Welding control system, device and method based on artificial intelligence

Abstract

The invention relates to a welding control system, device and method based on artificial intelligence. A welding control system based on artificial intelligence according to one embodiment of the present disclosure may include: a machine vision system that captures at least a portion of a welding object through a camera and transmits the captured image (hereinafter, the captured image) to a relay system; the relay system is used for receiving the shot image from the machine vision system, identifying the welding position of the welding object from the received shot image based on a pre-trained first artificial intelligence model, confirming whether the identified welding position is included in a specified position range or not, and transmitting the identified welding position to the machine vision system; and transmitting the identified welding position to a welding system when the identified welding position is included within the specified position range; and a welding system which receives the welding position from the relay system and performs welding at the received welding position.

Description

Technical Field

[0001] The present disclosure relates to an artificial intelligence-based welding control system, device, and method. Background Art

[0002] A secondary battery is a battery that can be repeatedly charged and discharged. With the development of the information communication and display industries, secondary batteries are widely used as power sources for portable electronic communication devices such as cameras, mobile phones, smartphones, tablet computers (personal computers, PCs), and laptop computers. In addition, in recent years, in order to use secondary batteries as power sources for eco-friendly vehicles such as electric vehicles, battery packs including multiple battery modules are being developed.

[0003] When manufacturing the battery module or the battery pack, some components may be welded (e.g., laser welding, ultrasonic welding, etc.). For example, when the battery cells are connected in series, the positive electrode of one battery cell and the negative electrode of another battery cell can be welded together. Or, when the battery cells are connected in parallel, the positive electrodes of the battery cells can be welded together, and the negative electrodes of the battery cells can be welded together.

[0004] On the other hand, the welding can be automatically performed by a welding control system. For example, the welding control system can capture an image of a welding object (e.g., the electrode part of a battery cell) through a camera, identify (recognize) a welding position (e.g., the edge of an electrode) from the captured image through machine vision technology, and perform welding at the identified welding position. However, the problem with machine vision technology is that scratches, etc. may be misidentified as welding positions. In addition, the problem with the machine vision technology of the welding control system is that the image quality varies depending on the shooting environment (e.g., brightness, degree of contamination of the camera lens, state of the welding object (e.g., tolerance), installation state of the assembly mechanism (jig)), which may lead to misidentification of the welding position. Summary of the Invention

[0005] (I) Technical Problems to be Solved

[0006] According to one aspect of the present disclosure, an artificial intelligence-based welding control system, device, and method that improve the recognition performance of welding positions can be provided.

[0007] According to another aspect of the present disclosure, an artificial intelligence-based welding control system, device, and method that can improve the recognition rate of an artificial intelligence model can be provided.

[0008] According to another aspect of the present disclosure, an artificial intelligence-based welding control system, device, and method for improving the efficiency of welding positions can be provided.

[0009] (II) Technical Solutions

[0010] An artificial intelligence-based welding control system according to an embodiment of the present disclosure may include: a machine vision system that captures at least a part of a welding object through a camera and transmits the captured image (hereinafter, the captured image) to a relay system; the relay system that receives the captured image from the machine vision system, identifies the welding position of the welding object from the received captured image based on a pre-trained first artificial intelligence model, determines whether the identified welding position is included within a specified position range, and when the identified welding position is included within the specified position range, transmits the identified welding position to a welding system; and the welding system that receives the welding position from the relay system and performs welding at the received welding position.

[0011] According to an embodiment, when the identified welding position is not included within the specified position range, or when no welding position is identified from the captured image, the relay system may issue an alarm.

[0012] According to an embodiment, when the identified welding position is not included within the specified position range, or when no welding position is identified from the captured image, the relay system may request the machine vision system to identify the welding position. When the machine vision system is requested to identify the welding position, it may identify the welding position from the captured image, determine whether the identified welding position is included within the specified position range, and when the identified welding position is included within the specified position range, transmit the identified welding position to the welding system.

[0013] According to an embodiment, when the identified welding position is not included within the specified position range, or when no welding position is identified from the captured image, the machine vision system may issue an alarm.

[0014] According to an embodiment, the machine vision system may attempt to identify the welding position from the captured image after capturing at least a part of the welding object, and when no welding position is identified from the captured image, or when the identified welding position is not included within the specified position range, transmit the captured image to the relay system.

[0015] According to an embodiment, when the identified welding position is included within the specified position range, the machine vision system may transmit the identified welding position to the welding system.

[0016] According to one embodiment, the relay system may extract a region of interest from the captured image based on a pre-trained second artificial intelligence model, and transmit information on the extracted region of interest to the machine vision system. The machine vision system may identify a welding position from the region of interest of the captured image, and transmit the identified welding position to the welding system.

[0017] According to one embodiment, the welding control system may further include: an inspection system that, after performing the welding, checks whether there is a welding defect, and when there is the welding defect, transmits welding defect information to the relay system. The relay system may update the first artificial intelligence model based on the welding defect information.

[0018] According to one embodiment, the relay system may include: an artificial intelligence server that includes the first artificial intelligence model; and a relay server that relays communication between the machine vision system, the welding system, and the artificial intelligence server.

[0019] According to one embodiment, the welding object may include a secondary battery. The welding position may include an edge of an electrode of the secondary battery.

