Method for generating trained model, learning device, welding robot control device, system, and program

Abstract

To enable extraction of learning data by taking into account accuracy of the learning data to be used for machine learning processing in which welding fields are assumed.SOLUTION: A method for generating a learnt model includes a generation step for generating a first learnt model according to machine learning, in which the input consists of a combination of data obtained during welding and characteristic information specified from the data, and the output consists of a category of correctness of the combination.SELECTED DRAWING: Figure 10

Description

[Technical field]

[0001] The present invention relates to a method for generating a trained model, a learning device, a welding robot control device, a system, oh and programs. [Background technology]

[0002] In recent years, in the field of welding work, automation technologies that automatically create or adjust optimal welding conditions have been realized. As a method for creating or adjusting parameters as welding conditions without human intervention, a trained model generated by performing machine learning processing using a large amount of training data is used. On the other hand, in order to improve the accuracy of the output obtained from the trained model, the effort and cost for preparing training data for machine learning processing will inevitably increase.

[0003] For example, Patent Document 1 discloses an automatic welding system that uses a machine learning model to output welding state related information and controls a robot based on the output result. Patent Document 2 discloses a method for reducing the cost of collecting teacher data for additional learning in machine learning. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2018-192524 A [Patent Document 2] JP 2020-135494 A Summary of the Invention [Problem to be solved by the invention]

[0005] When the target of feature extraction includes at least one of a fluid or plasma gas, such as in a welding work site, these often undergo abrupt changes due to disturbances such as the welding work conditions. Therefore, highly accurate feature extraction that can handle abrupt changes is required. In addition to welding, applications that target fluids and plasma gas include, for example, additive manufacturing. In this way, when welding work that requires more accurate feature extraction is assumed, more learning data is required, and the quality of the learning data, that is, accuracy suitable for generating a more accurate trained model, is required. Cited Document 2 is concerned with facilitating the additional work of extracting features of a human figure, and does not take into account the accuracy of the data extracted by the trained model. For the accuracy of the training data, manual work is also possible, but the effort and cost involved would be significant.

[0006] In view of the above-mentioned problems, the present invention aims to make it possible to extract learning data in consideration of the accuracy of the learning data used in machine learning processing assumed in the welding field. oh The purpose of this website is to provide information and programs. [Means for solving the problem]

[0007] In order to solve the above problems, the present invention has the following configuration. That is, a method for generating a trained model, A generation process for generating a first trained model by performing machine learning using a combination of data obtained during welding and feature information identified from the data as an input and a classification of the correctness or incorrectness of the combination as an output. has.

[0008] According to another aspect of the present invention, there is provided a learning device, comprising: A first generation means for generating a first trained model by performing machine learning using a combination of data obtained during welding and feature information identified from the data as an input and a classification of whether the combination is correct or incorrect as an output; A second generation means for performing machine learning to generate a second trained model by using a combination classified as positive in the first trained model as training data, inputting data obtained during welding, and outputting feature information corresponding to the data; has.

[0009] Another aspect of the present invention has the following configuration: A control device having a first trained model and a second trained model generated by a learning device, An acquisition means for acquiring data related to welding; A control means for outputting control information for a welding robot or a welding power source; having The control means Feature information output by inputting the data acquired by the acquisition means into the second trained model, or A combination of the data acquired by the acquisition means and feature information output by inputting the data into the second trained model is input into the first trained model, and feature information of the combination classified as positive is The control information is output based on any one of the above.

[0010] Another aspect of the present invention has the following configuration. That is, the system includes: A control device; A welding robot A detection means for detecting welding by the welding robot; A welding power source; having The control device includes: A first trained model and a second trained model generated by a learning device, An acquisition means for acquiring data related to welding; a control means for outputting control information for the welding robot or the welding power source; having The control means Feature information output by inputting the data acquired by the acquisition means into the second trained model, or inputting a combination of the data acquired by the acquisition means and feature information output by inputting the data into the second trained model into the first trained model, and obtaining feature information of the combination classified as positive; The control information is output based on any one of the above.

[0011] In addition, the present invention has the following configuration as another aspect thereof: A trained model is generated by machine learning using a combination of data obtained during welding and feature information identified from the data as input, and a classification of the combination as correct or incorrect as output, On the computer, When a combination of data obtained during welding and feature information specified from the data is input, the device is caused to function to output a classification of whether the combination is correct or incorrect.

[0012] In addition, another aspect of the present invention has the following configuration. That is, a program comprising: Computer, a first generation means for generating a first trained model by performing machine learning using a combination of data obtained during welding and feature information identified from the data as an input and a classification of whether the combination is correct or incorrect as an output; a second generation means for performing machine learning using the combination classified as positive in the first trained model as training data, with data obtained during welding as input, and feature information corresponding to the data as output, to generate a second trained model; Function as. Effect of the Invention

[0013] According to the present invention, it is possible to generate a trained model capable of extracting training data, taking into account the accuracy of training data used in machine learning processing intended for the welding field. [Brief description of the drawings]

[0014] [Figure 1] 1 is a schematic diagram showing an example of the configuration of a welding system according to an embodiment of the present invention; [Diagram 2] FIG. 2 is a perspective view for explaining an arrangement position of a visual sensor according to an embodiment of the present invention. [Diagram 3] 1 is a schematic diagram showing an example of the configuration of a robot control device according to an embodiment of the present invention; [Figure 4] 1 is a schematic diagram showing an example of the configuration of a data processing device according to an embodiment of the present invention; [Diagram 5] A table for explaining various trained models according to one embodiment of the present invention. [Figure 6] FIG. 1 is an explanatory diagram for conceptually explaining the generation process of trained models A and C according to an embodiment of the present invention. [Figure 7] FIG. 4 is a schematic diagram for explaining an example of a screen used for a teaching operation according to an embodiment of the present invention. [Figure 8] FIG. 2 is an explanatory diagram showing an example of an image that can be used in a learning process according to an embodiment of the present invention. [Figure 9] FIG. 2 is an explanatory diagram showing an example of an image that can be used in a learning process according to an embodiment of the present invention. [Figure 10] FIG. 2 is an explanatory diagram for explaining a process flow according to an embodiment of the present invention. [Figure 11] 4 is a flowchart of the processing of a robot control device and a power supply device according to an embodiment of the present invention. [Figure 12] 5 is a flowchart showing a process for generating correction information in the data processing device according to the embodiment of the present invention. [Figure 13] FIG. 11 is a graph showing a comparison of results of a learning process according to an embodiment of the present invention. [Figure 14] FIG. 11 is a graph showing a comparison of results of a learning process according to an embodiment of the present invention. [Figure 15] A table for explaining the accuracy of trained model B according to one embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] Hereinafter, the embodiment for carrying out the present invention will be described with reference to the drawings. Note that the embodiment described below is one embodiment for explaining the present invention, and is not intended to be interpreted as limiting the present invention, and all configurations described in each embodiment are not necessarily essential configurations for solving the problems of the present invention. In addition, in each drawing, the same components are given the same reference numbers to indicate their correspondence.

