JP2021137848A - Bead appearance inspection device and bead appearance inspection system - Google Patents

Abstract

To increase efficiency of a setting operation by a user in bead appearance inspection of a workpiece manufactured by final welding.SOLUTION: A bead appearance inspection device includes: an acquisition part for acquiring input data concerning a weld bead; a first determination part for storing a first determination reference used in inspection of defects of the weld bead, and executing first inspection determination for the weld bead; k pieces (k: integer of 1 or larger) of second determination parts for storing a second determination reference used in inspection of defects of the weld bead, and executing second inspection determination for the weld bead; and an overall determination part for determining whether or not the weld bead has defects. The first determination part outputs a first determination result whether or not a first inspection result satisfies a first determination reference to the overall determination result, k pieces of the second determination parts output a second determination result whether or not a second inspection result satisfies a second determination reference to the overall determination part, and the overall determination part creates an appearance inspection result of the weld bead based on the first determination result and the second determination result, and outputs the appearance inspection result.SELECTED DRAWING: Figure 5

Description

The present disclosure relates to a bead visual inspection device and a bead visual inspection system.

In Patent Document 1, slit light is projected onto the weld bead, shape lines sequentially formed on the weld bead are imaged by scanning the slit light, and the weld bead is based on the imaging data of the sequentially formed shape lines. A shape inspection device that acquires the three-dimensional shape of the above as point cloud data is disclosed. This shape inspection device corresponds to the cutting line by setting an arbitrary cutting line different from the shape line formed by scanning the slit light on the welding bead displayed based on the point cloud data according to the input. The cross-sectional shape of the weld bead at the cutting line is calculated from the point cloud data. Further, the shape inspection device compares the various feature amounts calculated according to the calculated cross-sectional shape with the permissible range of the various feature amounts registered in advance, and determines the quality of the feature amount.

Japanese Unexamined Patent Publication No. 2012-37487

The present disclosure provides a bead appearance inspection device and a bead appearance inspection system that make the setting operation by the user more efficient with respect to the bead appearance inspection of the workpiece produced by the present welding.

The present disclosure stores an acquisition unit for acquiring input data regarding a welding bead and a first determination criterion set by a user for inspecting a defect of the welding bead, and executes a first inspection determination for the welding bead. 1 Judgment unit and k (k (an integer of k: 1 or more) pieces that store the second judgment criterion used for the inspection of the defect of the welding bead set by the user and execute the second inspection judgment for the welding bead. A second determination unit and a comprehensive determination unit for determining whether or not the welding bead has the defect are provided, and the first determination unit is based on the first inspection result obtained by the first inspection determination. The first determination result of whether or not the first determination criterion is satisfied is output to the comprehensive determination unit, and in the k second determination units, the second inspection result obtained by the second inspection determination is the first. The second judgment result of whether or not the two judgment criteria are satisfied is output to the comprehensive judgment unit, and the comprehensive judgment unit outputs the appearance inspection result of the weld bead based on the first judgment result and the second judgment result. Provides a bead visual inspection device that creates and outputs a bead visual inspection device.

Further, the present disclosure stores an acquisition unit for acquiring input data regarding the welding bead and a first determination criterion set by the user for inspecting the defect of the welding bead, and executes the first inspection determination for the welding bead. The first determination unit to be used and the second determination criterion used for the inspection of the defect of the welding bead set by the user are stored, and the second inspection determination for the welding bead is executed k (k: an integer of 1 or more). ) A second determination unit and a comprehensive determination unit for determining whether or not the welding bead has the defect, and the first determination unit is a first inspection acquired by the first inspection determination. The first judgment result of whether or not the result satisfies the first judgment criterion is output to the comprehensive judgment unit, and the k second judgment units are obtained by the second inspection result obtained by the second inspection judgment. The second determination result of whether or not the second determination criterion is satisfied is output to the comprehensive determination unit, and the comprehensive determination unit outputs the appearance of the weld bead based on the first determination result and the second determination result. A bead appearance inspection system that creates and outputs inspection results is provided.

According to the present disclosure, it is possible to make the setting operation by the user more efficient with respect to the bead appearance inspection of the work produced by the main welding.