[0020] An artificial intelligence-based welding control method according to an embodiment of the present disclosure may include the following steps: a machine vision system captures at least a part of a welding object through a camera and transmits the captured image (hereinafter referred to as the captured image) to a relay system; the relay system receives the captured image from the machine vision system, identifies a welding position of the welding object from the received captured image based on a pre-trained first artificial intelligence model, determines whether the identified welding position is included in a specified position range, and when the identified welding position is included in the specified position range, transmits the identified welding position to a welding system; and the welding system receives the welding position from the relay system and performs welding at the received welding position.

[0021] According to one embodiment, the welding control method may further include the following step: when the identified welding position is not included in the specified position range, or when no welding position is identified from the captured image, the relay system issues an alarm.

[0022] According to one embodiment, the welding control method may further include the following steps: when the identified welding position is not included in the specified position range, or when the welding position is not identified from the captured image, the relay system requests the machine vision system to identify the welding position; and when the machine vision system is requested to identify the welding position, identify the welding position from the captured image, confirm whether the identified welding position is included in the specified position range, and when the identified welding position is included in the specified position range, transmit the identified welding position to the welding system.

[0023] According to one embodiment, the welding control method may further include the following steps: when the identified welding position is not included in the specified position range, or when the welding position is not identified from the captured image, the machine vision system issues an alarm.

[0024] According to one embodiment, the welding control method may further include the following steps: after the machine vision system captures at least a part of the welding object, attempt to identify the welding position from the captured image. When the welding position is not identified from the captured image, or when the identified welding position is not included in the specified position range, the step of transmitting the captured image to the relay system may be executed.

[0025] According to one embodiment, the welding control method may further include the following steps: when the identified welding position is included in the specified position range, the machine vision system transmits the identified welding position to the welding system.

[0026] According to one embodiment, the welding control method may further include the following steps: the relay system extracts the region of interest from the captured image based on a pre-trained second artificial intelligence model, and transmits the information of the extracted region of interest to the machine vision system; and the machine vision system identifies the welding position from the region of interest of the captured image and transmits the identified welding position to the welding system.

[0027] According to one embodiment, the welding control method may further include the following steps: after performing the welding, check whether there are welding defects through an inspection system; when there are welding defects, transmit the welding defect information to the relay system; and the relay system updates the first artificial intelligence model based on the welding defect information.

[0028] An artificial intelligence-based welding control device according to an embodiment of the present disclosure may include: a camera; a memory containing a pre-trained artificial intelligence model; a welding module; and a processor that controls the camera to capture at least a part of a welding object, identifies a welding position from an image captured by the camera, i.e., a captured image, based on the artificial intelligence model, determines whether the identified welding position is included within a specified position range, and controls the welding module to weld the identified welding position when the identified welding position is included within the specified position range.

[0029] According to an embodiment, the welding control device may further include an inspection module. After performing the welding, the processor may inspect whether there are welding defects through the inspection module, and when there are welding defects, update the artificial intelligence model based on the welding defect information.

[0030] (III) Beneficial Effects

[0031] According to an embodiment of the present disclosure, the recognition performance of the welding position can be improved. For example, the present disclosure can identify the welding position from the captured image and determine whether the identified welding position is included within a specified position range. The present disclosure can accurately detect the welding position.

[0032] In addition, the present disclosure can extract a region of interest (e.g., a welding part) from the captured image and identify the welding position from the extracted region of interest. The present disclosure can accurately detect the welding position (e.g., improve the accuracy) and minimize (or prevent) misidentification.

[0033] In addition, the present disclosure can prevent misidentification of the welding position due to the shooting environment. For example, the present disclosure can train the artificial intelligence model based on images captured in various environments, thereby minimizing the influence caused by the shooting environment.

[0034] In addition, the present disclosure can improve (or increase) the recognition rate of the artificial intelligence model. For example, the present disclosure can perform a welding defect inspection after the welding is completed and further train the artificial intelligence model based on the captured images with welding defects, continuously updating the artificial intelligence model, thereby improving (or increasing) the recognition rate of the artificial intelligence model.

[0035] In addition, the present disclosure can effectively identify the welding position by combining machine vision recognition and artificial intelligence recognition. For example, the present disclosure can use machine vision recognition and artificial intelligence recognition alone or in combination to identify the welding position in a more suitable manner for the situation. Description of the Drawings

[0036] Figure 1Schematic diagram of an artificial intelligence-based welding control system according to an embodiment of the present disclosure.

[0037] Figure 2 Structural diagram of an artificial intelligence model according to an embodiment of the present disclosure.

[0038] Figure 3 Flowchart for illustrating a welding control method of an artificial intelligence-based welding control system according to an embodiment of the present disclosure.

[0039] Figure 4 Flowchart for illustrating a welding control method of an artificial intelligence-based welding control system according to another embodiment of the present disclosure.

[0040] Figure 5 Flowchart for illustrating a welding control method of an artificial intelligence-based welding control system according to yet another embodiment of the present disclosure.

[0041] Figure 6a Flowchart for illustrating a welding control method of an artificial intelligence-based welding control system according to yet another embodiment of the present disclosure.

[0042] Figure 6b Example diagram for illustrating a welding control method of an artificial intelligence-based welding control system according to yet another embodiment of the present disclosure.

