A kind of automatic scanning painting method

Abstract

An automated scanning and painting method is provided in the embodiment of the present invention, which belongs to the field of operation technology, and specifically includes: using a depth camera installed on a mechanical arm to perform a rough scan of a target object, obtain initial point cloud data and pre-process it to generate a reference model; calculating the point cloud density and concavity of the reference model and merging them after normalization to obtain the evaluation index of each triangular facet area in the reference model and determine the secondary scanning attention points accordingly; performing a secondary scan according to the reference model and the secondary scanning attention points to obtain the secondary scanning results; performing precise modeling according to the secondary scanning results to generate a precise model corresponding to the target object; according to the generated precise model, for a variable curvature surface, optimizing and compensating the height, speed and stroke spacing of the spraying point, generating a motion path of the spraying equipment and spraying the target object accordingly. Through the scheme of the present invention, the spraying efficiency, real-time performance and adaptability are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of operation technology, and in particular to an automatic scanning painting method. Background Art

[0002] At present, among the applications of industrial robots today, painting robots are widely used in industrial production. Painting robots generally have multiple degrees of freedom, which gives the robot enough movement space to deal with various complex spraying targets. The uniformity of the paint on the surface of the product to be sprayed is not only related to its own performance, but also determined by the planned spraying trajectory. The application of spraying robots avoids harming the health of workers and improves economic benefits and spraying quality.

[0003] In the application of spray robot programming, the spraying effect of the traditional manual teaching method depends on the experience of the painter, is only applicable to a single product, is complex, time-consuming and has poor stability; the existing offline programming method requires inputting model information in advance and using specific algorithms to plan the spraying path. It can plan paths for workpieces of different shapes and different spraying requirements, but it cannot plan paths for special surfaces with large curvature changes, and it relies on prior models.

[0004] It can be seen that there is an urgent need for an automated scanning painting method that can improve spraying efficiency, real-time performance and adaptability. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides an automated scanning painting method, which at least partially solves the problems of poor painting efficiency, real-time performance and adaptability in the prior art.

[0006] An embodiment of the present invention provides an automated scanning painting method, comprising:

[0007] Step 1: Use the depth camera installed on the robotic arm to roughly scan the target object, obtain the initial point cloud data and pre-process it to generate a reference model;

[0008] Step 2, calculate the point cloud density and concavity of the reference model and normalize and merge them to obtain the evaluation index of each triangular patch area in the reference model and determine the attention points for the second scan based on it;

[0009] Step 3, performing a secondary scan according to the reference model and the secondary scan attention points to obtain a secondary scan result;

[0010] Step 4: Perform precise modeling based on the secondary scanning results to generate a precise model corresponding to the target object;

[0011] Step 5: Based on the generated precise model, the height, speed and stroke spacing of the spraying point are optimized and compensated for the variable curvature surface, and the motion path of the spraying equipment is generated to paint the target object accordingly.

[0012] According to a specific implementation of the embodiment of the present invention, step 1 specifically includes:

[0013] Step 1.1, control the robotic arm to drive the depth camera to move along the initial scanning trajectory to obtain initial point cloud data;

[0014] Step 1.2, convert the initial point cloud data to the base coordinate system;

[0015] Step 1.3, after downsampling and filtering the point cloud data in the base coordinate system, it is converted into a triangular mesh model using a preset algorithm and used as a reference model corresponding to the target object.

[0016] According to a specific implementation of the embodiment of the present invention, step 2 specifically includes:

[0017] Step 2.1, subdividing the reference model into n triangular facets;

[0018] Step 2.2: Evaluate the point cloud density of each point in the reference model within a spherical range with a given radius r, and obtain the point cloud density of each point as ;

[0019] Step 2.3, taking the local normal vector of each patch and the area near the triangular patch as the patch area, calculating and summing the consistency between the normal vectors in the patch area to obtain the concavity of the patch area surface;

[0020] Step 2.4, normalize the point cloud density and concavity and convexity and merge them to obtain the evaluation index of the patch area;

[0021] In step 2.5, the patch area with the evaluation index greater than the threshold is used as the attention point for the second scan.