[0016] In addition, the method for measuring welding behavior according to the present invention is useful not only for welding but also for additive manufacturing technology using GMAW, specifically, for metal additive manufacturing (WAAM: Wire and Arc Additive Manufacturing). The term additive manufacturing may be used broadly as the term additive manufacturing or rapid prototyping, but in the present invention, the term additive manufacturing is used uniformly. When the method according to the present invention is used for additive manufacturing technology, "welding" can be rephrased as "deposition," "additive manufacturing," or "additive manufacturing." For example, when treated as welding, it becomes "welding behavior," but when the present invention is used as additive manufacturing, it can be rephrased as "deposition behavior," or when treated as welding, it becomes "welding system," but when the present invention is used as additive manufacturing, it can be rephrased as "additive manufacturing system."

[0017] In the following description, "learning" or "machine learning" refers to generating a "trained model" by performing learning using training data and an arbitrary learning algorithm. The trained model is updated from time to time as learning progresses using multiple training data, and its output changes even for the same input. Therefore, the trained model is not limited to a state at any point in time. Here, a model used in learning is described as a "learned model", and a learning model that has undergone a certain degree of learning is described as a "trained model". In addition, specific examples of "trained data" will be described later, but the configuration may be changed depending on the learning algorithm used. In addition, the training data may include teacher data used for learning itself, verification data used to verify the trained model, and test data used to test the trained model. In the following description, when data related to learning is collectively referred to, it is described as "trained data", and when data when performing learning itself is referred to, it is described as "trained data". Note that it is not intended to clearly classify the teacher data, verification data, and test data contained in the training data, and for example, depending on the methods of learning, verification, and testing, all of the training data may also be teacher data.

[0018] <First embodiment> Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0019] [Welding system configuration] Fig. 1 shows an example of the configuration of a welding system 1 according to this embodiment. The welding system 1 shown in Fig. 1 includes a welding robot 10, a robot control device 20, a power supply device 30, a visual sensor 40, and a data processing device 50. When the method according to the present invention is applied to additive manufacturing, for example, the welding system 1 may be read as an additive manufacturing system, and the welding robot 10 may be read as an additive manufacturing robot.

[0020] The welding robot 10 shown in FIG. 1 is composed of a six-axis articulated robot, and a welding torch 11 for GMAW is attached to the tip of the robot. GMAW includes, for example, MIG (Metal Inert Gas) welding and MAG (Metal Active Gas) welding, and in this embodiment, MAG welding will be described as an example. The welding robot 10 is not limited to a six-axis articulated robot, and may be, for example, a portable small robot. An example of a portable small robot is an orthogonal robot with three or less axes. A welding device having a drive unit such as a cart may also be used. The welding method may be electrogas welding or electroslag welding in addition to GMAW.

[0021] Welding wire 13 is supplied to welding torch 11 from wire feeder 12. Welding wire 13 is fed from the tip of welding torch 11 toward the welding point. Power supply unit 30 supplies power to welding wire 13. This power applies an arc voltage between welding wire 13 and workpiece W, generating an arc. Power supply unit 30 is provided with a current sensor (not shown) that detects the welding current flowing from welding wire 13 to workpiece W during welding, and a voltage sensor (not shown) that detects the arc voltage between welding wire 13 and workpiece W.

[0022] The power supply device 30 has a processing unit and a storage unit, both of which are not shown. The processing unit is, for example, configured with a CPU (Central Processing Unit). The storage unit is, for example, configured with a volatile or non-volatile memory, such as a HDD (Hard Disk Drive), a ROM (Read Only Memory), or a RAM (Random Access Memory). The processing unit controls the power applied to the welding wire 13 by executing a computer program for power supply control stored in the storage unit. The power supply device 30 is also connected to the wire feeder 12, and the processing unit controls the feed speed and feed amount of the welding wire 13. The composition and type of the welding wire 13 are selected according to the object to be welded.

[0023] Visual sensor 40 is configured, for example, by a CCD (Charge Coupled Device) camera. The location of visual sensor 40 is not particularly important, and it may be attached directly to welding robot 10, or may be fixed to a specific location in the vicinity as a surveillance camera. When visual sensor 40 is attached directly to welding robot 10, visual sensor 40 moves in accordance with the operation of welding robot 10 so as to capture images of the periphery of the tip of welding torch 11. The number of cameras constituting visual sensor 40 may be multiple.

[0024] The direction in which the visual sensor 40 takes the image is not particularly limited, and for example, if the welding proceeds in the forward direction, the visual sensor 40 may be disposed to take the image of the front side, the side side, or the rear side. Therefore, the imaging range of the visual sensor 40 may be appropriately determined. Note that, in order to suppress interference with the welding torch 11, it is preferable to take the image from the front side.

[0025] In this embodiment, a visual sensor 40 is directly attached and fixed to the welding robot 10, and a moving image is captured as a welding image so that the imaging range includes at least the workpiece W, the welding wire 13, and the arc. Note that various imaging settings related to the welding image may be specified in advance, or may be switched according to the operating conditions of the welding system 1. Examples of imaging settings include the frame rate, the number of pixels of the image, the resolution, and the shutter speed.

[0026] Each part constituting the welding system 1 is communicatively connected by various wired / wireless communication methods. The communication method here is not limited to one, and a combination of multiple communication methods may be used for connection.

[0027] FIG. 2 is a perspective view for explaining the arrangement position of the visual sensor 40. In the present embodiment, the workpiece W is a butt joint. The workpiece W is two metal plates, which are butted together across a groove. A ceramic backing material 14 is attached to the back side of the two butted metal plates. A metal backing material may be used on the back side, or no backing material may be used. In a butt joint, arc welding is performed in one direction along the groove. Hereinafter, the direction in which the welding proceeds is referred to as "forward". In FIG. 2, the direction in which the welding proceeds is indicated by an arrow. For this reason, the welding torch 11 is located behind the visual sensor 40. In the present embodiment, the side where the right hand is located when looking backward from the visual sensor 40 is referred to as "right side", and the side where the left hand is located when looking forward is referred to as "left side".

[0028] In this embodiment, the workpiece W is placed horizontally so that the surface on the side to be welded faces vertically upward. Therefore, the welding robot 10 welds the workpiece W from above the workpiece W. As shown in FIG. 2, the visual sensor 40 may be placed obliquely upward with respect to the welding position of the workpiece W. The imaging range of the visual sensor 40 includes the welding position of the workpiece W, and captures an image of the welding position during arc welding. This image includes the molten pool, the welding wire 13, and the arc. The visual sensor 40 in this embodiment can capture still images of, for example, 1024×768 pixels continuously. In other words, the visual sensor 40 can capture the welding image as a moving image.

[0029] [Robot control device configuration] 3 shows an example of the configuration of robot control device 20 that controls the operation of welding robot 10. Robot control device 20 includes a CPU 201 that controls the entire device, a memory 202 that stores data, an operation panel 203 including a plurality of switches, a teaching pendant 204 used in teaching work, an input / output unit 205, and a communication unit 206. Memory 202 is formed of, for example, a volatile or non-volatile storage device such as a ROM, a RAM, or a HDD. Memory 202 stores a control program 202A used to control welding robot 10. CPU 201 controls various operations performed by welding robot 10 by executing control program 202A.

