A welding method, system, device and storage medium for block components

By acquiring node and point cloud data of components, welding points and paths are determined based on the fitted surface, solving the problem of cumbersome and inefficient welding process for block components, achieving precise welding and intelligent control, and improving welding quality and efficiency.

CN118513721BActive Publication Date: 2024-11-15安徽工布智造工业科技有限公司 +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410997547.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-24
Publication Date
2024-11-15
Estimated Expiration
2044-07-24

AI Technical Summary

Technical Problem

When welding a large number of small-sized block components, existing technologies suffer from cumbersome processes and low efficiency, making it difficult to accurately identify welding points and intelligently control welding posture.

Method used

By acquiring the node data and point cloud data of the components, the welding points are determined based on the fitted surface, a welding point table array is generated, the weld set is determined and the welding path is planned, and intelligent control is achieved using a processor.

Benefits of technology

It enables precise welding of block components, improving welding quality and efficiency, and enhancing adaptability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118513721B_ABST
    Figure CN118513721B_ABST
Patent Text Reader

Abstract

The present invention provides a welding method, system, device and storage medium for block components. The method is implemented based on a processor and includes: obtaining node data, the node data including component size; obtaining point cloud data of the component; the way of obtaining the point cloud data is related to the component size; obtaining a fitting surface based on the point cloud data; determining the welding points of the component based on the fitting surface, and generating a welding point table array; determining a weld set based on the welding point table array, the weld set including at least one weld, the weld type of the weld and the target point on the weld; determining a welding scheme based on the weld set and the welding point table array, the welding scheme including a welding path.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of welding technology, and in particular to a welding method, system, device and storage medium for block components. Background Art

[0002] With the development of industrial technology and intelligent manufacturing, the manufacturing industry is gradually realizing a high degree of automation and intelligence in the production process. As a basic process widely used in the manufacturing industry, welding efficiency and quality directly affect the overall performance and manufacturing cost of the product. In order to improve the accuracy, efficiency and adaptability of welding operations, the introduction of advanced robot welding technology, computer-aided design and programming, and visual recognition systems has become an inevitable trend in the development of the industry. However, at present, when welding small-sized and large-quantity block components, there are problems such as cumbersome processes and low efficiency.

[0003] Therefore, it is necessary to provide a welding method, system, device and storage medium for block components, which can accurately identify welding points, accurately plan the welding path of block components, and intelligently control the welding posture, so as to effectively improve the welding quality and welding efficiency. Summary of the invention

[0004] One or more embodiments of the present specification provide a method for welding a block component, which is implemented based on a processor and includes: obtaining node data, the node data including component size; obtaining point cloud data of the component; the method of obtaining the point cloud data is related to the component size; obtaining a fitting surface based on the point cloud data; based on the fitting surface, determining the welding points of the component and generating a welding point table array; based on the welding point table array, determining a weld set, the weld set including at least one weld, the weld type of the weld and a target point on the weld; based on the weld set and the welding point table array, determining a welding scheme, the welding scheme including a welding path.

[0005] One or more embodiments of the present specification provide a welding system for a block component, the system comprising: a first acquisition module, configured to acquire node data, the node data including component size; a second acquisition module, configured to acquire point cloud data of the component; the method of acquiring the point cloud data is related to the component size; a first determination module, configured to obtain a fitting surface based on the point cloud data; a second determination module, configured to determine the welding points of the component based on the fitting surface, and generate a welding point table array; a third determination module, configured to determine a weld set based on the welding point table array, the weld set including at least one weld, the weld type of the weld and a target point on the weld; a fourth determination module, configured to determine a welding scheme based on the weld set and the welding point table array, the welding scheme including a welding path.

[0006] One or more embodiments of the present specification provide a welding device for block components, which includes at least one memory and at least one processor, wherein the at least one memory is used to store computer instructions, and the at least one processor executes computer instructions or part of the instructions to implement the aforementioned welding method for block components.

[0007] One or more embodiments of the present specification provide a computer-readable storage medium, which stores computer instructions. When a computer reads the computer instructions, the computer executes the aforementioned method for welding a block component. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0009] Figure 1 is an exemplary module diagram of a welding system for a block-shaped component according to some embodiments of the present specification;

[0010] Figure 2 is an exemplary flow chart of a method for welding a block component according to some embodiments of the present specification;

[0011] Figure 3 is an exemplary flow chart of obtaining point cloud data of a component according to some embodiments of this specification;

[0012] Figure 4 is another exemplary flow chart of obtaining point cloud data of a component according to some embodiments of this specification;

[0013] Figure 5 is an exemplary flow chart for determining welding points of components according to some embodiments of this specification;

[0014] Figure 6 is an exemplary schematic diagram of a path generation model according to some embodiments of this specification;

[0015] Fig. 7A is an exemplary schematic diagram of center position data or weld seams of a component without missing corners according to some embodiments of the present specification;

[0016] Figure 7B is an exemplary schematic diagram of center position data of a chamfered component or a weld according to some embodiments of the present specification;

[0017] Figure 7C 1 is an exemplary schematic diagram of center position data of a filleted component or a weld according to some embodiments of the present specification. DETAILED DESCRIPTION

[0018] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0019] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0020] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0021] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0022] Figure 1 is an exemplary module diagram of a welding system for a block-shaped component according to some embodiments of the present specification.

[0023] The block component welding system 100 refers to a system for determining a welding scheme to implement a welding method for a block component. In some embodiments, the block component welding system 100 may include a first acquisition module 110, a second acquisition module 120, a first determination module 130, a second determination module 140, a third determination module 150, and a fourth determination module 160.

[0024] In some embodiments, the block component welding system 100 may include a processor. The processor may be used to process data related to the system. One or more of the first acquisition module 110, the second acquisition module 120, the first determination module 130, the second determination module 140, the third determination module 150, and the fourth determination module 160 may be integrated into the processor.

[0025] In some embodiments, the processor may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor may be local or remote. In some embodiments, the processor may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc., or any combination thereof.

[0026] The first acquisition module 110 is a module for acquiring node data. In some embodiments, the first acquisition module 110 may be configured to acquire node data, where the node data includes component size.

[0027] The second acquisition module 120 is a module for acquiring point cloud data. In some embodiments, the second acquisition module 120 may be configured to acquire point cloud data of a component, and the acquisition method of the point cloud data is related to the size of the component.

[0028] The first determination module 130 is a module for determining a fitting surface. In some embodiments, the first determination module 130 may be configured to obtain a fitting surface based on point cloud data.

[0029] The second determination module 140 is a module for determining welding points. In some embodiments, the second determination module 140 can be configured to determine welding points of a component based on the fitting surface and generate a welding point table array.

[0030] The third determination module 150 refers to a module for determining a weld set. In some embodiments, the third determination module 150 can be configured to determine a weld set based on a welding point table array. The weld set can include at least one weld, a weld type of the weld, and a target point on the weld.

[0031] The fourth determination module 160 is a module for determining a welding scheme. In some embodiments, the fourth determination module 160 may determine a welding scheme based on a weld set and a welding point table array, and the welding scheme may include a welding path.

