Welding robot operation system and method

Abstract

The present invention provides a welding robot operation system and method. The system includes: a welding planning module configured to generate a welding path set based on a desired weld, wherein multiple welding robots collaboratively weld the welded parts based on multiple preset welding paths in the welding path set; a visual processing module configured to calculate welding deviations based on each welding robot's actual scanning path and the preset welding paths; a welding process module configured to control the multiple welding robots to collaboratively weld the welded parts based on the preset welding path set and welding parameters; and updating the welding path set based on welding deviations. This method improves welding accuracy and efficiency by enabling collaborative welding of multiple robots and updating the path set based on real-time calculated welding deviations. This method achieves adaptive adjustment of the welding process to meet different welding requirements and enhances the stability and reliability of welding quality.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to a welding robot operation system and method. Background Art

[0002] With the development of computer and artificial intelligence technologies, welding robots have been widely used in manufacturing industries such as shipbuilding, bridge construction, and steel structures. For welding robots, vision systems and intelligent welding systems are key components for intelligent welding. However, current welding robots cannot accurately identify differences in welding environments (such as ambient temperature) and welding materials (such as thermal deformation). This results in significant fluctuations in robot welding quality for specific environments or welded parts, leading to serious quality issues and safety risks.

[0003] Therefore, it is necessary to provide an improved welding robot operation system and method to improve efficiency and quality. Summary of the Invention

[0004] The present invention provides a welding robot operation system, which includes: a welding planning module, which is configured to generate a welding path set based on a required weld, wherein the welding path set includes multiple preset welding paths, and the welding path set instructs multiple welding robots to collaboratively weld the welded parts based on the multiple preset welding paths in the welding path set; a visual processing module, which is configured to calculate the welding deviation based on the actual scanning path of each welding robot and the preset welding path during the collaborative welding; a welding process module, which is configured to: control the multiple welding robots to collaboratively weld the welded parts based on the preset welding path set and welding parameters; and update the welding path set based on the welding deviation.

[0005] The present invention provides a welding robot operation method, which includes: generating a welding path set based on a required weld, the welding path set including multiple preset welding paths, and multiple welding robots collaboratively welding the weldment based on the multiple preset welding paths in the welding path set; during the collaborative welding, calculating the welding deviation based on the actual scanning path of each welding robot and the preset welding path; based on the preset welding path set and welding parameters, controlling the multiple welding robots to collaboratively weld the weldment; and updating the welding path set based on the welding deviation.

[0006] The present invention provides a welding robot operation device, comprising a processor, wherein the processor is used to execute a welding robot operation method.

[0007] The present invention provides a computer-readable storage medium, which stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a welding robot operation method. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] This specification will be further described in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, like numbers represent like structures, wherein:

[0009] Figure 1 is a module diagram of a welding robot operation system according to some embodiments of this specification;

[0010] Figure 2 is an exemplary operation interface diagram of a welding planning module according to some embodiments of this specification;

[0011] Figure 3 is an exemplary operation interface diagram of the visual processing module according to some embodiments of this specification;

[0012] Figure 4 is an exemplary operation interface diagram of a welding process module according to some embodiments of this specification;

[0013] Figure 5 is an exemplary operation interface diagram of the model parsing module according to some embodiments of this specification;

[0014] Figure 6 is an exemplary operation interface diagram of a robot module according to some embodiments of this specification;

[0015] Figure 7 is an exemplary operation interface diagram of a graphics module according to some embodiments of this specification;

[0016] Figure 8 is an exemplary flow chart of a welding robot operation method according to some embodiments of this specification;

[0017] Figure 9 is an exemplary flow chart of updating a welding path set according to some embodiments of the present invention;

[0018] Figure 10 is an exemplary flow chart for updating welding parameters according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly describes the drawings required for describing the embodiments. Obviously, the drawings described below are merely examples or embodiments of the present invention. Those skilled in the art can apply the present invention to other similar scenarios based on these drawings without inventive effort. Unless otherwise apparent from the context or otherwise noted, the same reference numerals in the figures represent the same structure or operation.

[0020] It should be understood that the terms "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 terms can achieve the same purpose, the terms may be replaced by other expressions.

[0021] As used herein, unless the context clearly indicates otherwise, the terms "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list; a method or apparatus may also include other steps or elements.

[0022] Flowcharts are used in this disclosure to illustrate the operations performed by systems according to embodiments of the present invention. It should be understood that the preceding and following operations do not necessarily need to be performed in exact order. Instead, the steps may be processed in reverse order or simultaneously. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0023] Figure 1 It is a module diagram of a welding robot operation system according to some embodiments of this specification.

[0024] In some embodiments, the welding robot operation system 100 may include a welding planning module 110 , a vision processing module 120 , and a welding process module 130 .

[0025] In some embodiments, the welding planning module 110 is configured to generate a welding path set based on the required weld, the welding path set including multiple preset welding paths, and multiple welding robots collaboratively weld the weldment based on the multiple preset welding paths in the welding path set.

[0026] The welding planning module includes path generation, obstacle avoidance planning, and dynamic adjustment functions. The path generation function automatically generates the welding robot's welding path without manual instruction. The obstacle avoidance function prevents collisions between the welding robot and workpieces, fixtures, or other equipment, ensuring the safety of the welding process through obstacle avoidance algorithms. The dynamic path adjustment function dynamically corrects the robot's path using a local adjustment algorithm based on feedback from the visual processing module 120 to adapt to workpiece errors and environmental changes.

