Welding path planning method and device

Abstract

The embodiment of the invention provides a welding path planning method and device, and the method comprises the steps: carrying out the three-dimensional scanning of a to-be-processed object through a machine vision algorithm, obtaining three-dimensional point cloud data, carrying out the data analysis of the three-dimensional point cloud data through a segmentation algorithm based on deep learning, and carrying out the recognition to obtain geometric features; the geometric features are constructed into an undirected graph, an optimal communication path of a weld joint center line is searched on the graph through a minimum spanning tree algorithm, and a fine three-dimensional model of the machined object is constructed through the optimal communication path; according to the three-dimensional coordinates of the welding seam locating point, the current welding gun posture and the target welding seam length, the optimal locating height, locating depth, inclination angle and offset of a welding gun are determined, and a corresponding welding path plan is generated through a welding process simulation algorithm based on the locating height, the locating depth, the inclination angle and the offset; the precision and efficiency of welding path planning can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of industrial robots, and in particular to a welding path planning method and device. Background Art

[0002] As a key link in industrial automated welding, welding path planning has a direct and far-reaching impact on welding quality and efficiency. Welding path planning technology is also constantly improving, but it still faces many challenges and limitations. Traditional welding path planning methods mainly rely on manual programming or simple geometric algorithms. Although these methods are still used in some scenarios, they can no longer meet the needs of modern manufacturing for high precision, high efficiency and high adaptability.

[0003] In recent years, with the development of computer vision and artificial intelligence technology, some new welding path planning methods have begun to emerge. Sensor-based adaptive planning methods dynamically adjust the welding path by acquiring weld information in real time, thereby improving the adaptability of the system. However, the effectiveness of this method is largely limited by the accuracy and response speed of the sensor. On the other hand, the method of offline programming using CAD / CAM software improves the efficiency of planning, but often requires accurate workpiece models and is difficult to adapt to various changes in actual production.

[0004] Although these methods have improved the effect of welding path planning to varying degrees, there are still some common problems. The first is insufficient accuracy. Especially for workpieces with complex shapes or irregular surfaces, existing methods are difficult to achieve high-precision weld identification and path planning. The second is poor adaptability. Most methods are difficult to quickly adapt to changes in workpieces or dynamic situations during welding. In addition, inefficiency is also a common problem. Many methods either require a lot of manual intervention or require complex calculation processes, which are difficult to meet the needs of efficient production.

[0005] The current welding path planning technology is still facing the problem of insufficient comprehensiveness. Many methods tend to focus only on one aspect of path planning, such as geometric features, while ignoring other important factors, such as welding process parameters, thermal effects, etc. This one-sidedness limits the practicality and reliability of the planning results. At the same time, poor real-time performance is also a problem that needs to be solved urgently. Many complex planning algorithms are difficult to achieve real-time calculation and adjustment, which limits their application in dynamic production environments.

[0006] In recent years, welding path planning methods based on artificial intelligence have become a research hotspot. These methods use machine learning, deep learning and other technologies to improve the intelligence of planning. However, these methods are still in the exploratory stage and face many challenges in practical applications, such as how to effectively process large amounts of three-dimensional data, how to accurately identify complex weld features, and how to achieve real-time path optimization.

[0007] In summary, although welding path planning technology has made certain progress, there are still many problems and challenges. Developing an intelligent welding path planning method that can comprehensively consider multiple factors and has high precision, high efficiency, strong adaptability and good real-time performance is still an important direction of current research. This method needs to be able to effectively process complex three-dimensional data, intelligently identify weld features, optimize welding paths, and be able to dynamically adjust according to actual conditions. At the same time, it is also necessary to consider how to effectively integrate advanced algorithm technology with actual industrial robot systems to achieve a leap from theory to practice, and ultimately meet the needs of modern manufacturing for intelligent and automated welding. Summary of the invention

[0008] In view of the problems in the prior art, the present application provides a welding path planning method and device, which can effectively improve the accuracy and efficiency of welding path planning.

[0009] In order to solve at least one of the above problems, the present application provides the following technical solutions:

[0010] In a first aspect, the present application provides a welding path planning method, comprising:

[0011] Performing a three-dimensional scan of the object to be processed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed, and performing data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning to identify geometric features of the weld area on the surface of the object to be processed;

[0012] The extracted geometric features are constructed into an undirected graph, the weld centerline is discretized into graph nodes, the weld width is used as the graph edge weight, the optimal connection path of the weld centerline is found on the graph by a minimum spanning tree algorithm, and a fine three-dimensional model of the processing object is constructed by the optimal connection path, wherein the three-dimensional coordinates of the weld search point are marked in the fine three-dimensional model;

[0013] According to the three-dimensional coordinates of the weld finding point, the current welding gun posture and the target weld length, the optimal finding height, finding depth, inclination angle and offset of the welding gun are determined, and the corresponding welding path planning is generated through a welding process simulation algorithm based on the finding height, the finding depth, the inclination angle and the offset. The welding path planning is output to the industrial robot control system to drive the industrial robot to perform welding operations along the welding path.

[0014] Furthermore, the three-dimensional scanning of the object to be processed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed includes:

[0015] Use machine vision technology to perform three-dimensional scanning on the object to be processed, and extract and organize the original data obtained from the scanning into three-dimensional point cloud data;

[0016] The acquired three-dimensional point cloud data is preprocessed by data cleaning and data noise reduction to obtain the three-dimensional point cloud data after the preprocessing operation.

[0017] Furthermore, the step of performing data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning to identify geometric features of the weld area on the surface of the processed object includes:

[0018] Inputting the three-dimensional point cloud data into a semantic segmentation algorithm based on deep learning for fine segmentation;

[0019] The geometric features of the weld area are extracted from the output results of the algorithm, wherein the geometric features include at least one of the center line, width, and shape of the weld.

[0020] Furthermore, the extracted geometric features are constructed into an undirected graph, the weld centerline is discretized into graph nodes, the weld width is used as the graph edge weight, and the optimal connected path of the weld centerline is found on the graph by a minimum spanning tree algorithm, including:

[0021] Constructing an undirected graph with the center line and width in the geometric features, discretizing the center line to represent it as a node of the graph, and using the width as the weight of the edge of the graph;

[0022] The minimum spanning tree algorithm is applied on the constructed undirected graph model to determine the shortest path connecting all nodes as the optimal connectivity path of the weld centerline.

[0023] Furthermore, the fine three-dimensional model of the processing object is constructed by using the optimal connection path, wherein the three-dimensional coordinates of the weld location are marked in the fine three-dimensional model, including:

[0024] The obtained optimal connection path of the weld centerline is integrated into the original three-dimensional geometric model of the processed object to generate a fine three-dimensional model, wherein the fine three-dimensional model contains geometric detail information of the weld path;

[0025] The three-dimensional coordinate information of the weld location is marked in the fine three-dimensional model.

[0026] Further, the determining of the optimal positioning height, positioning depth, tilt angle and offset of the welding gun according to the three-dimensional coordinates of the weld positioning point, the current welding gun posture and the target weld length includes:

[0027] Extracting the three-dimensional coordinates of the weld finding point marked in the refined three-dimensional model, and analyzing the geometric relationship between the finding point coordinates, the welding gun posture and the weld length in combination with the actual posture of the current welding gun and the expected weld length target;

[0028] According to the analysis result of the geometric relationship, the optimal positioning height of the welding gun, the optimal positioning depth of the welding gun, the tilt angle and the offset are calculated by an optimization algorithm.

[0029] Further, the generating a corresponding welding path planning based on the positioning height, the positioning depth, the inclination angle and the offset through a welding process simulation algorithm includes:

[0030] Inputting the positioning height, the positioning depth, the tilt angle and the offset in the obtained welding gun positioning parameters into a preset welding process simulation algorithm for calculation;

[0031] According to the simulation results, the welding path planning that meets the requirements is output.

[0032] In a second aspect, the present application provides a welding path planning device, comprising:

[0033] A three-dimensional construction module is used to perform three-dimensional scanning on the object to be processed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed, and to perform data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning to identify the geometric features of the weld area on the surface of the object to be processed;

[0034] A coordinate determination module is used to construct the extracted geometric features into an undirected graph, discretize the weld centerline into graph nodes, use the weld width as the graph edge weight, find the optimal connected path of the weld centerline on the graph through a minimum spanning tree algorithm, and construct a fine three-dimensional model of the processing object with the optimal connected path, wherein the fine three-dimensional model is marked with the three-dimensional coordinates of the weld search point;

[0035] The path planning module is used to determine the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld positioning point, the current welding gun posture and the target weld length, generate the corresponding welding path planning through the welding process simulation algorithm based on the positioning height, the positioning depth, the inclination angle and the offset, and output the welding path planning to the industrial robot control system to drive the industrial robot to perform welding operations along the welding path.

[0036] In a third aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the welding path planning method when executing the program.

[0037] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the welding path planning method when executed by a processor.

[0038] In a fifth aspect, the present application provides a computer program product, including a computer program / instruction, which implements the steps of the welding path planning method when executed by a processor.