[0030] An operation panel 203 and a teaching pendant 204 are used to input instructions to the robot control device 20. An operator can input a teaching program using the teaching pendant 204. The robot control device 20 controls the welding robot 10 according to the teaching program input from the teaching pendant 204. Note that the teaching program can also be automatically created based on CAD (Computer-Aided Design) information or the like using a computer (not shown), for example. The operation contents defined in the teaching program are not particularly limited, and may differ depending on the specifications of the welding robot 10 and the welding method.

[0031] The input / output unit 205 is connected to the power supply device 30 and a drive circuit of the welding robot 10. For example, the current value of the welding current detected by a current sensor (not shown) and the voltage value of the arc voltage detected by a voltage sensor (not shown) are provided to the CPU 201 via the input / output unit 205 from the power supply device 30. The CPU 201 also outputs a control signal based on the control program 202A to a drive circuit (not shown) provided in the welding robot 10.

[0032] The communication unit 206 is a communication module for wired or wireless communication. The communication unit 206 is used for data communication with the power supply device 30 and the data processing device 50. The communication method or standard used by the communication unit 206 is not particularly limited, and a combination of multiple methods may be used.

[0033] The robot controller 20 also controls the moving speed and protruding direction of the welding torch 11 by controlling each axis of the welding robot 10. The robot controller 20 also controls the weaving operation of the welding robot 10 according to the set period, amplitude, and welding speed. The weaving operation means alternately swinging the welding torch 11 in a direction intersecting the welding proceeding direction. The robot controller 20 executes weld line tracing control in addition to the weaving operation. The weld line tracing control is an operation of controlling the left and right positions of the welding torch 11 with respect to the proceeding direction so that a bead is formed along the weld line. The robot controller 20 also controls the feeding speed of the welding wire 13 by controlling the wire feeder 12 via the power supply 30.

[0034] [Data processing device configuration] 4 is an explanatory diagram for explaining a configuration example of the data processing device 50. The data processing device 50 is configured by, for example, a computer. The computer is configured to include a main body 510, an input unit 520, and a display unit 530. The main body 510 is configured to include a CPU 511, a ROM 512, a RAM 513, a non-volatile storage device 514, an input / output interface 515, a communication interface 516, and a video output interface 517. The CPU 511, the ROM 512, the RAM 513, the non-volatile storage device 514, the input / output interface 515, the communication interface 516, and the video output interface 517 are connected to each other so as to be able to communicate with each other via a bus or a signal line.

[0035] The non-volatile storage device 514 stores a learning program 514A that executes deep learning using predetermined learning data, a trained model 514B that is generated through the execution of the learning program 514A, an information generating program 514C that generates characteristic information related to welding using the trained model 514B, and image data 514D. Although details will be described later, in this embodiment, three trained models are used. Therefore, the learning programs for generating each trained model and the training data used in the learning process are also different from each other. In addition, an operating system and application programs are also installed in the non-volatile storage device 514.

[0036] The data processing device 50 realizes various functions by the execution of the program by the CPU 511. In the case of the present embodiment, the data processing device 50 realizes a function of generating each trained model by machine learning and a function of performing various processes during actual welding using the trained model. The contents of these functions will be described later. The data processing device 50 may be divided according to the function of generating a trained model and the function of executing control processing based on information output from the trained model during actual welding. From the viewpoint of versatility, it is more preferable to divide the data processing device 50 according to each function. The ROM 512 stores a BIOS (Basic Input Output System) and the like executed by the CPU 511. The RAM 513 is used as a work area for the program read from the non-volatile storage device 514.

[0037] The input / output interface 515 is connected to an input unit 520 constituted by a keyboard, a mouse, etc. The visual sensor 40 is also connected to the input / output interface 515. Image data output from the visual sensor 40 is provided to the CPU 511 via the input / output interface 515. The communication interface 516 is a communication module for wired or wireless communication. The video output interface 517 is connected to a display unit 530 constituted by, for example, a liquid crystal display or an organic EL (Electro-Luminescence) display, and outputs a video signal corresponding to the video data provided from the CPU 511 to the display unit 530.

[0038] [Welding system operation] The operation of the welding system 1 according to this embodiment will be described below.

[0039] [Trained model] First, the generation of the trained model according to this embodiment will be described. The trained model is generated by executing a learning program. In this embodiment, three trained models are handled. For convenience, the three trained models are described as "trained model A" (also referred to as "third trained model"), "trained model B" (also referred to as "first trained model"), and "trained model C" (also referred to as "second trained model"). For convenience, the training data used to generate each of the three trained models will be described as "trained data A", "trained data B", and "trained data C". In addition, when the teacher data needs to be separately described, they will be similarly referred to with reference symbols. FIG. 5 is a table summarizing an outline of each trained model according to this embodiment.

[0040] The trained model A is a trained model generated by performing training using training data A consisting of a pair of a welding image and welding information that is added to the welding image by manually specifying a region. The welding information here corresponds to feature information indicating a region corresponding to various elements included in the welding image, which will be described in detail later, and includes, for example, coordinate information of each region in the image. Elements included in the welding image include a flow part and a plasma gas part generated by welding. The trained model A inputs a welding image and outputs the welding information of the welding image. As described above, the task of manually specifying a region in a welding image and adding welding information generally involves a high load and cost. Therefore, in this embodiment, the training data A used in the training process for generating the trained model A is performed using a relatively small amount of training data. In this embodiment, as data obtained during welding, data of a welding image that is easy to capture the characteristics of a welding phenomenon is used as a preferred example, but for example, sound data such as an arc sound, numerical data such as current and voltage may be used, or these data may be used in combination.

[0041] The trained model B is a trained model generated by performing training using training data B consisting of a combination of a welding image and welding information of the welding image, and correct / incorrect information for the combination. The correct / incorrect information here is information indicating whether or not the combination of a welding image and welding information of the welding image is correct, as will be described in detail later. Therefore, the trained model B receives the combination of a welding image and welding information of the welding image as input, and outputs a result indicating whether or not the combination is correct.

[0042] The trained model C is a trained model generated by performing training using training data C consisting of combinations of welding images and their welding information obtained using the trained model A that are determined to be correct by the trained model B. In other words, the trained model C and the trained model A have the same type of input information and the same type of output information, but use different data as training data. In particular, when training the trained model A, a small amount of training data is used, taking into account the effort and cost of preparing the training data. On the other hand, when training the trained model C, a large amount of highly accurate and unattended training data output using the trained model A and the trained model B is used.

[0043] 6 is a diagram conceptually explaining the generation process of trained model A and trained model C, that is, the training process. The training process in this embodiment is performed using a convolutional neural network (CNN), which includes multiple convolutional layers, multiple pooling layers, and multiple fully connected layers. Note that the configuration of the convolutional neural network is not limited to the above, and the number of layers and the configuration may be different.