[0027] Figure 2 1 is an exemplary operation interface diagram of the welding planning module according to some embodiments of this specification. In some embodiments, the welding planning module 110 can be integrated into the Smart Weld (SW) software. The welding planning module 110 can be operated through the first operation interface of the SW. For example, Figure 2 As shown, the first operation interface includes panels such as line laser, operation information, and path queue. The path queue panel can display the preset welding path of the welding robot. The line laser panel can view the images and points acquired by the line laser in real time, and the operation information panel can view the points acquired by the line laser and write them to the variables in real time. The device can be started online normally in the operation bar. After clicking Start Tracking, the tracking program starts to load and run the laser tracking function. Click Stop Tracking to stop the laser tracking. L1 and L2 are used to adjust the distance between the tip of the welding gun and the weld. The starting distance is used to adjust the position of the starting point of the weld. When tracking welds with large angle changes, turning on the angle change function can keep the welding gun angle consistent with the weld angle.

[0028] In some embodiments, the vision processing module 120 is configured to calculate welding deviations based on the actual scanning path of each welding robot and the preset welding path during collaborative welding.

[0029] In some embodiments, the vision processing module 120 is further configured to generate a sequence of thermal images of the weldment, where the sequence of thermal images includes a plurality of thermal images at a plurality of time points.

[0030] In some embodiments, the vision processing module 120 is further configured to generate a sequence of optical images of the weld.

[0031] In some embodiments, the visual processing module 120 may include 2D vision functions, 3D vision functions, and real-time processing functions. The 2D vision function includes acquiring a two-dimensional image of the weld area using an image acquisition device (e.g., a camera), and identifying weld boundaries using image processing algorithms such as edge detection (e.g., the Canny algorithm) and feature matching (e.g., SIFT / SURF). The 3D vision function includes generating three-dimensional point cloud data of the weld using a laser sensor or depth camera, and segmenting and registering the three-dimensional point cloud data using algorithms such as RANSAC or ICP. The real-time processing function includes extracting the actual scanning path of the welding robot in real time using an image processing algorithm, and determining the deviation between the actual scanning path and the preset welding path.

[0032] Figure 3 This is an exemplary operation interface diagram of the visual processing module according to some embodiments of this specification. In some embodiments, the visual processing module 120 can be integrated into the SW software. The visual processing module 120 can be operated through the second operation interface of the SW to achieve the above functions. For example, Figure 3 As shown, the second operation interface includes panels such as scene point cloud, scene image, program content, scene path, and posture control. The scene point cloud panel can display the three-dimensional point cloud data of the weld. The scene image panel can display the two-dimensional image of the welding area. The scene path panel can display the actual welding path of the welding robot. The posture control panel can be used to control the image acquisition device and the posture of the image acquisition device. The program content panel can be used to view the real-time welding content generated after the scan calculation. The common operation bar of the second operation interface also includes multiple options. Figure 3 As shown, "Position Point" is used to switch the corresponding auxiliary point, "Robot" is used to disconnect and power on the robot, and "Component" can be used to manage components and select the corresponding components for welding. "3D Camera" and "Positioning Single Shot" are used to control the camera's connection, disconnection, and shooting mode. "Multi-point", "Straight Line", and "Curve" are used to determine the starting and ending points of the weld, thereby determining the corresponding nodes. Fixed-point scanning welding allows for continuous photo-taking and welding of the same type of components within the camera's fixed operating range. 3D scanning is to scan the weld position using the camera after selecting the corresponding node.

[0033] In some embodiments, the visual processing module 120 can scan the workpiece and automatically identify the actual weld position, adapt to the actual workpiece installation error, and ensure the accuracy of the welding position; when the workpiece deviates due to thermal expansion and contraction or processing errors, the welding path can be adjusted in real time to adapt to the deformation of the workpiece; it can also detect in real time whether the weld meets the process requirements, such as no pores, uniform weld height, etc., to achieve welding quality monitoring.

[0034] In some embodiments, the welding process module 130 is configured to: control multiple welding robots to collaboratively weld the weldment based on a preset welding path set and welding parameters; and update the welding path set based on welding deviations.

[0035] In some embodiments, the welding process module 130 is further configured to: during collaborative welding, for each welding robot, determine the weighted cumulative deviation of the welding robot based on the actual scanning path of the welding robot and the temperature zone passed by the actual scanning path, and use the weighted cumulative deviation as the welding deviation; based on multiple thermal imaging images and welding deviations in the thermal imaging image sequence, determine the update time point, and update the welding path set at the update time point based on the welding deviation.

[0036] In some embodiments, the welding process module 130 is further configured to: determine multiple target points based on the welding path set; determine the thermal distortion rate of each target point based on the optical image sequence; and select an update time point from the update time point set based on the welding deviation and the thermal distortion rate.

[0037] In some embodiments, the welding process module 130 is further configured to: identify at least one welding joint based on the welding path set; for each welding joint: determine the first object and the second object corresponding to the welding joint, the first object refers to the welding robot that is currently welding the welding joint, and the second object refers to the welding robot that is not currently welding the welding joint; based on the average residence time of the first object in the adjacent area of ​​the welding joint, determine the temperature influence domain corresponding to the welding joint; based on the average residence time and the temperature influence domain, update the welding parameters of the second object in the temperature influence domain.

[0038] In some embodiments, the welding process module 130 can perform collaborative welding based on the control of the robot module. For details about the robot module, please refer to the relevant description below.

[0039] In some embodiments, the welding process module 130 can store and manage welding-related process parameters, such as welding current, voltage, welding speed, and welding posture. The welding process module 130 also includes a material database and welding specifications. The material database includes welding characteristics of welding materials (e.g., carbon steel, stainless steel, and aluminum alloy). Welding specifications include various welding process standards, such as AWS and ISO.