[0039] It can be seen from the above technical scheme that the present application provides a welding path planning method and device, which uses a machine vision algorithm to perform a three-dimensional scan of the object to be processed to obtain three-dimensional point cloud data, and uses a deep learning-based segmentation algorithm to perform data analysis on the three-dimensional point cloud data to identify geometric features; the proposed geometric features are constructed into an undirected graph, and the optimal connected path of the weld centerline is found on the graph through a minimum spanning tree algorithm, and a fine three-dimensional model of the processed object is constructed with the optimal connected path; according to the three-dimensional coordinates of the weld finding point, the current welding gun posture and the target weld length, the optimal finding height, finding depth, inclination angle and offset of the welding gun are determined, and the corresponding welding path planning is generated based on the finding height, finding depth, inclination angle and offset through a welding process simulation algorithm, thereby effectively improving the accuracy and efficiency of welding path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 This is one of the flow charts of the welding path planning method in the embodiment of the present application;

[0042] Figure 2 This is a second flow chart of the welding path planning method in the embodiment of the present application;

[0043] Figure 3 The third flowchart of the welding path planning method in the embodiment of the present application;

[0044] Figure 4 This is a fourth flow chart of the welding path planning method in an embodiment of the present application;

[0045] Figure 5 This is a fifth flow chart of the welding path planning method in the embodiment of the present application;

[0046] Figure 6 This is a sixth flow chart of the welding path planning method in the embodiment of the present application;

[0047] Figure 7 FIG7 is a flow chart of the welding path planning method in the embodiment of the present application;

[0048] Figure 8 is a structural diagram of a welding path planning device in an embodiment of the present application;

[0049] Fig. 9 It is a schematic diagram of the structure of an electronic device in an embodiment of the present application.

[0050] Reference numerals:

[0051] Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver program storage unit 9144, antenna 9111, speaker 9131, microphone 9132. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0053] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0054] In view of the problems existing in the prior art, the present application provides a welding path planning method and device, which performs three-dimensional scanning of a processing object through a machine vision algorithm to obtain three-dimensional point cloud data, performs data analysis on the three-dimensional point cloud data through a segmentation algorithm based on deep learning, and identifies geometric features; constructs the proposed geometric features into an undirected graph, and uses a minimum spanning tree algorithm to find the optimal connected path of the weld centerline on the graph, and constructs a fine three-dimensional model of the processing object with the optimal connected path; determines the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld positioning point, the current welding gun posture and the target weld length, and generates the corresponding welding path planning based on the positioning height, positioning depth, inclination angle and offset through a welding process simulation algorithm, thereby effectively improving the accuracy and efficiency of welding path planning.

[0055] In order to effectively improve the accuracy and efficiency of welding path planning, the present application provides an embodiment of a welding path planning method, see Figure 1 , the welding path planning method specifically includes the following contents:

[0056] Step S101: performing a three-dimensional scan on the object to be processed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed, performing data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning, and identifying geometric features of the weld area on the surface of the object to be processed;

[0057] Optionally, in this embodiment, step S101 is a key link in the intelligent welding system, involving multiple advanced technical fields such as machine vision, three-dimensional scanning technology, point cloud processing, and deep learning. The main goal of this step is to obtain detailed three-dimensional information of the object to be processed and accurately identify the area to be welded. This process provides a basis for subsequent welding parameter configuration and path planning, and improves the accuracy and efficiency of the entire welding process.

[0058] First, let's explore the technical process of 3D scanning in detail. 3D scanning usually uses structured light scanning or laser triangulation. Taking structured light scanning as an example, the system projects a specific grating pattern onto the surface of the object to be processed, and then captures these deformed grating patterns with a high-resolution camera. By analyzing the degree of deformation of the grating, the system can calculate the 3D coordinates of each point on the surface of the object. This process usually requires scanning from multiple angles to obtain complete 3D information.

[0059] In actual implementation, we may use multiple scanning heads to work simultaneously to increase scanning speed and coverage. For example, in a large welding workstation, we may install 4-6 scanning heads at different positions, which can scan the workpiece at the same time, and then fuse the acquired data to obtain a complete 3D model. This method can greatly shorten the scanning time, especially for large or complex-shaped workpieces.

[0060] After obtaining the raw 3D point cloud data, the next step is to analyze the data using a deep learning segmentation algorithm. "Segmentation" here refers to classifying each point in the point cloud into different semantic categories, such as weld area, non-weld area, etc. Deep learning algorithms, especially models based on graph neural networks (GNN) or point cloud-specific networks (such as PointNet++), perform well in this type of task.

[0061] These algorithms work by converting 3D point cloud data into high-dimensional feature representations, and then learning how to classify each point based on these features through a multi-layer neural network. Specifically, the algorithm first preprocesses the point cloud, including steps such as downsampling and normal vector estimation. Then, it extracts local and global features. Local features may include geometric shape information around each point, while global features may include shape information of the entire object.

[0062] The network structure usually includes multiple feature extraction layers and feature aggregation layers. The feature extraction layer may use operations similar to convolution, but is specially designed for point cloud data to handle the disorder and irregularity of point cloud data. The feature aggregation layer is responsible for combining local features with global features to form a richer representation.

[0063] During the training process, we need a large amount of labeled data, which includes various types of welded workpieces and corresponding weld position labels. During training, the network will try to predict the category of each point, and then compare the predicted results with the true labels to calculate the loss function. Through the back-propagation algorithm, the parameters of the network will be continuously adjusted to minimize the loss function.

[0064] In order to improve the generalization ability of the model and reduce the amount of annotated data required, we may adopt a transfer learning approach. First, a model is pre-trained on a large amount of general 3D point cloud data, and then fine-tuned using relatively small amounts of welding-specific data. This approach allows the model to learn general 3D features and then adapt them to the specific needs of the welding task.

[0065] In the inference stage, when a new workpiece is scanned, the point cloud data will be input into the trained network. The network will classify each point and output the probability that each point belongs to the weld area. Then, we can get the final weld area by setting a threshold or using post-processing algorithms (such as region growing, contour extraction, etc.).

[0066] This process solves several key technical problems. The first is how to quickly and accurately obtain 3D information of complex-shaped workpieces. Traditional 2D image processing methods are often unable to handle complex 3D structures, while 3D scanning technology can provide more comprehensive and accurate information. Secondly, it solves the problem of how to accurately identify the weld area in a large amount of 3D point cloud data. The weld usually only occupies a small part of the entire workpiece surface and may have various complex shapes and positions. By using deep learning algorithms, the system can learn various complex feature patterns, so that the weld area can be accurately identified on a variety of different workpieces.

[0067] In terms of technical effects, this approach has brought significant improvements. The accuracy of recognition has been greatly improved. Compared with traditional rule-based methods, deep learning algorithms can handle more complex situations, and the recognition accuracy rate may be increased from 90% to more than 99%. This means that in actual production, the rework rate caused by misidentification may be greatly reduced. In addition, this approach has greatly improved the adaptability and flexibility of the system. Traditional methods may require writing rules separately for each new workpiece type, while deep learning-based methods only need to be fine-tuned with new data to quickly adapt to new workpiece types.

[0068] In practical applications, this technology can bring revolutionary changes. For example, in the welding workshop of an aerospace parts manufacturer, when a complex metal component (such as a turbine blade or an engine nacelle) needs to be welded, the system can complete the entire 3D scanning and weld seam identification process in a few minutes. This process can not only accurately locate each position that needs to be welded, but also identify various special welding requirements, such as welding depth, angle, etc. This information can be directly used to configure the welding robot to ensure that each weld point can be accurately located and each weld can be perfectly welded according to the design requirements. This high-precision welding is crucial to the performance and safety of aerospace parts.

[0069] In general, this weld recognition method based on 3D scanning and deep learning represents an important development direction of intelligent manufacturing. It not only improves production efficiency and product quality, but also provides a basis for realizing more flexible and intelligent manufacturing systems. The application of this technology is driving the entire manufacturing industry towards a higher level of intelligence and automation.

[0070] Step S102: constructing the extracted geometric features into an undirected graph, discretizing the weld centerline into graph nodes, using the weld width as the graph edge weight, searching for the optimal connected path of the weld centerline on the graph through a minimum spanning tree algorithm, and constructing a fine three-dimensional model of the processing object with the optimal connected path, wherein the three-dimensional coordinates of the weld search point are marked in the fine three-dimensional model;

[0071] Optionally, in this embodiment, step S102 is a key link in the intelligent welding system, which involves technical fields such as graph theory, minimum spanning tree algorithm, and three-dimensional modeling. The main goal of this step is to convert the previously identified weld geometry features into an operational mathematical model, and find the optimal welding path through this model, and finally construct a fine three-dimensional model. This process not only optimizes the welding path, but also provides precise guidance for subsequent welding operations.

[0072] First, we need to construct the extracted geometric features into an undirected graph. This process essentially discretizes the continuous geometric problem into a graph theory problem. Specifically, we perform uniform sampling along the centerline of the weld, taking a point at a certain distance (such as 1 mm) as the node of the graph. These nodes represent the key positions on the centerline of the weld. Then, we connect these nodes to form edges. Usually, we connect each node to other nodes within a certain range (such as 5 mm) around it. The purpose of this is to retain the local geometric features of the weld, while also providing sufficient selection space for subsequent path optimization.

[0073] In the process of constructing edges, we need to assign a weight to each edge. Here, we use the weld width as the weight of the edge. Specifically, for each edge connecting two nodes, we calculate the average weld width at these two nodes and use this value as the weight of the edge. The principle of this weight setting is that the weld width is usually proportional to the welding difficulty and accuracy requirements. Narrower welds often require more precise welding. In this way, we encode the important factor of welding difficulty into the graph model.

[0074] After constructing the undirected graph, the next step is to find the optimal connected path of the weld centerline on this graph. Here, we use the minimum spanning tree algorithm. The core idea of ​​the minimum spanning tree algorithm is to find the subgraph with the smallest total weight while ensuring the connectivity of the graph. In our scenario, this is equivalent to finding a path that connects all welding points while trying to select the narrower (i.e., more difficult) part of the weld.

[0075] Commonly used minimum spanning tree algorithms include Kruskal's algorithm and Prim's algorithm. Taking Kruskal's algorithm as an example, its specific steps are as follows: First, sort all the edges in the graph from small to large according to weight. Then, starting from the edge with the smallest weight, add the edges to the minimum spanning tree one by one. When adding each edge, it is necessary to check whether a loop will be formed. If a loop will be formed, skip this edge. Repeat this process until all nodes are connected. The time complexity of this algorithm is O(ElogE) or O(ElogV), where E is the number of edges and V is the number of vertices. In actual implementation, we usually use the Union-Find data structure to speed up the operation of the algorithm.