[0022] According to a specific implementation of the embodiment of the present invention, the expression of the concavity is:

[0023] ;

[0024] in, is the local normal vector of the triangle patch, is the average normal vector of the triangle patch area, is the absolute value of the angle between the two normal vectors, It is the sum of the angles between the normal vector of the triangle patch and the normal vectors of the surrounding patches, and is used to measure the flatness and concavity of the surface of the patch area.

[0025] The expression of the evaluation index is:

[0026] ;

[0027] in is the weight coefficient, is the maximum and minimum regional point cloud density, are the maximum and minimum area convexity, Indicates the convexity of the area.

[0028] According to a specific implementation of the embodiment of the present invention, step 3 specifically includes:

[0029] Step 3.1, determine the number of sections and sampling intervals of the reference model;

[0030] Step 3.2, calculate the intersection lines of each section with the adjacent sections to obtain a continuous intersection line set;

[0031] Step 3.3, uniformly sampling at sampling intervals on the continuous intersection line set to obtain basic scanning points;

[0032] Step 3.4, based on the patch area corresponding to each secondary scanning attention point, equidistant sampling is performed on the triangular patch sampling points in the same patch area to obtain supplementary scanning points;

[0033] In step 3.5, the basic scanning points and the supplementary scanning points are combined to generate a secondary scanning trajectory and the robot arm is controlled to drive the depth camera to scan accordingly to obtain the secondary scanning result.

[0034] According to a specific implementation of the embodiment of the present invention, step 5 specifically includes:

[0035] Step 5.1, establishing a spray gun model and a curved paint film surface deposition model;

[0036] Step 5.2, optimizing the spraying height according to the surface curvature of the precise model;

[0037] Step 5.3, according to the spray gun model and the curved paint film deposition model, the objective function is established with the paint film thickness uniformity as the optimization target, the spraying speed and stroke spacing are randomly initialized, and the gradient method is used to solve the maximum value of the objective function to obtain the spraying speed and stroke spacing that can obtain the best paint layer uniformity;

[0038] Step 5.4, taking the solved spraying speed as the basic speed, the spraying speed is compensated according to the influence of the paint film formation rate caused by the optimization of the spraying height;

[0039] Step 5.5, taking the optimized spraying height, compensated spraying speed and stroke spacing as the optimal spraying parameters and accordingly performing spraying path slicing planning, generating the motion path of the spraying equipment and spraying the target object accordingly.

[0040] According to a specific implementation of an embodiment of the present invention, the expression for optimizing the spraying height is:

[0041] ;

[0042] in, For the optimized spraying height, Substitute the height h of the actual spray gun at the maximum curvature to meet the spraying approximation plane operation into the above formula to determine the calculated value, which is determined by actual experiments. represents the maximum curvature of the part, and are the maximum and minimum values ​​of the spray gun’s operating height range, and are the maximum and minimum values ​​of the curvature of the sprayed part surface respectively;

[0043] The expression of the objective function is

[0044] ;

[0045] in, is the spraying speed, is the spraying stroke spacing, is the long semi-axis of the boundary ellipse of the spray gun model at the minimum spray height of the spray route. is the paint film thickness distribution function at a certain point in the adjacent overlapping area, is the expected film thickness.

[0046] According to a specific implementation of the embodiment of the present invention, after step 5, the method further includes:

[0047] Obtain the parameters of the spraying results, analyze the average and variance of the paint film thickness, compare the expected results with the actual results, calculate the error data according to user requirements, and generate a spraying report.

[0048] The automated scanning and painting scheme in the embodiment of the present invention includes: step 1, using a depth camera installed on a robotic arm to perform a rough scan of the target object, obtain initial point cloud data and pre-process the data to generate a reference model; step 2, calculating the point cloud density and convexity of the reference model and merging them after normalization to obtain evaluation indicators for each triangular patch area in the reference model and determining the secondary scanning attention points accordingly; step 3, performing a secondary scan based on the reference model and the secondary scanning attention points to obtain the secondary scanning results; step 4, performing precise modeling based on the secondary scanning results to generate a precise model corresponding to the target object; step 5, based on the generated precise model, for variable curvature surfaces, optimizing and compensating the height, speed and stroke spacing of the spraying point, generating a motion path of the spraying equipment and spraying the target object accordingly.