[0040] Figure 4 1 is an exemplary operation interface diagram of the welding process module according to some embodiments of this specification. In some embodiments, the welding process module 130 can be integrated into the SW software. The welding process module 130 can be operated through the third operation interface of the SW. For example, Figure 4As shown, the third operation interface includes panels such as the Weld Task and Nodes. A welding task includes a workstation, working range, and processed model files, allowing for intuitive and efficient welding task allocation. The Node panel displays real-time information about welding nodes, allowing for better judgment of welding information and tasks. In the operation bar, "Load Workstation" and "Unload Workstation" control the addition and deletion of workstations, while "Load Component" and "Unload Component" control the addition and deletion of components. Together, these two functions allow for efficient and rapid commissioning of welding tasks, allowing the robot to complete the job. "Refresh Scene" allows refreshing scene content and welding tasks.

[0041] In some embodiments, the welding process module can automatically load the optimal welding control parameters based on the material properties of the welding material and the weld information to avoid manual debugging errors; dynamically adjust the welding parameters to adapt to the characteristics of different welded parts and provide real-time feedback; and provide standardized welding processes to ensure consistency in welding quality.

[0042] In some embodiments, the welding robot operation system may further include a model parsing module.

[0043] In some embodiments, the model parsing module includes preprocessing the model. The model includes a 3D model such as a STEP, IGES, or STL model. Preprocessing functions include repairing model defects such as holes, duplicate faces, and non-manifold geometry. The model parsing module can automatically read the model and analyze the weld locations of the required welds.

[0044] Figure 5 This is an exemplary operation interface diagram of the model analysis module according to some embodiments of this specification. In some embodiments, the model analysis module can be integrated into the SW software. The model analysis module can be operated through the fourth operation interface of the SW to achieve the above functions. For example, Figure 5 As shown, the fourth operation interface includes panels such as posture control, scene path, and properties. The posture control panel can intuitively see the real-time status and point information of the robot, thereby controlling the posture of the image acquisition device. The scene path panel can display the actual welding path of the welding robot. The property panel can see the workstation called in the current state and the current posture. In the operation bar, "Select Project" can select the corresponding workstation, process package and robot, and "Multi-point" and "Straight Line" are used to determine the starting and ending points of the weld, thereby determining the corresponding node. After the node is confirmed, click "Laser Scan" to scan the weld with a line laser to obtain point information.

[0045] In some embodiments, the model parsing module can automatically identify welding areas in large and complex workpieces, reduce manual operations, and improve work efficiency; convert the three-dimensional model into geometric data executable by the robot, providing a basis for subsequent path planning, and ensuring the accuracy and reliability of path planning.

[0046] In some embodiments, the welding robot operation system may further include a robot module.

[0047] In some embodiments, the robot module includes functions such as motion control, path execution, tool management, and communication. Motion control involves precise multi-axis control of multiple welding robots. Path execution involves coordinated welding based on welding parameters. Tool management involves configuring and managing various tools, such as welding guns and laser sensors. Communication involves communicating with the welding robot controller via an industrial bus (e.g., EtherCAT or PROFINET).

[0048] Figure 6 This is an exemplary operation interface diagram of the robot module according to some embodiments of this specification. In some embodiments, the robot module can be integrated into the SW software. The robot module can be operated through the fifth operation interface of the SW to achieve the above functions. For example, Figure 6 As shown, the fifth operation interface includes panels such as scene point cloud, program content, scene path, and posture control. The scene point cloud panel can display the three-dimensional point cloud data of the weld. The program content can see the real-time welding content generated after scanning and calculation. The scene path panel can display the actual welding path of the welding robot. The posture control panel can be used to control the image acquisition device and the posture of the image acquisition device. The functions in the operation bar are the same as Figure 2-Figure 5 The functions of medium concentration are similar, please refer to the relevant instructions above for details.

[0049] The robot module can accurately perform welding operations based on visual feedback and path planning; adapt to the multi-faceted and multi-angle welding requirements of complex workpieces; and adjust welding parameters such as welding speed and position in real time based on feedback during the welding process.

[0050] In some embodiments, the welding robot operation system may further include a graphics module.

[0051] In some embodiments, the image module may include functions such as 3D rendering, simulation and verification, and result feedback. 3D rendering uses OpenGL, DirectX, and other technologies to render the desired weld's 3D point cloud data in real time, providing real-time 3D visualization of the weld, path, and robot motion. Simulation and verification can simulate robot motion and the welding process before actual welding. Result feedback can intuitively display welding status, parameter changes, and welding results.

[0052] Figure 7 This is an exemplary operation interface diagram of the graphics module according to some embodiments of this specification. In some embodiments, the graphics module can be integrated into the SW software. The graphics module can be operated through the sixth operation interface of the SW to achieve the above functions. For example, Figure 7 As shown, the sixth operation interface includes panels such as dock panel, stacking path, graphics, and point table. The dock panel can intuitively see the real-time status and point information of the robot to control the posture of the image acquisition device. The trajectory path can display the actual welding path of the welding robot. The graphics panel is used to determine whether the calculated graphics are consistent with the actual situation. The point table panel can see the scanning and welding point information. In the operation bar, "Start online" and "Stop online" are used to disconnect the robot connection. In controlled mode, you can select the welding mode (such as direct welding or scanning welding). "Step-by-step operation" splits the entire set of scanning welding actions to optimize the scanning welding process.