[0032] For more information about the first acquisition module 110 , the second acquisition module 120 , the first determination module 130 , the second determination module 140 , the third determination module 150 and the fourth determination module 160 , please refer to the relevant description below.

[0033] It should be noted that the above description of the block component welding system 100 and its modules is only for the convenience of description and does not limit the present specification to the scope of the embodiments. It is understandable that for those skilled in the art, after understanding the principle of the system, it is possible to arbitrarily combine the modules or form a subsystem to connect with other modules without deviating from the principle. In some embodiments, Figure 1 The first acquisition module 110, the second acquisition module 120, the first determination module 130, the second determination module 140, the third determination module 150 and the fourth determination module 160 disclosed in the specification can be different modules in a system, or one module can realize the functions of two or more modules mentioned above. For example, each module can share a storage module, or each module can have its own storage module. Such variations are all within the protection scope of this specification.

[0034] Figure 2 is an exemplary flow chart of a method for welding a block component according to some embodiments of this specification. Figure 2 As shown, the process 200 includes the following steps. In some embodiments, the process 200 can be implemented based on a processor.

[0035] Step 210, obtaining node data.

[0036] A node refers to a process node in the component welding process. In some embodiments, the complete process of welding a component can be referred to as a node. For example, if there are 10 components to be welded, the complete process of welding the second component can be recorded as the second node.

[0037] A component is a single part or assembly that makes up an overall structure or system.

[0038] In this specification, the component is a quadrilateral block structure. The block structure may be in the shape of a cuboid or a cube, etc. For example, the block component may be a quadrilateral tile (e.g., a rectangular tile of 90 cm×60 cm×1 cm, a square tile of 80 cm×80 cm×0.8 cm, etc.).

[0039] This manual can achieve accurate welding of block components to main parts (such as flat metal plates, etc.) by determining the welding plan.

[0040] In some embodiments, the node data may include component dimensions.

[0041] Component size refers to the linear size of the component. In some embodiments, component size can refer to the size of the welding surface of the block component (the contact surface between the block component and the main part). For example, the length of the component and the width of the component (e.g., the length of a rectangular tile is 90 cm and the width is 60 cm), or the side length of the component (e.g., the side length of a square tile is 80 cm).

[0042] In some embodiments, the node data may also include node operating parameters.

[0043] The node operating parameters refer to the operating parameters of the relevant equipment in the welding process corresponding to the node. In some embodiments, the node operating parameters may include at least scanning parameters, visual recognition parameters, and weld parameters.

[0044] Scanning parameters refer to working parameters of a point cloud acquisition device used to scan a component. In some embodiments, the scanning parameters may include a scanning range, a scanning number, and a scanning angle.

[0045] Point cloud acquisition device refers to a device used to scan components to acquire point cloud data, such as a laser radar scanner, a depth camera, etc.

[0046] For more information about obtaining point cloud data, please refer to the relevant description below.

[0047] The scanning range refers to the spatial range that the point cloud acquisition device can scan and cover, for example, 100cm×100cm.

[0048] The number of scans refers to the number of times the point cloud acquisition device scans the component. In some embodiments, the point cloud acquisition device needs to complete a complete scan of the component within the number of scans.

[0049] The scanning angle refers to the angle at which the point cloud acquisition device scans the component. In some embodiments, the scanning angle can be the pitch angle and yaw angle of the point cloud acquisition device relative to the component.

[0050] Visual recognition parameters refer to parameters related to the operation of the visual system. In some embodiments, the visual recognition parameters may include edge sensitivity thresholds and image processing parameters.

[0051] The edge sensitivity threshold refers to a critical value of the sensitivity used to extract edge points in an image. In some embodiments, if the intensity change of a pixel point of an image exceeds the edge sensitivity threshold, the pixel point is an edge point.

[0052] Image processing parameters refer to the control parameters involved in the image processing process, such as filtering parameters.

[0053] Weld parameters refer to parameters related to the weld in the welding process. In some embodiments, the weld parameters may include weld size and weld groove form.

[0054] The weld size may include weld thickness, leg size of fillet welds, weld width of butt welds, etc. In some embodiments, the weld size must comply with the design drawings of component welding or relevant standard requirements.

[0055] The weld groove form may include a V-shaped groove, a U-shaped groove, etc. In some embodiments, the weld groove form may be selected and determined by a technician based on factors such as joint thickness, welding method, and material type.

[0056] In some embodiments of the present specification, by setting the working parameters of related nodes, users can conveniently call to meet process requirements and ensure welding quality.

[0057] In some embodiments, the processor can retrieve node data uploaded manually or set by system default, wherein the component size is manually input and uploaded, and the node working parameters can be manually input or set by system default.

[0058] In some embodiments, obtaining node data may further include: determining whether the node data satisfies a preset condition; and in response to not satisfying the preset condition, triggering a user warning instruction.

[0059] The preset condition refers to the preset relevant condition for judging whether the node data is correct. In some embodiments, the preset condition can be that the data type and parameter range of the node data are correct, and the required parameters are filled in completely.

[0060] The data type may be a data type in a programming language, such as a numeric type, a string type, etc.

[0061] The parameter range refers to the interval range of the numeric node data. For example, the parameter range of the scan number can be [1, 50].

[0062] Required parameters refer to the necessary parameters in the welding process, such as component size, etc.

[0063] In some embodiments, in response to the data type and parameter range of the node data being correct and the required parameters being fully filled in, the processor may determine that the node data meets the preset conditions; otherwise, the processor may trigger a user warning instruction.

[0064] User warning instructions are instructions that issue warnings to users. User warning instructions can be used to remind users that there are errors or missing node data. For example, when a required parameter is missing in the node data, the processor can trigger a user warning instruction to prompt the user to supplement the required parameter.

[0065] In some embodiments of the present specification, when there is node data that does not meet the preset conditions, a warning is issued to the user, and the user can be promptly reminded to check, verify and modify the supplementary node data, which can effectively prevent the situation where the node data cannot be used due to user input errors, and is beneficial to ensuring welding efficiency and welding quality.

[0066] Step 220: Acquire point cloud data of the component.

[0067] Point cloud data refers to a data set consisting of discrete points obtained by scanning a component by a point cloud acquisition device. In some embodiments, the point cloud data may be the three-dimensional position coordinates of multiple discrete points of the component in the coordinate system of the point cloud acquisition device itself (e.g., the radar coordinate system of a laser radar scanner).

[0068] In some embodiments, the point cloud data is acquired in a manner related to the size of the component.

[0069] In some embodiments, in response to the size of the component being greater than a size threshold, the component is divided into a plurality of sub-component regions based on the size of the component; based on the point cloud acquisition device, the plurality of sub-component regions are identified in a preset scanning order to obtain point cloud data of the component. For more information about this section, see Figure 3 and its related description.

[0070] In some embodiments, in response to the size of a component being smaller than a size threshold, a node legend is determined based on the size of the component; based on a collection device, the component is identified to obtain corner point position data of at least two corner points of the component, and the center position data of the component is further determined, and then the scanning position data is determined; based on a point cloud acquisition device, the component is identified according to the scanning position data to obtain the point cloud data of the component. For more information about this part, please refer to Figure 4 and its related description.

[0071] Step 230: obtaining a fitting surface based on the point cloud data.