[0049] The beneficial effects of the embodiments of the present invention are as follows: through the scheme of the present invention, a reference model acquisition link, a model analysis link, a secondary scanning trajectory planning link, a precise model building link, and a spray path planning link are set to jointly realize the full automation of the spraying process; a depth camera is used to roughly scan unknown objects and perform feature analysis to provide a scanning focus for subsequent laser scanner precise modeling; for variable curvature surfaces, combined with the selected spray gun model and paint surface generation model, the uniformity of the paint film thickness distribution is used as the objective function to optimize the spraying speed, spraying height, and spacing distance, thereby improving the spraying uniformity, and improving the spraying efficiency, real-time and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A schematic diagram of a process flow of an automated scanning painting method provided by an embodiment of the present invention;

[0052] Figure 2 A schematic diagram of a reference model acquisition process provided by an embodiment of the present invention;

[0053] Figure 3 A schematic diagram of a model analysis process provided by an embodiment of the present invention;

[0054] Figure 4 A schematic diagram of a secondary scanning path planning process provided by an embodiment of the present invention;

[0055] Figure 5 A schematic diagram of a spraying route planning process provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0057] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0058] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present invention, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this device and / or practice this method.

[0059] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention. The illustrations only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the form, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0060] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0061] An embodiment of the present invention provides an automated scanning painting method, which can be applied to the object spraying process in an industrial generation scene.

[0062] See also Figure 1 , is a flow chart of an automated scanning painting method provided by an embodiment of the present invention. Figure 1As shown, the method mainly comprises the following steps:

[0063] Step 1: Use the depth camera installed on the robotic arm to roughly scan the target object, obtain the initial point cloud data and pre-process it to generate a reference model;

[0064] Furthermore, the step 1 specifically includes:

[0065] Step 1.1, control the robotic arm to drive the depth camera to move along the initial scanning trajectory to obtain initial point cloud data;

[0066] Step 1.2, convert the initial point cloud data to the base coordinate system;

[0067] Step 1.3, after downsampling and filtering the point cloud data in the base coordinate system, it is converted into a triangular mesh model using a preset algorithm and used as a reference model corresponding to the target object.

[0068] In a specific implementation, the method of the embodiment of the present invention may be applied to a scanning painting system, wherein the scanning painting system may include a scanner module, an end effector motion module, a turntable, a scanned object, a measuring space, and a computer.

[0069] The above-mentioned scanner module includes: a scanner body, a depth camera, and a visible light source.

[0070] (a) Depth cameras (RGB-D cameras) have both depth detection and visible light imaging capabilities. Laser scanners have high-precision and high-resolution distance detection capabilities.

[0071] (b) Depth cameras and laser scanners have complementary detection capabilities. Specifically, depth cameras have a large measurement range, but poor measurement accuracy, which is still far from the accuracy required for the measurement task. Laser scanners can provide higher-precision scanning, but have a smaller measurement range.

[0072] like Figure 2 As shown in the figure, in the reference model acquisition link, the unknown object is scanned roughly for the first time using the depth camera installed on the robot arm, and the preliminary point cloud data is preprocessed, denoised, downsampled, smoothed, etc. to generate a preliminary object model. The point cloud obtained by multiple shots of the depth camera around the measurement space is spliced ​​and triangulated to obtain the reference model of the scanned object. The reference model provides basic geometric shape and structural information for subsequent analysis and processing. The specific process can be shown as follows:

[0073] (a) The robot arm drives the RGB-D camera at the end to move to point O at the bottom of the side of the measurement space cylinder, and the camera optical axis coincides with the radial direction of the measurement space cylinder. tableDefined as {P(x,y,z)|0 <z<h, x 2 +y 2 <d 2}, the starting scanning position of the scanning camera is O scaner_in_table =[-RL mid_DOF ,0,0], where R is the radius of the spraying table, L mid_DOF is the median depth of field of the RGB-D camera.

[0074] (b) The turntable rotates one circle and the point cloud {P in_scaner,time_stamp,0~n}.

[0075] (c) The robotic arm drives the RGB-D camera to rise along the axis of the measurement space cylinder (the distance at one end can be 0.8×sin(FOV RGB-D / 2)×L mid_DOF , where FOV RGB-D is the field of view of the RGB-D camera).

[0076] (d) Repeat the above two steps until the camera optical axis is above the top of the measurement space cylinder.