[0053] In some embodiments, through the graphics module, users can preview the path and welding actions through the graphical interface before starting welding, identify problems in advance and make adjustments; monitor data changes during the welding process in real time to facilitate timely optimization of welding strategies; provide dynamic simulation of the welding process to verify the feasibility of the welding plan and reduce risks and errors in actual welding.

[0054] In some embodiments, the welding planning module 110, the visual processing module 120, the welding process module 130, and the like of the welding robot operation system 100 may be fully or partially integrated into a processor. The processor is configured to process information and / or data related to the welding robot operation system 100. The processor may also process data obtained from other devices, such as user terminals, which may include personal computers, mobile devices, and the like. The processor may include a combination of one or more of a central processing unit (CPU), an application-specific integrated circuit (ASIC), and the like.

[0055] It should be noted that the above description of the welding robot operation system and its modules is for convenience only and does not limit this specification to the scope of the embodiments. It is understandable that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the modules or form subsystems connected with other modules without deviating from the principles. In some embodiments, Figure 1The welding planning module, visual processing module, and welding process module disclosed in the present disclosure may be separate modules within a single system, or a single module may implement the functions of two or more of these modules. For example, each module may share a storage module, or each module may have its own storage module. Such variations are within the scope of protection of this specification.

[0056] Figure 8 FIG. 1 is an exemplary flow chart of a welding robot operation method according to some embodiments of this specification. Figure 8 As shown, the process 800 includes the following steps 810-840. Steps 810-840 can be executed by a processor.

[0057] Step 810 : Generate a welding path set based on the required weld.

[0058] The desired weld refers to the target weld to be welded. In some embodiments, the desired weld includes a weld location, which includes geometric features such as the weld's start point, end point, length, and angle. The processor can obtain the weld location of the desired weld based on the model parsing module.

[0059] In some embodiments, the welding path set includes multiple preset welding paths. The preset welding paths are the optimal welding paths that each welding robot needs to execute. In some embodiments, the preset welding paths can be generated by a path planning algorithm, such as an A-Star (A algorithm) or a Rapidly Exploring Random Tree (RRT) algorithm.

[0060] In some embodiments, the welding path set instructs multiple welding robots to collaboratively weld the weldment based on multiple preset welding paths in the welding path set.

[0061] In some embodiments, the processor can perform global path planning based on the desired weld seam and welding requirements using the A algorithm and the RRT algorithm to generate multiple preset welding paths. Welding requirements include requirements for welding time, welding energy consumption, welding quality, etc. The processor can obtain user-entered welding requirements through a client, which can include a mobile phone, personal computer, laptop, etc.

[0062] Weldments are workpieces that require welding operations.

[0063] Step 820 , during collaborative welding, calculating welding deviations based on the actual scanning path of each welding robot and the preset welding path.

[0064] The real scanning path refers to the actual welding path of the welding robot during the actual welding process. The processor can obtain the real scanning path of each welding robot based on the image processing algorithm through the visual processing module 120.

[0065] Welding deviation is the distance difference between the actual scanning path of the welding robot and the preset welding path. In some embodiments, the welding deviation includes the deviation amplitude of the welding robot at multiple points. The multiple points may include the locations of a preset number of welding points on the desired weld seam. The preset number can be set based on demand or experience. For example, the preset number is positively correlated with the desired weld seam length.

[0066] In some embodiments, for each welding robot, the processor may calculate the deviation between the actual scanning path of multiple points and the preset welding path as the welding deviation of the welding robot.

[0067] Step 830 : Based on the welding path set and the welding parameters, control multiple welding robots to perform collaborative welding on the weldment.

[0068] Welding parameters refer to parameters related to welding by the welding robot, such as welding gun movement speed, welding gun temperature, etc. In some embodiments, the processor can determine the welding parameters based on the material of the welded part by searching the material database and welding specifications of the welding process module 130. For more information about the material database and welding specifications, please refer to Figure 1-7 And related instructions.

[0069] Collaborative welding involves multiple welding robots working together to weld the desired weld. A welded part can consist of one or more plates. Collaborative welding allows multiple robots to work together to weld the same plate. Alternatively, multiple robots can work together to weld multiple plates together. Multiple robots can weld simultaneously or in a different order. The welding sequence can be preset based on requirements or experience.

[0070] In some embodiments, the processor can control each welding robot to weld the weldment through a corresponding preset welding path in the welding path set based on the welding parameters.

[0071] In step 840 , the welding path set is updated based on the welding deviation.

[0072] In some embodiments, for each welding robot, the processor may re-plan the path for a deviation region consisting of points where the welding deviation exceeds a welding deviation threshold, thereby updating the preset welding path for the deviation region. For example, the processor may re-plan the preset welding path for the deviation region using a path planning algorithm (such as the A algorithm or the RRT algorithm), and use the multiple preset welding paths re-planned by the multiple welding robots as a welding path set. For another example, if the direction of the welding deviation is a first direction, the processor may re-plan the path for the deviation region so that the preset welding path moves away from the first direction within the deviation region, with the amount of movement being positively correlated with the welding deviation.

[0073] The welding deviation threshold can be set according to needs or experience. For more information about welding deviation threshold, please refer to Figure 9 And related instructions.

[0074] According to some embodiments of the present specification, by generating a preset welding path set based on the required welds and allowing multiple robots to collaborate in welding, and combining the path set with real-time calculated welding deviations to update the path set, it is possible to improve welding accuracy and efficiency, achieve adaptive adjustment of the welding process, meet different welding requirements, and improve the stability and reliability of welding quality.