[0072] The fitting surface refers to a plane fitted by the point cloud data. In some embodiments, the fitting surface may include a first fitting surface and a second fitting surface.

[0073] In some embodiments, the processor may determine a first fitting surface based on the point cloud data; and offset the first fitting surface based on the offset parameter to obtain a second fitting surface. For more information about this section, please refer to Figure 5 Related description.

[0074] Step 240: determine the welding points of the component based on the fitting surface, and generate a welding point table array.

[0075] Welding points refer to the corner points where components need to be welded to the main parts.

[0076] In some embodiments, the processor can determine the welding point based on the second fitting surface. For more information about this part, please refer to Figure 5 Related description.

[0077] The welding point table array refers to an array including information related to welding points. In some embodiments, the welding point table array may include a plurality of welding points and corresponding three-dimensional position coordinates.

[0078] In some embodiments, the welding point table array may include welding tip posture parameters.

[0079] The welding nozzle posture parameter refers to the posture parameter of the welding nozzle of the welding equipment (such as a welding gun) when welding. For example, the welding nozzle posture parameter may include the welding nozzle height and the welding nozzle angle.

[0080] The welding tip height refers to the height of the welding tip from the contact gap between the component and the main part. For example, 3mm.

[0081] The welding nozzle angle refers to the tilt angle of the welding nozzle. The welding nozzle angle can be expressed by the angle between the welding nozzle and the main part. For example, 45°.

[0082] In some embodiments, the processor may construct a welding feature vector based on weld size, material parameters, welding working parameters, main component inclination, and component thickness; and determine the welding nozzle posture parameters by retrieving a vector database based on the welding feature vector.

[0083] Material parameters refer to parameters related to the material of the welding workpiece, such as the material type of the component and the material type of the main part.

[0084] Welding operating parameters refer to operating parameters of welding equipment during welding. In some embodiments, welding operating parameters may include welding power source (e.g., arc welding power source, resistance welding power source, etc.), operating mode (e.g., pulse mode, constant current mode, constant voltage mode, etc.), welding power (e.g., 100W, etc.), welding gun type (e.g., arc welding gun, resistance welding gun, flame welding gun, etc.), etc.

[0085] In some embodiments, material parameters and welding process parameters may be manually input in advance.

[0086] The main part inclination refers to the inclination of the main part placement position. The main part inclination can be expressed by the angle between the main part placement surface and the horizontal direction. For example, 10°, etc.

[0087] In some embodiments, the processor may determine the body inclination based on the component image acquired by the acquisition device through a related algorithm (such as Hough transform, RANSAC, etc.). For more information about the acquisition device acquiring component images, please refer to Figure 4 Related description.

[0088] In some embodiments, the component thickness at each position of the component may be the same or different. For components with the same thickness at each position, the component thickness may be manually measured and input; for components with different thickness at each position, the component thickness may be obtained by the processor based on the point cloud data of the component.

[0089] The welding feature vector refers to a vector that can characterize welding features. In some embodiments, the welding feature vector may include weld size, material parameters, welding work parameters, main component inclination, and component thickness.

[0090] In some embodiments, the processor constructs a vector database based on historical data, and the vector database includes a plurality of reference welding feature vectors and corresponding reference welding nozzle posture parameters. A reference welding feature vector is constructed based on a historical weld size, historical material parameters, historical welding working parameters, historical main part inclination, and historical component thickness of a historical component. The reference welding nozzle posture parameters may be corresponding historical actual welding nozzle posture parameters (including historical actual welding nozzle height and historical actual welding nozzle angle).

[0091] In some embodiments, the processor may retrieve a reference welding feature vector having the smallest vector distance to the welding feature vector from a vector database, and determine the reference welding tip posture parameter corresponding thereto as the current welding tip posture parameter. The vector distance may be any one of Euclidean distance, cosine distance, etc. The distance threshold may be determined manually based on prior experience or historical data.

[0092] In some embodiments of the present specification, based on multiple parameters, the welding nozzle posture parameters are determined by vector database retrieval, and the weld size, material parameters, welding working parameters and other data selected by the user are fully considered, and the designed welding nozzle height and welding nozzle angle can achieve better welding effects.

[0093] In some embodiments, the processor may store the welding points and corresponding welding nozzle posture parameters to generate a welding point table.

[0094] In some embodiments, the processor may generate a welding point table array based on welding points, three-dimensional position coordinates of the welding points, and welding nozzle posture parameters.

[0095] Step 250, determining a weld set based on the welding point table array.

[0096] A weld set refers to a collection of welds. In some embodiments, the weld set includes at least one weld, a weld type, and a target point on the weld.

[0097] In some embodiments, the processor may determine the line connecting adjacent welding points as the location of the weld. Fig. 7A As shown, if point A, point B, point C, and point D represent welding points, the welding point table array is {[welding point A, ( ), (3mm, 45°)], [welding point B, ( ), (3mm, 45°)], [welding point C, ( ), (3mm, 45°)], [welding point D, ( ), (3mm, 45°)], there are 4 welds, represented by welds [AB), [BD), [DC), and [CA) respectively.

[0098] Among them, the weld type may include a fillet weld and a butt weld. In some embodiments, the processor may determine the weld type according to the connection method between the component and the main part. For example, the component is placed on the main part, the side surface of the component and the surface of the main part are perpendicular to each other to form a 90° angle, and the weld corresponding to the angle is a fillet weld; the component is placed on the main part and is located at the edge of the main part, the side surface of the component and the side surface of the main part form an end face, and the weld corresponding to the junction of the two surfaces is a butt weld.

[0099] For example, Fig. 7A As shown, the side surfaces of the components at the positions AB, DC, and CA all form an angle with the connection of the main component, and the side surface of the component at the position BD forms an end face with the connection of the main component. The weld types corresponding to the welds [AB), [BD), [DC), and [CA) can be fillet weld, butt weld, fillet weld, and fillet weld, respectively.

[0100] The target point refers to the point on the weld that needs to be welded. In some embodiments, the processor can automatically extract the point cloud data at the position of the weld as the target point.

[0101] In some embodiments, the target locations may include welding locations.

[0102] In some embodiments, the processor may determine a weld based on the welding points in the welding point table array, and determine a weld set based on the weld, the weld type, and the target point on the weld.

[0103] like Fig. 7A As shown in the figure, the target points on the weld [AB) include the welding point A and other points between point A and point B (e.g., A1, A2, ..., A t , t is a positive integer), excluding welding point B; the same is true for other welds, so the weld set can be expressed as {[AB)- fillet weld-(A, A1, A2, …, A t ), [BD)-butt welding-(B, B1, B2, ..., B v ), [DC)- fillet welding-(D, D1, D2, ..., D s ), [CA)- fillet welding-(C, C1, C2, ..., C w )}.

[0104] Step 260, determining a welding plan based on the weld set and the welding point table array.

[0105] A welding scheme refers to a scheme for welding a component to a main component. In some embodiments, a welding scheme may include a welding path.

[0106] The welding path refers to the movement path of the welding equipment for automated welding of components.

[0107] In some embodiments, the processor can connect the welds in the weld set corresponding to the same welding points end to end based on the weld set and the welding point table array, thereby determining the continuous order of the target points and automatically generating a welding path.