[0077] (e) For each moment, the point cloud collected by the RGB-D camera is transformed from F scaner Convert to F table Next: {P in_table,time_stamp,0~n =T world_in_table,time_stamp T robot_base_in_world T scaner_in_robot_base,time_stamp {P in_scaner,time_stamp,0~n},in:

[0078] {P in_scaner,time_stamp,0~n} is the point cloud at time_stamp in F scaner The coordinates below;

[0079] T world_in_table,time_stamp is time_stamp moment F world to F table SE3;

[0080] T robot_base_in_world is time_stamp moment F robot_base to F world SE3;

[0081] T scaner_in_robot_base,time_stamp is time_stamp moment F scaner to F robot_base SE3;

[0082] {P in_table,time_stamp,0~n} is the point cloud at time_stamp in F table The coordinates of the point cloud at each moment are then merged to obtain {P in_table,0~m}

[0083] (f) After downsampling and filtering the point cloud, the Poisson reconstruction algorithm, Marching Cubes algorithm, Ball Pivoting algorithm and other methods are used to convert {P in_table,0~m}Convert it into a triangular facet model, that is, obtain the reference model of the scanned object.

[0084] Step 2, calculate the point cloud density and concavity of the reference model and normalize and merge them to obtain the evaluation index of each triangular patch area in the reference model and determine the attention points for the second scan based on it;

[0085] Based on the above embodiment, step 2 specifically includes:

[0086] Step 2.1, subdividing the reference model into n triangular facets;

[0087] Step 2.2: Evaluate the point cloud density of each point in the reference model within a spherical range with a given radius r, and obtain the point cloud density of each point as ;

[0088] Step 2.3, taking the local normal vector of each patch and the area near the triangular patch as the patch area, calculating and summing the consistency between the normal vectors in the patch area to obtain the concavity of the patch area surface;

[0089] Step 2.4, normalize the point cloud density and concavity and convexity and merge them to obtain the evaluation index of the patch area;

[0090] In step 2.5, the patch area with the evaluation index greater than the threshold is used as the attention point for the second scan.

[0091] Furthermore, the expression of the concavity is:

[0092] ;

[0093] in, is the local normal vector of the triangle patch, is the average normal vector of the triangle patch area, is the absolute value of the angle between the two normal vectors, It is the sum of the angles between the normal vector of the triangle patch and the normal vectors of the surrounding patches, and is used to measure the flatness and concavity of the surface of the patch area.

[0094] The expression of the evaluation index is:

[0095] ;

[0096] in is the weight coefficient, is the maximum and minimum regional point cloud density, are the maximum and minimum area convexity, Indicates the convexity of the area.

[0097] When implementing it, Figure 3 As shown, in the model analysis phase, the system extracts key features such as edges, concave and convex areas, important geometric structures and texture information based on the reference model. These features help identify complex parts of objects and provide a basis for subsequent detailed scanning and processing. Based on the features extracted in the model analysis phase, a path strategy for laser scanning is formulated. This phase involves planning the scanning path of the laser scanner based on the geometric features and complexity of the object to ensure comprehensive and efficient acquisition of detailed data on the surface of the object. The path planning strategy needs to take into account factors such as scanning accuracy, coverage, and avoiding blind spots. The specific steps can be as follows:

[0098] (a) The generated triangular mesh model consists of n triangular facets, represented by {F 1 ~F n} represents a set of triangular faces, and {A 1 ~A n}、{B 1 ~B n}、{C 1 ~C n} represents the set of vertices of a triangular patch, and {P 1 ~P n} represents the set of centroids of triangular facets, and {N 1 ~N n} represents the set of normal vectors of the triangle patch.

[0099] (b) Evaluate the point cloud density of each point in the generated model within a sphere with a given radius r, and record the point cloud density of each point as .

[0100] (c) Perform a concavity analysis on the model. Perform a consistency analysis on the normal vectors of the mesh facets within a certain range. The consistency between the local normal vector of each facet and the normal vectors of the area near the facet is calculated and summed to determine the concavity and convexity of the surface of the area. The definition is as follows:

[0101] ,

[0102] ;

[0103] in is the local normal vector of the patch, is the average normal vector of the patch area, is the absolute value of the angle between the two normal vectors, It is the sum of the angles between the normal vector of a face and the normal vectors of the surrounding faces, and is used to measure the flatness and concavity of the surface of the area.