[0075] Figure 9 FIG. 1 is an exemplary flow chart of updating a welding path set according to some embodiments of the present invention. Figure 9 As shown, the process 900 includes the following steps 910 to 930. Steps 910 to 930 may be executed by a processor.

[0076] Step 910: Generate a sequence of thermal images of the weldment.

[0077] The thermal image sequence includes multiple thermal images at multiple time points. The multiple time points can be preset based on demand or experience.

[0078] A thermographic image is a visual representation of the surface temperature distribution of a weld.

[0079] In some embodiments, the temperatures of different areas of a weld can be the same or different. Due to the asynchronous welding times of multiple welding robots, different areas of the weld may be in the process of welding, cooled, or unwelded, resulting in temperature differences between these areas. During the welding process, differences in welding current and speed between different welding robots can lead to temperature differences between different areas. After welding, varying cooling times can also create temperature gradients across different areas of the weld.

[0080] In some embodiments, the processor may collect the surface temperature of the weldment at multiple time points using a thermal imaging device (such as an infrared thermal imaging sensor), generate thermal imaging images at multiple time points, and form a thermal imaging image sequence.

[0081] Step 920: During collaborative welding, for each welding robot, based on the actual scanning path of the welding robot and the temperature zone passed by the actual scanning path, determine the weighted cumulative deviation of the welding robot, and use the weighted cumulative deviation as the welding deviation.

[0082] The temperature region is a continuous region divided according to the temperature gradient in the thermal image. In some embodiments, each region is characterized by an average temperature or a center temperature.

[0083] In some embodiments, the processor can perform hierarchical segmentation on the thermal image based on a dynamic threshold. For example, the processor can extract temperatures from the thermal image to form a temperature matrix, and then dynamically segment the temperature matrix of the thermal image using a multi-level Otsu threshold segmentation algorithm to obtain multiple temperature regions.

[0084] The weighted cumulative deviation refers to the cumulative deviation of the welding robot during the entire welding process.

[0085] In some embodiments, the processor can divide the welding robot into at least one sub-segment based on at least one temperature zone passed by the actual scanning path, with each temperature zone corresponding to a sub-segment; determine the sub-welding deviation of each sub-segment; and perform weighted summation of the sub-welding deviations corresponding to all sub-segments to obtain a weighted cumulative deviation.

[0086] In some embodiments, the sub-welding deviation of each sub-segment can be determined by the visual processing module 120. For example, the processor can obtain the welding deviation of at least one welding point included in the temperature region corresponding to each sub-segment through the visual processing module 120, and use the statistical value (such as the average value, variance, etc.) of the welding deviation of the at least one welding point as the sub-welding deviation.

[0087] In high-temperature zones, distortion is more likely to occur during welding, and the impact of welding deviation is greater. Therefore, the weight of the sub-welding deviation is positively correlated with the temperature of the temperature zone.

[0088] Step 930 : determining an update time point based on the plurality of thermal images in the thermal image sequence and the welding deviation, and updating the welding path set at the update time point based on the welding deviation.

[0089] The update time point is the time point at which the preset welding paths are updated synchronously for all welding robots.

[0090] In some embodiments, in response to the presence of welding deviations greater than a welding deviation threshold in multiple thermal imaging images, the processor can use the current time point as an update time point; in response to the absence of welding deviations greater than the welding deviation threshold in multiple thermal imaging images, based on the welding deviations at multiple time points, with time as the independent variable and the welding deviation as the dependent variable, a linear function of the welding deviation and time is constructed, and the independent variable value when the dependent variable is the welding deviation threshold is calculated as the update time point.

[0091] In some embodiments, the welding deviation threshold can be preset based on experience or demand.

[0092] In some embodiments, the processor may determine a weld deviation threshold based on the material of the current weld and the weld duration.

[0093] The material of the current weldment refers to the material of the weldment being welded at the current time, such as carbon steel, stainless steel, or aluminum alloy. The welding duration refers to the total duration of the current weld. The welding duration can be the time from the first welding robot starting welding to the last welding robot completing welding in a welding process.

[0094] In some embodiments, the processor may determine a standard welding error based on the material of the welding plate; determine a correlation coefficient based on the welding duration and the average welding time; and use the product of the standard welding error and the correlation coefficient as a welding deviation threshold.

[0095] In some embodiments, the processor may query the welding specifications of the welding process module 130 based on the material of the welding plate to determine the standard welding error.

[0096] The average welding time refers to the average welding duration of a particular weld. In some embodiments, the processor may search historical data for all welding records of similar welds to the current weld and calculate the average welding time as the average welding time. In some embodiments, the processor may use the ratio of the duration of the collaborative welding to the average welding time as a correlation coefficient.

[0097] In some embodiments of the present specification, the welding deviation threshold is dynamically determined by combining material properties and welding duration, and the welding deviation threshold can be flexibly adjusted according to actual working conditions.

[0098] In some embodiments, the processor can also generate an optical image sequence of the weld; determine multiple target points based on the welding path set; determine the thermal distortion rate of each target point based on the optical image sequence; and select an update time point from the update time point set based on the welding deviation and the thermal distortion rate.

[0099] The optical image sequence includes optical images of the weldment at multiple time points.

[0100] The optical image refers to an optical image of the welded part, and the optical image can reflect the thermal deformation amplitude of the welded part surface. In some embodiments, the processor can obtain the optical image through the visual processing module 120.

[0101] Target points are key points on welded parts that require focused observation.

[0102] In some embodiments, the processor can randomly select a preset number of points for each preset welding path in the welding path set; cluster the points based on the horizontal and vertical coordinates of the points to obtain multiple cluster centers; for each cluster center, the point closest to the cluster center is used as the target point.

