Training support system, training support method, and program

Abstract

To provide a training support system which enables mastering of even appropriate postures while welding.SOLUTION: A training support system is provided, comprising: virtual welding means having a display unit and an operating unit; measurement means for measuring postures of a user as three-dimensional data; learning means configured to machine-learn the posture data measured by the measurement means during actual welding work done in a site environment by a first user to build a first welding posture model; and comparison means configured to compare the first welding posture model built by the learning means with the posture data measured by the measurement means during virtual welding work done by a second user using the virtual welding means and output a comparison result.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a training support system, a training support method, and a program for supporting training on welding techniques. [Background technology]

[0002] In training to learn welding techniques, actual equipment has traditionally been used, and in recent years, training using VR (Virtual Reality) and AR (Augmented Reality) has also been on the rise. Patent Document 1 discloses a technology in which a user undergoing training wears a helmet equipped with a small display device and manipulates a simulated welding object (welding coupon) and mock welding tool, and the display device displays a simulated welding process in a virtual reality space. Using such a mechanism makes it possible to train in a safe environment to improve welding techniques. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-223966 A Summary of the Invention [Problem to be solved by the invention]

[0004] In the above-mentioned welding training using VR, it is possible to simulate the ideal welding state of the welding part, but this is only evaluated from the distance between the welding base material and the torch and the movement of the torch. In actual welding technology, it is known that the welding accuracy, beauty of the finish, work efficiency, and work safety differ greatly between an expert and an unexperienced person depending on the welding posture, etc. The object of the present invention is to provide a training support system that makes it possible to learn the appropriate welding posture. [Means for solving the problem]

[0005] The training support system of the present invention is characterized by having a virtual welding means having a display unit and an operation unit, a measurement means for measuring a user's posture as three-dimensional data, a learning means for creating a first welding posture model by machine learning posture data measured by the measurement means during actual welding work performed by a first user in an on-site environment, and a comparison means for comparing the first welding posture model created by the learning means with posture data measured by the measurement means during virtual welding work performed by a second user using the virtual welding means, and outputting a comparison result. Effect of the Invention

[0006] According to the present invention, it is possible to provide a training support system that enables a person to learn the proper posture for welding. [Brief description of the drawings]

[0007] [Figure 1] 1 is a diagram illustrating an example of a hardware configuration of a training support system according to an embodiment of the present invention. [Diagram 2] 1 is a diagram illustrating an example of a functional configuration of a training assistance system according to an embodiment of the present invention. [Diagram 3] FIG. 2 is a diagram illustrating virtual welding in the present embodiment. [Figure 4] FIG. 2 is a diagram illustrating the creation of each model in this embodiment. [Diagram 5] FIG. 4 is a diagram illustrating torch operation data in the present embodiment. [Figure 6] 1A to 1C are diagrams illustrating the movement angle and operation angle of a torch in a welding operation. [Figure 7] 4A and 4B are diagrams illustrating posture data of a user in the present embodiment. [Figure 8] FIG. 11 is a diagram illustrating an evaluation of welding work in this embodiment. [Figure 9] FIG. 11 is a diagram showing an example of display of evaluation results in the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the following description, actual welding performed in an on-site environment is also referred to as "on-site welding", and virtual welding performed in a simulated virtual space is also referred to as "virtual welding".

[0009] 1 is a diagram showing an example of a hardware configuration of a training support system 100 according to an embodiment of the present invention. The training support system 100 according to the present embodiment is a training support system that provides training in welding techniques based on the movement of a torch and the posture of a worker during welding work.

[0010] The training support system 100 includes a CPU 101, a main memory device 102, an auxiliary memory device 103, an output device 104, an HMD 105, a controller 106, a motion sensor 107, a camera 108, and a torch motion detection device 109. The CPU 101, the main memory device 102, the auxiliary memory device 103, the output device 104, the HMD 105, the controller 106, the motion sensor 107, the camera 108, and the torch motion detection device 109 are connected to each other so as to be able to communicate with each other.

[0011] The CPU (Central Processing Unit) 101 is a central processing unit that controls the training support system 100. The main memory 102 is a memory device that functions as a work area for the CPU 101 and a temporary storage location for data. The main memory 102 is implemented using, for example, a RAM (Random Access Memory) or the like. The auxiliary memory 103 is a memory device that stores various data, various programs, and the like. The auxiliary memory 103 is implemented using, for example, a ROM (Read Only Memory), a hard disk drive (HDD), a solid state drive (SSD), and the like. The CPU 101 executes processing based on a program stored in, for example, the auxiliary memory 103, thereby realizing various processing operations in the training support system 100. The output device 104 is a device that outputs various information such as an evaluation result of a welding operation, and is used to present various information to a user, etc.