[0043] Figure 7 Schematic diagram of an artificial intelligence-based welding control system according to another embodiment of the present disclosure.

[0044] Figure 8 Flowchart for illustrating a welding control method of an artificial intelligence-based welding control system according to another embodiment of the present disclosure.

[0045] Figure 9 Block diagram showing the configuration of an artificial intelligence-based welding control device according to an embodiment of the present disclosure. Detailed implementation manners

[0046] Hereinafter, the present disclosure will be described in detail with reference to the accompanying drawings. However, this is merely exemplary, and the present disclosure is not limited to the specific implementation manners illustrated exemplarily.

[0047] Although terms such as "first", "second", etc. are used to describe various elements, components, and / or parts, these elements, components, and / or parts are not limited by these terms. These terms are merely used to distinguish one element, component, or part from another element, component, or part. Therefore, the first element, first component, or first part mentioned below may also be the second element, second component, or second part within the technical concept scope of the present disclosure.

[0048] The terms used in this specification are intended to describe embodiments and not to limit the present disclosure. Unless otherwise specifically mentioned in the text, the singular forms in this specification include the plural forms. The use of "comprises" and / or "made of" in the specification means that the mentioned components, steps, operations, and / or elements do not exclude the existence or addition of one or more other components, steps, operations, and / or elements.

[0049] Unless otherwise defined, all terms (including technical terms and scientific terms) used in this specification are used in the meanings commonly understood by those of ordinary skill in the technical field to which the present disclosure pertains. Additionally, unless otherwise specifically defined, terms defined in common dictionaries should not be idealized or over-interpreted.

[0050] Figure 1 is a schematic diagram of an artificial intelligence-based welding control system according to an embodiment of the present disclosure. Figure 2 is a structural diagram of an artificial intelligence model according to an embodiment of the present disclosure.

[0051] Referring to Figure 1 and Figure 2 , an artificial intelligence-based welding control system 100 according to an embodiment of the present disclosure may include a machine vision system 110, a relay system 120, and a welding system 130.

[0052] The machine vision system 110 may capture at least a part (hereinafter referred to as the welding part 10) of a welding object (e.g., a secondary battery) through a camera 111 and transmit the captured image (hereinafter referred to as the captured image 101) to the relay system 120. The welding part 10 may include, for example, electrodes 10a, 10b, 10c of a secondary battery, the housing of the secondary battery (e.g., the housing of a battery module, the housing of a prismatic battery, the housing of a cylindrical battery), and the foil of the secondary battery. In certain embodiments, the machine vision system 110 may use machine vision technology to identify the welding position (e.g., the edge of the electrode of a secondary battery) from the captured image 101 and transmit the identified welding position to the welding system 130. The operation of the machine vision system 110 according to various embodiments of the present disclosure will be described in detail later with reference to Figures 3 to 6b the detailed description of the operation of the machine vision system 110 according to various embodiments of the present disclosure.

[0053] The relay system 120 is communicatively connected to the machine vision system 110 and the welding system 130 and may control (or manage) the overall operation of the welding control system 100. According to one embodiment, the relay system 120 may include a relay server 121 and an artificial intelligence server 122.

[0054] The relay server 121 can control (e.g., relay) the communication (e.g., data transmission and reception) between the machine vision system 110, the welding system 130, and the artificial intelligence server 122. The relay server 121 can control the overall operation of the relay system 120 (e.g., the loading and unloading of welding objects, artificial intelligence-based welding position recognition, machine vision-based welding position recognition, issuing alarms, and initialization, etc.). In addition, the relay server 121 can receive user inputs (e.g., operation commands) for the operation of the relay system 120 and output the status of the relay system 120.

[0055] The artificial intelligence server 122 can include an artificial intelligence model (hereinafter referred to as the first artificial intelligence model 20). As Figure 2 shown, when the captured image 101 is input, the first artificial intelligence model 20 can generate a feature map through a convolutional neural network (e.g., Convolution Neural Network: CNN), can estimate candidate regions of interest (ROI) through a Region Proposal Network (RPN), classify the class of the estimated candidate region of interest (ROI), identify the welding position based on the classification result, and output the identified welding position after mask processing. On the other hand, Figure 2 the structure of the first artificial intelligence model 20 is only an example and does not limit the present disclosure.

[0056] The first artificial intelligence model 20 can be trained based on images captured in various environments. In this way, the present disclosure can minimize (or prevent) the influence (e.g., misidentification) caused by the shooting environment (e.g., brightness, degree of contamination of the camera lens, state of the welding object (e.g., tolerance), installation state of the assembly mechanism (jig), etc.).

[0057] The first artificial intelligence model 20 can be further trained based on the captured images of the welding objects with welding defects (e.g., welding is performed at the wrong position). This will be described in detail later with reference to Figure 7 and Figure 8 for details.

[0058] On the other hand, the artificial intelligence server 122 can include a second artificial intelligence model (not shown), and the second artificial intelligence model can extract the region of interest for identifying the welding position from the captured image. This will be described in detail later with reference to Figure 6a and Figure 6b for details.