[0044] In the learning process for the trained model A and the trained model C, the learning model receives a welding image as an input and outputs welding information indicating various characteristics that appear in the welding image. The welding image input to the learning model includes a molten pool, a welding wire 13, and an arc. In the case of this embodiment, welding is performed with different welding condition settings, and welding images are prepared by capturing images of the state of each welding. The welding conditions in this embodiment include, for example, the extension length of the welding wire 13, the brightness (F value) of the image data, the shape of the workpiece W, the shape of the contact tip, the welding current, the arc voltage, the welding speed, the weaving conditions, and the attitude (welding attitude) of the welding wire 13 relative to the workpiece W. The shape of the contact tip includes, for example, a straight nozzle and a tapered nozzle. In addition, the information that specifies the welding attitude includes, for example, a retreat angle, a forward angle, and a deviation of the welding line relative to the groove.

[0045] In this embodiment, the welding information used includes the tip of welding wire 13 (hereinafter also referred to as the wire tip), the center point of the arc (hereinafter also referred to as the arc center), the positions of the left and right tips of the molten pool, and the positions of the left and right ends of the molten pool. The welding information used as teacher data A when generating trained model A is input by an operator, for example, by following instructions on an operation screen that supports a teaching operation and specifying a specific position on a welding image.

[0046] Fig. 7 is an explanatory diagram for explaining an example of a screen used for a teaching operation. The welding image shown in Fig. 7 includes a molten pool 15, a welding wire 13, and an arc 16. In Fig. 7, the molten pool 15 is shown shaded. Fig. 8 is an explanatory diagram showing a specific example of a welding image obtained by welding and an example of welding information in the welding image. Here, for ease of explanation, positions corresponding to the coordinates indicated in the welding information are drawn on the welding image.

[0047] In the case of this embodiment, the operator teaches the coordinate position of the arc center (ArcX, ArcY), the coordinate position of the wire tip (WireX, WireY), the coordinate position of the left end of the tip of the molten pool (Pool_Lead_Lx, Pool_Lead_Ly), the coordinate position of the right end of the tip of the molten pool (Pool_Lead_Rx, Pool_Lead_Ry), the coordinate position of the left end of the molten pool (Pool_Lx), and the coordinate position of the right end of the molten pool (Pool_Rx) as welding information constituting the teacher data A used for learning the learned model A. The operator inputs welding information by specifying a specific position on the screen. The coordinates that give the boundary between the welding wire 13 and the arc are an example of the position coordinates of the wire tip. In addition, the left end of the tip of the molten pool, the right end of the tip of the molten pool, the left end of the molten pool, and the right end of the molten pool are an example of information on the behavior of the molten pool 15. If the left end and the right end of the molten pool are known, the width of the molten pool 15 can be calculated.

[0048] In the learning process for generating the trained model A, a welding image prepared as training data A is input, and various welding information is output. Then, the output welding information is compared with the welding information prepared as training data A, and the error is calculated using an error function and fed back. This feedback adjusts various parameters of each layer of the learning model, and learning progresses. This process is repeated the number of times as many times as the training data A, thereby generating the trained model A. The learning process for generating the trained model C is basically the same as the trained model A, except that the training data used and the amount of data used are different.

[0049] The learning process for generating the trained model B is performed using a convolutional neural network (CNN) like trained model A and trained model C, and includes multiple convolution layers, multiple pooling layers, and multiple fully connected layers. In the learning process for trained model B, the learning model receives a combination of a welding image and welding information as input, and outputs correct / incorrect information. The correct / incorrect information here is classified into two categories: OK and NG. That is, when "OK" is output, it indicates that the combination of the input welding image and welding information is correct. On the other hand, when "NG" is output, it indicates that the combination of the input welding image and welding information is incorrect. Note that the correct / incorrect information is not limited to two categories, and three or more categories may be used depending on the accuracy of the classification.

[0050] FIG. 9 is an explanatory diagram for explaining data used as teacher data B in a learning process for generating a trained model B. The teacher data used for generating the trained model B is prepared by, for example, manually labeling a combination of a welding image and welding information as correct or not. Teacher data 901 shows an example of data treated as NG. In the combination of a welding image and its welding information shown in teacher data 901, a ceramic joint is identified as a position on the tip of a molten pool. Since this is an error in the coordinate position shown as the welding information, this combination of a welding image and welding information is labeled as "NG" and set as teacher data B. In the combination of a welding image and its welding information shown in teacher data 902, a position on the tip of a molten pool is not identified. Since the position on the tip of a molten pool is not visible due to the arc light, it is correct that it is not identified. Therefore, this combination of a welding image and welding information is labeled as "OK" and set as teacher data B.

[0051] [Overview] FIG. 10 is a diagram for explaining an overall outline of the process according to the present embodiment. The process according to the present embodiment is divided into two phases: a learning phase and a control phase. The learning phase is a phase in which three trained models are generated, and the control phase is a phase in which actual welding control is performed using the trained model C generated in the learning phase. The process of each phase is realized by the CPU 511 of the data processing device 50 reading and executing various programs stored in the non-volatile storage device 514 or the like. Note that, in each learning process step described later, a configuration may be adopted in which a verification process or a test process is further performed on the obtained trained model. In that case, each learning data may include verification data or test data.

[0052] First, the learning phase will be described. In S1001, the data processing device 50 performs a learning process for generating a trained model A by using training data A prepared in advance. The learning process here is as described with reference to FIG. 6. In this embodiment, the training data A is a relatively small amount of data that has been prepared manually in advance.

[0053] In S1002, the data processing device 50 acquires welding information for the welding image by inputting the welding image to be processed to the trained model A generated in S1001. The number of welding images input here is not particularly limited, but it is preferable to use more welding images in order to generate a trained model C with higher accuracy.

[0054] In S1003, the data processing device 50 performs a learning process for generating a trained model B by using previously prepared learning data B. The learning process here is as described with reference to FIG. 9 and the like. Note that when preparing the learning data B, the data may be created by manually labeling, i.e., assigning "OK" or "NG", using a combination of the welding image and the welding information obtained in the process of S1002.

[0055] In S1004, the data processing device 50 inputs the combination of the welding image and welding information output in S1002 to the trained model B generated in S1003, and acquires correct / incorrect information for the combination.

[0056] In S1005, the data processing device 50 extracts learning data C from the combination of the welding image and the welding information based on the correct / incorrect information obtained as a processing result of S1004. More specifically, the data processing device 50 extracts a combination of the welding image and the welding information that is determined to be correct, i.e., "OK", in the correct / incorrect information, and identifies it as learning data C. Note that, in the flow of FIG. 10, the correct / incorrect information and the combination corresponding to the correct / incorrect information are separately input to the processing of S1005, but this is not limited thereto. For example, in the processing of S1004, the combination of the obtained welding image and the welding information, which is the input data of the trained model B, and the correct / incorrect information, which is the output data, may be associated with each other, held, and managed, and the data may be output collectively to the processing of S1005.