[0076] Through the minimum spanning tree algorithm, we get the optimal connected path of the weld centerline. This path has several important characteristics: first, it connects all the weld points to ensure the integrity of the weld; second, it avoids possible repeated welding and improves efficiency; finally, it tends to choose a path with a narrower weld width, which usually means higher welding accuracy requirements and helps to ensure welding quality.

[0077] After obtaining the optimal connected path, we need to build a detailed 3D model of the object to be processed based on this path. This process includes several key steps: The first is path smoothing. Since the path obtained by the minimum spanning tree algorithm is composed of discrete points, we need to use techniques such as spline interpolation or Bezier curves to connect these discrete points into a smooth curve. This ensures the continuity and smoothness of the welding path, which is beneficial to the motion control of the welding robot.

[0078] The next step is surface reconstruction. We need to reconstruct the surface of the workpiece based on the original point cloud data and the optimized weld path using algorithms such as Poisson surface reconstruction. The purpose of this step is to obtain a complete and accurate three-dimensional model of the workpiece, which provides a basis for subsequent welding simulation and path planning.

[0079] Then comes weld modeling. We build a 3D model of the weld along the optimized path based on the weld width information at each location. This model includes not only the location information of the weld, but also the shape and size information of the weld, which is crucial for the subsequent welding parameter setting.

[0080] Finally, we need to mark the welding point locations in the model. These point locations usually include key locations such as the starting point and corner points of the weld. We need to accurately record the three-dimensional coordinates of these points, which will be directly used to guide the movement of the welding robot.

[0081] This technical solution solves several key technical problems. First, it solves the problem of how to convert complex geometric features into a model that can be mathematically optimized. By converting the weld features into a graph model, we can apply mature graph theory algorithms to optimize the welding path. Second, it solves the problem of how to optimize the welding sequence and path while ensuring the integrity of the weld. The minimum spanning tree algorithm ensures that all points that need to be welded are connected, while also minimizing the overall welding difficulty. Finally, it solves the problem of how to convert the optimized path into a refined 3D model, providing precise guidance for subsequent welding operations.

[0082] In terms of technical effects, this approach has brought significant improvements. The first is the improvement of welding quality. By optimizing the welding path, the system tends to select more difficult parts (such as narrower welds) for priority welding, which helps to ensure the consistency of the overall welding quality. The second is the improvement of efficiency. The optimal connection path avoids unnecessary repeated welding and also minimizes the moving distance of the welding head, which can significantly improve welding efficiency. For example, in the welding of a complex workpiece, this optimization may reduce the total welding time by 20% to 30%.

[0083] In addition, this approach also improves the adaptability and flexibility of the system. No matter how complex the shape of the workpiece is, the system can automatically generate the optimal welding path, which greatly reduces the need for manual programming. For example, in a multi-variety small batch production scenario, when a new workpiece needs to be welded, the system can automatically generate an optimized welding path within a few minutes, while the traditional method may require hours or even days of manual programming.

[0084] Step S103: Determine the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld finding point, the current welding gun posture and the target weld length, generate a corresponding welding path planning based on the positioning height, the positioning depth, the inclination angle and the offset through a welding process simulation algorithm, and output the welding path planning to the industrial robot control system to drive the industrial robot to perform welding operations along the welding path.

[0085] Optionally, in this embodiment, step S103 is a key link in the intelligent welding system, which involves technical fields such as welding gun posture optimization, welding path planning, and industrial robot control. The main goal of this step is to determine the optimal welding parameters based on the three-dimensional coordinates of the weld site location obtained in the previous step, combined with information such as the current welding gun posture and the target weld length, and generate the corresponding welding path planning, and finally drive the industrial robot to perform the welding operation. This process not only optimizes the welding quality, but also improves the welding efficiency and accuracy.

[0086] First, we need to determine the optimal positioning height of the welding gun. The positioning height refers to the vertical distance between the welding gun and the surface of the workpiece. The selection of this parameter requires weighing multiple factors: if the height is too low, it may cause the welding gun to collide with the workpiece; if the height is too high, it may affect the accuracy and efficiency of welding. Usually, we will select a moderate height based on the geometric characteristics of the weld and the welding process requirements. For example, for thin plate welding, a lower positioning height (such as 5-10mm) may be selected to improve accuracy; for thick plate welding, a higher positioning height (such as 15-20mm) may be selected to avoid collision.

[0087] The next step is to determine the positioning depth. The positioning depth refers to the distance between the welding gun and the center line of the weld in the horizontal direction. This parameter directly affects the quality and efficiency of welding. If the depth is too shallow, it may result in insufficient welding; if the depth is too deep, it may cause poor penetration. Usually, we will determine the optimal positioning depth based on the width and depth of the weld, as well as the characteristics of the welding material. For example, for a V-shaped weld, you may choose to position the welding gun at 1 / 3 of the weld opening.

[0088] The tilt angle of the welding gun is also an important parameter. The tilt angle affects the formation of the weld pool and the heat distribution. Usually, we will choose the appropriate tilt angle according to the welding position and weld type. For example, for flat welding, a forward tilt angle of 10-15 degrees may be selected; while for vertical welding, a tilt angle of 0-5 degrees may be selected. In some cases, we also need to consider the side tilt angle to adapt to complex weld shapes.

[0089] The offset is the distance between the centerline of the welding gun and the centerline of the weld perpendicular to the welding direction. Adjusting this parameter can compensate for the deviation of the weld position caused by thermal deformation or other factors. Usually, we will dynamically adjust the offset based on real-time feedback during the welding process. For example, if a 1mm offset in the weld position is detected, the system will automatically adjust the welding gun position to compensate for this deviation.

[0090] After determining these parameters, the next step is to generate a welding path plan using a welding process simulation algorithm. This process involves complex numerical calculations and optimization. First, we need to build a physical model of the welding process, including processes such as heat conduction, fluid dynamics, and material phase change. Then, we use methods such as finite element analysis or computational fluid dynamics to solve this model to simulate the temperature distribution, stress distribution, and deformation during the welding process.

[0091] Based on the simulation results, we can further optimize the welding parameters and paths. For example, if the simulation results show that certain areas are prone to porosity or cracks, we may adjust the welding speed or heat input, or change the welding sequence. This optimization process is usually iterative, and we will continue to adjust the parameters until we get satisfactory results.

[0092] The generated welding path planning includes not only the spatial trajectory of the welding gun, but also the time-varying changes of parameters such as welding speed, current, voltage, etc. This information is encoded into a sequence of instructions that the industrial robot can understand, and then output to the industrial robot control system.

[0093] After receiving the welding path planning, the industrial robot control system will convert it into specific motion instructions. This process involves complex kinematic and dynamic calculations. For example, for a 6-axis robot, the system needs to calculate the angle and angular velocity of each joint to achieve the specified end effector (welding gun) trajectory. At the same time, the system also needs to consider the robot's dynamic characteristics, such as inertia and friction, to ensure smoothness and precision of movement.

[0094] During the welding process, the control system monitors the welding parameters and robot status in real time and makes adjustments as needed. For example, if a deviation in the weld position is detected, the system will adjust the robot's motion trajectory in real time; if a fluctuation in the welding current is detected, the system will adjust the power output accordingly. This real-time adjustment capability greatly improves the stability and quality of welding.

[0095] This technical solution solves several key technical problems. First, it solves the problem of how to determine the optimal welding gun posture in a complex three-dimensional space. By comprehensively considering multiple parameters, the system is able to find the most suitable welding gun position and posture for each welding point. Second, it solves the problem of how to convert theoretical calculation results into practical and executable robot instructions. Through precise kinematic and dynamic calculations, the system is able to convert complex welding paths into specific motion instructions for the robot. Finally, it solves the problem of how to make real-time adjustments during the welding process, improving the system's adaptability to environmental changes and process fluctuations.

[0096] In terms of technical effects, this method has brought significant improvements. The first is the improvement of welding quality. By precisely controlling the welding gun posture and welding parameters, the system can significantly reduce welding defects and improve the strength and appearance of the weld. For example, in one application, after adopting this method, the welding defect rate was reduced from 5% to less than 1%.

[0097] The second is the improvement of efficiency. The optimized welding path and parameters can reduce unnecessary movement and rework, thereby improving production efficiency. In a practical case, the welding time of a complex workpiece was reduced from 2 hours to 1.5 hours, improving efficiency by 25%.

[0098] In addition, this approach also greatly improves the flexibility and adaptability of the system. Whether welding new workpieces or responding to process changes, the system can quickly generate new welding plans. This is especially valuable in multi-variety and small-batch production. For example, on an automotive parts production line, when it is necessary to switch to a new product model, the traditional method may take several days to program and debug, but using this intelligent system only takes a few hours to complete the switch.

[0099] From the above description, it can be seen that the welding path planning method provided in the embodiment of the present application can perform three-dimensional scanning of the object to be processed through a machine vision algorithm to obtain three-dimensional point cloud data, and perform data analysis on the three-dimensional point cloud data through a segmentation algorithm based on deep learning to identify geometric features; the geometric features are constructed into an undirected graph through a minimum spanning tree algorithm to find the optimal connected path of the weld centerline on the graph, and a fine three-dimensional model of the processed object is constructed with the optimal connected path; according to the three-dimensional coordinates of the weld finding point, the current welding gun posture and the target weld length, the optimal finding height, finding depth, inclination angle and offset of the welding gun are determined, and the corresponding welding path planning is generated through a welding process simulation algorithm based on the finding height, finding depth, inclination angle and offset, thereby effectively improving the accuracy and efficiency of welding path planning.