[0012] The HMD (Head Mounted Display) 105 is a head-mounted display device that displays images (VR images) in a virtual space. The HMD 105 may be a non-transparent HMD that prevents the user from checking the outside when worn, or a transparent (see-through) HMD that displays the outside when necessary. The controller 106 is a controller that accompanies the HMD 105 and is used to perform operations in a virtual space displayed on the HMD 105. For example, in virtual welding (virtual welding) performed in a virtual space, a VR image showing the welding state, such as various information and an image during welding, is displayed on the HMD 105, and the virtual welding is performed using the controller 106. At this time, the controller 106 is used to detect, for example, a pseudo torch operation. The HMD 105 is an example of a display unit of a virtual welding means, and the controller 106 is an example of an operation unit of a virtual welding means.

[0013] The motion sensor 107 is a sensor for three-dimensionally measuring (motion capture) the user's posture (movement). The motion sensor 107 is not limited to a type of motion capture as long as it can measure the user's posture, etc. In the following, the motion sensor 107 will be described taking as an example a wearable motion capture suit type, also called smart apparel, incorporating an IMU sensor (inertial measurement unit). The motion sensor 107 may be a wearable suit type motion sensor with a marker that is detected by a tracker, for example. The motion sensor 107 may also be a photographing type (camera type) that detects the distance and position of each part of the subject by capturing the subject, for example.

[0014] Camera 108 is a camera that captures images of welding phenomena during actual welding (on-site welding) performed in an on-site environment. Torch motion detection device 109 detects torch motion during actual welding (on-site welding) performed in an on-site environment. Torch motion detection device 109 detects the movement of the torch, for example, based on a marker attached to the torch. Note that torch motion detection device 109 only needs to be able to detect the motion of the torch during welding, and there are no limitations on the detection method.

[0015] In actual welding (on-site welding) performed in an on-site environment, various data during welding are acquired, for example, using a motion sensor 107, a camera 108, and a torch motion detection device 109. In virtual welding (virtual welding) performed in a virtual space, various data during welding are acquired, for example, using a motion sensor 107 and a controller 106.

[0016] 2 is a diagram showing an example of the functional configuration of the training support system 100 in this embodiment. The training support system 100 includes an acquisition unit 201, a storage unit 202, a display unit 203, an operation unit 204, an analysis unit 205, an evaluation unit 206, a learning processing unit 207, an image generation unit 208, a display control unit 209, and a control unit 210.

[0017] The acquisition unit 201 acquires various data used for training support of welding techniques. For example, the acquisition unit 201 acquires various data related to the operation of a torch in on-site welding, the posture of a user (worker), and welding phenomena such as a molten pool and a weld mark (bead). The various data related to on-site welding are acquired, for example, from the motion sensor 107, the camera 108, and the torch operation detection device 109. Also, for example, the acquisition unit 201 acquires various data related to the operation of a torch and the posture of a user (worker) in virtual welding. The various data related to virtual welding are acquired, for example, from the motion sensor 107 and the controller 106. Also, the molten pool and the weld mark (bead) can be acquired as an image by photographing them with the camera 108.

[0018] The storage unit 202 stores various data acquired by the acquisition unit 201 and various models created by the learning processing unit 207. The storage unit 202 may also store data such as processing results acquired by the analysis unit 205 and the evaluation unit 206. The display unit 203 is a device that outputs various information such as an evaluation result of a welding operation, and displays various information to be presented to a user or the like according to the control of the display control unit 209. For example, the display unit 203 displays a result screen showing the evaluation result of the welding operation by the evaluation unit 206, and displays various information when performing virtual welding and VR images showing the state of welding. The operation unit 204 is used to perform operations in a virtual space. The operation unit 204 is used, for example, when performing various operations in virtual welding. The display unit 203 is an example of a display unit of a virtual welding means, and the operation unit 204 is an example of an operation unit of a virtual welding means.

[0019] The analysis unit 205 analyzes the torch movement during welding, the user (worker) posture, welding phenomena, and the like, based on various data acquired by the acquisition unit 201. For example, the analysis unit 205 analyzes the torch movement (speed, position, and angle) during welding, based on data acquired from the torch movement detection device 109 during on-site welding and data acquired from the controller 106 during virtual welding. The analysis unit 205 also analyzes the posture (movement) of the user (worker) during welding, based on data acquired from the motion sensor 107 during on-site welding and virtual welding.