[0108] For example, the welds [AB), [BD), [DC), and [CA) corresponding to the same welding points are connected end to end. If the connection order of the welding points is ABDC starting from the weld [AB), the welding path can be: [A-A1-A2-…-A t -B-B1-B2-…-B v -D-D1-D2-…-D s -C-C1-C2-…-C w , (3mm, 45°)]; if starting from the weld [BD), the connection sequence of the welding points is BDCA, then the welding path can be: [B-B1-B2-…-B v -D-D1-D2-…-D s -C-C1-C2-…-C w -A-A1-A2-…-A t , (3mm, 45°)] etc.

[0109] In some embodiments, the processor can determine the welding path through the path generation model based on the material parameters of the point cloud data, the welding nozzle posture parameters, the welding gun type, the welding nozzle size parameters, the obstacle image, the environmental data, the attribute parameters of the target point and the weld type. For more information about this part, please refer to Figure 6 and its related description.

[0110] In some embodiments, process 200 may further include step 270 .

[0111] Step 270 , visually display the welding plan.

[0112] In some embodiments, the processor may use a drawing tool to automatically generate a legend for the welding path and parameter (eg, welding nozzle posture parameters) annotations to achieve a visual display of the welding plan.

[0113] In some embodiments of the present specification, the visual display of the welding plan allows the user to observe the welding plan in a timely and convenient manner, so as to further ensure the smooth progress of the welding work.

[0114] In some embodiments of the present specification, point cloud data of components is acquired through a point cloud acquisition device, welding positions are accurately determined, a welding point table array is determined based on the welding positions, and a welding path is further determined to achieve accurate identification of the welding position of the component, and fully consider the influence of weld size, material parameters, welding working parameters, etc. on the welding nozzle posture, so as to achieve accurate planning of the component welding path, so as to improve the accuracy, efficiency and adaptability of the welding operation.

[0115] Figure 3 FIG. 1 is an exemplary flow chart of obtaining point cloud data of a component according to some embodiments of this specification. Figure 3 As shown, the process 300 includes steps 310 to 320.

[0116] In some embodiments, in response to the component size being greater than a size threshold, the processor may perform steps 310 - 320 .

[0117] The size threshold refers to the maximum size of a single scanning component of the point cloud acquisition device. For example, the size threshold may include a preset length threshold and width threshold (or side length threshold) that are smaller than the scanning range. For example, the length threshold is 80 cm and the width threshold is 60 cm (or the side length threshold is 90 cm).

[0118] A component size greater than a size threshold means that the length of the component is greater than a length threshold or the width of the component is greater than a length threshold.

[0119] Step 310 , dividing the component into a plurality of sub-component regions based on the component size.

[0120] The sub-component region refers to a sub-region of a component. In some embodiments, the size of the sub-component region is less than a size threshold.

[0121] In some embodiments, the processor may divide the component into multiple sub-component regions smaller than a size threshold based on the component size. For example, if the scanning range is 100 cm×100 cm, the component size is 90 cm×60 cm, and the size threshold is 80 cm×60 cm, the processor may divide the component into two sub-component regions of 40 cm×30 cm.

[0122] Step 320 , based on the point cloud acquisition device, multiple sub-component areas are identified according to a preset scanning order to obtain point cloud data of the component.

[0123] The preset scanning order refers to the preset order in which the point cloud acquisition device scans the sub-component area, for example, from left to right, from top to bottom.

[0124] For more information about point cloud acquisition devices and point cloud data, see Figure 2 Related description.

[0125] In some embodiments, the processor can use a point cloud acquisition device to scan and identify multiple sub-component areas in a preset scanning order, acquire point cloud data of the multiple sub-component areas, and splice the point cloud data of the multiple sub-component areas to obtain point cloud data of the complete component.

[0126] In some embodiments, in response to the existence of overlapping areas among multiple sub-component areas, the processor can extract a first common feature point corresponding to the first sub-component area and a second common feature point corresponding to the second sub-component area in the overlapping area; determine a matching feature point pair based on the first common feature point and the second common feature point through a feature matching algorithm; determine a homography matrix based on the matching feature point pair; based on the homography matrix, convert the first feature point to the second coordinate system to obtain a third feature point; merge the second feature point and the third feature point to determine point cloud data.

[0127] The first sub-component area and the second sub-component area are two adjacent sub-component areas with an overlapping area. For example, sub-component area A and sub-component area B are two adjacent sub-component areas, and there is an overlapping area C, that is, the overlapping area C is included in both sub-component area A and sub-component area B, then sub-component area A can be the first sub-component area, sub-component area B can be the second sub-component area, or sub-component area B can be the first sub-component area, and sub-component area A can be the second sub-component area.

[0128] Common feature points refer to feature points in the overlapping area. For example, the pixel points whose grayscale value changes in the overlapping area C exceed the preset threshold are recorded as , ,……, , Is a positive integer.

[0129] The first common feature point refers to the common feature point in the first coordinate system corresponding to the first sub-component area. For example, the common feature point in the first coordinate system corresponding to the sub-component area ① (the first sub-component area) is recorded as , where the first common feature point Represents the common feature points in the first coordinate system , Can be 1 to Partial value within.

[0130] The first coordinate system refers to the coordinate system corresponding to the first sub-component area. For example, the first coordinate system can be a rectangular coordinate system constructed with the physical center of the first sub-component area as the coordinate origin and any three mutually perpendicular rays as the X axis, Y axis and Z axis.

[0131] The second common feature point refers to the common feature point in the second coordinate system corresponding to the second sub-component area. For example, the common feature point in the second coordinate system corresponding to the sub-component area ② (the second sub-component area) is recorded as Among them, the second common feature point Represents the common feature points in the second coordinate system , Think 1 to Partial value within.

[0132] The second coordinate system refers to the coordinate system corresponding to the second sub-component area. For example, the second coordinate system can be a rectangular coordinate system constructed with the physical center of the second sub-component area as the coordinate origin and any three mutually perpendicular rays as the X axis, Y axis and Z axis.

[0133] In some embodiments, the first common feature point and the second common feature point are common feature points of different coordinate systems of different sub-component regions in the same overlapping region, and the common feature points corresponding to the first common feature point and the second common feature point may be partially the same, that is, the first common feature point and the second common feature point In and The values ​​of can be partially the same (or completely the same). For example, To take 1, 2, etc. It can be 1, 3, etc., that is, the first common feature point and the second common feature point All correspond to common feature points ; Common feature points There is a corresponding first common feature point , no second common feature point; common feature point There is a corresponding second common feature point , no first common feature point, etc.

[0134] In some embodiments, the processor can extract point cloud data in the overlapping area of ​​the first sub-component area, determine its coordinates in the first coordinate system, as the first common feature point; extract point cloud data in the overlapping area of ​​the second sub-component area, determine its coordinates in the second coordinate system, as the second common feature point.

[0135] Feature matching algorithms refer to algorithms used in the field of computer vision to find and identify the same feature points in different images for matching, such as Scale-invariant Feature Transform (SIFT), Speeded Up Robust Features (SURF), OrientedFAST and Rotated BRIEF (ORB), etc.