[0104] (d) The two index parameters are normalized and combined to obtain the evaluation index of the area:

[0105] ;

[0106] in is the weight coefficient, is the maximum and minimum regional point cloud density, is the maximum and minimum regional point cloud density;

[0107] (e) For the value function In the subsequent secondary scanning, priority is given to the areas with lower point cloud density and higher concavity.

[0108] Step 3, performing a secondary scan according to the reference model and the secondary scan attention points to obtain a secondary scan result;

[0109] Based on the above embodiment, step 3 specifically includes:

[0110] Step 3.1, determine the number of sections and sampling intervals of the reference model;

[0111] Step 3.2, calculate the intersection lines of each section with the adjacent sections to obtain a continuous intersection line set;

[0112] Step 3.3, uniformly sampling at sampling intervals on the continuous intersection line set to obtain basic scanning points;

[0113] Step 3.4, based on the patch area corresponding to each secondary scanning attention point, equidistant sampling is performed on the triangular patch sampling points in the same patch area to obtain supplementary scanning points;

[0114] In step 3.5, the basic scanning points and the supplementary scanning points are combined to generate a secondary scanning trajectory and the robot arm is controlled to drive the depth camera to scan accordingly to obtain the secondary scanning result.

[0115] When implementing it, Figure 4As shown, in the secondary scanning path planning stage, the laser scanning path is designed and optimized according to the geometric characteristics and complexity of the object to ensure full coverage and efficient acquisition of detailed point cloud data, while avoiding scanning blind areas. According to the planned path strategy, the laser scanner is used to perform a secondary detailed scan of the object. In this stage, the laser scanner obtains high-density and high-precision point cloud data according to the predetermined path to supplement the detailed areas that were not covered by the initial scan. The specific process can be shown as follows:

[0116] 3.1 Scan point generation:

[0117] (a) A 3D OBB bounding box is established for the rough triangulated model, and the long axis of the bounding box is selected as the extension direction of the section.

[0118] (b) Determine the slice interval, i.e., the sampling pitch of the scanner, based on the scanner's field of view, scanner performance, and desired results.

[0119] (c) Calculate the number of bands = roundup((1 + overlap ratio) × distribution width / section interval).

[0120] (d) Calculate the (number of strips + 1) equal division point of the distribution interval of the projection points on the section normal vector, which is the intersection point of the section and the section normal vector.

[0121] (e) Find the intersection line between the section obtained in the previous step and the face to which the voxel instance belongs. The continuous intersection lines obtained by intersecting a section with multiple adjacent face patches are called a continuous intersection line set.

[0122] (f) Perform uniformly spaced sampling on the continuous intersection line set obtained in the previous step to obtain the basic scanning point P basic .

[0123] (g) For those who meet the screening conditions The patch area is sampled equidistantly on the triangular patch sampling points in the same area to obtain the supplementary scanning points P replenish .

[0124] 3.2 Scanning trajectory generation and implementation:

[0125] (a) The default shooting posture is the normal vector direction of the triangle patch, and the shooting distance is 1 / 2×(maximum depth of field + minimum depth of field).

[0126] (b) Using A Or Dijkstra and other path planning algorithms sort the scanning points to generate the scanning trajectory {T scaner_in_reference_model,}.

[0127] (c) Use Movit software to control the robotic arm to scan along the planned scanning path.

[0128] Step 4: Perform precise modeling based on the secondary scanning results to generate a precise model corresponding to the target object;

[0129] In specific implementation, the precise modeling phase uses the detailed point cloud data obtained from the secondary scan to accurately model the object. This phase includes further processing and fusion of the point cloud data to generate a high-precision 3D model. The precise modeling phase ensures that the detailed features and geometric structures of the object are accurately captured and reconstructed.

[0130] The same measures as those used in the reference model acquisition process (e) and (f) to obtain the rough point cloud are taken to construct an accurate triangulated mesh model of the unknown object.

[0131] Step 5: Based on the generated precise model, the height, speed and stroke spacing of the spraying point are optimized and compensated for the variable curvature surface, and the motion path of the spraying equipment is generated to paint the target object accordingly.