[0057] In S1006, the data processing device 50 performs a learning process for generating a trained model C using the training data C identified as a result of the process in S1005. The learning process here is as described with reference to FIG. 6. The trained model C here may be trained based on a trained model A that has already been generated, or may be trained completely separately from the trained model A. In addition, the training data C used in the learning process in this step may include the training data A used in S1001.

[0058] Next, the control phase will be described. In this control phase, actual welding is performed by the welding robot 10. In this embodiment, the welding speed is mainly controlled as the automatic welding control, so a control signal is input to the welding robot 10 to control the movement of the robot. However, when controlling the welding current or arc voltage, a control signal may be input to the welding power source to control the welding conditions such as the welding current and the arc voltage. In S1011, the data processing device 50 inputs the welding image to be processed to the learned model C generated in the learning phase, thereby acquiring welding information for the welding image. The welding image here corresponds to a welding image captured when the welding robot 10 is operating in real time.

[0059] In S1012, the data processing device 50 uses the welding information obtained in the process of S1011 to control the welding robot 10 and cause it to perform welding.

[0060] (Welding Control) The welding of the welding system 1 during actual welding will be described below. This operation is performed in the control phase shown in FIG. 10. When performing arc welding, an operator starts the robot control device 20, the power supply device 30, and the data processing device 50. The robot control device 20 controls the movement of the welding robot 10, and the power supply device 30 performs welding according to preset welding conditions. The data processing device 50 also acquires welding images captured by the visual sensor 40, and sequentially calculates welding information related to arc welding using the trained model C. In this embodiment, as described above, the position of the wire tip, the arc center, the right and left end positions of the tip of the molten pool, and the right and left end positions of the molten pool are used as the welding information.

[0061] 11 is a flowchart explaining the processing operations of robot controller 20 and power supply device 30. When starting arc welding, an operator operates teach pendant 204 provided on robot controller 20 to input a teaching program, various setting values, and a welding start command to robot controller 20. The teaching program here specifies the movement of welding robot 10, etc.

[0062] In S1101, the robot control device 20 receives a teaching program, various setting values, and a welding start instruction.

[0063] In S1102, the robot controller 20 commands the power supply device 30 to start welding (to apply voltage).

[0064] In S1103, power supply device 30 receives a command from robot control device 20 to start welding.

[0065] In S1104, power supply device 30 controls a built-in power supply circuit (not shown) to supply power, thereby starting welding. As a result, a voltage is applied between welding wire 13 (see FIG. 1) and workpiece W (see FIG. 1), and an arc is generated.

[0066] In S1105, the robot control device 20 transmits a control signal to the power supply device 30 or the welding robot 10 to perform welding control. The welding control includes automatic welding control (S1120), control of a weaving operation (S1121), and welding line tracking control (S1122). In the automatic welding control, the data processing device 50 transmits a correction signal for controlling at least one of the welding speed, the welding current, and the arc voltage to the welding robot 10 or the power supply device 30 while automatically moving the welding torch 11 in the welding direction, and the welding robot 10 or the power supply device 30 performs welding according to the correction signal. From the viewpoint of ease of control, it is preferable that the automatic welding control includes the control of the welding speed, and in this embodiment, only the welding speed is controlled.

[0067] In S1106, robot controller 20 determines whether or not welding needs to be stopped. For example, it may determine that welding needs to be stopped when an instruction to stop welding is received from an operator, when a sensor provided on welding robot 10 detects a welding end position, or when a welding abnormality is detected, or when both are detected. If welding does not need to be stopped (NO in S1106), robot controller 20 proceeds to S1107. On the other hand, if welding needs to be stopped (YES in S1106), robot controller 20 proceeds to S1108.

[0068] In S1107, the robot control device 20 receives the welding information from the data processing device 50. The welding information received here is the welding information output by the data processing device 50 using the trained model C. Details of the generation of the welding information received in this process will be described later with reference to Fig. 12. Then, the process returns to S1105, and the robot control device 20 repeats the process using the received welding information.

[0069] In S1108, the robot controller 20 stops the welding control.

[0070] In S1109, the robot controller 20 commands the power supply device 30 to stop welding. The welding is stopped by stopping the supply of welding power.

[0071] In S1110, the robot control device 20 commands the data processing device 50 to stop generating the welding information.

[0072] In S1111, power supply device 30 receives a command from robot controller 20 to stop welding.

[0073] In S1112, the power supply device 30 controls the power supply circuit by a CPU (not shown) to stop welding, whereby the operations of the robot control device 20 and the power supply device 30 end.

[0074] (Welding information generation process) The generation process of the welding information received by the robot control device 20 in S1107 of Fig. 11 will be described with reference to Fig. 12. Fig. 12 is a flowchart explaining the generation process of the welding information by the data processing device 50, the execution of welding control based on the welding information, and the transmission of correction signals for each control factor to the power supply device 30. The control factors refer to welding conditions, such as the welding speed, welding current, arc voltage, weaving operation, and weld line tracking.

[0075] In S1201, data processing device 50 receives a welding image of a welding position captured by visual sensor 40 while automatic welding is being performed.

[0076] In S1202, the data processing device 50 reduces the received welding image and converts it into a grayscale image. For example, the data processing device 50 reduces the size of the welding image to 120×100 pixels. The image processing here is an example, and other processing may be further performed, or part of the processing may be omitted depending on the processing load, etc.

[0077] In S1203, the data processing device 50 inputs the reduced welding image to the trained model C generated in the above-mentioned learning phase, and obtains the welding information output as a result. In this embodiment, the trained model C outputs, as the welding information, the coordinate position (ArcX, ArcY) of the arc center, the coordinate position (WireX, WireY) of the wire tip, the coordinate position (Pool_Lead_Lx, Pool_Lead_Ly) of the left end of the tip of the molten pool, the coordinate position (Pool_Lead_Rx, Pool_Lead_Ry) of the right end of the tip of the molten pool, the coordinate position (Pool_Lx) of the left end of the molten pool, and the coordinate position (Pool_Rx) of the right end of the molten pool.

[0078] In S1204, data processing device 50 performs welding control based on the acquired welding information, and calculates correction signals for each control factor. In the case of this embodiment, the correction signals calculated are a left-right position correction amount ΔX of welding torch 11, a speed correction amount ΔV of welding torch 11, and a set value w of a width of a weaving operation (hereinafter referred to as "weaving width"). Left-right position correction amount ΔX of welding torch 11 is calculated by the following formula (1). PoolCenX=(Pool_Lx+Pool_Rx) / 2 ΔX=PoolCenX-X0 (1) Here, X0 is a reference value for the left-right position of the center of the molten pool, and is a preset value. The calculated left-right position correction amount ΔX is used to move the welding torch 11 left-right so that the center PoolCenX of the molten pool image coincides with X0.