[0136] A matching feature point pair refers to a matching pair consisting of a first common feature point and a second common feature point corresponding to the same common feature point. , ), representing the same common feature point The first common feature point and the second common feature point The matching feature point pairs are formed. No more than and not more than .

[0137] In some embodiments, the processor determines, through a feature matching algorithm, a first common feature point and a second common feature point representing the same common feature point to form a plurality of matching feature point pairs, and a first common feature point and a second common feature point representing the same common feature point to form a matching feature point pair.

[0138] A homography matrix is ​​a matrix that describes the position mapping relationship of an object between two different coordinate systems. In some embodiments, the coordinates of a point in a first coordinate system may be converted to a second coordinate system by applying the homography matrix.

[0139] In some embodiments, the processor may eliminate erroneous matching feature point pairs through a random sample consensus algorithm (RANSAC algorithm), and determine the homography matrix through coordinate calculation (such as direct linear transformation).

[0140] The first feature point refers to the feature point in the first coordinate system corresponding to the first sub-component area. The first feature point includes the first common feature point and other feature points of the first sub-component area in the non-overlapping area. For example, the feature point in the first coordinate system corresponding to the sub-component area ① (the first sub-component area) is recorded as , where the first feature point Represents the feature points in the first coordinate system , the first feature point Including the first common feature point , Greater than .

[0141] The second feature point refers to the feature point in the second coordinate system corresponding to the second sub-component area. The second feature point includes the second common feature point and other feature points of the second sub-component area in the non-overlapping area. For example, the feature point in the second coordinate system corresponding to the sub-component area ② (the second sub-component area) is recorded as , where the second feature point Represents the feature points in the second coordinate system , the second feature point Including the second common feature point , Greater than .

[0142] The third feature point refers to the feature point obtained by converting the first feature point to the second coordinate system, that is, the feature point in the second coordinate system corresponding to the first sub-component area. Convert the first coordinate system to the second coordinate system to obtain the third feature point .

[0143] In some embodiments, the third feature point and the corresponding first feature point represent the same feature point, but belong to different coordinate systems. and the first feature point Corresponding feature points , but the two belong to the second coordinate system and the first coordinate system respectively.

[0144] In some embodiments, the processor may transform the first feature point from the first coordinate system to the second coordinate system based on the homography matrix to obtain the third feature point.

[0145] In some embodiments, the third feature point and the second feature point are feature points in the same coordinate system (the second coordinate system), and the corresponding feature points may be partially identical. and the second feature point ,in, and The value part is the same.

[0146] In some embodiments, the processor may merge the second feature point and the third feature point to obtain all feature points of the first sub-component area and the second sub-component area in the second coordinate system (for the second feature point and the third feature point with the same coordinates or similar coordinates, representing the same feature point, either one of them may be retained), and determine them as point cloud data.

[0147] Similar coordinates may refer to coordinates with very small differences. For example, if the coordinate distance between the second feature point and the third feature point is less than a preset threshold (a very small distance preset manually), they may be identified as similar coordinates.

[0148] In some embodiments of the present specification, when there are overlapping areas among multiple sub-component areas, a feature matching algorithm is used to find common feature points in the overlapping area to determine matching feature point pairs and homography matrices, and then transform the coordinate system of one sub-component area to the coordinate system of another sub-component area, thereby achieving precise matching and splicing of points, forming gapless continuous contour data, and effectively improving the accuracy of point cloud data.

[0149] In some embodiments of the present specification, components whose sizes are greater than a size threshold are divided and based on a point cloud acquisition device, multiple sub-component areas are identified in a preset scanning order, fully taking into account the situation where a single scan cannot acquire point cloud data of large-sized components, making the method of acquiring point cloud data more complete and reliable.

[0150] Figure 4 FIG. 1 is another exemplary flow chart of obtaining point cloud data of a component according to some embodiments of this specification. Figure 4 As shown, process 400 includes steps 410 to 450.

[0151] In some embodiments, in response to the component size being less than a size threshold, the processor may perform steps 410 - 450 .

[0152] For an explanation of the size threshold, see Figure 3 Related description.

[0153] Step 410, determining a node legend based on the component size.

[0154] A node legend refers to a legend used as a visual identifier of a node. In some embodiments, a node legend can be configured to indicate the placement of a component and / or a collection device. For example, a node legend can be a quadrilateral outline.

[0155] In some embodiments, a plurality of node legends may be manually preset and uploaded to the processor.

[0156] The acquisition device refers to a device used to acquire component image data (including ordinary image data or point cloud data), such as a digital camera, a depth camera, etc.

[0157] In some embodiments, the node legend may include text annotations to guide the component and / or the acquisition device to a suitable position so that the acquisition device can obtain a complete component image. For example, the text annotation may be "preliminarily adjust the placement of the component or the acquisition device so that the component outline is included in the quadrilateral outline."

[0158] In some embodiments, the processor may automatically retrieve an appropriate node legend based on the size of the component.

[0159] Step 420: Based on the acquisition device, the component is identified to obtain corner point position data of at least two corner points of the component.

[0160] Component corners are points in the component image that are used to determine the location of the component. For example, the four intersection points formed by the four contour edges of the surface of a tiled quadrilateral (the four vertices of the quadrilateral).

[0161] Corner point position data refers to data that can reflect the position of the corner point of the component. In some embodiments, the corner point position data can be represented by the three-dimensional position coordinates of the corner point of the component in the camera coordinate system, which is expressed as ( ).

[0162] The camera coordinate system refers to the coordinate system used by the acquisition device when acquiring data. In some embodiments, the origin of the camera coordinate system can be the center of the acquisition device (the optical center of the camera), the X-axis and the Y-axis are respectively parallel to the two vertical sides of the component image acquired by the acquisition device, and the Z-axis is the axis perpendicular to the plane where the component image is located (the optical axis of the camera).

[0163] In some embodiments, the processor can photograph or scan the component based on the acquisition device to obtain a component image, and can obtain corner point position data of at least two corner points of the component based on the component image.

[0164] Step 430, determining the center position data of the component based on the corner point position data of at least two corner points of the component.

[0165] The center position data refers to data that can reflect the center point position of a component. In some embodiments, the center position data can be represented by the position coordinates of the geometric center of the component.

[0166] In some embodiments, the processor can connect two component corner points to determine the first line segment; translate the first line segment to determine the second line segment and two other component corner points of the component, the first line segment and the second line segment are parallel to each other and equal in length; connect the component corner points corresponding to the first line segment and the second line segment to determine the third line segment and the fourth line segment, the third line segment and the fourth line segment are parallel to each other and equal in length; connect the midpoints of the first line segment and the second line segment to determine the fifth line segment; connect the midpoints of the third line segment and the fourth line segment to determine the sixth line segment; determine the position data of the intersection of the fifth line segment and the sixth line segment as the center position data of the component.

[0167] The first line segment refers to the line segment formed by connecting two component corner points corresponding to a contour edge of the component. For example, if the component is a quadrilateral tile, the first line segment is the edge formed by connecting two vertices (component corner points) on one side of the rectangular surface of the quadrilateral tile (such as Fig. 7AAs shown in , two component corner points A and B are connected to determine the first line segment AB).