[0132] Based on the above embodiment, step 5 specifically includes:

[0133] Step 5.1, establishing a spray gun model and a curved paint film surface deposition model;

[0134] Step 5.2, optimizing the spraying height according to the surface curvature of the precise model;

[0135] Step 5.3, according to the spray gun model and the curved paint film deposition model, the objective function is established with the paint film thickness uniformity as the optimization target, the spraying speed and stroke spacing are randomly initialized, and the gradient method is used to solve the maximum value of the objective function to obtain the spraying speed and stroke spacing that can obtain the best paint layer uniformity;

[0136] Step 5.4, taking the solved spraying speed as the basic speed, the spraying speed is compensated according to the influence of the paint film formation rate caused by the optimization of the spraying height;

[0137] Step 5.5, taking the optimized spraying height, compensated spraying speed and stroke spacing as the optimal spraying parameters and accordingly performing spraying path slicing planning, generating the motion path of the spraying equipment and spraying the target object accordingly.

[0138] Further, the expression for optimizing the spraying height is:

[0139] ;

[0140] in, For the optimized spraying height, Substitute the height h of the actual spray gun at the maximum curvature to meet the spraying approximation plane operation into the above formula to determine the calculated value, which is determined by actual experiments. represents the maximum curvature of the part, and are the maximum and minimum values ​​of the spray gun’s operating height range, and are the maximum and minimum values ​​of the curvature of the sprayed part surface respectively;

[0141] The expression of the objective function is

[0142] ;

[0143] in, is the spraying speed, is the spraying stroke spacing, is the long semi-axis of the boundary ellipse of the spray gun model at the minimum spray height of the spray route. is the paint film thickness distribution function at a certain point in the adjacent overlapping area, is the expected film thickness.

[0144] Furthermore, after step 5, the method further includes:

[0145] Obtain the parameters of the spraying results, analyze the average and variance of the paint film thickness, compare the expected results with the actual results, calculate the error data according to user requirements, and generate a spraying report.

[0146] When implementing it, Figure 5 As shown in the figure, in the spraying path planning stage, according to the generated accurate model, for the variable curvature surface, the height, speed, and interval distance parameters of the spraying point are optimized and compensated, and the motion path of the spraying equipment is generated to ensure the uniformity of the paint film thickness during the spraying process. Based on the generated accurate model, the spraying path strategy is planned. This stage involves determining the motion path of the spraying equipment to ensure uniform and comprehensive spraying on the surface of the object to avoid missing or overlapping areas.

[0147] 5.1 Spraying model modeling

[0148] (a) The spray paint model is selected as the elliptical double β model, and the specific description form is:

[0149] ;

[0150] Where x and y are located in an elliptical area, that is, their boundary values ​​satisfy the following relationship:

[0151] ;

[0152] in, That is, the maximum value of the coating thickness per unit time, a and b are the major and minor semi-axes of the boundary ellipse respectively, and β 1 , β 2The different cumulative rates of coating thickness in the major and minor axis directions within the elliptical spraying area are described respectively, and the spray gun is selected to move along the minor axis direction of the ellipse.

[0153] (b) The surface deposition rate model is constructed using the geometric transformation method. The paint film thickness distribution model on the free surface is:

[0154] ;

[0155] in is the distance from the target point to the center of the spray gun, and the angle between the connecting line and the vertical direction , for The vertical distance from the reference plane constructed above the point to the spray gun, for The vertical distance from the point to the center of the spray gun, for The normal vector of the plane where the point lies and The angle between the point and the line connecting the center of the spray gun.

[0156] (c) Take the spraying trajectory and spray at a constant speed v along the Y-axis direction. The paint film thickness distribution function of a point S in a single-pass spraying stroke on the free-form surface is:

[0157] ;

[0158] in: It is the coating accumulation time at point S during the spraying process; That is, it is the maximum value of the coating thickness per unit time, a and b are the major and minor semi-axes of the boundary ellipse respectively. point

[0159] The different accumulation rates of coating thickness in the major and minor axis directions within the elliptical spraying area are described separately.

[0160] The paint film thickness distribution function for adjacent overlapping areas is as follows, where the paint film thicknesses produced by two adjacent sprayings at point S are q 1 ,q 2 , the spraying stroke interval is recorded as D, and the paint film thickness at point S can be obtained as:

[0161] ;

[0162] ;

[0163] ;

[0164] in , , It is the time required for two spraying tracks to pass through this point.