[0079] When calculating the speed correction amount ΔV, first, ΔL is calculated by the following formula (2). LeadY=(Pool_Lead_Ly+Pool_Lead_Ry) / 2-ArcY ΔL=LeadY-L0 (2) Here, L0 is the distance between the molten pool tip position and the arc center, and is preset to a value that can form the best quality back bead. LeadY is also the distance between the molten pool tip position and the arc center. Here, LeadY is calculated by formula (2) using Pool_Lead_Ly, Pool_Lead_Ry, and ArcY, which are image feature information extracted from the welding image.

[0080] ΔL is the difference between LeadY and L0. Once the distance ΔL is calculated, the speed correction amount ΔV is calculated by the following equation (3) using a conversion coefficient β. ΔV=ΔL×β (3)

[0081] The set value w of the weaving width is calculated by the following formula (4) or (5). Leadw=Pool_Lead_Rx-Pool_Lead_Lx IF Leadw <w0 THEN w=0 (4) IF Leadw≧w0 THEN w=(Leadw-α) (5) Here, w0 is the gap width of the groove, and is preset to the minimum value at which weaving is required.

[0082] Leadw also indicates the gap width, and is a value calculated as described above since the coordinate positions (Pool_Lead_Lx, Pool_Lead_Rx) of the welding information extracted from the welding image correspond to the left and right positions of the groove, respectively. According to the above formula (4), when the gap width Leadw is less than w0, the weaving width setting value w becomes 0, and no weaving operation is performed. On the other hand, according to the above formula (5), when the gap width Leadw is w0 or more, the weaving width setting value w is set to a value smaller than the gap width Leadw by α. Note that α is a value that is set in advance.

[0083] The process of S1204 may be performed by using a trained model generated by a training process different from the training process for generating the trained models A to C described above. For example, a trained model may be generated by using information that associates welding information with welding result information, etc., as teacher data. Then, the welding information may be input to the trained model, and a correction signal for each control factor may be output so as to obtain an optimal welding result. Here, the welding result information may include waveform data of the welding current, waveform data of the arc voltage, arc sound data during welding, bead shape data, spatter generation data, fume generation data, welding defect data, etc., and it is preferable to handle at least one of these data as the welding result information. In this case, the training process may be implemented by a training process using reinforcement learning in addition to the above-mentioned supervised learning.

[0084] In S1205, the data processing device 50 transmits the correction signal calculated in S1204 to the robot control device 20.

[0085] In S1206, the data processing device 50 determines whether or not a stop command has been received from the robot control device 20. The stop command here corresponds to the command to stop generating welding information transmitted from the robot control device 20 in the process of S1110 in Fig. 11. If a stop command has not been received (NO in S1206), the process of the data processing device 50 returns to S1201. On the other hand, if a stop command has been received (YES in S1206), this process flow ends.

[0086] [verification] The operation verification was performed by applying the method according to the present embodiment. The verification results will be described below with reference to the drawings. In the verification of the present embodiment, first, 3700 welding images were prepared as training data A for generating a trained model A, and welding information was manually added to the training data A. Then, a trained model A was generated by a learning process using the training data A. Next, the welding images were input to the trained model A to obtain the welding information of the welding images, and 17000 combinations of the welding images were obtained.

[0087] Separately, as training data B for generating trained model B, combinations of welding images and welding information were prepared, and correct / incorrect information for the combinations was added. In this embodiment, 397 combinations with correct / incorrect information of "OK" and 274 combinations with correct / incorrect information of "NG" were prepared, and these 671 pieces of data were used as training data B. Then, using the training data B, trained model B was generated by a learning process. Next, the 17,000 combinations acquired by using trained model A were input to trained model B, and correct / incorrect information for each combination was acquired. Based on the results, 14,400 combinations judged to be "OK", i.e., correct, could be acquired.

[0088] Furthermore, as the training data C for generating the trained model C, the 14,400 combinations determined to be correct in the trained model B and the 3,700 training data A used for generating the trained model A were used. Then, the trained model C was generated by performing a learning process using a total of 18,100 training data C.

[0089] In other words, in the case of the conventional learning method, 18,100 pieces of training data must be manually created to generate a trained model to be used in actual welding control. However, with the method of the present embodiment, it is sufficient to prepare 3,700 pieces of such training data. As a result, the cost of creating training data can be reduced by as much as 80%. If more training data is required to further improve the accuracy of the trained model, the cost can be further reduced.

[0090] Next, an example of improved accuracy of welding control according to this embodiment will be described. Fig. 13 shows the recognition result of the weld pool tip width in a tapered workpiece in which the groove gap width changes continuously from 3 mm to 10 mm. The weld pool tip width is the Y coordinate above the weld pool tip - the Y coordinate below the weld pool tip, that is, a feature value corresponding to the gap width. In Fig. 13, the vertical axis indicates the weld pool tip width [mm], and the horizontal axis indicates the passage of time [s]. Here, the focus is on the latter half of welding, from 100 seconds to 170 seconds.

[0091] Line 1301 shows the recognition result when welding is performed using the trained model C generated by the learning method according to the present embodiment. Line 1302 shows the recognition result when welding is performed using a trained model generated by a conventional learning method, i.e., a model equivalent to the trained model A described above. The hatching around each of the lines 1301 and 1302 indicates the variation range, i.e., the variation, of the molten pool tip width of each line. When the trained model generated by the conventional learning method is used, as shown by line 1302, the variation in the molten pool tip width shown on the vertical axis is large. Since the groove width of an actual welded joint cannot vary in this way, it can be seen that the molten pool tip position cannot be correctly recognized and the recognition accuracy is unstable. On the other hand, when the trained model C generated by the learning method according to the present embodiment is used, the line 1301 has less variation in the molten pool tip width compared to line 1302, and the molten pool tip position is stably recognized.

[0092] In addition, the gap width at the end of welding at around 170 seconds is expected to be about 10 mm. Referring to line 1302, the conventional learning method recognizes the gap as about 8.5 mm, which is narrower than the actual width. On the other hand, when trained model C generated by the learning method according to this embodiment is used, the gap width is recognized as about 10.5 mm, which is closer to the actual width of the tip of the molten pool. In light of the above, the use of the trained model generated by the learning method according to this embodiment enables more stable and accurate recognition than the conventional learning method.

[0093] Fig. 14 is a diagram showing the error between the weld pool tip width value obtained manually and the weld pool tip width recognition result obtained using the trained model in the welding images taken around 120 seconds and 160 seconds in the graph shown in Fig. 13. In Fig. 14, the vertical axis shows the recognition error [mm] of the weld pool tip width. The error between the recognition result obtained using the trained model according to the conventional method and the manual value is about 1.4 mm. On the other hand, the error between the recognition result obtained using the trained model C according to this embodiment and the manual value is about 0.4 mm, which means that the weld pool tip width can be recognized more accurately than the conventional method.