[0168] In some embodiments, when a component has missing corners, the processor can determine whether at least two corner points obtained are component corner points based on the component image. If the number of component corner points is less than two, the component is re-identified based on the acquisition device, and the corner point position data of at least two component corner points are re-obtained, the two component corner points are connected, and the first line segment is determined.

[0169] like Figure 7B As shown, the quadrilateral tile is chamfered to form a missing corner in the upper right corner of the surface. The image recognition is used to determine whether the angle corresponding to the acquired corner point (the inferior angle formed by the contour edge constituting the corner point) exceeds 90°. If so, it is an angle formed by the chamfer, that is, a non-component corner point (e.g., corner E and corner F are non-component corner points). The component corner points are determined to determine whether the number of component corner points among the collected corner points is not less than two. If less than two, the component corner points are collected again until there are at least two component corner points among the collected corner points. The first line segment is determined based on the two component corner points (e.g., the corner point position data of component corner point A and component corner point B are re-identified to determine the first line segment AB).

[0170] The second line segment is a line segment that is parallel to the first line segment and overlaps another contour edge of the component. In some embodiments, the processor can translate the first line segment, determine the second line segment, and determine the two endpoints of the second line segment as the other two component corner points of the component. The first line segment and the second line segment are parallel to each other and have equal lengths. Fig. 7A As shown, the first line segment AB is translated to obtain the second line segment CD, and point C and point D are the other two corner points of the component.

[0171] In some embodiments, the moving distance of translating the first line segment may be determined by the size of the component. For example, the moving distance of translating the first line segment may be the distance between the contour edge and the opposite contour edge of the component corresponding to the first line segment. Fig. 7A As shown, the first line segment AB actually corresponds to a contour edge a on the surface of the quadrilateral tile, and the opposite side of edge a is edge d. The moving distance is the distance between edge a and edge d (i.e., the length of contour edge c or contour edge b).

[0172] In some embodiments, when a component has a missing corner, the second line segment does not completely coincide with the contour edge of the component, that is, the length of the second line segment is greater than the length of the corresponding contour edge. Figure 7B As shown, the first line segment AB is translated to obtain the second line segment CD, the translation distance is the length of the contour edge c, the length of the second line segment CD is greater than the length of the corresponding contour edge d, and the two endpoints of the second line segment CD are the other two component corner points C and D of the component, wherein the component corner point D is the intersection of the extension lines of the contour edge d and the contour edge b.

[0173] In some embodiments, the processor may connect the component corner points corresponding to the first line segment and the second line segment to determine the third line segment and the fourth line segment, wherein the third line segment and the fourth line segment are parallel to each other and have the same length. Fig. 7A or Figure 7B As shown, the component corner point A of the first line segment AB and the corresponding component corner point C of the second line segment CD are connected to form a third line segment AC; the component corner point B of the first line segment AB and the corresponding component corner point D of the second line segment CD are connected to form a fourth line segment BD. The third line segment AC and the fourth line segment BD are parallel and equal.

[0174] In some embodiments, the processor may connect the midpoints of the first line segment and the second line segment to determine a fifth line segment; connect the midpoints of the third line segment and the fourth line segment to determine a sixth line segment, and the fifth line segment and the sixth line segment intersect. Fig. 7A or Figure 7B or Figure 7C As shown, connect the midpoint of the first line segment AB and the midpoint of the second line segment CD , forming the fifth line segment ; Connect the midpoint of the third line segment AC and the midpoint of the fourth line segment BD , forming the sixth line segment ; The fifth line segment and the sixth line segment Intersect at the intersection .

[0175] In some embodiments, the processor can determine the position data of the intersection of the fifth line segment and the sixth line segment by using a coordinate formula based on the corner point position data of at least two component corner points as the center position data of the component. For example, by using the position coordinates of component corner points A and B in the camera coordinate system, the position coordinates of component corner points C and D are calculated by using a coordinate formula to further determine the midpoint , , , The position coordinates of the intersection are finally calculated The position coordinates in the camera coordinate system are determined as the center position data of the component and recorded as ( ).

[0176] In some embodiments of the present specification, based on the corner point position data of at least two component corner points, the center position data of the component can be quickly and accurately determined through point and line calculation, which is conducive to the precise determination of subsequent scanning position data.

[0177] Step 440, determining scanning position data based on the central position data.

[0178] The scanning position data refers to the position data of the point cloud acquisition device when scanning the component. In some embodiments, the scanning position data can be represented by the three-dimensional position coordinates of the point cloud acquisition device in the world coordinate system, which is recorded as ( ).

[0179] The world coordinate system refers to an absolute coordinate system that remains fixed after being specified. For example, the world coordinate system can be an existing coordinate system such as the Universal Transverse Mercator (UTM) coordinate system, the geocentric inertial coordinate system, or a fixed coordinate system constructed with any point in space as the origin and any three mutually perpendicular directions as the X-axis, Y-axis, and Z-axis, or other coordinate systems pre-constructed by humans or the system.

[0180] In some embodiments, the processor may determine the scanning position data through coordinate conversion based on the center position data. For example, the processor may convert the center position data from the camera coordinate system to the world coordinate system through camera calibration parameters, and determine the center position data in the world coordinate system as the scanning position data. Among them, the camera calibration parameters may include camera extrinsic parameters (parameters related to the position and direction of the camera in the world coordinate system, such as rotation matrix, translation matrix, etc.), and the camera calibration parameters may be directly retrieved by the processor from the acquisition device.

[0181] Step 450 , based on the point cloud acquisition device, the component is identified according to the scanning position data to obtain the point cloud data of the component.

[0182] In some embodiments, the processor can control the point cloud acquisition device to move to the scanning position data, directly scan and identify the component, and acquire point cloud data.

[0183] In some embodiments, the processor may identify the component according to the scanning angle, scanning path and scanning position data based on the point cloud acquisition device.

[0184] For more information about scanning angles, see Figure 2 Related description.

[0185] The scanning path refers to the moving path of the point cloud acquisition device to scan and identify the component. For example, the midpoint of the scanning position data and the contour edge of the component is selected as the scanning point, and the scanning path is formed by moving to the scanning points in sequence. The movement methods include but are not limited to linear movement, smooth movement, spiral movement, etc. Fig. 7A As shown, the scan path can be - - - - Move smoothly in sequence.

[0186] In some embodiments, the processor may utilize a path planning algorithm to smoothly connect adjacent component corner points to generate a scanning path.

[0187] like Fig. 7A As shown, the processor can smoothly connect the component corner points ABDC to generate a scanning path.

[0188] The path planning algorithm may include but is not limited to linear interpolation, spline curve interpolation, etc.

[0189] In some embodiments of the present specification, a scanning path generated by smoothly connecting adjacent component corner points using a path planning algorithm can ensure collision-free and smooth motion during the scanning process to obtain point cloud data of high-quality points.

[0190] In some embodiments of the present specification, the point cloud acquisition device identifies the component according to the scanning angle, scanning path and scanning position data, and can observe the same component area from different angles with a fixed angle posture and a specific trajectory. It can collect depth information from multiple perspectives, thereby improving the accuracy of the point cloud data and facilitating the acquisition of more relevant features of the weld.