[0059] According to an embodiment, the relay system 120 may identify the welding position 11 from the captured image 101 based on a pre-trained first artificial intelligence model 20 and transmit the identified welding position 11 to the welding system 130. For example, the relay system 120 may receive the captured image 101 from the machine vision system 110, analyze the received captured image 101 based on the first artificial intelligence model 20 to identify the welding position 11, and confirm whether the identified welding position 11 is included within a specified position range 12. When the identified welding position 11 is included within the specified position range 12, the relay system 120 transmits the identified welding position 11 to the welding system 130. Details thereof will be described later with reference to Figure 3 for a detailed description.

[0060] According to an embodiment, when the machine vision system 110 fails to identify the welding position or the identified welding position is not included within the specified position range, the relay system 120 may receive the captured image from the machine vision system 110 and identify the welding position from the captured image. Details thereof will be described later with reference to Figure 4 for a detailed description.

[0061] According to an embodiment, when the identified welding position is not included within the specified position range or the welding position is not identified from the captured image, the relay system 120 may request the machine vision system 110 to identify the welding position. Details thereof will be described later with reference to Figure 5 for a detailed description.

[0062] According to an embodiment, the relay system 120 may extract a region of interest from the captured image based on a second artificial intelligence model different from the first artificial intelligence model 20 and transmit information on the extracted region of interest to the machine vision system 110. At this time, the machine vision system 110 may identify the welding position from the region of interest. Details thereof will be described later with reference to Figure 6a and Figure 6b for a detailed description.

[0063] The welding system 130 may receive the welding position 11 from the relay system 120 or the machine vision system 110 and perform welding (e.g., laser welding, ultrasonic welding) at the received welding position 11. For example, the welding system 130 may weld at least a part of the boundary between the first electrode 10a and the second electrode 10b and at least a part of the boundary between the second electrode 10b and the third electrode 10c.

[0064] Figure 3 is a flowchart for explaining a welding control method of an artificial intelligence-based welding control system according to an embodiment of the present disclosure.

[0065] Refer to Figure 3, the welding control method of the artificial intelligence-based welding control system according to an embodiment of the present disclosure may include step S301 of photographing a welding object. For example, the machine vision system 110 may photograph at least a part (e.g., a welding part) of the welding object (e.g., a secondary battery) through the camera 111. Here, the welding part may include the electrodes of the secondary battery.

[0066] The welding control method may include step S303 of transmitting the photographed image to the relay server 121. For example, the machine vision system 110 may transmit the photographed image to the relay server 121 of the relay system 120 through wired or wireless communication. The welding control method may include step S305 of transmitting the photographed image to the artificial intelligence server 122. For example, the relay server 121 may transmit the photographed image to the artificial intelligence server 122 through wired or wireless communication.

[0067] The welding control method may include step 307 of identifying the welding position. For example, the artificial intelligence server 122 may identify (recognize) the welding position from the photographed image based on a pre-trained first artificial intelligence model.

[0068] The welding control method may include step S309 of transmitting the welding position. For example, the artificial intelligence server 122 may transmit the identified welding position (e.g., coordinates) to the relay server 121 through wired or wireless communication. On the other hand, when the welding position is not identified, the artificial intelligence server 122 may transmit the non-identification information to the relay server 121.

[0069] The welding control method may include step S311 of confirming whether the welding position is appropriate. For example, when the welding position is within the specified position range, the relay server 121 of the relay system 120 may determine that the welding position is appropriate. On the contrary, when the welding position is not identified, or the identified welding position is not within the specified position range, the relay server 121 may determine that the welding position is inappropriate. The position range may be the area where the component to be welded (e.g., the electrodes of the secondary battery) is estimated to be in the photographed image, and may be specified in advance by the user. Thus, the present disclosure can suppress (or prevent) the problem of welding defects caused by misidentifying the welding position.

[0070] As a result of the confirmation in step S311, when the welding position is appropriate, the welding control method may execute step S313 of transmitting the welding position to the welding system 130. For example, the relay server 121 of the relay system 120 may transmit the welding position to the welding system 130 through wired or wireless communication. The welding control method may include step S315 of performing welding. For example, the welding system 130 may perform welding at the welding position.

[0071] On the other hand, for the confirmation result of step S311, when the welding position is inappropriate, the welding control method can enter step S317 of issuing an alarm. For example, the relay server 121 of the relay system 120 can issue an alarm in a specified variety of ways (such as auditory, visual, tactile, etc.) (such as outputting sound effects, emitting light, generating vibration, etc.). On the other hand, although not shown, the relay server 121 can also issue an alarm through the machine vision system 110 and / or the welding system 130.

[0072] Figure 4 is a flowchart for explaining a welding control method of an artificial intelligence-based welding control system according to another embodiment of the present disclosure.

[0073] Refer to Figure 4 , the welding control method of the artificial intelligence-based welding control system according to another embodiment of the present disclosure may include step S401 of photographing a welding object. For example, the machine vision system 110 can photograph at least a part (such as a welding part) of the welding object (such as a secondary battery) through the camera 111. Here, the welding part may include the electrodes of the secondary battery.

[0074] The welding control method may include step S403 of identifying the welding position from the photographed image. For example, the machine vision system 110 can identify the welding position from the photographed image based on machine vision technology.

[0075] The welding control method may include step S405 of confirming whether the welding position is appropriate. For example, when the welding position is included within a specified position range, the machine vision system 110 can determine that the welding position is appropriate. On the contrary, when the welding position is not recognized, or the recognized welding position is not included within the specified position range, the machine vision system 110 can determine that the welding position is inappropriate.