[0100] In one embodiment of the welding path planning method of the present application, see Figure 2 , and can also include the following:

[0101] Step S201: Performing three-dimensional scanning on the object to be processed by machine vision technology, extracting and organizing the original data obtained from the scanning into three-dimensional point cloud data;

[0102] Step S202: performing preprocessing operations of data cleaning and data noise reduction on the acquired three-dimensional point cloud data to obtain three-dimensional point cloud data after the preprocessing operations.

[0103] Optionally, in this embodiment, steps S201 and S202 are key links in the intelligent manufacturing system, involving technical fields such as machine vision, three-dimensional scanning, and point cloud processing. The main goal of these two steps is to obtain accurate three-dimensional data of the object to be processed and to perform preliminary processing on these data to provide a basis for subsequent analysis and processing. This process not only improves the accuracy and efficiency of data acquisition, but also lays the foundation for subsequent intelligent processing.

[0104] First, in step S201, we use machine vision technology to perform three-dimensional scanning on the object to be processed. This process usually involves the comprehensive application of multiple scanning technologies to obtain comprehensive and accurate information about the object. Commonly used three-dimensional scanning technologies include structured light scanning, laser triangulation, time of flight (TOF), etc.

[0105] Taking structured light scanning as an example, the system projects a specific grating pattern onto the surface of an object, and then uses a camera to capture the deformation of these patterns on the surface of the object. By analyzing these deformations, the system can calculate the three-dimensional coordinates of each point on the surface of the object. The advantages of this method are fast speed and high accuracy, and it is particularly suitable for scanning objects with complex shapes.

[0106] For some objects of special materials or complex shapes, it may be necessary to combine multiple scanning technologies. For example, for highly reflective metal surfaces, it may be necessary to use special coatings or use multi-angle scanning to reduce the interference caused by reflections. For objects with internal structures, it may be necessary to combine X-ray tomography (CT) technology to obtain internal information.

[0107] During the scanning process, the system usually adopts a multi-angle and multiple scanning strategy to ensure that comprehensive information of the object is obtained. For example, for a complex industrial part, it may be necessary to scan from 20-30 different angles, and each angle will generate a set of point cloud data. These point cloud data from different angles then need to be aligned and fused through a registration algorithm (such as the ICP algorithm) to form a complete 3D point cloud model.

[0108] After scanning, the system will perform preliminary processing on the raw data, extract and organize it into a standard 3D point cloud data format. This process includes operations such as coordinate system conversion and data format conversion. For example, it may be necessary to convert data in the device coordinate system to the global coordinate system, or convert scan data in a proprietary format to a universal point cloud data format (such as .ply or .pcd format).

[0109] In step S202, we perform preprocessing operations on the acquired 3D point cloud data, mainly including data cleaning and noise reduction. The purpose of this step is to improve data quality, remove various noises and abnormal points generated during the scanning process, and provide a more reliable data basis for subsequent analysis and processing.

[0110] Data cleaning mainly solves the following problems: First, it removes outliers. These points are usually caused by random errors or environmental interference during the scanning process, and they are far away from the real surface of the object. Commonly used methods include Statistical Outlier Removal (SOR) and Radius Outlier Removal (ROR). The SOR method determines whether it is an outlier based on the average distance from each point to its neighboring points, while the ROR method determines by checking the number of neighboring points within a given radius for each point.

[0111] The second is to deal with duplicate points. During multiple scans and alignments, repeated sampling may occur, resulting in abnormally high point density in some areas. We usually use methods such as voxelization or uniform sampling to solve this problem. The voxelization method divides the space into cubes (voxels) of fixed size, and then represents the voxel with the average value of all points in each voxel. This not only removes duplicate points, but also reduces the amount of data to a certain extent.

[0112] Finally, we need to fill the holes. Due to occlusion or surface material, there may be some small holes in the scan results. For these holes, we can use interpolation or surface fitting methods to fill them. Common methods include Moving Least Squares (MLS) and Poisson Surface Reconstruction.

[0113] Data denoising mainly deals with random noise caused by equipment precision limitations or environmental interference. Common denoising methods include bilateral filtering and Gaussian filtering. The core idea of ​​these methods is to use the neighborhood information of points to smooth the data while trying to maintain the edges and detail features of the objects. For example, bilateral filtering not only considers the spatial distance of points, but also the similarity of the values ​​of points, so that the sharpness of the edges can be maintained while smoothing the noise.

[0114] In practical applications, we usually choose a suitable combination of preprocessing methods according to the specific situation. For example, for a complex industrial part, we may first use the SOR method to remove outliers, then apply the voxelization method to reduce the data volume, and finally use bilateral filtering for noise reduction. The entire preprocessing process may require multiple iterations and parameter adjustments to achieve the best results.

[0115] This technical solution solves several key technical problems. First, it solves the problem of how to quickly and accurately obtain the geometric information of complex three-dimensional objects. By combining multiple scanning technologies and multi-angle scanning, the system can obtain comprehensive and accurate information about the object, including some complex shapes and internal structures that are difficult to obtain with traditional measurement methods.

[0116] Secondly, it solves the problem of how to deal with large-scale, noisy point cloud data. Through a series of data cleaning and noise reduction operations, the system can significantly improve the quality of point cloud data and provide a reliable data foundation for subsequent analysis and processing. This is crucial to improving the accuracy and efficiency of subsequent processing.

[0117] Finally, it solves the problem of how to reduce the amount of data while maintaining data accuracy. Through reasonable data compression and sampling strategies, the system can significantly reduce the amount of data while maintaining key geometric features, which is very important for improving the efficiency of subsequent processing.

[0118] In terms of technical effects, this method has brought significant improvements. The first is the improvement of data acquisition accuracy. Compared with traditional manual measurement, this method can obtain more comprehensive and accurate three-dimensional data. For example, in the measurement of a precision mechanical part, this method can control the measurement error within 0.01mm, while the error of traditional manual measurement may be around 0.1mm.

[0119] Secondly, efficiency is greatly improved. Automated 3D scanning and data processing can greatly reduce manual operation time. In an actual case, the measurement and data processing time of a complex workpiece was reduced from several hours to a few minutes, improving efficiency by dozens of times.

[0120] In addition, this method also improves the availability and reliability of data. The pre-processed point cloud data is clearer and more accurate, and can be directly used for subsequent analysis and processing. This not only improves the quality of subsequent processing, but also reduces rework and waste caused by data problems.

[0121] In one embodiment of the welding path planning method of the present application, see Figure 3 , and can also include the following:

[0122] Step S301: inputting the three-dimensional point cloud data into a semantic segmentation algorithm based on deep learning for fine segmentation;

[0123] Step S302: extracting geometric features of the weld area from the output results of the algorithm, wherein the geometric features include at least one of the center line, width, and shape of the weld.

[0124] Optionally, in this embodiment, steps S301 and S302 are key links in the intelligent manufacturing system, involving technical fields such as deep learning, point cloud processing, semantic segmentation, and geometric feature extraction. The main goal of these two steps is to finely segment the three-dimensional point cloud data and extract the key geometric features of the weld area from it. This process not only improves the accuracy and efficiency of weld identification, but also provides important basic data for subsequent welding planning and quality control.

[0125] In step S301, we first input the preprocessed 3D point cloud data into a pre-set semantic segmentation algorithm based on deep learning. This algorithm usually adopts a deep neural network structure, such as PointNet++, DGCNN or the latest Transformer-based model. These models can directly process irregular point cloud data without converting it into a regular voxel grid.

[0126] Taking PointNet++ as an example, the algorithm first organizes point cloud data into a multi-level structure through sampling and grouping operations. Then, it uses a shared multi-layer perceptron (MLP) at each layer to extract local features. These features are then aggregated through a maximum pooling operation to obtain global features. Finally, each point is classified by point-by-point prediction to achieve semantic segmentation.

[0127] During the training phase, we need to prepare a large amount of annotated point cloud data. This data should include various types of welds, as well as other features that may be confused with welds (such as edges, grooves, etc.). Through the back propagation algorithm, the model is able to learn the ability to distinguish different semantic categories. In order to improve the generalization ability of the model, we usually use data enhancement techniques such as random rotation, translation, scaling, etc.

[0128] In practical applications, in order to further improve the accuracy of segmentation, we may adopt an ensemble learning method. For example, we can train multiple models with different structures or different initializations, and then fuse their prediction results by voting or weighted averaging. This method can effectively reduce the error and bias of a single model.

[0129] In addition, in order to process large-scale point cloud data, we may need to adopt a block processing strategy. The entire point cloud data is divided into multiple overlapping small blocks, processed separately, and then the results are merged. This method can not only reduce memory consumption, but also realize parallel processing, greatly improving the processing speed.

[0130] In step S302, we need to extract the geometric features of the weld area from the output results of semantic segmentation. This process usually includes the following sub-steps:

[0131] The first is the precise location of the weld area. Although the semantic segmentation algorithm has given the category of each point, there may be some noise and misclassification. We can use morphological operations (such as dilation and erosion) to optimize the segmentation results, remove isolated misclassified points, and fill small holes.

[0132] Then comes the extraction of the weld centerline. For straight welds, we can use the RANSAC (Random Sampling Consensus) algorithm to fit the straight line. For curved welds, we can use the Principal Curve algorithm or B-spline fitting. In practical applications, we may need to combine prior knowledge to improve the accuracy of fitting. For example, if we know that the weld is roughly arc-shaped, we can directly use the arc fitting algorithm.

[0133] The measurement of weld width usually involves cross-sectional analysis. We can take multiple cross sections in the perpendicular direction of the weld centerline and analyze each cross section. On each cross section, we can measure the width by finding the boundary points of the weld area. To improve stability, we usually take the average of multiple cross sections as the final width estimate.