[0020] The evaluation unit 206 compares the torch movements, working postures, and welding phenomena between the skilled and trainees in the welding work, and evaluates the welding work. The evaluation unit 206 is an example of a comparison means. For example, the evaluation unit 206 compares the trainee's torch movements with a torch model, which will be described later and is created in advance regarding the skilled torch movements, to evaluate the trainee's torch movements during welding. The evaluation unit 206 also compares the trainee's working posture based on the three-dimensional measurement results with a welding posture model, which will be described later and is created in advance regarding the skilled's working posture, to evaluate the trainee's posture during welding. The evaluation unit 206 also compares the welding phenomenon simulated based on the trainee's torch movements, etc., with a welding phenomenon model, which will be described later and is created in advance regarding the welding phenomenon in welding by a skilled person, to evaluate the welding phenomenon in welding by the trainee.

[0021] The learning processing unit 207 receives various data obtained during welding work by an expert, performs machine learning using the input data as teacher data, and creates a model corresponding to the data. For example, the learning processing unit 207 creates a torch model related to the operation of the torch, a welding posture model related to the posture during welding work, and a welding phenomenon model related to the welding phenomenon (for example, a molten pool model related to the molten pool) based on data obtained during on-site welding or virtual welding work by an expert.

[0022] The image generation unit 208 generates images to be displayed in the training support system 100. For example, the image generation unit 208 generates VR images to be displayed when performing virtual welding, images showing evaluation results of welding work, etc. The display control unit 209 controls the display on the display unit 203. For example, the display control unit 209 controls the display position, size, etc. of the image generated by the image generation unit 208 on the display unit 203. The control unit 210 controls each functional unit of the training support system 100.

[0023] FIG. 3 is a diagram for explaining virtual welding (virtual welding) performed in a virtual space in this embodiment. FIG. 3(a) shows an example of a VR image 301 displayed on the HMD 105 when virtual welding is performed. In the VR image 301, a torch (welding rod) 302 simulating actual welding (on-site welding) performed in a field environment, a welding mark (bead) 303, and a molten pool 304 formed by a generated arc are displayed as elements related to the welding work itself. In addition, the VR image 301 can also display an evaluation 306 of the welding work (for example, a four-level display, etc.), operation information 307 of the torch (welding rod) 302, and posture information 308 during the welding work. In addition, in virtual welding, it is also possible to output sounds during welding, sounds that beat rhythms, etc. together with the VR image 301. In addition, precautions (reminders) regarding the operation and posture of the torch during virtual welding may be displayed by image or output by sound.

[0024] The display (output) of each element in virtual welding can be set to be displayed (output) or not via a setting screen such as that shown in Fig. 3(b). Fig. 3(b) shows an example in which it is possible to set whether to display the arc light, spatter, fume, and molten pool as elements to be displayed in virtual welding, and whether to display (output) the ruler, metronome (rhythmic sound), molten pool frame, and torch movement angle / operation angle as guides to support the welding work, and whether to display real-time evaluation. Note that the settings shown in Fig. 3(b) are merely examples and are not limited to these.

[0025] The training support system 100 in this embodiment evaluates the welding by the trainee by comparing a model of the welding by the expert, which is created based on data acquired during the welding work by the expert, with data acquired during the welding work by the trainee. The operation of the training support system 100 related to this evaluation will be described below.

[0026] The training support system 100 creates models for each piece of data based on various data acquired during welding work by an expert. FIGS. 4(a) to 4(c) are diagrams for explaining the creation of models in the training support system 100. In the example shown in this embodiment, the training support system 100 creates a molten pool model related to the molten pool, which is a welding phenomenon, a torch model related to the movement of the torch during welding work, and a welding posture model related to the posture of the expert during welding work. Note that the models to be created are not limited to these, and other models related to welding phenomena may be created, for example.

[0027] 4(a) is a diagram for explaining the creation of a molten pool model. The learning processing unit 207 receives as input time-series molten pool images 401 captured by the camera 108 during on-site welding by an expert, performs machine learning, and creates a molten pool model 402. For example, the learning processing unit 207 performs binarization processing on the molten pool images 401 to calculate the molten pool area, and creates the molten pool model 402 using the time-series molten pool images 401 and the moving variance value of the molten pool area as explanatory variables.