[0103] In some embodiments, the clustering features further include point temperature, welding deviation, etc.

[0104] Thermal distortion rate is a parameter that measures the growth rate of thermal deformation at a target point. The unit of thermal distortion rate is the same as the unit of welding deviation.

[0105] In some embodiments, the processor can extract the thermal deformation amplitude of each target point in the optical image sequence at multiple time points; construct a linear function with time as the independent variable and the thermal deformation amplitude as the dependent variable, and use the slope of the linear function as the thermal distortion rate.

[0106] The update time point set is a set of time points at which the welding robots perform common updates. In some embodiments, the update time point set can be preset based on experience.

[0107] In some embodiments, the processor may sample and determine the update time point set based on the welding path set.

[0108] In some embodiments, the processor can determine the non-stop time of each weld based on the welding path set; determine the stoppable time point based on the non-stop time of each weld and the estimated welding time period; and sample all stoppable time points to obtain an updated time point set.

[0109] In some embodiments, the processor can query the welding process standards in the welding process module 130 for each weld in the set welding path set to determine the continuous welding locations (such as the weld pool area and critical transition section). Based on the welding path set and the torch speed of the current weld, the processor calculates the time range for the welding robot to pass through the continuous welding locations as the non-stop time. The estimated welding time period is the estimated time range from the start to the end of welding the weld. The estimated welding time period can be determined based on the torch speed and the weld location of the desired weld.

[0110] In some embodiments of the present specification, by determining an update time point set through a welding path set, the continuity of the welding process can be ensured, and weld quality defects caused by random pauses can be avoided.

[0111] In some embodiments, the sampling frequency is related to the network upload rate.

[0112] The network upload rate refers to the rate at which the welding robot operation system uploads data through the network. In some embodiments, the processor can obtain the network upload rate through a gateway or a network device (such as a router).

[0113] In some embodiments, the sampling frequency is positively correlated with the network upload rate.

[0114] The sampling frequency is dynamically adjusted according to the network upload rate. The sampling density can be increased when the network rate is high, and reduced when the network rate is low, so as to adapt to different network environments and ensure data transmission efficiency and stability.

[0115] In some embodiments, the processor selects the update time point from the set of update time points in a variety of ways based on the welding deviation and the thermal distortion rate.

[0116] For example, in response to a welding deviation greater than a welding deviation threshold or a thermal distortion rate greater than a rate threshold, the processor may select the time point closest to the current time point from the update time point set as the update time point. For another example, in response to the absence of a welding deviation greater than the welding deviation threshold and the absence of a thermal distortion rate greater than the rate threshold, the processor may determine a reference time point based on the most recent time point at which a welding deviation occurred at the target location, the corresponding welding deviation, thermal distortion rate, and welding deviation threshold at that time point, and select the time point closest to the reference time point from the update time point set as the update time point.

[0117] In some embodiments, the processor may calculate the reference time point based on formula (1). Formula (1) is as follows:

[0118]

[0119] in, Indicates a reference time point; Indicates the most recent time point when welding deviation occurs at the target point; Indicates that the target point is at the time point Welding deviation; Indicates that the target point is at the time point Thermal distortion rate; Indicates the welding deviation threshold.

[0120] In some embodiments, for each time point in the update time point set, the processor can determine the welding deviation through a deviation determination model based on the welding deviation, thermal distortion rate and the time point in the update time point set; and determine the update time point based on multiple welding deviations corresponding to multiple time points in the update time point set.

[0121] The deviation determination model is a model used to determine welding deviation. In some embodiments, the deviation determination model can be a machine learning model, such as a deep neural network (DNN) model.

[0122] The input of the deviation determination model includes the welding deviation, the thermal distortion rate and the time point of the update time point concentration, and the output of the deviation determination model includes the welding deviation corresponding to the time point.

[0123] In some embodiments, the deviation determination model can be trained using a large number of training samples with training labels. For example, the processor can input multiple predicted training samples into the initial deviation determination model, construct a loss function based on the output of the initial deviation determination model and the training labels, and iteratively update the parameters of the initial deviation determination model based on the loss function. When an iteration completion condition is met, the iteration ends, resulting in a trained deviation determination model. The iterative update method includes, but is not limited to, gradient descent, and the iteration completion condition can be when the loss function converges or the number of iterations reaches a threshold.

[0124] The training samples include the historical update time points of the updated welding path set, the historical welding deviations before the update, and the historical thermal distortion rates. The training labels are the actual welding deviations after the training sample is subsequently updated. The training labels can be obtained through the visual processing module.

[0125] In some embodiments, the processor may select a time point corresponding to a minimum welding deviation from a plurality of welding deviations corresponding to a plurality of time points in the update time point set as the determined update time point.

[0126] In some embodiments of this specification, the rationality of selecting the update time point is improved through model analysis, avoiding decision deviation caused by single parameter judgment.

[0127] In some embodiments of this specification, the thermal distortion rate of target points is analyzed through optical image sequences. Dynamically selecting update times based on a set of update time points balances welding accuracy and process continuity, effectively reducing welding errors caused by accumulated thermal deformation and improving welding consistency for complex workpieces. Determining the set of update time points ensures that each pause does not affect the smoothness of subsequent welding, thereby improving welding quality.

[0128] For the update method of welding path set, please refer to Figure 8 Step 840 and related instructions.