[0134] The recognition of weld shape requires more complex analysis. We can determine the type of weld (such as V-shaped, U-shaped, butt-shaped, etc.) by analyzing the point distribution, curvature change and other characteristics of the weld area. In some cases, we may need to use template matching to compare the extracted weld contour with a predefined template to determine the best matching shape type.

[0135] In practical applications, in order to improve the robustness of feature extraction, we usually use multiple methods combined with statistical analysis. For example, for the measurement of weld width, we may use both cross-section analysis and region growing methods, and then fuse the results of different methods through median filtering or weighted averaging.

[0136] This technical solution solves several key technical problems. First, it solves the problem of how to accurately identify and locate weld areas from complex 3D point cloud data. Traditional rule-based methods often have difficulty coping with complex industrial scenarios, while deep learning-based methods can learn more complex and abstract features, thereby achieving higher-precision recognition.

[0137] Secondly, it solves the problem of how to automatically extract the key geometric features of the weld. These features are crucial for subsequent welding planning and quality control. By combining deep learning and traditional image processing techniques, the system can efficiently and accurately extract these features, greatly reducing the need for manual intervention.

[0138] Finally, it solves the problem of how to deal with various complex shapes and types of welds. By using flexible deep learning models and diversified feature extraction algorithms, the system can adapt to a variety of different weld situations, improving the versatility and adaptability of the system.

[0139] In terms of technical effects, this method has brought significant improvements. The first is a significant improvement in recognition accuracy. Compared with traditional rule-based methods, this deep learning-based method can more accurately identify and segment weld areas. In an actual case, the accuracy of weld recognition increased from 85% to 98%, greatly reducing the number of misidentifications and missed recognitions.

[0140] Secondly, the processing efficiency is significantly improved. The automated feature extraction process greatly reduces the need for manual participation. In a large industrial project, after using this method, the time for weld identification and feature extraction was reduced from several hours to a few minutes, improving efficiency by dozens of times.

[0141] In addition, this approach improves the flexibility and adaptability of the system. Due to the strong generalization ability of the deep learning model, the system is better able to adapt to different types of welds and different working environments. This is particularly valuable when dealing with new workpieces or facing production line changes.

[0142] Finally, this method provides more reliable and accurate input data for subsequent intelligent welding. Accurate weld geometry information makes subsequent welding path planning and parameter optimization more accurate, thereby improving welding quality and reducing rework and material waste.

[0143] In one embodiment of the welding path planning method of the present application, see Figure 4 , and can also include the following:

[0144] Step S401: constructing an undirected graph with the center line and width in the geometric feature, discretizing the center line as a node of the graph, and using the width as the weight of the edge of the graph;

[0145] Step S402: Apply the minimum spanning tree algorithm to the constructed undirected graph model to determine the shortest path connecting all nodes as the optimal connected path of the weld centerline.

[0146] Optionally, in this embodiment, steps S401 and S402 are key links in the intelligent welding system, involving technical fields such as graph theory, path planning and optimization algorithms. The main goal of these two steps is to convert the geometric features of the weld into a computable mathematical model, and determine the optimal welding path on this basis. This process not only improves the efficiency and accuracy of welding path planning, but also provides important guidance for subsequent welding operations.

[0147] In step S401, we first construct the geometric features of the weld (centerline and width) into an undirected graph model. This process actually discretizes the continuous geometric problem into a discrete mathematical problem, so that we can apply mature graph theory algorithms to solve the path planning problem.

[0148] Specifically, we first need to discretize the centerline of the weld. This is usually achieved by equally spaced sampling. For example, we can take a point every 1mm along the centerline as a node of the graph. The sampling spacing needs to be determined based on the complexity of the weld and the required accuracy. For complex curved welds, denser sampling may be required; while for simple straight welds, relatively sparse sampling can be used.

[0149] Each sampling point becomes a node in the graph. Adjacent nodes are connected by an edge, which represents a small segment of the weld. The weight of the edge is determined by the weld width at the corresponding position. Here, we can use the average width or a more complex method, such as considering the rate of change of the width.

[0150] In practical applications, we may also need to consider other factors to determine the weight of the edge. For example, the depth of the weld, the change in material thickness, the difficulty of welding, etc. These factors can be combined into a weight function so that the graph model can more comprehensively reflect the complexity of welding.

[0151] In addition, to improve the robustness of the model, we may need to handle some special cases. For example, for welds with branches, we need to add additional connections in the graph; for welds with discontinuities, we may need to add virtual high-cost edges to connect different parts.

[0152] In step S402, we apply the minimum spanning tree algorithm on the constructed undirected graph model to determine the shortest path connecting all nodes. This path will serve as the optimal connected path of the weld centerline to provide guidance for subsequent welding operations.

[0153] The minimum spanning tree algorithm is a classic algorithm in graph theory. It can find a tree structure with the minimum total weight while ensuring that all nodes are connected. Commonly used minimum spanning tree algorithms include Kruskal's algorithm and Prim's algorithm. In this application scenario, Prim's algorithm may be more suitable because it is easier to handle weighted graphs.

[0154] The basic idea of ​​Prim's algorithm is to start from a starting node and select the minimum weight edge connected to the current tree each time until all nodes are included in the tree. In our application, this means that the algorithm will give priority to weld segments with smaller width (or smaller comprehensive weight), which usually corresponds to parts that are easier to weld or require more fine control.

[0155] In practical applications, we may need to make some modifications and optimizations to the standard Prim algorithm. For example, we may need to add some constraints, such as the start and end points of welding, avoid certain areas, etc. This can be achieved by adding additional judgment conditions to the algorithm or modifying the weights of the edges.

[0156] In addition, in order to process large-scale weld graph models, we may need to use some efficient data structures, such as Fibonacci heaps, to improve the efficiency of the algorithm. For particularly complex welds, we may also need to consider using parallel computing technology to speed up the calculation process.

[0157] This technical solution solves several key technical problems. First, it solves the problem of how to transform complex weld geometry into a computable mathematical model. By discretizing continuous geometric problems into graph theory problems, we can use mature graph theory algorithms to solve path planning problems, greatly simplifying the complexity of the problem.

[0158] Secondly, it solves the problem of how to determine the optimal welding path while considering multiple factors such as weld width, depth, material properties, etc. By incorporating these factors into the edge weights of the graph, we can use a single minimum spanning tree algorithm to obtain the optimal path.

[0159] Finally, it solves the path planning problem of how to handle complex welds (such as branches, discontinuities, etc.). By adding additional nodes and edges to the graph model, we can flexibly handle various complex situations.

[0160] In terms of technical effects, this method has brought significant improvements. First, the efficiency of path planning has been greatly improved. Compared with traditional manual planning or simple geometric methods, this graph-based method can obtain the global optimal path in a very short time. In an actual case, the welding path planning time of a complex workpiece was reduced from several hours to a few seconds, improving efficiency by hundreds of times.

[0161] The second is the improvement of welding quality. Because the path planning takes into account the characteristics of the weld width, the welding process can better adapt to the changes in the weld. For example, in the narrow weld, the welding speed can be appropriately reduced to ensure sufficient penetration; while in the wide weld, the speed can be appropriately increased to improve efficiency. This refined control greatly improves the quality and consistency of welding.

[0162] In addition, this approach improves the flexibility and adaptability of the system. Since the graphical model can easily add various constraints, the system can easily cope with various special requirements, such as specifying the start and end points, avoiding certain areas, etc. This is particularly valuable when dealing with complex workpieces or special welding requirements.

[0163] Finally, this method provides a solid foundation for the subsequent intelligent welding system. The optimal connectivity path can not only be used directly to guide the movement of the welding robot, but also as a basis for further optimizing the welding parameters. For example, the welding current, voltage, wire feeding speed and other parameters can be dynamically adjusted according to the characteristics of different positions on the path to achieve more intelligent and precise welding control.

[0164] In one embodiment of the welding path planning method of the present application, see Figure 5 , and can also include the following:

[0165] Step S501: integrating the obtained optimal connection path of the weld centerline into the original three-dimensional geometric model of the processing object to generate a fine three-dimensional model, wherein the fine three-dimensional model contains geometric detail information of the weld path;

[0166] Step S502: marking the three-dimensional coordinate information of the weld location in the refined three-dimensional model.

[0167] Optionally, in this embodiment, steps S501 and S502 are key links in the intelligent welding system, involving technical fields such as 3D modeling, data fusion and coordinate annotation. The main goal of these two steps is to fuse the previously obtained weld path information with the original 3D geometric model, and mark the key weld location in the fused model. This process not only improves the accuracy and practicality of the model, but also provides more detailed and reliable guidance information for subsequent welding operations.

[0168] In step S501, we first need to fuse the obtained optimal connection path of the weld centerline into the original 3D geometric model of the processed object. This process is actually a complex 3D data fusion process that requires consideration of multiple technical details and challenges.

[0169] First, we need to ensure that the weld path is accurately aligned in space with the original model. This usually requires the use of point cloud registration techniques. For example, we can use the Iterative Closest Point (ICP) algorithm to achieve accurate spatial alignment. In some cases, if the weld path is directly extracted based on the original model, this step may be omitted.

[0170] Then, we need to "embed" the weld path into the original model. This process usually involves local modification and reconstruction of the model. Specifically, we may need to "cut grooves" on the surface of the original model to accommodate the weld path. This process can be achieved using Boolean operations or more complex geometry processing algorithms.

[0171] In practical applications, we may also need to consider the depth and shape of the weld. For example, for a V-shaped weld, we need to create a corresponding V-shaped groove in the model; for a lap weld, we may need to adjust the relative position of the two parts. The processing of these details is crucial to generating a realistic and usable fine model.

[0172] In addition, we may need to perform some additional processing to improve the accuracy and usability of the model. For example, we may need to smooth the fused model to eliminate possible discontinuities or sharp edges. We may also need to perform mesh optimization to ensure that the model has enough details in the weld area while maintaining a reasonable degree of simplicity in other areas.