[0165] 5.2 Spraying path optimization strategy

[0166] (a) The height is adjusted according to the curvature of the workpiece. When the central angle is α=4°, the arc length is approximately equal to the chord length. Let the semi-major axis of the spraying area be a, then:

[0167] ;

[0168] The relationship between spraying height and curvature is as follows:

[0169] ;

[0170] Where P max is the maximum curvature of the part, Substitute the height h of the actual spray gun at the maximum curvature to meet the requirements of spraying approximately in a flat surface into the above formula and calculate the value determined by actual experiments.

[0171] Consider the operating height range of the spray gun ( ), the mapping relationship between spraying height and curvature is established as follows:

[0172] ;

[0173] in, For the optimized spraying height, Substitute the height h of the actual spray gun at the maximum curvature to meet the spraying approximation plane operation into the above formula to determine the calculated value, which is determined by actual experiments. represents the maximum curvature of the part, and are the maximum and minimum values ​​of the spray gun’s operating height range, and They are the maximum and minimum values ​​of the curvature of the sprayed part surface, respectively.

[0174] (b) Since the spraying speed of the nozzle is constant, the moving speed of the nozzle is proportionally compensated while the height changes to ensure the accuracy of the paint film deposition accumulation model of the generated surface, that is, .

[0175] (c) Taking the uniformity of paint film thickness as the optimization target, the other two key parameters affecting the coating thickness: spraying speed v and stroke spacing D are optimized.

[0176] Establish an objective function with average paint layer thickness as the optimization target:

[0177] ;

[0178] Establish an objective function with the standard deviation of coating thickness distribution as the optimization target:

[0179] ;

[0180] The overall objective function is defined as:

[0181] ;

[0182] in, is the spraying speed, is the spraying stroke spacing, is the long semi-axis of the boundary ellipse of the spray gun model at the minimum spray height of the spray route. is the paint film thickness distribution function at a certain point in the adjacent overlapping area, is the expected film thickness, v and D The selection of needs to satisfy the constraints.

[0183] With random initial v and D, the gradient method is used to solve the maximum value of the value function and obtain the parameters that can achieve the best paint uniformity. and .

[0184] (d) The speed obtained is is the basic speed, according to The speed of each spraying point is compensated by the change, and the speed curve of the spraying trajectory is interpolated and fitted by using interpolation methods such as polynomial interpolation, spline curve interpolation, and Hermite curve.

[0185] 5.3 Spraying path planning

[0186] The details are similar to the second scanning path planning step in step 3 and will not be repeated here.

[0187] 5.4 Control the robot arm to spray parts according to the planned path.

[0188] At the same time, after the spraying is completed, the parameters of the spraying results can be obtained experimentally, and the average and variance of the paint film thickness can be analyzed. The expected results and actual results can be compared, and the error data can be calculated according to user requirements to generate a spraying report.

[0189] The automated scanning and painting method provided in this embodiment improves the spraying quality in the spraying stage by setting a reference model acquisition link, a model analysis link, a secondary scanning trajectory planning link, a precise model building link, and a spraying path planning link, so that the robot's offline path planning does not require prior model information, and together realizes the full automation of the spraying process; adds an algorithm for scanning unknown objects and outputting unknown object features, including identifying areas with sparse point cloud density and areas with large surface unevenness of unknown objects, and providing reference for secondary planning scanning; an improved spraying trajectory optimization algorithm, for variable curvature surfaces, combined with a selected spray gun model and a paint surface generation model, takes the uniformity of paint film thickness distribution as the objective function, optimizes the spraying speed, spray gun height, and spraying interval distance, thereby improving spraying uniformity, spraying efficiency, real-time and adaptability.

[0190] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof.

[0191] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. An automated scanning painting method, characterized in that: include: Step 1: Use the depth camera installed on the robotic arm to roughly scan the target object, obtain the initial point cloud data and pre-process it to generate a reference model; Step 2, calculate the point cloud density and concavity of the reference model and normalize and merge them to obtain the evaluation index of each triangular patch area in the reference model and determine the attention points for the second scan based on it; Step 3, performing a secondary scan according to the reference model and the secondary scan attention points to obtain a secondary scan result; Step 4: Perform precise modeling based on the secondary scanning results to generate a precise model corresponding to the target object; Step 5: Based on the generated precise model, the height, speed and stroke spacing of the spraying point are optimized and compensated for the variable curvature surface, and the motion path of the spraying equipment is generated to paint the target object accordingly.