[0129] In some embodiments of this specification, thermal image sequence analysis and weighted deviation calculation significantly reduce welding errors caused by workpiece thermal deformation. A synchronized update mechanism avoids robot motion conflicts, preventing untimely robot updates or outdated welding parameters from affecting overall welding quality. Furthermore, dynamic threshold segmentation, combined with the definition of different temperature zones, enables the welding robot system to maintain optimal welding performance under various operating conditions, demonstrating broad engineering application value.

[0130] Figure 10 FIG. 1 is an exemplary flow chart of updating welding parameters according to some embodiments of the present invention. Figure 10 As shown, the process 1000 includes the following steps 1010 to 1040. Steps 1010 to 1040 may be executed by a processor.

[0131] Step 1010 : identifying at least one welding joint based on the welding path set.

[0132] A weld intersection is a point where the welding paths of two or more welding robots intersect. This section uses the example of two robots' welding paths intersecting as an example.

[0133] In some embodiments, the processor may extract a point where any two welding paths intersect in the welding path set as a welding intersection point.

[0134] In some embodiments, the processor may perform steps 1020 - 1040 for each weld joint.

[0135] Step 1020: Determine the first object and the second object corresponding to the welding intersection.

[0136] The first object refers to the welding robot that has currently welded the welding joint, that is, the welding robot that welded the welding joint first. If there is no welding robot that has completed welding the welding joint, the first object can be marked as empty.

[0137] The second object refers to the welding robot that is currently needed but has not yet welded the welding joint, that is, the welding robot that will weld the welding joint later. If all welding robots that need to weld the welding joint have completed welding, the second object can be marked as empty.

[0138] In some embodiments, the processor can query the welding path set based on the current time point, determine the welding robot whose preset welding path includes a welding joint, and calculate and determine whether the welding robot has passed the welding joint based on the current time point and the welding gun movement speed; the welding robot that has completed welding on the welding joint is taken as the first object, and the welding robot that has not yet completed welding on the welding joint is taken as the second object.

[0139] Step 1030 : Determine a temperature influence domain corresponding to the weld junction based on an average residence time of the first object in an area adjacent to the weld junction.

[0140] The adjacent area refers to the area within a first preset range around the weld intersection. The preset range can be set based on demand or experience. For example, the first preset range can be a circular area with a radius R1 centered at the weld intersection. R1 can be set based on demand or experience.

[0141] In some embodiments, the adjacent regions are determined based on current welding parameters of the first object.

[0142] In some embodiments, the radius R1 of the adjacent region is positively correlated with the current welding temperature of the welding gun.

[0143] Determining the adjacent region based on the current welding parameters of the first object can better determine the extent of the first object's impact on the weld joint during the current welding process. For example, the higher the welding torch temperature of the first object, the greater the range of the weld joint's temperature impact, and the larger the adjacent region that needs to be considered.

[0144] The average dwell time refers to the dwell time per unit area of ​​the first object within the adjacent area of ​​the weld joint. The processor may obtain the total dwell time of the first object within the adjacent area through the camera and calculate the average of the total dwell time and the area of ​​the adjacent area as the average dwell time.

[0145] The temperature influence zone refers to the area within a second preset range around the weld intersection. This preset range can be set based on demand or experience. For example, the second preset range can be a circular area with a radius R2 centered at the weld intersection. R2 can be set based on demand or experience. For example, R2 is positively correlated with the average dwell time, and R2 is greater than R1.

[0146] In some embodiments, the temperature impact domain is related to the network upload rate. For more information about the network upload rate, see Figure 9 Related instructions.

[0147] In some embodiments, the temperature impact domain is negatively correlated with the network upload rate.

[0148] The slower the network upload rate, the greater the data processing delay, and the current data is likely not the data presented on the current weldment; the greater the heat accumulation caused by the decision delay accompanying the data delay. By appropriately increasing the temperature influence domain, the decision range can be expanded, thereby offsetting the decision delay.

[0149] Step 1040: Update the welding parameters of the second object in the temperature influence domain based on the average dwell time and the temperature influence domain.

[0150] In some embodiments, in response to the average dwell time being greater than the dwell time threshold, the processor may query the welding process standard through the welding process module 130 to determine whether the welding temperature of the weld is adjustable; in response to the welding temperature being adjustable, the processor may lower the welding temperature (or adjust the welding temperature by adjusting the welding gun current); in response to the welding temperature being unadjustable, the processor may increase the moving speed of the welding gun.

[0151] In some embodiments, the adjustment range of the welding temperature, welding gun current, and moving speed is positively correlated with R2 of the temperature influence domain.

[0152] In some embodiments of the present specification, by determining a first object and a second object and updating the welding parameters of the second object based on an average dwell time and a temperature influence domain, the welding parameters of a welding robot that passes through a welding joint can be optimized, thereby reducing thermal distortion and improving welding quality.

[0153] It should be noted that the above description of the relevant processes is for illustration and purpose only and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to the processes under the guidance of this specification. However, such modifications and changes are still within the scope of this specification.

[0154] Some embodiments of the present specification provide a welding robot operation device, including a processor, and the processor is used to execute a welding robot operation method.

[0155] Some embodiments of the present specification provide a computer-readable storage medium, which stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a welding robot operation method.

[0156] While the basic concepts have been described above, it will be apparent to those skilled in the art that the detailed disclosure is merely illustrative and does not limit the present invention. Although not explicitly described herein, those skilled in the art may make various modifications, improvements, and revisions to the present invention. Such modifications, improvements, and revisions are suggested in the present invention and remain within the spirit and scope of the exemplary embodiments of the present invention.