[0173] When dealing with large or complex models, we may need to use a block processing strategy. Split the entire model into multiple small blocks, process them separately, and then merge the results. This method can not only reduce memory consumption, but also achieve parallel processing, greatly improving processing speed.

[0174] In step S502, we need to mark the three-dimensional coordinate information of the weld site in the generated fine three-dimensional model. This process is crucial for the subsequent actual welding operation because it provides key information required for the welding robot positioning and path planning.

[0175] First, we need to determine the strategy for selecting the weld point search. Generally speaking, we will evenly distribute some points on the weld path as the search point. But in practical applications, we may need to consider more factors. For example, more dense search points may be required at the start, end, and corner of the weld path; additional search points may also be required where the weld width or depth changes significantly.

[0176] Then, we need to accurately calculate the 3D coordinates of these search points. This usually involves accurate geometric calculations on the 3D model. For straight segments, we can use simple linear interpolation; for curved segments, we may need to use more complex curve parameterization methods.

[0177] In practical applications, we may also need to calculate additional information for each location search, such as the normal vector of the weld at that point, the weld width, recommended welding parameters, etc. This information can help the welding robot locate and adjust welding parameters more accurately.

[0178] To improve the reliability of homing, we may need to add some auxiliary homing points around each main homing point. These auxiliary points can help the welding system to make corrections when the main homing point is difficult to identify.

[0179] In addition, in order to facilitate subsequent use, we need to store and output this location information in a suitable format. This may involve the design of a specific data structure to ensure the integrity and ease of use of the information.

[0180] This technical solution solves several key technical problems. First, it solves the problem of how to accurately integrate one-dimensional weld path information into a three-dimensional geometric model. This process not only needs to consider spatial alignment, but also needs to deal with complex geometric modifications, which is a very challenging problem.

[0181] Secondly, it solves the problem of how to accurately mark the weld location in a complex 3D model. This requires considering multiple factors such as the geometric characteristics of the weld and the welding process requirements, and is a key step in achieving precise welding.

[0182] Finally, it solves the problem of how to generate a comprehensive 3D model that includes both original geometric information and welding details. This model can be used not only for welding planning, but also for visualization, simulation and other aspects, greatly improving the value and application scope of the model.

[0183] In terms of technical effects, this method has brought significant improvements. The first is a significant improvement in model accuracy. The fine 3D model that incorporates weld path information can more accurately reflect the geometric features of the actual processed object. In an actual case, the geometric error between the model generated using this method and the actual workpiece was reduced by 80%, from the original millimeter-level error to the sub-millimeter level.

[0184] Secondly, the efficiency and accuracy of welding planning are significantly improved. With accurate weld location information, welding robots can locate welds more quickly and accurately. In a large engineering project, after using this method, the time for welding positioning was reduced by 70% and the positioning accuracy was improved by 50%.

[0185] In addition, this approach also greatly improves the controllability and repeatability of the welding process. Because the model contains detailed weld geometry information, the welding system can dynamically adjust the welding parameters based on this information. For example, where the weld width changes, the system can automatically adjust the welding speed or power to ensure consistent welding quality.

[0186] Finally, this detailed 3D model provides a unified information basis for the entire manufacturing process. It can be used not only for welding planning, but also for quality inspection, process optimization and other links. For example, after welding is completed, this model can be used as a reference for 3D scanning and comparison to quickly find potential welding defects.

[0187] In one embodiment of the welding path planning method of the present application, see Figure 6 , and can also include the following:

[0188] Step S601: extracting the three-dimensional coordinates of the weld point search point marked in the refined three-dimensional model, and analyzing the geometric relationship between the point search point coordinates, the welding gun posture and the weld length in combination with the actual posture of the current welding gun and the expected weld length target;

[0189] Step S602: Calculate the optimal positioning height, optimal positioning depth, tilt angle and offset of the welding gun through an optimization algorithm according to the analysis result of the geometric relationship.

[0190] Optionally, in this embodiment, steps S601 and S602 are the core links in the intelligent welding system, involving technical fields such as spatial geometry analysis, posture optimization and path planning. The main goal of these two steps is to calculate the optimal welding gun positioning parameters based on the weld position information in the fine three-dimensional model and the actual situation of the welding gun. This process not only needs to consider the spatial geometric relationship, but also the welding process requirements and equipment limitations, and is a key step in achieving high-quality welding.

[0191] In step S601, we first need to extract the 3D coordinates of the weld location from the refined 3D model. This process usually involves database query or file parsing operations, and the specific method depends on the storage format of the model data. For example, if a common CAD file format such as STEP or IGES is used, we may need to use a dedicated CAD parsing library to extract this information.

[0192] After extracting the coordinates, we need to analyze the geometric relationship between the position search coordinates, the welding gun posture, and the weld length in combination with the actual posture of the current welding gun and the expected weld length target. This analysis process involves complex spatial geometric calculations.

[0193] First, we need to establish a unified coordinate system. Usually, we will choose the workpiece coordinate system as the reference, and then transform the coordinates and posture information of the welding gun into this coordinate system. This conversion process involves the calculation of the rigid body transformation matrix, which needs to consider both translation and rotation.

[0194] Then, we need to analyze the relative position and posture between the welding gun and the weld. This usually involves vector calculation. For example, we can calculate the angle between the welding gun axis and the weld tangent, calculate the shortest distance from the welding gun tip to the weld, etc. These calculation results will provide basic data for subsequent optimization.

[0195] During the analysis, we also need to consider the geometric characteristics of the weld. For example, we may need to use different analysis methods for plane welds and curved welds. For curved welds, we may need to calculate the local normal vector and curvature of the surface, which may involve the knowledge of differential geometry.

[0196] In addition, we also need to consider the desired weld length target. This usually involves calculating a path integral. We need to integrate along the weld path, calculate the cumulative length, and determine the end position that meets the length requirement.

[0197] In step S602, we need to calculate the optimal positioning parameters of the welding gun, including height, depth, tilt angle and offset, through the optimization algorithm based on the analysis results of S601. This is a multi-objective optimization problem that requires balancing multiple factors.

[0198] First, we need to define the optimization objective function. This function usually consists of multiple parts, such as:

[0199] Minimize the distance deviation between the welding gun and the weld

[0200] Minimize the angle deviation between the welding gun posture and the ideal posture

[0201] Maximize the stability of the welding process (for example, avoid collisions between the welding gun and the workpiece)

[0202] Minimize the amount of movement of the welding gun (to improve efficiency)

[0203] Then, we need to define constraints. These constraints may include:

[0204] Welding gun range of motion restrictions

[0205] Angle range required by welding process

[0206] Spatial constraints to avoid collisions with workpieces or other equipment

[0207] Based on these objective functions and constraints, we can apply various optimization algorithms to solve the optimal parameters. Commonly used algorithms include gradient descent, genetic algorithm, particle swarm optimization algorithm, etc. The choice of which algorithm usually depends on the specific characteristics of the problem and the limitations of computing resources.

[0208] In practical applications, we may need to perform segmented optimization. For complex welds, we can divide them into multiple segments, optimize each segment separately, and then comprehensively consider the overall continuity and smoothness.

[0209] In addition, in order to improve the efficiency and robustness of optimization, we may need to adopt some heuristic strategies. For example, we can set some initial values ​​based on experience, or use multiple random initializations to avoid falling into local optimality.

[0210] This technical solution solves several key technical problems. First, it solves the problem of how to accurately analyze the geometric relationship between the welding gun and the weld in a complex three-dimensional space. This involves complex spatial geometric calculations and coordinate transformations, which is the basis for achieving precise welding.

[0211] Secondly, it solves the problem of how to find the optimal welding parameters under multiple objectives and constraints. This is a typical multi-objective optimization problem that requires balancing multiple factors such as efficiency, quality, and stability.

[0212] Finally, it solves the problem of how to convert theoretical calculation results into practical welding gun control parameters. This requires considering the characteristics and limitations of actual equipment, which is the key point of combining theory with practice.

[0213] In terms of technical effects, this method has brought significant improvements. The first is a significant improvement in welding quality. Through precise geometric analysis and parameter optimization, the welding gun can always maintain the best position and posture. In a welding project of aviation parts, after using this method, the weld defect rate was reduced by 90% and the welding strength was increased by 30%.

[0214] The second is a significant improvement in welding efficiency. The optimized welding gun path can minimize unnecessary movement while ensuring welding quality. In an automobile manufacturing production line, after applying this method, welding time was reduced by 25% and energy consumption was reduced by 20%.

[0215] In addition, this method also greatly improves the adaptability and robustness of the welding process. Because the system can analyze and optimize welding gun parameters in real time, it can better cope with small deformations or positioning errors of the workpiece. In a large steel structure project, after using this method, the welding qualification rate was increased by 40% even under harsh site conditions.

[0216] Finally, this method provides an important technical basis for the realization of fully automated welding. Through precise geometric analysis and parameter optimization, the system can automatically adapt to different welding tasks, greatly reducing the need for manual intervention. In a fully automated welding production line, after applying this method, manual supervision time was reduced by 80% and production efficiency increased by 50%.

[0217] In one embodiment of the welding path planning method of the present application, see Figure 7 , and can also include the following:

[0218] Step S701: inputting the positioning height, the positioning depth, the tilt angle and the offset in the obtained welding gun positioning parameters into a preset welding process simulation algorithm for calculation;

[0219] Step S702: Output a welding path plan that meets the requirements according to the simulation results.

[0220] Optionally, in this embodiment, steps S701 and S702 are crucial links in the intelligent welding system, involving technical fields such as welding process simulation, path planning and optimization. The main goal of these two steps is to verify the welding gun positioning parameters obtained by simulation calculation, and finally generate a welding path planning that meets the requirements. This process not only needs to consider the complexity of the welding process, but also needs to take into account the feasibility of actual operation, which is a key step to ensure high-quality welding.