[0028] FIG. 4(b) is a diagram for explaining the creation of a torch model. The learning processing unit 207 performs machine learning using torch operation data 411 obtained by analyzing data acquired during welding work by an expert and obtained by the analysis unit 205 as an input, and creates a torch model 412. The torch operation data 411 is data showing the operation of the torch in a time series during welding work. For example, as shown in FIG. 5, the torch operation data 411 is composed of data showing the position (x, y, z) of the torch on the x-axis, y-axis, and z-axis at each time and the rotation angle (φ, θ, ψ) on the x-axis, y-axis, and z-axis. The learning processing unit 207 calculates the distance between the torch and the welding base material, which is the welding object, the torch, the moving speed of the torch, and the moving angle and operating angle of the torch based on the torch operation data 411, and creates a torch model 412 using the respective moving variance values ​​as explanatory variables. Here, as shown in FIG. 6, when welding base material 601 with torch (welding rod) 602, the angle shown at 603 is the movement angle and the angle shown at 604 is the operation angle, and it is preferable that movement angle 603 is 70 to 80 degrees and the operation angle is 90 degrees.

[0029] 4(c) is a diagram for explaining the creation of a welding posture model. Learning processing unit 207 takes as input on-site welding posture data 421 obtained by analysis by analysis unit 205 based on data acquired by motion sensor 107 during on-site welding by an expert, and performs machine learning to create on-site welding posture model 422. Learning processing unit 207 also takes as input virtual welding posture data 431 obtained by analysis by analysis unit 205 based on data acquired by motion sensor 107 during virtual welding by an expert, and performs machine learning to create virtual welding posture model 432.

[0030] The posture data 421, 431 are data showing the posture of an expert during welding work in a time series. The posture data 421, 431 are the result of three-dimensional measurement of the posture, and are composed of data showing the positions (x, y, z) of each of the preset measurement points on the x-axis, y-axis, and z-axis at each time and the rotation angles (φ, θ, ψ) on the x-axis, y-axis, and z-axis, as shown in Fig. 7. Here, examples of the measurement points include joints and extremities of the body.

[0031] The learning processing unit 207 creates welding posture models 422 and 432 based on the posture data 421 and 431, using the movement variance values ​​of the positions of predetermined measurement points, such as the pelvis, right foot, left foot, head, right shoulder, left shoulder, right elbow, and right hand of the worker, as explanatory variables. Although an example in which the learning processing unit 207 creates the on-site welding posture model 422 and the virtual welding posture model 432 has been shown, it may be configured to create either the on-site welding posture model 422 or the virtual welding posture model 432. In on-site welding and virtual space welding, the postures may differ even when the same worker performs welding due to, for example, differences in the environment or differences in operability between an actual torch and a pseudo torch. In particular, when such differences are expected, it is preferable to create both the on-site welding posture model 422 and the virtual welding posture model 432 and use them as the welding posture model described later. Also, one welding posture model may be created without dividing into the on-site welding posture model 422 and the virtual welding posture model 432.

[0032] Each model related to welding by an expert, which is created based on data acquired during welding work by the expert, is stored, for example, in the storage unit 202. When there are multiple experts, each model related to welding is created for each expert.

[0033] After each model relating to welding by an expert is created as described above, the training support system 100 evaluates the welding technique of the trainee based on data acquired during the trainee's virtual welding work when the trainee performs virtual welding. Figs. 8(a) to 8(c) are diagrams for explaining the evaluation of welding work by the training support system 100. In the example shown in this embodiment, the training support system 100 evaluates the molten pool, which is a welding phenomenon, the movement of the torch during welding work, and the trainee's posture during welding work. For example, the training support system 100 divides the welding work into a plurality of sections and performs evaluation for each section. Note that the items to be evaluated are not limited to these, and other items may also be evaluated.

[0034] Fig. 8(a) is a diagram for explaining the evaluation of the molten pool during welding. The evaluation unit 206 compares the molten pool according to the torch operation data 801 obtained by the analysis unit 205 based on the data acquired during virtual welding by the trainee with the molten pool model 402 created as shown in Fig. 4(a), evaluates the molten pool during welding by the trainee, and outputs the evaluation result. The evaluation unit 206 simulates the molten pool based on the torch operation data 801 (802), and performs molten pool evaluation (803) by comparing the area and shape of the simulated molten pool with the molten pool model 804.

[0035] Fig. 8(b) is a diagram for explaining the evaluation of the torch movement during welding. The evaluation unit 206 compares the torch movement data 811 obtained by the analysis unit 205 based on the data acquired during virtual welding by the trainee with the torch model 412 created as shown in Fig. 4(b), evaluates the torch movement (speed, position, and angle) during welding by the trainee, and outputs the evaluation result. Note that the torch movement data 801, 811 are configured in the same manner as in Fig. 5, and can be obtained based on data detected by the controller 106 during virtual welding, for example.