[0076] For the confirmation result of step S405, when the welding position is appropriate, the welding control method can execute step S407 of transmitting the welding position to the welding system. For example, the machine vision system 110 can transmit the welding position to the welding system 130 through wired or wireless communication.

[0077] On the contrary, for the confirmation result of step S405, when the welding position is inappropriate, the welding control method can enter step S409. Here, Figure 4 steps S409, S411, S413, S415, S417, S419, S421, S423 are respectively the same as Figure 3 steps S303, S305, S307, S309, S311, S313, S315, S317. Therefore, the detailed description will be omitted.

[0078] Figure 5 It is a flowchart for explaining a welding control method of an artificial intelligence-based welding control system according to another embodiment of the present disclosure.

[0079] Referring to Figure 5 , the "Yes" of steps S501, S503, S505, S507, S509, S511, S513, S523 of the welding control method of the artificial intelligence-based welding control system according to another embodiment of the present disclosure is the same as the "Yes" of steps S301, S303, S305, S307, S309, S311, S313, S315 of Figure 3 . Therefore, the detailed description will be omitted.

[0080] On the other hand, for the confirmation result of step S511, when the welding position is inappropriate, the welding control method may execute step S515 of requesting the machine vision system 110 to identify the welding position. For example, the relay server 121 of the relay system 120 may request the machine vision system 110 to identify the welding position through wired or wireless communication.

[0081] The welding control method may include step S517 of identifying the welding position from the captured image. For example, the machine vision system 110 may identify the welding position from the captured image based on machine vision technology.

[0082] The welding control method may include step S519 of confirming whether the welding position is appropriate. For example, when the welding position is within the specified position range, the machine vision system 110 may determine that the welding position is appropriate. On the contrary, when the welding position is not recognized or the recognized welding position is not within the specified position range, the machine vision system 110 may determine that the welding position is inappropriate.

[0083] For the confirmation result of step S519, when the welding position is appropriate, the welding control method may execute step S521 of transmitting the welding position to the welding system and step S523 of performing welding. For example, the machine vision system 110 may transmit the welding position to the welding system 130 through wired or wireless communication. At this time, the welding system 130 may perform welding at the received welding position.

[0084] On the contrary, for the confirmation result of step S519, when the welding position is inappropriate, the welding control method may execute step S525 of issuing an alarm. For example, the machine vision system 110 may issue an alarm (such as outputting a sound effect, emitting light, generating vibration, etc.) in specified various ways (such as auditory, visual, tactile, etc.). On the other hand, although not shown, the machine vision system 110 may also issue an alarm through the relay system 120 and / or the welding system 130.

[0085] The above-mentioned Figure 4 and Figure 5 Embodiments of can use traditional machine vision-based welding position recognition and artificial intelligence-based welding position recognition alone or in combination, and effectively recognize the welding position in a more suitable manner, further improving the accuracy of recognition.

[0086] Figure 6a is a flowchart for explaining a welding control method of an artificial intelligence-based welding control system according to another embodiment of the present disclosure.

[0087] Referring to Figure 6a , steps S601, S603, and S605 of the welding control method of the artificial intelligence-based welding control system according to another embodiment of the present disclosure are the same as Figure 3 steps S301, S303, and S305 of. Therefore, detailed descriptions will be omitted.

[0088] The welding control method may include step S607 of extracting a region of interest. For example, the artificial intelligence server 122 may extract a region of interest from the captured image using a pre-trained second artificial intelligence model. The region of interest may include a partial configuration (or partial region) to be welded in the configuration (or entire region) of the welding object (e.g., the electrode of a secondary battery). For example, the second artificial intelligence model may extract a partial region of the captured image (e.g., a region including the electrode of the secondary battery) as the region of interest.

[0089] The welding control method may include step S609 of transmitting the region of interest information to the machine vision system 110. For example, the artificial intelligence server 122 may transmit the region of interest information to the relay server 121 (S609), and the relay server 121 may transmit the region of interest information to the machine vision system 110 (S611).

[0090] The welding control method may include step S613 of identifying a welding position from the region of interest of the captured image. For example, the machine vision system 110 may use machine vision technology to identify a welding position (e.g., the edge of the electrode of a secondary battery) from the region of interest of the captured image.

[0091] The welding control method may include step S615 of transmitting the welding position to the welding system and step S617 of performing welding. For example, the machine vision system 110 may transmit the welding position to the welding system 130 via wired or wireless communication. When receiving the welding position from the machine vision system 110, the welding system 130 may perform welding at the received welding position.

[0092] Figure 6bIt is an exemplary diagram for explaining a welding control method of an artificial intelligence-based welding control system according to another embodiment of the present disclosure.

[0093] Referring to Figure 6b , the setting state (or shooting position) of the welding object may not be constant due to various reasons (e.g., assembly tolerances, installation state of the assembly mechanism, etc.). For example, the welding position of the welding object may be located at the upper end portion of the captured image as shown by reference numeral 610 in Figure 6b , or may be located in the middle portion of the captured image as shown by reference numeral 615. Hereinafter, for the sake of convenience of explanation, the setting state shown by reference numeral 610 in Figure 6b will be referred to as the normal state, and the setting state shown by reference numeral 615 will be referred to as the abnormal state.