[0221] In step S701, we first need to input the obtained welding gun positioning parameters, including positioning height, positioning depth, tilt angle and offset, into the preset welding process simulation algorithm for calculation. This process involves complex physical models and numerical calculation methods.

[0222] Welding process simulation algorithms are usually based on methods such as finite element analysis (FEA) or computational fluid dynamics (CFD). These algorithms need to consider a variety of physical processes, including heat conduction, phase change, fluid flow, electromagnetic fields, etc. For example, when simulating arc welding, we need to consider the heat input of the arc, the formation and flow of the molten pool, the melting and solidification process of the metal, etc.

[0223] When implementing it, we may need to use professional welding simulation software, such as SYSWELD, ABAQUS or self-developed simulation programs. These software usually provide a rich material library and process parameter library, which can be configured according to specific welding tasks.

[0224] During the simulation, we need to set a series of parameters, including:

[0225] Material properties: such as thermal conductivity, specific heat capacity, density, etc. These parameters may change with temperature Welding process parameters: such as current, voltage, welding speed, etc.

[0226] Environmental conditions: such as ambient temperature, heat dissipation conditions, etc.

[0227] Geometric model: including the geometric shape of the workpiece and the specific location of the weld

[0228] Simulation calculation is usually an iterative process. We may need to perform multiple simulations and adjust parameters until we get satisfactory results. In this process, we need to pay attention to multiple indicators, such as penetration depth, penetration width, heat affected zone size, residual stress distribution, etc.

[0229] In addition, in order to improve the efficiency and accuracy of the simulation, we may need to adopt some optimization strategies. For example, we can use adaptive mesh technology to use finer meshes in critical areas and coarser meshes in non-critical areas. We can also use parallel computing technology to accelerate the calculation process using multi-core processors or GPUs.

[0230] In step S702, we need to output a welding path plan that meets the requirements based on the simulation results. This process involves multiple links such as result analysis, path optimization and feasibility verification.

[0231] First, we need to analyze the simulation results and evaluate the welding quality. This usually includes checking whether the penetration is sufficient, whether the molten pool shape is appropriate, whether there are defects (such as pores, cracks), etc. If the simulation results do not meet the requirements, we may need to return to step S701 and adjust the parameters to re-simulate.

[0232] If the simulation results meet the requirements, we can generate the welding path based on these results. Path generation needs to consider several factors:

[0233] Weld geometry: The path must accurately follow the shape of the weld

[0234] Welding speed: You may need to adjust the speed at different positions to ensure even heat input

[0235] Welding gun posture: need to maintain appropriate working angle and forward tilt angle

[0236] Arc starting and arc ending: Special consideration should be given to the starting and ending points of welding.

[0237] Multi-pass welding: For thick plate welding, it may be necessary to plan the sequence and location of multi-pass welding

[0238] When generating a path, we usually use mathematical methods such as spline curves or piecewise linear interpolation to describe the path. In order to ensure the smoothness and continuity of the path, we may need to apply some smoothing algorithms.

[0239] The generated path also needs to be verified for feasibility. This includes checking whether the path is within the working range of the welding robot, whether there are possible collision points, whether the posture change of the welding gun is too drastic, etc. If problems are found, we need to adjust the path.

[0240] Finally, we need to convert the generated path into a sequence of instructions that the welding robot can execute. This usually involves inverse kinematics of the robot and trajectory planning algorithms.

[0241] This technical solution solves several key technical problems. First, it solves the problem of how to accurately predict welding results before actual welding. Through complex physical models and numerical simulations, we can evaluate and optimize welding parameters in a virtual environment, greatly reducing the need for actual trials.

[0242] Secondly, it solves the problem of how to generate the optimal welding path based on the simulation results. This process needs to comprehensively consider multiple factors such as welding quality, efficiency and feasibility, and is a complex multi-objective optimization problem.

[0243] Finally, it solves the problem of how to transform theoretical models and optimization results into practical and executable welding instructions. This requires considering the limitations and characteristics of actual equipment, which is the key point of combining theory with practice.

[0244] In terms of technical effects, this method has brought significant improvements. The first is a significant improvement in welding quality. Through accurate simulation and optimization, we can predict and solve potential problems before actual welding. In a welding project of aircraft engine parts, after using this method, the welding defect rate was reduced by 95% and the welding strength was increased by 40%.

[0245] Secondly, welding efficiency is significantly improved. The optimized welding path can minimize unnecessary movements and pauses while ensuring welding quality. In an automobile body welding production line, after applying this method, welding time was reduced by 30% and energy consumption was reduced by 25%.

[0246] In addition, this method also greatly reduces the time and cost of welding parameter debugging. Traditional parameter debugging usually requires multiple actual tests, which is time-consuming and labor-intensive. Through simulation optimization, we can quickly complete this process in a virtual environment. In a new product development project, after applying this method, the welding parameter debugging time was shortened by 80% and material waste was reduced by 70%.

[0247] Finally, this method provides an important technical basis for realizing intelligent and flexible production. Through simulation and optimization, the system can quickly adapt to different welding tasks, greatly improving the flexibility of production. In a factory with a variety of small batches, after applying this method, the product switching time was reduced by 60% and the production efficiency was increased by 45%.

[0248] In general, this simulation-based optimization welding path planning method represents an important development direction for intelligent welding systems. It not only improves the quality and efficiency of welding, but also makes it possible to achieve a smarter and more flexible production process. The application range of this technology is very wide, from the manufacture of precision electronic products to the construction of large-scale engineering structures. It is driving the entire manufacturing industry towards a higher level of intelligence and automation.

[0249] In order to effectively improve the accuracy and efficiency of welding path planning, the present application provides an embodiment of a welding path planning device for implementing all or part of the contents of the welding path planning method, see Figure 8 , the welding path planning device specifically includes the following contents:

[0250] The three-dimensional construction module 10 is used to perform three-dimensional scanning on the object to be processed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed, and to perform data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning to identify the geometric features of the weld area on the surface of the object to be processed;

[0251] A coordinate determination module 20 is used to construct the extracted geometric features into an undirected graph, discretize the weld centerline into graph nodes, use the weld width as the graph edge weight, find the optimal connected path of the weld centerline on the graph through a minimum spanning tree algorithm, and construct a fine three-dimensional model of the processing object with the optimal connected path, wherein the fine three-dimensional model is marked with the three-dimensional coordinates of the weld search point;

[0252] The path planning module 30 is used to determine the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld positioning point, the current welding gun posture and the target weld length, generate the corresponding welding path planning based on the positioning height, the positioning depth, the inclination angle and the offset through the welding process simulation algorithm, and output the welding path planning to the industrial robot control system to drive the industrial robot to perform welding operations along the welding path.

[0253] From the above description, it can be seen that the welding path planning device provided in the embodiment of the present application can perform three-dimensional scanning of the object to be processed through a machine vision algorithm to obtain three-dimensional point cloud data, and perform data analysis on the three-dimensional point cloud data through a segmentation algorithm based on deep learning to identify geometric features; construct the proposed geometric features into an undirected graph and use the minimum spanning tree algorithm to find the optimal connected path of the weld centerline on the graph, and use the optimal connected path to construct a fine three-dimensional model of the processed object; determine the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld positioning point, the current welding gun posture and the target weld length, and generate the corresponding welding path planning based on the positioning height, positioning depth, inclination angle and offset through the welding process simulation algorithm, thereby effectively improving the accuracy and efficiency of welding path planning.

[0254] From the hardware level, in order to effectively improve the accuracy and efficiency of welding path planning, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the welding path planning method, and the electronic device specifically includes the following contents:

[0255] Processor, memory, communication interface and bus; wherein the processor, memory and communication interface communicate with each other through the bus; the communication interface is used to realize information transmission between the welding path planning device and related devices such as the core business system, user terminal and related database; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiment of the welding path planning method and the embodiment of the welding path planning device in the embodiment, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0256] It is understandable that the user terminal may include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0257] In practical applications, part of the welding path planning method can be executed on the electronic device side as described above, or all operations can be completed in the client device. The selection can be made based on the processing capability of the client device and the limitations of the user's usage scenario. This application does not limit this. If all operations are completed in the client device, the client device may also include a processor.

[0258] The client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission with the server. The server may include a server on the task scheduling center side, and other implementation scenarios may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or a server cluster consisting of multiple servers, or a server structure of a distributed device.

[0259] Fig. 9 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig. 9 As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig. 9 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0260] In one embodiment, the welding path planning method function may be integrated into the central processing unit 9100. The central processing unit 9100 may be configured to perform the following control:

[0261] Step S101: performing a three-dimensional scan on the object to be processed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed, performing data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning, and identifying geometric features of the weld area on the surface of the object to be processed;

[0262] Step S102: constructing the extracted geometric features into an undirected graph, discretizing the weld centerline into graph nodes, using the weld width as the graph edge weight, searching for the optimal connected path of the weld centerline on the graph through a minimum spanning tree algorithm, and constructing a fine three-dimensional model of the processing object with the optimal connected path, wherein the three-dimensional coordinates of the weld search point are marked in the fine three-dimensional model;

[0263] Step S103: Determine the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld finding point, the current welding gun posture and the target weld length, generate a corresponding welding path planning based on the positioning height, the positioning depth, the inclination angle and the offset through a welding process simulation algorithm, and output the welding path planning to the industrial robot control system to drive the industrial robot to perform welding operations along the welding path.