[0036] Fig. 8(c) is a diagram for explaining the evaluation of the posture of the trainee during welding work. The evaluation unit 206 compares the trainee's virtual welding posture data 821 obtained by the analysis unit 205 based on the data acquired by the motion sensor 107 during virtual welding by the trainee with the welding posture model 422 or 432 created as shown in Fig. 4(c), evaluates the posture (movement) of the trainee during welding work in the virtual space, and outputs the evaluation result. The evaluation unit 206 compares all the measurement points or a predetermined part of the measurement points (e.g., the pelvis, the right toe, the left toe, the head, the right shoulder, the left shoulder, the right elbow, and the right hand) in the trainee's posture data 821 obtained by three-dimensional measurement by the motion sensor 107 with the welding posture model 422 or 432, respectively, and evaluates the posture of the trainee during welding work.

[0037] The posture during welding work is influenced by various factors such as the worker's bone structure and habits, or the height and direction of the welding point, which may cause the bead shape to be not as desired. Therefore, it is important to evaluate not only the movement of the torch itself during welding work but also the trainee's posture, but this has not been used in conventional welding technology training using VR, making it difficult for trainees to learn the ideal posture. As described above, the present invention is characterized by comparing the welding posture model 422 or 432 based on data obtained from an expert with the trainee's posture data 821, and the difference components, such as bending too much or twisting the body too much, are calculated and evaluated using numerical values, etc.

[0038] The evaluation results of the welding work by the trainee obtained as described above can be displayed, for example, on HMD 105 or output device 104. The evaluation results of the welding work by the trainee may also be stored, for example, in storage unit 202. In this manner, the evaluation results of the welding work by the trainee in past virtual welding may be stored in storage unit 202, so that the evaluation results of the current virtual welding and the evaluation results of the past virtual welding by the same trainee can be compared.

[0039] In addition, when there are multiple experts and a model is created for each of them, for example, a model of an expert having attributes similar to that of the trainee may be selected based on the trainee's attributes (such as the trainee's body shape, period of experience, habits, etc.) to evaluate the welding work of the trainee. Also, the welding work of the trainee may be evaluated using models of multiple experts, and in subsequent evaluations, a model of an expert having a high similarity to the trainee may be selected for evaluation.

[0040] In the above-described embodiment, the evaluation is performed based on the posture during welding in addition to the welding phenomenon and torch movement during welding, but the evaluation may also be performed by measuring and comparing vital data such as pulse rate (heart rate).

[0041] Fig. 9 is a diagram showing an example of the display of the evaluation results of welding work by a trainee. Fig. 9 shows an example in which evaluation is made on torch operation, bead shape (weld mark), molten pool, posture during welding work, and sense of rhythm, and each evaluation result and an overall evaluation result are displayed. In the example shown in Fig. 9, the evaluation results for each section are shown together with the torch trajectory, and the evaluation results for the entire welding work are shown in a band chart and a radar chart.

[0042] By showing the evaluation for each section of the torch trajectory, it is also possible to correlate the evaluation of the posture when, for example, the bead shape or molten pool shape is undesirable.

[0043] Furthermore, in addition to the display of the evaluation results of the welding work shown in Fig. 9, the postures of the expert and the trainee during the welding work may be displayed in an overlapping or side-by-side manner so that the difference (deviation) in posture can be recognized. Furthermore, the posture and welding state during virtual welding by the trainee may be recorded and these may be reproduced and displayed as images. The posture during virtual welding may further be displayed as a free viewpoint image.

[0044] The training support system 100 in this embodiment evaluates not only the torch movement and welding phenomenon in the virtual welding work performed by the trainee, but also the posture during the welding work by comparing it with each model of the welding work performed by an expert. In this way, by comparing the posture during the welding work performed by the trainee with the posture during the welding work performed by an expert and outputting the comparison result, the trainee can easily recognize the ideal posture of an expert and can even learn the appropriate posture during welding.