[0094] FIG. 15 is a confusion matrix regarding the judgment accuracy of the trained model B generated by the learning method according to the present embodiment. When a combination of a welding image labeled with "OK" as correct / incorrect information and the welding information thereof is input, the trained model B correctly judges that the welding image is "OK" at a rate of 0.9, i.e., 90%. When a combination of a welding image labeled with "NG" as correct / incorrect information and the welding information thereof is input, the trained model B correctly judges that the welding image is "NG" at a rate of 0.7, i.e., 70%. That is, in this example, a trained model B capable of accurate judgment can be generated using a relatively small number of training data B, 671 pieces. In particular, the training data B for generating the trained model B can be prepared by only relatively simple labeling of OK / NG, and therefore the effort involved can be reduced.

[0095] As described above, this embodiment makes it possible to generate a trained model capable of extracting trained data, taking into consideration the accuracy of the trained data used in machine learning processing assuming a welding site. In particular, it is possible to determine whether or not a combination of a welding image and welding information is correct, and to extract appropriate trained data, so that the burden and cost of manually creating trained data can be significantly reduced. Then, by using the trained model generated using the trained data, it becomes possible to realize welding with higher accuracy.

[0096] <Other embodiments> In the above embodiment, the learning process is performed using a welding image. However, the present invention is not limited to this configuration. In addition to the welding image, the data obtained by welding may include audio information and numerical information as data during learning. For example, the learning data for generating the learned model A and the learned model C may include the control parameters of the welding robot 10 when the welding image is acquired. Examples of the control parameters include the welding posture of the welding robot 10, welding conditions, welding materials, and shielding gas information. Examples of the welding posture include information such as whether the welding torch 11 is facing upward, downward, or sideways. Examples of the welding conditions include current, voltage, and welding speed. Examples of the audio information include arc sound. In addition, the welding information output from the learned model A and the learned model C is not limited to the welding information shown in FIG. 6 and the like, and may be other characteristic information that is correlated with the data obtained by welding.

[0097] Furthermore, each trained model described in the above embodiment may be updated as appropriate. That is, after a certain amount of welding images is accumulated, trained model A or trained model B may be used again to extract trained data C and update trained model C. This method reduces the cost of generating trained data C, and therefore makes it easy to retrain trained model C to improve its accuracy.

[0098] In the above embodiment, the trained model B is configured to extract training data from the output result of the trained model A. This is not limited to the configuration, and may also be applied to the output result of the trained model C. With this configuration, for example, appropriate training data C can be selected when re-training the trained model C. As a more specific example, the data selected here may be used as verification data or test data for calculating the accuracy of the trained model C.

[0099] In the present invention, a program or application for realizing the functions of one or more of the above-mentioned embodiments can be supplied to a system or device via a network or a storage medium, etc., and one or more processors in a computer of the system or device can read and execute the program.

[0100] Alternatively, the function may be realized by a circuit that realizes one or more functions, such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0101] As described above, the present specification discloses the following: (1) A generation step of generating a first trained model by performing machine learning using a combination of data obtained during welding and feature information identified from the data as an input and a classification of whether the combination is correct or incorrect as an output. A method for generating a trained model, comprising: According to this configuration, it is possible to extract learning data taking into consideration the accuracy of the learning data used in machine learning processing assuming a welding site.

[0102] (2) the data is a welding image, The feature information is welding information indicating elements included in the welding image. A method for generating a trained model according to (1). According to this configuration, it is possible to carry out learning processing using the welding image and the welding information shown therein.

[0103] (3) A method for generating a trained model described in (1) or (2), characterized in that the feature information is information relating to at least one of a flow portion generated by welding and a plasma gas portion. According to this configuration, it is possible to perform a learning process using information relating to at least one of the flow section and the plasma gas section.

[0104] (4) the characteristic information regarding the flow portion includes positions of both ends of the width of the molten pool and positions of both ends of the tip of the molten pool; The characteristic information relating to the plasma gas portion includes a center position of the arc. A method for generating a trained model according to (3). According to this configuration, it is possible to perform a learning process using information relating to at least one of the flow section and the plasma gas section.

[0105] (5) In the training data for generating the first trained model, The combinations we label as positive include Data identifying at least the positions of both ends of the molten pool width, the positions of both ends of the molten pool tip, the center position of the arc, and the position of the wire tip, or The data includes data in which at least one of the positions of both ends of the molten pool width or the positions of both ends of the molten pool tip cannot be recognized due to the influence of the arc light, The combinations labeled as "no" include data in which the positions of both ends of the molten pool were misidentified due to the influence of a certain material. A method for generating a trained model according to (4). According to this configuration, the accuracy of the first trained model can be improved in response to actual conditions in welding.

[0106] (6) further comprising a second generation step of performing machine learning using the combination classified as positive in the first trained model as training data, with data obtained during welding as input, and feature information corresponding to the data as output, to generate a second trained model; A method for generating a trained model according to any one of (1) to (5). According to this configuration, it is possible to use the first trained model to generate a second trained model for controlling the welding robot with high accuracy.

[0107] (7) The combination classified using the first trained model is generated using a third trained model generated by performing machine learning using data obtained during welding as input and feature information identified from the data as output. A method for generating a trained model according to any one of (1) to (6). With this configuration, it is possible to generate data to which the first trained model is applied without human intervention.

[0108] (8) a first generation means for generating a first trained model by performing machine learning using a combination of data obtained during welding and feature information identified from the data as an input and a classification of whether the combination is correct or incorrect as an output; A second generation means for performing machine learning using a combination classified as positive in the first trained model as training data, with data obtained during welding as input, and feature information corresponding to the data as output, to generate a second trained model; having A learning device characterized by: With this configuration, it is possible to generate a first trained model and then use this to generate a second trained model for controlling a welding robot with high accuracy.

[0109] (9) Further comprising a third generation means for generating a third trained model by performing machine learning using data obtained during welding as an input and feature information identified from the data as an output, The learning device described in (8), characterized in that a combination of input and output of the third trained model is a target to be classified using the first trained model. With this configuration, it is possible to generate data to which the first trained model is applied without human intervention.

[0110] (10) The learning device described in (9) is characterized in that the training data for generating the third trained model includes data obtained under different conditions for at least one element of the welding posture, welding conditions, welding material, and shielding gas in the welding operation. According to this configuration, by using training data corresponding to various conditions, it is possible to enable the third trained model to perform highly accurate training with a relatively small amount of data.

[0111] (11) The training data for generating the second trained model includes A combination of data obtained during welding and characteristic information manually set on the data; and A combination manually corrected for a combination classified as negative by the first trained model; The learning device according to (9) or (10), comprising at least one of the following: According to this configuration, by including highly accurate manually prepared training data, it is possible to improve the accuracy of the second trained model.

[0112] (12) A control device having a first trained model and a second trained model generated by the learning device according to any one of (8) to (11), An acquisition means for acquiring data related to welding; A control means for outputting control information for a welding robot or a welding power source; having The control means Feature information output by inputting the data acquired by the acquisition means into the second trained model, or A combination of the data acquired by the acquisition means and feature information output by inputting the data into the second trained model is input into the first trained model, and feature information of the combination classified as positive is The control device outputs the control information based on any one of the above. According to this configuration, it is possible to control the welding robot with high accuracy using a trained model that has been trained using highly accurate teacher data.