2. The method according to claim 1, characterized in that The step 1 specifically includes: Step 1.1, control the robotic arm to drive the depth camera to move along the initial scanning trajectory to obtain initial point cloud data; Step 1.2, convert the initial point cloud data to the base coordinate system; Step 1.3, after downsampling and filtering the point cloud data in the base coordinate system, it is converted into a triangular mesh model using a preset algorithm and used as a reference model corresponding to the target object.

3. The method according to claim 2, characterized in that The step 2 specifically includes: Step 2.1, subdividing the reference model into n triangular facets; Step 2.2: Evaluate the point cloud density of each point in the reference model within a spherical range with a given radius r, and obtain the point cloud density of each point as ; Step 2.3, taking the local normal vector of each patch and the area near the triangular patch as the patch area, calculating and summing the consistency between the normal vectors in the patch area to obtain the concavity of the patch area surface; Step 2.4, normalize the point cloud density and concavity and convexity and merge them to obtain the evaluation index of the patch area; In step 2.5, the patch area with the evaluation index greater than the threshold is used as the attention point for the second scan.

4. The method according to claim 3, characterized in that The expression of the concavity is: ； in, is the local normal vector of the triangle patch, is the average normal vector of the triangle patch area, is the absolute value of the angle between the two normal vectors, It is the sum of the angles between the normal vector of the triangle patch and the normal vectors of the surrounding patches, and is used to measure the flatness and concavity of the surface of the patch area. The expression of the evaluation index is: ； in is the weight coefficient, is the maximum and minimum regional point cloud density, are the maximum and minimum area convexity, Indicates the convexity of the area.

5. The method according to claim 4, characterized in that The step 3 specifically includes: Step 3.1, determine the number of sections and sampling intervals of the reference model; Step 3.2, calculate the intersection lines of each section with the adjacent sections to obtain a continuous intersection line set; Step 3.3, uniformly sampling at sampling intervals on the continuous intersection line set to obtain basic scanning points; Step 3.4, based on the patch area corresponding to each secondary scanning attention point, equidistant sampling is performed on the triangular patch sampling points in the same patch area to obtain supplementary scanning points; In step 3.5, the basic scanning points and the supplementary scanning points are combined to generate a secondary scanning trajectory and the robot arm is controlled to drive the depth camera to scan accordingly to obtain the secondary scanning result.

6. The method according to claim 5, characterized in that The step 5 specifically includes: Step 5.1, establishing a spray gun model and a curved paint film surface deposition model; Step 5.2, optimizing the spraying height according to the surface curvature of the precise model; Step 5.3, according to the spray gun model and the curved paint film deposition model, the objective function is established with the paint film thickness uniformity as the optimization target, the spraying speed and stroke spacing are randomly initialized, and the gradient method is used to solve the maximum value of the objective function to obtain the spraying speed and stroke spacing that can obtain the best paint layer uniformity; Step 5.4, taking the solved spraying speed as the basic speed, the spraying speed is compensated according to the influence of the paint film formation rate caused by the optimization of the spraying height; Step 5.5, taking the optimized spraying height, compensated spraying speed and stroke spacing as the optimal spraying parameters and accordingly performing spraying path slicing planning, generating the motion path of the spraying equipment and spraying the target object accordingly.

7. The method according to claim 6, characterized in that The expression for optimizing the spraying height is: ； in, For the optimized spraying height, Substitute the height h of the actual spray gun at the maximum curvature to meet the spraying approximation plane operation into the above formula to determine the calculated value, which is determined by actual experiments. represents the maximum curvature of the part, and are the maximum and minimum values ​​of the spray gun’s operating height range, and are the maximum and minimum values ​​of the curvature of the sprayed part surface respectively; The expression of the objective function is ； in, is the spraying speed, is the spraying stroke spacing, is the long semi-axis of the boundary ellipse of the spray gun model at the minimum spray height of the spray route. is the paint film thickness distribution function at a certain point in the adjacent overlapping area, is the expected film thickness.

8. The method according to claim 1, characterized in that After step 5, the method further comprises: Obtain the parameters of the spraying results, analyze the average and variance of the paint film thickness, compare the expected results with the actual results, calculate the error data according to user requirements, and generate a spraying report.

Images