system combines the robot's kinematic model to perform a spray gun reachability analysis. The robot's kinematic model describes the relationship between the robot's joint angles and the spray gun's spatial position and posture. Through the kinematic forward solution, the range of spatial positions that the spray gun can reach under different joint angle combinations is calculated. For each spray unit, determine whether there is at least one robot joint angle configuration that allows the spray gun to reach the center of the unit and cover the entire unit with a suitable posture. If there is an unreachable spray unit, the robot's workspace or the placement of the workpiece needs to be adjusted or replanned; Step 23: Path connection and optimization based on robot kinematic characteristics After all reachable spraying units are determined, the movement path of the spray gun between these units is planned. Specifically, a graph search-based algorithm is used, in which each spraying unit is regarded as a node in the graph. The connecting edges between the nodes represent the possible paths for the spray gun to move from one unit to another. The path search algorithm is used to find the shortest path sequence from the starting spraying unit to all other units in the constructed graph. The evaluation function of the path search algorithm is f(n) = g(n) + h(n), where g(n) represents the actual cost from the starting node to the current node n, and h(n) represents the heuristic estimated cost from the current node n to the target node. By continuously expanding the nodes and selecting the node with the smallest f(n) value for exploration, an approximate shortest spraying path sequence is finally obtained. Step 24: Calculate the movement speed of the spray gun According to the flow characteristics of the paint, the spray pattern of the spray gun and the area of ​​the spray unit, the moving speed of the spray gun on each unit is calculated; let the flow rate of the paint be Q, the spray width of the spray gun be ω, and the area of ​​the spray unit be A. In order to ensure uniform coating thickness, the moving speed v of the spray gun is given by the formula Calculated, where t is the target coating thickness.

4. The intelligent robot spraying method based on the spraying system according to claim 1 is characterized in that: The specific contents of the coating parameter setting and mixing in step 3 are as follows: Determine the coating type: The computer control system first reads the workpiece information to obtain the workpiece material information marked in the model building step. Different materials have different requirements for the adhesion, corrosion resistance, and wear resistance of the coating, so the appropriate coating can be selected; Paint viscosity setting: The computer control system determines the initial paint viscosity range based on the shape, size and spraying process of the workpiece; then determines the type of diluent required for a specific paint and the approximate dilution ratio range based on the information in the paint database; during the paint mixing process, the viscosity sensor monitors the viscosity of the paint in real time, compares the actual viscosity value with the set value, and automatically controls the diluent addition system to increase the amount of diluent added if the actual viscosity is higher than the set value; On the contrary, if the actual viscosity is lower than the set value, reduce the amount of diluent added or add thickener until the viscosity reaches the set target range; Paint flow setting: Calculate the paint flow based on the spraying area of ​​the workpiece, coating thickness requirements, and solid content of the paint. Assume that the total spraying area of ​​the workpiece is B, the required coating thickness is T, the solid content of the paint is C, the density of the paint is ρ, and the spraying time is t. The calculation formula for the paint flow Q is: During the spraying process, the flow sensor monitors the actual flow of the paint in real time; Paint mixing process control: After the computer control system determines the paint type, viscosity, flow rate and other parameters, it sends instructions to the paint mixing system; the paint mixing system first selects the corresponding paint raw material barrel and diluent barrel according to the instructions, and starts the delivery pump to deliver the paint and diluent to the mixing tank according to the set ratio; during the mixing process, the flow sensor and viscosity sensor continue to work, and feed back real-time monitoring data to the computer control system, and the computer control system makes dynamic adjustments based on the feedback data.

5. The intelligent robot spraying method based on the spraying system according to claim 1 is characterized in that: The specific execution steps of the robot spraying operation in step 4 are as follows: Step 41, the computer control system sends the planned spray path data to the controller of the intelligent robot in a specific instruction format. The data includes a series of target position coordinates of the spray gun in space and corresponding posture information, as well as the movement speed and acceleration requirements between each path point; Step 42, after receiving the command, the robot controller calculates the control signal required by each joint motor according to its own kinematic model and dynamic model, wherein the kinematic model is used to convert the target position and posture in space into the joint angle value, and the dynamic model is used to calculate the joint torque or force required to achieve the desired movement according to the mass, inertia, and friction factor of the robot; Step 43, the robot arm carrying the spray gun starts to move, and the position sensor, specifically including the joint encoder and the laser tracker, monitors the actual position information of the robot in real time; the joint encoder is installed on each joint axis of the robot, and indirectly obtains the position information of the end effector by measuring the rotation angle and speed of the joint; the laser tracker directly measures the absolute position coordinates of the spray gun in space; the actual position information is compared with the preset path position information, and the position deviation Δp=p is calculated actual -p target , where p catual is the actual position vector, p target is the target position vector; Step 44: According to the position deviation, the robot controller uses the PID control algorithm to calculate the corrected joint control signal. The output of the PID control algorithm is: Where K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient. Through the proportional, integral and differential coefficients, the path tracking performance of the robot is optimized so that the robot can respond to path changes quickly and accurately during movement and maintain the stability of movement. Step 45, the computer control system sends a control signal to the control system of the spray gun according to the set paint flow and pressure parameters, and when the robot moves along the spraying path, the spray gun sprays the paint evenly on the surface of the workpiece according to the set paint flow and pressure; Step 46: Visual sensors are installed on the surface of the spray gun and the surface of the robot arm, and the visual sensors monitor the spray surface in real time; the captured image data is transmitted to the computer control system in real time for processing and analysis, and image preprocessing is first performed, including image grayscale, filtering and noise reduction, and image enhancement operations; Then, the image processing algorithm is used to detect the thickness, uniformity, and whether there are any missed spraying or over-spraying defects in the coating. When the visual sensor detects defects in the coating, the defect information is fed back to the computer control system, and the computer control system takes appropriate measures based on the type and severity of the defects.

6. The intelligent robot spraying method based on the spraying system according to claim 1 is characterized in that: The specific steps of the spraying quality detection and feedback adjustment in step 6 are as follows: Step 51, when a region or the entire workpiece is sprayed, the intelligent robot moves the workpiece to a designated inspection region, or the inspection device moves to the location of the workpiece, and then the non-contact thickness measuring instrument starts to work, and the measuring instrument measures the thickness of the workpiece surface point by point or region by region according to a preset scanning path to obtain a large number of thickness data points; then the gloss meter is used to detect the glossiness of the coating surface, that is, the gloss meter measures at multiple locations on the workpiece surface to obtain glossiness value distribution data, and transmits the above data back to the computer control system through wired or wireless communication; Step 52, after receiving the thickness data points and the gloss value distribution data, the computer control system first reads the preset quality standard data for the workpiece from the database, and then compares and analyzes the test data with the preset standard point by point and area by area, and calculates the thickness deviation of each test point or area. If the average thickness deviation of a certain area exceeds the allowable range and the standard deviation is large, it means that not only the average thickness of the area does not meet the requirements, but also the thickness uniformity is poor; Step 53. Based on the deviation analysis results, the computer control system determines the spraying parameters that need to be adjusted; the computer control system sends the adjusted spraying parameters to the robot and the paint supply system; then the robot is directed to move to the unqualified area for re-spraying or re-spraying. During the re-spraying process, the robot operates according to the new parameters and the optimized path, and the visual sensor continues to monitor the re-spraying area in real time to ensure the re-spraying effect; after the re-spraying or re-spraying is completed, quality inspection and data comparison analysis are performed again, and the above steps are repeated until the spraying quality of the entire workpiece reaches the preset quality standard.

Images