[0094] When the setting state of the welding object is the normal state, as shown by reference numeral 620 in Figure 6b , the welding position of the welding object may be included in the designated region of interest 12a. However, when the setting state of the welding object is the abnormal state, as shown by reference numeral 625 in Figure 6b , the welding position of the welding object may not be included in the designated region of interest indicated by the dashed line. However, the artificial intelligence server 122 according to the present disclosure can extract the region of interest 12a through the second artificial intelligence model, rather than the region of interest at a designated position (e.g., Figures 3 to 5 embodiment). Therefore, the present disclosure can eliminate the inconvenience of having to set the welding object in the correct position. In addition, the present disclosure can prevent problems such as incorrect identification or failure to identify the welding position due to the welding object not being set in the correct position.

[0095] The machine vision system 110 according to the present disclosure can identify the welding position 11 from the region of interest 12a. As described above, even if the welding object is not set correctly, the present disclosure can extract the region of interest for identifying the welding position through the second artificial intelligence model and identify the welding position from the extracted region of interest, thereby enabling more accurate identification of the welding position.

[0096] On the other hand, although Figure 6a and Figure 6b illustrate and explain that the machine vision system 110 identifies the welding position from the region of interest, the artificial intelligence server 122 can also identify the welding position from the region of interest. For example, the artificial intelligence server 122 can extract the region of interest from the captured image through the second artificial intelligence model and identify the welding position from the extracted region of interest through the first artificial intelligence model.

[0097] Figure 7Schematic diagram of an artificial intelligence-based welding control system according to another embodiment of the present disclosure.

[0098] Referring to Figure 7 , an artificial intelligence-based welding control system 700 according to another embodiment of the present disclosure may include a machine vision system 710, a camera 711, a relay system 720 including a relay server 721 and an artificial intelligence server 722, a welding system 730, and an inspection system 740.

[0099] Hereinafter, the machine vision system 710, the relay system 720, and the welding system 730 of the welding control system 700 are similar to the machine vision system 110, the relay system 120, and the welding system 130 of Figure 1 the welding control system 100, and thus detailed descriptions thereof will be omitted.

[0100] The inspection system 740 may inspect whether there are welding defects after the welding of the welding object is completed by the welding system 730. When a welding defect is detected, the inspection system 740 may transmit the welding defect information to the relay system 720. The relay system 720 (e.g., the artificial intelligence server 722) may further train (e.g., update) the first artificial intelligence model based on the captured image of the welding object where the welding defect occurred. For example, the artificial intelligence server 722 may further train the first artificial intelligence model based on the marked image after an administrator of the artificial intelligence server 722 marks (e.g., marks the correct welding position) the captured image of the welding object where the welding defect occurred. As described above, the welding control system 700 of the present disclosure may further train the first artificial intelligence model based on the captured image in which the welding position is misidentified, thereby continuously updating the first artificial intelligence model. Thus, the present disclosure may improve (or enhance) the welding position recognition rate of the first artificial intelligence model. Details of the detailed operations for further training the first artificial intelligence model will be described later with reference to Figure 8 for the detailed operations for further training the first artificial intelligence model.

[0101] Figure 8 Flowchart for explaining a welding control method of an artificial intelligence-based welding control system according to another embodiment of the present disclosure.

[0102] Referring to Figure 8 , steps S801, S803, S805, S807, S809, S811, S813 of the welding control method of the artificial intelligence-based welding control system according to another embodiment of the present disclosure are the same as Figure 3 steps S301, S303, S305, S307, S309, S313, S315 of

[0103] The welding control method may include a step S815 of requesting a welding inspection. For example, the welding system 730 may request the inspection system 740 to inspect the welding of the welding object after welding has been completed.

[0104] The welding control method may include a step (S817) of inspecting for welding defects and a step (S819) of confirming for welding defects. For example, the inspection system 740 may inspect the welded object for welding defects (e.g., welding position defects) and confirm the presence of welding defects based on the inspection results. Specifically, the inspection system 740 may photograph the welded secondary battery using a separate camera or camera 711 of the machine vision system 710, analyze the captured image using an artificial intelligence model (not shown) or a separate machine vision system (not shown) (or the machine vision system 710) to measure the weld position and width, and inspect the presence of welding defects based on the measurement results. Alternatively, the inspection system 740 may further measure the weld height using a three-dimensional (3D) camera and inspect the presence of welding defects based on the measured weld height.

[0105] As a result of the confirmation in step S819 , when there is no welding defect, the welding control method may execute step S827 in which the inspection system 740 transmits welding completion information to the relay server 721 of the relay system 720 .

[0106] On the contrary, as a result of the confirmation of step S819, when there is a welding defect, the welding control method may execute step S821 in which the inspection system 740 transmits the welding defect information (e.g., a welding position defect) to the relay server 721 of the relay system 720 and step S823 in which the relay server 721 transmits the welding defect information to the artificial intelligence server 722.

[0107] The welding control method may include step S825 in which the artificial intelligence server 722 further trains (e.g., updates) the first artificial intelligence model. The artificial intelligence server 722 may further train (e.g., update) the first artificial intelligence model based on a captured image of the welding object in which the welding defect occurs.

[0108] On the other hand, although Figure 7 and Figure 8 The welding system 730 and the inspection system 740 are shown and described as being separate configurations, but the welding system 730 and the inspection system 740 may be integrated into one configuration.