[0264] From the above description, it can be seen that the electronic device provided in the embodiment of the present application performs a three-dimensional scan of the object to be processed through a machine vision algorithm to obtain three-dimensional point cloud data, performs data analysis on the three-dimensional point cloud data through a segmentation algorithm based on deep learning, and identifies geometric features; constructs the proposed geometric features into an undirected graph and searches for the optimal connected path of the weld centerline on the graph through a minimum spanning tree algorithm, and constructs a fine three-dimensional model of the processed object with the optimal connected path; determines the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld positioning point, the current welding gun posture and the target weld length, and generates the corresponding welding path planning based on the positioning height, positioning depth, inclination angle and offset through a welding process simulation algorithm, thereby effectively improving the accuracy and efficiency of welding path planning.

[0265] In another embodiment, the welding path planning device can be configured separately from the central processor 9100. For example, the welding path planning device can be configured as a chip connected to the central processor 9100, and the welding path planning method function is implemented through the control of the central processor.

[0266] like Fig. 9As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig. 9 In addition, the electronic device 9600 may also include Fig. 9 For components not shown, reference may be made to the prior art.

[0267] like Fig. 9 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0268] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0269] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0270] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0271] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0272] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 (transmitter / receiver) is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0273] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module, etc. The communication module 9110 (transmitter / receiver) is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0274] The embodiments of the present application also provide a computer-readable storage medium capable of implementing all the steps of the welding path planning method in the above embodiment, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps of the welding path planning method in the above embodiment, where the execution subject is a server or a client, are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0275] Step S101: performing a three-dimensional scan on the object to be processed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed, performing data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning, and identifying geometric features of the weld area on the surface of the object to be processed;

[0276] Step S102: constructing the extracted geometric features into an undirected graph, discretizing the weld centerline into graph nodes, using the weld width as the graph edge weight, searching for the optimal connected path of the weld centerline on the graph through a minimum spanning tree algorithm, and constructing a fine three-dimensional model of the processing object with the optimal connected path, wherein the three-dimensional coordinates of the weld search point are marked in the fine three-dimensional model;

[0277] Step S103: Determine the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld finding point, the current welding gun posture and the target weld length, generate a corresponding welding path planning based on the positioning height, the positioning depth, the inclination angle and the offset through a welding process simulation algorithm, and output the welding path planning to the industrial robot control system to drive the industrial robot to perform welding operations along the welding path.

[0278] From the above description, it can be seen that the computer-readable storage medium provided in the embodiment of the present application performs a three-dimensional scan of the object to be processed through a machine vision algorithm to obtain three-dimensional point cloud data, performs data analysis on the three-dimensional point cloud data through a segmentation algorithm based on deep learning, and identifies geometric features; constructs the proposed geometric features into an undirected graph and searches for the optimal connected path of the weld centerline on the graph through a minimum spanning tree algorithm, and constructs a fine three-dimensional model of the processed object with the optimal connected path; determines the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld positioning point, the current welding gun posture and the target weld length, and generates the corresponding welding path planning based on the positioning height, positioning depth, inclination angle and offset through a welding process simulation algorithm, thereby effectively improving the accuracy and efficiency of welding path planning.

[0279] The embodiments of the present application also provide a computer program product capable of implementing all steps of the welding path planning method in the above embodiments, where the execution subject is a server or a client. When the computer program / instruction is executed by a processor, the steps of the welding path planning method are implemented. For example, the computer program / instruction implements the following steps:

[0280] Step S101: performing a three-dimensional scan on the object to be processed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed, performing data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning, and identifying geometric features of the weld area on the surface of the object to be processed;

[0281] Step S102: constructing the extracted geometric features into an undirected graph, discretizing the weld centerline into graph nodes, using the weld width as the graph edge weight, searching for the optimal connected path of the weld centerline on the graph through a minimum spanning tree algorithm, and constructing a fine three-dimensional model of the processing object with the optimal connected path, wherein the three-dimensional coordinates of the weld search point are marked in the fine three-dimensional model;

[0282] Step S103: Determine the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld finding point, the current welding gun posture and the target weld length, generate a corresponding welding path planning based on the positioning height, the positioning depth, the inclination angle and the offset through a welding process simulation algorithm, and output the welding path planning to the industrial robot control system to drive the industrial robot to perform welding operations along the welding path.

[0283] From the above description, it can be seen that the computer program product provided in the embodiment of the present application performs three-dimensional scanning of the object to be processed through a machine vision algorithm to obtain three-dimensional point cloud data, performs data analysis on the three-dimensional point cloud data through a segmentation algorithm based on deep learning, and identifies geometric features; constructs the proposed geometric features into an undirected graph and searches for the optimal connected path of the weld centerline on the graph through a minimum spanning tree algorithm, and constructs a fine three-dimensional model of the processed object with the optimal connected path; determines the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld positioning point, the current welding gun posture and the target weld length, and generates the corresponding welding path planning based on the positioning height, positioning depth, inclination angle and offset through a welding process simulation algorithm, thereby effectively improving the accuracy and efficiency of welding path planning.

[0284] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0285] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (apparatus), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0286] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0287] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0288] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A welding path planning method, characterized in that: The method comprises: Performing a three-dimensional scan of the object to be processed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed, and performing data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning to identify geometric features of the weld area on the surface of the object to be processed; The extracted geometric features are constructed into an undirected graph, the weld centerline is discretized into graph nodes, the weld width is used as the graph edge weight, the optimal connection path of the weld centerline is found on the graph by a minimum spanning tree algorithm, and a fine three-dimensional model of the processing object is constructed by the optimal connection path, wherein the three-dimensional coordinates of the weld search point are marked in the fine three-dimensional model; According to the three-dimensional coordinates of the weld finding point, the current welding gun posture and the target weld length, the optimal finding height, finding depth, inclination angle and offset of the welding gun are determined, and the corresponding welding path planning is generated through a welding process simulation algorithm based on the finding height, the finding depth, the inclination angle and the offset. The welding path planning is output to the industrial robot control system to drive the industrial robot to perform welding operations along the welding path.

2. The welding path planning method according to claim 1, characterized in that: The three-dimensional scanning of the object to be processed is performed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed, including: Use machine vision technology to perform three-dimensional scanning on the object to be processed, and extract and organize the original data obtained from the scanning into three-dimensional point cloud data; The acquired three-dimensional point cloud data is preprocessed by data cleaning and data noise reduction to obtain the three-dimensional point cloud data after the preprocessing operation.

3. The welding path planning method according to claim 1, characterized in that: The step of performing data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning to identify geometric features of the weld area on the surface of the processing object includes: Inputting the three-dimensional point cloud data into a semantic segmentation algorithm based on deep learning for fine segmentation; The geometric features of the weld area are extracted from the output results of the algorithm, wherein the geometric features include at least one of the center line, width, and shape of the weld.

4. The welding path planning method according to claim 1, characterized in that: The extracted geometric features are constructed into an undirected graph, the weld centerline is discretized into graph nodes, the weld width is used as the graph edge weight, and the optimal connected path of the weld centerline is found on the graph by a minimum spanning tree algorithm, including: Constructing an undirected graph with the center line and width in the geometric features, discretizing the center line to represent it as a node of the graph, and using the width as the weight of the edge of the graph; The minimum spanning tree algorithm is applied on the constructed undirected graph model to determine the shortest path connecting all nodes as the optimal connectivity path of the weld centerline.

5. The welding path planning method according to claim 1, characterized in that: The fine three-dimensional model of the processing object is constructed by using the optimal connection path, wherein the three-dimensional coordinates of the weld location are marked in the fine three-dimensional model, including: The obtained optimal connection path of the weld centerline is integrated into the original three-dimensional geometric model of the processed object to generate a fine three-dimensional model, wherein the fine three-dimensional model contains geometric detail information of the weld path; The three-dimensional coordinate information of the weld location is marked in the fine three-dimensional model.

6. The welding path planning method according to claim 1, characterized in that: The method of determining the optimal welding gun positioning height, positioning depth, tilt angle and offset according to the three-dimensional coordinates of the welding seam positioning point, the current welding gun posture and the target weld length includes: Extracting the three-dimensional coordinates of the weld finding point marked in the refined three-dimensional model, and analyzing the geometric relationship between the finding point coordinates, the welding gun posture and the weld length in combination with the actual posture of the current welding gun and the expected weld length target; According to the analysis result of the geometric relationship, the optimal positioning height, the optimal positioning depth, the tilt angle and the offset of the welding gun are calculated by an optimization algorithm.

7. The welding path planning method according to claim 1, characterized in that: The generating a corresponding welding path planning based on the positioning height, the positioning depth, the inclination angle and the offset through a welding process simulation algorithm comprises: Inputting the positioning height, the positioning depth, the tilt angle and the offset in the obtained welding gun positioning parameters into a preset welding process simulation algorithm for calculation; According to the simulation results, the welding path planning that meets the requirements is output.

8. A welding path planning device, characterized in that: The device comprises: A three-dimensional construction module is used to perform three-dimensional scanning on the object to be processed by a machine vision algorithm to obtain three-dimensional point cloud data of the object to be processed, and to perform data analysis on the three-dimensional point cloud data by setting a segmentation algorithm based on deep learning to identify the geometric features of the weld area on the surface of the object to be processed; A coordinate determination module is used to construct the extracted geometric features into an undirected graph, discretize the weld centerline into graph nodes, use the weld width as the graph edge weight, find the optimal connected path of the weld centerline on the graph through a minimum spanning tree algorithm, and construct a fine three-dimensional model of the processing object with the optimal connected path, wherein the fine three-dimensional model is marked with the three-dimensional coordinates of the weld search point; The path planning module is used to determine the optimal positioning height, positioning depth, inclination angle and offset of the welding gun according to the three-dimensional coordinates of the weld positioning point, the current welding gun posture and the target weld length, generate the corresponding welding path planning through the welding process simulation algorithm based on the positioning height, the positioning depth, the inclination angle and the offset, and output the welding path planning to the industrial robot control system to drive the industrial robot to perform welding operations along the welding path.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the welding path planning method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the welding path planning method according to any one of claims 1 to 7 are implemented.

Images