[0045] It should be noted that the above-mentioned embodiments are merely examples of the implementation of the present invention, and the technical scope of the present invention should not be interpreted as being limited by these embodiments. In other words, the present invention can be implemented in various forms without departing from its technical concept or main features. [Explanation of symbols]

[0046] 100 Training Support System 201 Acquisition Department 202 Storage section 203 Display section 204 Operation section 205 Analysis Department 206 Evaluation Department 207 Learning Processing Unit 208 Image Generation Unit 209 Display control section 210 Control section

Claims

1. a virtual welding means having a display unit and an operation unit; A measuring means for measuring a posture of a user as three-dimensional data; a learning means for performing machine learning on posture data measured by the measuring means during actual welding work performed by a first user in an on-site environment to create a first welding posture model; a comparison means for comparing the first welding posture model created by the learning means with posture data measured by the measurement means in a virtual welding operation performed by a second user using the virtual welding means, and outputting a comparison result.

2. the learning means performs machine learning on posture data measured by the measuring means during a virtual welding operation performed by the first user using the virtual welding means to create a second welding posture model; 2. The training support system according to claim 1, wherein the comparison means compares the second welding posture model created by the learning means with posture data measured by the measurement means during a virtual welding operation performed by the second user using the virtual welding means, and outputs a comparison result.

3. 3. The training support system according to claim 1, wherein a result of the comparison by said comparison means is displayed on said display unit of said virtual welding means.

4. 3. The training support system according to claim 1, wherein the comparison means divides the virtual welding work into a plurality of sections and performs comparison for each of the sections.

5. 3. The training support system according to claim 1, wherein the comparison means evaluates the welding work performed by the second user based on a result of the comparison.

6. The training support system according to claim 1 or 2, characterized in that the comparison means further compares at least one of welding phenomena and torch operation in the welding work by the first user and the virtual welding work by the second user.

7. 3. The training support system according to claim 1, further comprising a presentation unit that displays guide information for the second user on the display unit of the virtual welding means based on a comparison result by the comparison unit.

8. A training support method using a training support system having a virtual welding means having a display unit and an operation unit, A measuring step of measuring a posture of a user as three-dimensional data; a learning process of creating a first welding posture model by machine learning posture data measured in the measuring process during actual welding work performed by a first user in an on-site environment; a comparison step of comparing the first welding posture model created in the learning step with posture data measured in the measuring step during a virtual welding operation by a second user using the virtual welding means, and outputting a comparison result.

9. A computer of a training support system having a virtual welding means having a display unit and an operation unit, A measuring step of measuring a posture of a user as three-dimensional data; a learning step of creating a first welding posture model by machine learning posture data measured in the measuring step during actual welding work performed by a first user in an on-site environment; a comparison step of comparing the first welding posture model created in the learning step with posture data measured in the measurement step during a virtual welding operation performed by a second user using the virtual welding means, and outputting a comparison result.

10. a virtual welding means having a display unit and an operation unit; A measuring means for measuring a posture of a user as three-dimensional data; a learning means for performing machine learning on posture data measured by the measuring means during a virtual welding operation performed by a first user using the virtual welding means to create a welding posture model; a comparison means for comparing the welding posture model created by the learning means with posture data measured by the measurement means during a virtual welding operation by a second user using the virtual welding means, and outputting a comparison result.

11. The training support system according to claim 10, wherein a result of the comparison by said comparison means is displayed on said display unit of said virtual welding means.

12. 11. The training support system according to claim 10, wherein the comparison means divides the virtual welding work into a plurality of sections and performs comparison for each of the sections.

13. The training support system according to claim 10 , wherein the comparison means evaluates the welding work performed by the second user based on the comparison result.

14. The training support system according to claim 10, wherein the comparison means further compares at least one of welding phenomena and torch operations in the virtual welding work by the first user and the virtual welding work by the second user.

15. The training support system according to claim 10, further comprising a presentation means for displaying guide information for the second user on the display unit of the virtual welding means based on a comparison result by the comparison means.

16. A training support method using a training support system having a virtual welding means having a display unit and an operation unit, A measuring step of measuring a posture of a user as three-dimensional data; a learning process for creating a welding posture model by machine learning posture data measured in the measuring process during a virtual welding operation performed by a first user using the virtual welding means; a comparison step of comparing the welding posture model created in the learning step with posture data measured in the measuring step during a virtual welding operation by a second user using the virtual welding means, and outputting a comparison result.

17. A computer of a training support system having a virtual welding means having a display unit and an operation unit, A measuring step of measuring a posture of a user as three-dimensional data; a learning step of creating a first welding posture model by machine learning posture data measured in the measuring step in a virtual welding work performed by a first user using the virtual welding means; a comparison step of comparing the welding posture model created in the learning step with posture data measured in the measurement step during a virtual welding operation by a second user using the virtual welding means, and outputting a comparison result.

Images