[0109] Figure 9 is a block diagram illustrating a configuration of an artificial intelligence-based welding control device according to one embodiment of the present disclosure.

[0110] Reference Figure 9, according to an embodiment of the present disclosure, the artificial intelligence-based welding control device 900 may include a memory 910, a processor 920, a camera 930, a welding module 940, an alarm module 950, and an inspection module 960.

[0111] The memory 910 may store a program for controlling the operation of the welding control device 900. Additionally, the memory 910 may store information required for controlling the operation of the welding control device 900. According to one embodiment, the memory 910 may include an artificial intelligence model 911 and a machine vision algorithm 912. The artificial intelligence model 911 (e.g., the first artificial intelligence model 20) may be pre-trained based on images captured in various environments and use artificial intelligence technology to identify the welding position from the image of the welding object. The machine vision algorithm 912 may use machine vision technology to identify the welding position from the captured image. On the other hand, although not shown, the memory 910 may further include other artificial intelligence models different from the artificial intelligence model 911, which extract the region of interest from the captured image.

[0112] The processor 920 may control the overall operation of the welding control device 900. For example, the processor 920 may control the camera 930 to capture at least a part of the welding object, analyze the captured image based on the artificial intelligence model 911 to identify the welding position, confirm whether the identified welding position is included within a specified position range, and when the identified welding position is included within the specified position range, control the welding module 940 to perform welding at the identified welding position.

[0113] When the artificial intelligence model 911 fails to identify the welding position, or the identified welding position is not included within the specified position range, the processor 920 may use the machine vision algorithm 912 to identify the welding position. Conversely, when the machine vision algorithm 912 fails to identify the welding position, or the identified welding position is not included within the specified position range, the processor 920 may use the artificial intelligence model 911 to identify the welding position. Or, the processor 920 may identify the welding position by combining the recognition results of the artificial intelligence model 911 and the machine vision algorithm 912 according to the situation. For example, the processor 920 may use the machine vision algorithm 912 to identify the welding position, use the artificial intelligence model 911 to identify the size (e.g., height) of the secondary battery, and apply a specified offset to the welding position according to the size of the secondary battery. The offset set according to the size of the secondary battery may be as shown in . On the other hand, is only an example and does not limit the present disclosure.

[0114] [Table 1]

[0115]

[0116] The processor 920 can extract regions of interest from the captured images based on other artificial intelligence models different from the artificial intelligence model 911, and identify the welding positions from the extracted regions of interest by using the artificial intelligence model 911 and / or the machine vision algorithm 912.

[0117] The processor 920 can control the inspection module 960 to inspect whether there are welding defects in the welded object that has been welded by the welding module 940. When welding defects are detected by the inspection module 960, the processor 920 can further train (e.g., update) the artificial intelligence model 911 based on the welding defect information (e.g., the captured image of the welded object where the welding defect occurs).

[0118] When the welding position is not recognized, or the recognized welding position is not within the specified position range, the processor 920 can control the alarm module 950 to issue an alarm.

[0119] The camera 930 can capture at least a part of the welded object.

[0120] The welding module 940 can perform welding (e.g., laser welding, ultrasonic welding) at the welding positions recognized by the artificial intelligence model 911 and / or the machine vision algorithm 912.

[0121] When the recognized welding position is inappropriate (e.g., the welding position is not recognized, or the recognized welding position is not within the specified position range), the alarm module 950 can issue an alarm. The alarm module 950 can provide at least one of a visual alarm (e.g., a light-emitting diode (LED) emits light, an icon is displayed, a pop-up window is displayed, etc.), an auditory alarm (e.g., an effect sound is output), and a tactile alarm (e.g., vibration is generated). The alarm module 950 can include at least one of a light-emitting diode, a display, a speaker, and a vibration motor.

[0122] The inspection module 960 can inspect whether there are welding defects after the welding of the welded object is completed by the welding module 940. When welding defects are detected, the inspection module 960 can transmit the welding defect information to the processor 920. The inspection module 960 is similar to the inspection system 740 of Figure 7 and Figure 8 so the detailed description will be omitted.

[0123] On the other hand, the welding control device 900 may not include some of the above configurations, or may further include other configurations. For example, the welding control device 900 may not include the alarm module 950 and the inspection module 960. As another example, the welding control device 900 may further include a communication module for communicating with an external device. Alternatively, some of the configurations of the welding control device 900 may be configured separately. For example, the welding module 940 and / or the inspection module 960 of the welding control device 900 may be configured as separate external devices. As another example, although Figure 9 it is shown that the artificial intelligence model 911 is included in the memory 910 of the welding control device 900, the artificial intelligence model 911 may be included in an external server (e.g., an artificial intelligence server). At this time, the welding control device 900 may further include a communication module (not shown) for wired or wireless communication with the external server.

[0124] The above description is only an example of applying the principles of the present disclosure, and other configurations may be included without departing from the scope of the present invention. For example, at least a part of the various embodiments of the present disclosure described above may be combined.