[0113] (13) The welding robot control device further includes a learning means for performing a learning process using the control information output by the control means and the data acquired by the acquisition means as a result of controlling the welding robot or the welding power source based on the control information, The control device according to (12) is further characterized in that the control means outputs control information for the welding robot or the welding power source based on the result of the learning process by the learning means. According to this configuration, a learned model that has been trained using highly accurate teacher data is used to control a welding robot or welding power source, and then the learning process can be further carried out to further improve the accuracy of control.

[0114] (14) The control device according to (12) or (13), wherein the control information is at least one of a welding posture of the welding robot, welding conditions, and a control amount for a shielding gas. According to this configuration, it is possible to improve the accuracy of the control amount for the welding posture of the welding robot, the welding conditions, and the shielding gas.

[0115] (15) A control device according to any one of (12) to (14), A welding robot A detection means for detecting welding by the welding robot; A welding power source; A system comprising: According to this configuration, it is possible to control the welding robot with high accuracy using a trained model that has been trained using highly accurate teacher data.

[0116] A trained model generated by machine learning using a combination of data obtained during welding and feature information identified from the data as input and a classification of whether the combination is correct or incorrect as output, On the computer, A trained model that functions to output a classification of whether a combination of data obtained during welding and feature information identified from the data is correct when the combination is input. According to this configuration, it is possible to extract learning data taking into consideration the accuracy of the learning data used in machine learning processing assuming a welding site.

[0117] Computer, a first generation means for generating a first trained model by performing machine learning using a combination of data obtained during welding and feature information identified from the data as an input and a classification of whether the combination is correct or incorrect as an output; a second generation means for performing machine learning using the combination classified as positive in the first trained model as training data, with data obtained during welding as input, and feature information corresponding to the data as output, to generate a second trained model; A program to function as a With this configuration, it is possible to generate a first trained model and then use this to generate a second trained model for controlling a welding robot with high accuracy. [Explanation of symbols]

[0118] 1 Welding system 10. Welding robot 11 Welding torch 12 Wire feeder 13 Welding wire 20 Robot control device 30 Power supply 40 Visual Sensor 50 Data processing device 510 Main unit 511 CPU 512 ROM 513 RAM 514 Non-volatile storage 514A Program of Study 514B Pre-trained model 514C Information Generation Program 514D Image data 515 Input / Output Interface 516 Communication Interface 517 Video Output Interface 520 Input section 530 Display section W work

Claims

1. a first generation step of performing machine learning using a combination of a welding image obtained during welding and feature information identified from the welding image as an input and a true / false classification indicating whether the combination is correct or not as an output to generate a first trained model; a second generation step of performing machine learning using the combination classified as positive in the first trained model as training data, a welding image obtained during the welding as an input, and the feature information corresponding to the welding image as an output to generate a second trained model; In addition to providing a third generation step of generating a combination to be classified using the first trained model by using a third trained model generated by performing machine learning using a welding image obtained during the welding as an input and the feature information identified from the welding image as an output; the characteristic information includes welding information indicating information related to at least one of a flow portion generated by welding and a plasma gas portion, which are elements included in the welding image; the welding information relating to the flow portion includes positions of both ends of a width of the molten pool and positions of both ends of a tip of the molten pool, and the welding information relating to the plasma gas portion includes a center position of an arc; A method for generating a trained model.

2. In the training data for generating the first trained model, The combinations we label as positive include Data identifying at least the positions of both ends of the molten pool width, the positions of both ends of the molten pool tip, the center position of the arc, and the position of the wire tip, or The data includes data in which at least one of the positions of both ends of the molten pool width or the positions of both ends of the molten pool tip cannot be recognized due to the influence of the arc light, The combinations labeled as "no" include data in which the positions of both ends of the molten pool were misidentified due to the influence of a certain material. The method for generating a trained model according to claim 1 .

3. a first generation means for performing machine learning using a combination of a welding image obtained during welding and feature information identified from the welding image as an input and a classification of correct or incorrect indicating whether the combination is correct or not as an output to generate a first trained model; a second generation means for performing machine learning using a combination classified as positive in the first trained model as training data, a welding image obtained during the welding as an input, and feature information corresponding to the welding image as an output to generate a second trained model; having A third generation means for generating a third trained model by performing machine learning using a welding image obtained during welding as an input and feature information identified from the welding image as an output, A combination of inputs and outputs of the third trained model is an object to be classified using the first trained model, The training data for generating the third trained model includes data obtained under different conditions for at least one element of a welding posture, a welding condition, a welding material, and a shielding gas in the welding, the characteristic information is welding information indicating information related to at least one of a flow portion and a plasma gas portion generated by welding, which are elements included in the welding image; A learning device characterized in that the welding information related to the flow portion includes the positions of both ends of the molten pool width and the positions of both ends of the molten pool tip, and the welding information related to the plasma gas portion includes the center position of the arc.

4. The training data for generating the second trained model includes: A combination of data obtained during welding and characteristic information manually set on the data; and A combination manually corrected for a combination classified as negative in the first trained model; 4. The learning device according to claim 3, further comprising at least one of:

5. A control device having a first trained model and a second trained model generated by the learning device according to claim 3 or 4, An acquisition means for acquiring data related to welding; A control means for outputting control information for a welding robot or a welding power source; having The control means Feature information output by inputting the data acquired by the acquisition means into the second trained model, or inputting a combination of the data acquired by the acquisition means and feature information output by inputting the data into the second trained model into the first trained model, and obtaining feature information of the combination classified as positive; The control device outputs the control information based on any one of the above.

6. a learning means for performing a learning process using the control information output by the control means and the data acquired by the acquisition means as a result of controlling the welding robot or the welding power source based on the control information; 6. The control device according to claim 5, wherein the control means further outputs control information for the welding robot or the welding power source based on a result of the learning process performed by the learning means.

7. 7. The control device according to claim 5, wherein the control information is at least one of a welding posture of the welding robot, welding conditions, and a control amount for a shielding gas.

8. A control device according to any one of claims 5 to 7, A welding robot, A detection means for detecting welding by the welding robot; A welding power source; A system comprising:

9. A program for causing a computer to generate a trained model, comprising: Computer, a first generation means for performing machine learning using a combination of a welding image obtained during welding and feature information identified from the welding image as an input and a true / false classification indicating whether the combination is correct or not as an output to generate a first trained model; a second generation means for performing machine learning using a combination classified as positive in the first trained model as training data, a welding image obtained during the welding as an input, and the feature information corresponding to the welding image as an output, thereby generating a second trained model; a third generation means for generating a combination to be classified using the first trained model by using a third trained model generated by performing machine learning using a welding image obtained during the welding as an input and the feature information identified from the welding image as an output; Function as a the characteristic information includes welding information indicating information related to at least one of a flow portion generated by welding and a plasma gas portion, which are elements included in the welding image; the welding information relating to the flow portion includes positions of both ends of a width of the molten pool and positions of both ends of a tip of the molten pool, and the welding information relating to the plasma gas portion includes a center position of an arc; A program characterized by:

Images