[0191] In some embodiments of the present specification, when the size of a component is smaller than a size threshold, the corner point position data and center position data of the component corner points are obtained with the help of a node legend and an acquisition device, and then the optimal scanning position of the point cloud acquisition device is determined, making the acquisition of point cloud data more accurate.

[0192] Figure 5 FIG. 1 is an exemplary flow chart for determining the welding points of components according to some embodiments of this specification. Figure 5 As shown, process 500 includes steps 510 to 530.

[0193] Step 510: Determine a first fitting surface based on the point cloud data.

[0194] The first fitting surface refers to the surface of the component opposite to the welding surface, for example, the upper surface of the fitted quadrilateral tile.

[0195] In some embodiments, the processor may extract point cloud data of the component surface opposite to the welding surface, and determine the first fitting surface according to a plane fitting algorithm (such as the least squares method, etc.).

[0196] Step 520: offset the first fitting surface based on the offset parameter to obtain a second fitting surface.

[0197] The offset parameter refers to a distance parameter for offsetting the first fitting surface. In some embodiments, the offset parameter can be determined based on point cloud data.

[0198] In some embodiments, the offset parameter may be the component thickness, or the offset parameter may be the difference between the component thickness and the welding tip height.

[0199] For more information about component thickness and welding nozzle height, please visit Figure 2 Related description.

[0200] In some embodiments of the present specification, the component thickness or the difference between the component thickness and the welding nozzle height is used as an offset parameter to accurately obtain the downward height of the first fitting surface.

[0201] The second fitting surface refers to a fitted surface that can characterize the actual welding position. In some embodiments, the second fitting surface is parallel to and equal to the first fitting surface.

[0202] In some embodiments, the processor may shift the first fitting surface downward according to the shift parameter to obtain the second fitting surface.

[0203] Step 530: Determine welding points based on the second fitting surface.

[0204] For more information about welding points, please refer to Figure 2 Related description.

[0205] In some embodiments, the processor may determine a corner point of the second fitting surface as a welding point.

[0206] In some embodiments of the present specification, a first fitting surface can be directly determined based on point cloud data; then the first fitting surface is offset downward based on an offset parameter to obtain a second fitting surface representing the welding position, thereby further determining the precise welding point.

[0207] It should be noted that the above descriptions of processes 200, 300, 400, and 500 are only for illustration and description, and do not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to processes 200, 300, 400, and 500 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0208] Figure 6 is an exemplary schematic diagram of a path generation model according to some embodiments of the present specification.

[0209] In some embodiments, the processor can determine the welding path through a path generation model based on point cloud data, material parameters, welding nozzle posture parameters, welding gun type, welding nozzle size parameters, obstacle images, environmental data, chamfer parameters, attribute parameters of the target point and weld type.

[0210] For more information about point cloud data, material parameters, nozzle posture parameters, welding gun types, target points and weld types, please visit Figure 2 Related description.

[0211] The welding tip size parameters refer to parameters related to the welding tip size, such as welding tip diameter, etc. The welding tip size parameters can be manually input.

[0212] The obstacle image refers to a three-dimensional image of an object that causes an obstacle to the welding of the component. In some embodiments, the processor can control other image acquisition devices around the component to take multi-angle photos of the component (including the main body, welding equipment, etc.), and obtain the obstacle image through image recognition.

[0213] Environmental data refers to the data related to the environment in which the components are welded. For example, the temperature and humidity of the welding workshop. Environmental data can be monitored and obtained by relevant monitoring equipment (such as thermometers, hygrometers, etc.).

[0214] Chamfer parameters refer to parameters related to chamfer processing. For example, chamfer type and points corresponding to chamfer processing. Points corresponding to chamfer processing refer to points formed by chamfering processing. Points corresponding to flat chamfer processing refer to the points of two new corner points formed by chamfering processing, and points corresponding to round chamfer processing refer to all points on the arc formed by chamfering processing.

[0215] Among them, the chamfer types can include no chamfer, flat chamfer and round chamfer, as shown in the figure below. Fig. 7A (Points without chamfering treatment), Figure 7B (The points corresponding to the chamfering process are points E and F) and Figure 7C (The points corresponding to the chamfering treatment are the points on the arc EF).

[0216] Property parameters refer to properties related to the heat treatment state, such as annealing state, etc. Property parameters can be manually input.

[0217] Welding control parameters refer to the parameters related to the welding process at each target point during the welding process, such as welding method, welding voltage, current welding speed, swing amplitude and speed when using swing welding method, etc.

[0218] The path generation model is a model for generating a welding path. In some embodiments, the path generation model can be a machine learning model, such as a neural network (NN), a deep neural network (DNN), etc.

[0219] In some embodiments, Figure 6As shown, the path generation model 620 includes a process determination layer 620 - 1 and a path generation layer 620 - 2 .

[0220] The process determination layer 620-1 refers to a model for determining welding control parameters. In some embodiments, the process determination layer 620-1 may be a machine learning model, such as a convolutional neural network (CNN).

[0221] In some embodiments, Figure 6 As shown, the input of the process determination layer 620-1 may include point cloud data 610-1, material parameters 610-2, environmental data 610-3, chamfer parameters 610-4, attribute parameters of the target point 610-5 and weld type 610-6, and the output is welding control parameters 630 of the target point.

[0222] The path generation layer 620-2 refers to a model for generating a welding path. In some embodiments, the path generation layer 620-2 may be a machine learning model, such as a recurrent neural network (RNN), a long short-term memory network (LSTM), and the like.

[0223] In some embodiments, Figure 6 As shown, the input of the path generation layer 620-2 may include a welding gun type 640-1, a welding nozzle size parameter 640-2, an obstacle image 640-3, a welding nozzle posture parameter 640-4 and a welding control parameter 630 of a target point, and the output may be a welding path 650.

[0224] In some embodiments, the processor may train the process determination layer and the path generation layer separately based on a plurality of labeled training samples.

[0225] The first training sample of the process determination layer may include sample point cloud data, sample material parameters, sample environmental data, sample chamfer parameters, sample attribute parameters of sample target points, and sample weld types. The first label corresponding to the first training sample of the process determination layer may be the welding control parameter of the target point with the best welding effect.

[0226] The first training sample of the process determination layer can be determined based on historical data. The processor can perform multiple experiments on the same or similar first training samples using multiple sets of welding control parameters, and determine the welding effects corresponding to each set of welding control parameters, and select the welding control parameters with the best welding effect manually evaluated as the first label corresponding to the first training sample. Among them, the best welding effect can refer to the best comprehensive evaluation of multiple effects such as smooth welds, weld sizes that meet requirements, and firm welds.

[0227] The second training sample of the path generation layer may include sample welding nozzle posture parameters, sample welding gun types, sample welding nozzle size parameters, sample obstacle images and sample welding control parameters. The second label corresponding to the second training sample of the path generation layer may be the welding path with the best welding effect.

[0228] The second training sample of the path generation layer can be determined based on historical data. The processor can perform multiple experiments on the same or similar second training sample using multiple welding paths, and determine the welding effect corresponding to each welding path, and select the welding path with the best welding effect manually evaluated as the second label corresponding to the second training sample.