Claims

1. A welding control system, which is an artificial intelligence-based welding control system, comprising: A machine vision system that captures at least a part of a welding object through a camera and transmits the captured image, i.e., the captured image, to a relay system; A relay system that receives the captured image from the machine vision system, identifies the welding position of the welding object from the received captured image based on a pre-trained first artificial intelligence model, determines whether the identified welding position is included within a specified position range, and when the identified welding position is included within the specified position range, transmits the identified welding position to a welding system; And A welding system that receives the welding position from the relay system and performs welding at the received welding position.

2. The welding control system according to claim 1, wherein When the identified welding position is not included within the specified position range, or when no welding position is identified from the captured image, the relay system issues an alarm.

3. The welding control system according to claim 1, wherein When the identified welding position is not included within the specified position range, or when no welding position is identified from the captured image, the relay system requests the machine vision system to identify the welding position. When the machine vision system is requested to identify the welding position, it identifies the welding position from the captured image, determines whether the identified welding position is included within the specified position range, and when the identified welding position is included within the specified position range, transmits the identified welding position to the welding system.

4. The welding control system according to claim 3, wherein When the identified welding position is not included within the specified position range, or when no welding position is identified from the captured image, the machine vision system issues an alarm.

5. The welding control system according to any one of claims 1 to 4, wherein After capturing at least a part of the welding object, the machine vision system attempts to identify the welding position from the captured image, and when no welding position is identified from the captured image, or when the identified welding position is not included within the specified position range, it transmits the captured image to the relay system.

6. The welding control system according to claim 5, wherein When the identified welding position is included within the specified position range, the machine vision system transmits the identified welding position to the welding system.

7. The welding control system according to any one of claims 1 to 4, wherein The relay system extracts a region of interest from the captured image based on a pre-trained second artificial intelligence model and transmits the information of the extracted region of interest to the machine vision system. The machine vision system identifies the welding position from the region of interest of the captured image and transmits the identified welding position to the welding system.

8. The welding control system according to any one of claims 1 to 4, further comprising: The inspection system, after performing the welding, checks for welding defects. When there are such welding defects, it transmits the welding defect information to the relay system. The relay system updates the first artificial intelligence model based on the welding defect information.

9. The welding control system according to any one of claims 1 to 4, wherein The relay system includes: An artificial intelligence server that contains the first artificial intelligence model; and A relay server that relays the communication between the machine vision system, the welding system, and the artificial intelligence server.

10. The welding control system according to any one of claims 1 to 4, wherein The welding object includes a secondary battery, The welding position includes the edge of the electrode of the secondary battery.

11. A welding control method, which is an artificial intelligence-based welding control method, comprising the following steps: The machine vision system captures at least a part of the welding object through a camera and transmits the captured image, i.e., the captured image, to the relay system; The relay system receives the captured image from the machine vision system, identifies the welding position of the welding object from the received captured image based on a pre-trained first artificial intelligence model, determines whether the identified welding position is included within a specified position range, and when the identified welding position is included within the specified position range, transmits the identified welding position to the welding system; and The welding system receives the welding position from the relay system and performs welding at the received welding position.

12. The welding control method according to claim 11, further comprising the following steps: When the identified welding position is not included within the specified position range, or when no welding position is identified from the captured image, the relay system issues an alarm.

13. The welding control method according to claim 11, further comprising the following steps: When the identified welding position is not included within the specified position range, or when no welding position is identified from the captured image, the relay system requests the machine vision system to identify the welding position; And When the machine vision system is requested to identify the welding position, it identifies the welding position from the captured image, determines whether the identified welding position is included within the specified position range, and when the identified welding position is included within the specified position range, transmits the identified welding position to the welding system.

14. The welding control method according to claim 13, further comprising the following steps: When the identified welding position is not included within the specified position range, or when no welding position is identified from the captured image, the machine vision system issues an alarm.

15. The welding control method according to any one of claims 11 to 14, further comprising the following steps: After the machine vision system captures at least a part of the welding object, it attempts to identify the welding position from the captured image. When the welding position is not recognized from the captured image, or the recognized welding position is not included in the specified position range, perform the step of transmitting the captured image to the relay system.

16. The welding control method according to claim 15, further comprising the following steps: When the recognized welding position is included in the specified position range, the machine vision system transmits the recognized welding position to the welding system.

17. The welding control method according to any one of claims 11 to 14, further comprising the following steps: The relay system extracts the region of interest from the captured image based on a pre-trained second artificial intelligence model, and transmits the information of the extracted region of interest to the machine vision system; and The machine vision system recognizes the welding position from the region of interest of the captured image, and transmits the recognized welding position to the welding system.

18. The welding control method according to any one of claims 11 to 14, further comprising the following steps: After performing the welding, check whether there are welding defects through an inspection system; When there are welding defects, transmit the welding defect information to the relay system; and The relay system updates the first artificial intelligence model based on the welding defect information.

19. A welding control device, which is an artificial intelligence-based welding control device, comprising: A camera; A memory containing a pre-trained artificial intelligence model; A welding module; and A processor that controls the camera to capture at least a part of the welding object, recognizes the welding position from the image captured by the camera, i.e., the captured image, based on the artificial intelligence model, determines whether the recognized welding position is included in the specified position range, and when the recognized welding position is included in the specified position range, controls the welding module to weld the recognized welding position.

20. The welding control device according to claim 19, further comprising an inspection module, After performing the welding, the processor checks whether there are welding defects through the inspection module, and when there are welding defects, updates the artificial intelligence model based on the welding defect information.

Images