[0157] At the same time, the present invention uses specific terms to describe embodiments of the present invention. For example, "one embodiment," "an embodiment," and / or "some embodiments" refer to a certain feature, structure, or characteristic associated with at least one embodiment of the present invention. Therefore, it should be emphasized and noted that the use of "one embodiment," "an embodiment," or "an alternative embodiment" mentioned twice or more in different places in the present invention does not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of the present invention may be appropriately combined.

[0158] In addition, unless expressly stated in the invention, the order of the processing elements and sequences, the use of alphanumeric characters, or the use of other names described in the present invention are not intended to limit the order of the processes and methods of the present invention. Although the above disclosure discusses some embodiments of the invention that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the present invention is intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of the present invention. 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.

[0159] Similarly, it should be noted that in order to simplify the description of the present invention and facilitate understanding of one or more embodiments of the invention, the foregoing description of the embodiments of the present invention sometimes combines multiple features into a single embodiment, figure, or description thereof. In reality, the features of an embodiment may be less than all the features of a single embodiment disclosed above.

[0160] In some embodiments, the numerical parameters used in the present invention are approximate values, which may vary depending on the desired characteristics of individual embodiments. In some embodiments, numerical parameters should take into account the specified number of significant digits and adopt a general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of the present invention are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0161] Finally, it should be understood that the embodiments described herein are intended only to illustrate the principles of the present invention. Other variations may also fall within the scope of the present invention. Therefore, by way of example and not limitation, alternative configurations of the embodiments of the present invention may be considered consistent with the teachings of the present invention. Accordingly, the embodiments of the present invention are not limited to the embodiments explicitly described and illustrated herein.

Claims

1. A welding robot operation system, characterized in that: include: a welding planning module configured to generate a welding path set based on a required weld, the welding path set including a plurality of preset welding paths, the welding path set instructing a plurality of welding robots to collaboratively weld a weldment based on the plurality of preset welding paths in the welding path set; a visual processing module configured to calculate a welding deviation based on a real scanning path of each welding robot and the preset welding path during the collaborative welding; The welding process module is configured to: Based on the preset welding path set and welding parameters, controlling the multiple welding robots to perform the collaborative welding on the weldment; updating the welding path set based on the welding deviation; The visual processing module is further configured to: generate a sequence of thermal images of the weldment, the sequence of thermal images comprising a plurality of thermal images at a plurality of time points; The welding process module is further configured to: During the collaborative welding, for each welding robot, based on the actual scanning path of the welding robot and the temperature zone passed by the actual scanning path, a weighted cumulative deviation of the welding robot is determined, and the weighted cumulative deviation is used as the welding deviation; determining an update time point based on a plurality of the thermal imaging images in the thermal imaging image sequence and the welding deviation, and updating the welding path set at the update time point based on the welding deviation; The visual processing module is further configured to: generate a sequence of optical images of the weldment; The welding process module is further configured to: Based on the welding path set, determining a plurality of target points; determining a thermal distortion rate of each of the target points based on the optical image sequence; The update time point is selected from a set of update time points based on the welding deviation and the thermal distortion rate.

2. The system according to claim 1, wherein The welding process module is further configured to: identifying at least one weld intersection based on the set of weld paths; For each weld junction: Determining a first object and a second object corresponding to the welding joint, wherein the first object refers to the welding robot that is currently welding the welding joint, and the second object refers to the welding robot that is currently not welding the welding joint; determining a temperature influence domain corresponding to the weld junction based on an average residence time of the first object in an area adjacent to the weld junction; Based on the average dwell time and the temperature influence domain, the welding parameters of the second object within the temperature influence domain are updated.

3. A welding robot operation method, characterized in that: include: generating a welding path set based on the required weld, the welding path set including a plurality of preset welding paths, the welding path set instructing a plurality of welding robots to collaboratively weld the weldment based on the plurality of preset welding paths in the welding path set; During the collaborative welding, a welding deviation is calculated based on the actual scanning path of each welding robot and the preset welding path; Based on the preset welding path set and welding parameters, controlling the multiple welding robots to perform the collaborative welding on the weldment; updating the welding path set based on the welding deviation; The method further comprises: generating a sequence of thermal images of the weldment, wherein the sequence of thermal images comprises a plurality of thermal images at a plurality of time points; During the collaborative welding, for each welding robot, based on the actual scanning path of the welding robot and the temperature zone passed by the actual scanning path, a weighted cumulative deviation of the welding robot is determined, and the weighted cumulative deviation is used as the welding deviation; determining an update time point based on a plurality of the thermal imaging images in the thermal imaging image sequence and the welding deviation, and updating the welding path set at the update time point based on the welding deviation; The method further comprises: generating a sequence of optical images of the weldment; Based on the welding path set, determining a plurality of target points; determining a thermal distortion rate of each of the target points based on the optical image sequence; The update time point is selected from a set of update time points based on the welding deviation and the thermal distortion rate.

4. The method according to claim 3, wherein The method further comprises: identifying at least one weld intersection based on the set of weld paths; For each weld junction: Determining a first object and a second object corresponding to the welding joint, wherein the first object refers to the welding robot that is currently welding the welding joint, and the second object refers to the welding robot that is currently not welding the welding joint; determining a temperature influence domain corresponding to the weld junction based on an average residence time of the first object in an area adjacent to the weld junction; Based on the average dwell time and the temperature influence domain, the welding parameters of the second object within the temperature influence domain are updated.

5. A welding robot operation device, comprising a processor, wherein the processor is configured to execute the welding robot operation method according to claim 3.

6. A computer-readable storage medium storing computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the welding robot operation method according to claim 3.

Images