[0229] The first loss function is constructed by the welding control parameters of the target point output by the process determination layer and the first label, and the parameters of the process determination layer are updated based on the first loss function by gradient descent or other iterative methods. When the preset update condition is met, the process determination layer training is completed and the trained process determination layer is obtained. Among them, the preset update condition can be that the first loss function is less than the first threshold, the first loss function converges, the number of iterations reaches the threshold, etc. or any combination thereof.

[0230] The training process of the path generation layer is the same as that of the process determination layer, which will not be described in detail here.

[0231] In some embodiments of the present specification, based on material parameters, welding nozzle posture parameters, welding gun type, and other parameters, a machine learning model can accurately determine the welding path with the best welding effect to ensure welding efficiency and welding quality.

[0232] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0233] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0234] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0235] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0236] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining the digits. Although the numerical domains and parameters used to confirm the breadth of the range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0237] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, documents, etc., cited in this specification is hereby incorporated by reference in its entirety. Except for application history documents that are inconsistent with or conflicting with the contents of this specification, documents that limit the broadest scope of the claims of this specification (currently or later attached to this specification) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the materials attached to this specification are inconsistent or conflicting with the contents described in this specification, the descriptions, definitions, and / or use of terms in this specification shall prevail.

[0238] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A method for welding a block component, characterized in that: The method is implemented based on a processor and includes: Obtain node data, which includes component size; Acquire point cloud data of the component; the point cloud data is acquired in a manner related to the size of the component, including: In response to the component size being greater than a size threshold, Based on the size of the component, the component is divided into a plurality of sub-component regions; wherein the size of the sub-component region is smaller than the size threshold; Based on the point cloud acquisition device, the plurality of sub-component regions are identified according to a preset scanning order to obtain the point cloud data of the component; In response to the plurality of sub-component regions having overlapping regions, the method further comprises: Extracting a first feature point corresponding to a first sub-component region and a second feature point corresponding to a second sub-component region in the overlapping region; wherein the first sub-component region and the second sub-component region are two adjacent sub-component regions having the overlapping region, the first feature point is a common feature point in a first coordinate system corresponding to the first sub-component region, and the second feature point is a common feature point in a second coordinate system corresponding to the second sub-component region; Based on the first feature point and the second feature point, determining a matching feature point pair by a feature matching algorithm; Based on the matching feature point pairs, determining a homography matrix; the homography matrix is ​​a matrix describing the position mapping relationship between the common feature points in the first coordinate system and the second coordinate system; Based on the homography matrix, transform the first feature point to the second coordinate system to obtain a third feature point; Merging the second feature point and the third feature point to determine the point cloud data of the overlapping area; In response to the component size being less than the size threshold, Determine a node legend based on the size of the component; the node legend is configured to indicate the placement position of the component and / or the acquisition device; Based on the acquisition device, the component is identified to obtain corner point position data of at least two corner points of the component; Determining the center position data of the component based on the corner point position data of the at least two component corner points; Based on the central position data, determining scanning position data; Based on the point cloud acquisition device, the component is identified according to the scanning position data to obtain the point cloud data of the component; Obtaining a fitting surface based on the point cloud data; Based on the fitting surface, the welding points of the component are determined, and a welding point table array is generated; the welding point table array includes welding nozzle posture parameters, and the method for determining the welding nozzle posture parameters includes: Construct welding feature vector based on weld size, material parameters, welding working parameters, main part inclination and component thickness; Based on the welding feature vector, determining the welding nozzle posture parameter by searching a vector database; Determine a weld set based on the welding point table array, wherein the weld set includes at least one weld, a weld type of the weld, and a target point on the weld; A welding plan is determined based on the weld set and the welding point table array, wherein the welding plan includes a welding path.

2. The method according to claim 1, characterized in that The obtaining of node data further comprises: Determining whether the node data meets a preset condition; In response to not meeting the preset condition, a user warning instruction is triggered.

3. The method according to claim 1, characterized in that The step of determining the welding position of the component based on the fitting surface comprises: Based on the point cloud data, determining a first fitting surface; Based on the offset parameter, the first fitting surface is offset to obtain a second fitting surface; the second fitting surface is parallel and equal to the first fitting surface; the offset parameter is determined based on the point cloud data; Based on the second fitting surface, the welding point is determined.

4. The method according to claim 1, characterized in that The method also includes visually displaying the welding plan.

5. A welding system for block components, characterized in that: The system comprises: A first acquisition module is configured to acquire node data, wherein the node data includes component size; The second acquisition module is configured to acquire point cloud data of the component; the acquisition method of the point cloud data is related to the size of the component, and the second acquisition module is further configured to: In response to the component size being greater than a size threshold, Based on the size of the component, the component is divided into a plurality of sub-component regions; wherein the size of the sub-component region is smaller than the size threshold; Based on the point cloud acquisition device, the plurality of sub-component regions are identified according to a preset scanning order to obtain the point cloud data of the component; In response to the presence of overlapping areas among the plurality of sub-component areas, the second acquisition module is further configured to: Extracting a first feature point corresponding to a first sub-component region and a second feature point corresponding to a second sub-component region in the overlapping region; wherein the first sub-component region and the second sub-component region are two adjacent sub-component regions having the overlapping region, the first feature point is a common feature point in a first coordinate system corresponding to the first sub-component region, and the second feature point is a common feature point in a second coordinate system corresponding to the second sub-component region; Based on the first feature point and the second feature point, determining a matching feature point pair by a feature matching algorithm; Based on the matching feature point pairs, determining a homography matrix; the homography matrix is ​​a matrix describing the position mapping relationship between the common feature points in the first coordinate system and the second coordinate system; Based on the homography matrix, transform the first feature point to the second coordinate system to obtain a third feature point; Merging the second feature point and the third feature point to determine the point cloud data of the overlapping area; In response to the component size being less than the size threshold, Determine a node legend based on the size of the component; the node legend is configured to indicate the placement position of the component and / or the acquisition device; Based on the acquisition device, the component is identified to obtain corner point position data of at least two corner points of the component; Determining the center position data of the component based on the corner point position data of the at least two component corner points; Based on the central position data, determining scanning position data; Based on the point cloud acquisition device, the component is identified according to the scanning position data to obtain the point cloud data of the component; A first determination module is configured to obtain a fitting surface based on the point cloud data; The second determination module is configured to determine the welding points of the component based on the fitting surface and generate a welding point table array; the welding point table array includes welding nozzle posture parameters, and the second determination module is further configured to: Construct welding feature vector based on weld size, material parameters, welding working parameters, main part inclination and component thickness; Based on the welding feature vector, determining the welding nozzle posture parameter by searching a vector database; A third determination module is configured to determine a weld set based on the welding point table array, wherein the weld set includes at least one weld, a weld type of the weld, and a target point on the weld; The fourth determination module is configured to determine a welding plan based on the weld set and the welding point table array, wherein the welding plan includes a welding path.

6. A welding device for block components, characterized in that: The device comprises at least one memory and at least one processor, wherein the at least one memory is used to store computer instructions, and the at least one processor executes the computer instructions or part of the instructions to implement the welding method of the block component according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that: The storage medium stores computer instructions, and when a computer reads the computer instructions, the computer executes the welding method for a block component according to any one of claims 1 to 4.