Schematic diagram showing a system configuration example of a welding system The figure which shows the internal structure example of the inspection control apparatus, the robot control apparatus and the higher-level apparatus which concerns on Embodiment 1. A sequence diagram showing a series of processing procedure examples including main welding, bead appearance inspection, and repair welding by the welding system according to the first embodiment. A table showing appropriate examples of the first inspection judgment and the second inspection judgment for each of a plurality of inspection items. A sequence diagram showing an example of processing procedures for the first inspection determination (point cloud comparison) and the second inspection determination to the Nth inspection determination (AI determination) according to the first embodiment. The figure which shows the internal structure example of the inspection control apparatus, the robot control apparatus and the higher-order apparatus which concerns on Embodiment 2. A sequence diagram showing a series of processing procedure examples including main welding, bead appearance inspection, and repair welding by the welding system according to the second embodiment. A sequence diagram showing an example of processing procedures for the first inspection determination (point cloud comparison) and the second inspection determination to the Nth inspection determination (AI determination) according to the second embodiment.

(Background to this disclosure)
As in Patent Document 1, when the calculated value of the feature amount (for example, bead width, bead height, etc.) related to the shape of the weld bead of the work produced by the main welding is within the permissible range, it is judged to be a good product. , A device configuration for automatically inspecting the appearance and shape of a weld bead has been conventionally known. However, in an actual welding site, an operator often visually inspects the appearance of the weld bead to determine whether or not the main welding of the work has been successful.

In the visual inspection of the weld bead, in addition to the feature amount related to the shape of the weld bead described above, there may be a wide variety of inspection items such as the position shift of the weld bead, the presence or absence of holes, and the presence or absence of welding defects such as spatter, and the user. Depending on the case, the criteria for judging whether or not a product is non-defective are often not uniform. For this reason, in the visual inspection of weld beads, not only the inspection items are different for each user, but also the quality of the finished workpiece is different for each user, and the inspection items can be adjusted arbitrarily. It is considered that there was room for improvement over the conventional technology in that the usability of the visual inspection was further required.

Therefore, in the following embodiment, an example of the bead appearance inspection device and the bead appearance inspection system that makes the setting operation by the user more efficient with respect to the bead appearance inspection of the work produced by the main welding will be described.

Hereinafter, embodiments in which the bead appearance inspection apparatus and the bead appearance inspection system according to the present disclosure are specifically disclosed will be described in detail with reference to the drawings as appropriate. However, more detailed explanation than necessary may be omitted. For example, detailed explanations of already well-known matters and duplicate explanations for substantially the same configuration may be omitted. This is to avoid unnecessary redundancy of the following description and to facilitate the understanding of those skilled in the art. It should be noted that the accompanying drawings and the following description are provided for those skilled in the art to fully understand the present disclosure, and are not intended to limit the subject matter described in the claims.

(Embodiment 1)
The bead appearance inspection device according to the first embodiment inputs input data regarding the weld bead of the work produced by the main welding, and uses the input data and the master data of the non-defective work to compare the input data with the master data. Based on this, the first inspection judgment regarding the shape of the weld bead is performed, and k (k: 1 or more integer) types of artificial intelligence are installed, and the weld bead is processed based on the processing of k types of artificial intelligence that targets the input data. The second inspection judgment regarding the welding defect of is performed. Weld defects in weld beads are, for example, perforations, pits, undercuts, spatters, and protrusions. Welding defects are not limited to those described above. The bead visual inspection device outputs the result of the visual inspection of the weld bead to the output device based on the judgment results of the first visual inspection and the k second inspection judgments.

Hereinafter, the object to be welded (for example, metal) is the "original work", the object produced (manufactured) by the main welding is the "work", and the welding defect detected by the visual inspection of the "work". An object whose location is repair-welded is defined as "repair work".

The process of producing a work by joining the original work and other original works by a welding robot is defined as "main welding", and the process of repairing defective parts of the work by a welding robot is defined as "repair welding". do.

The "work" or "repair work" is not limited to the work produced by one main welding, but may be a composite work produced by two or more main weldings.

(Welding system configuration)
FIG. 1 is a schematic view showing a system configuration example of the welding system 100. The welding system 100 includes a higher-level device 1 connected to each of the external storage ST, the input interface UI1 and the monitor MN1, a robot control device 2, an inspection control device 3, a sensor 4, a main welding robot MC1a, and repair welding. The configuration includes the robot MC1b. The welding robot MC1a and the repair welding robot MC1b may be configured as separate robots, or may be configured as the same welding robot MC1. In order to make the following description easier to understand, it will be described that the main welding and repair welding steps are executed by the welding robot MC1. Although FIG. 1 shows only one pair of one robot control device 2, the main welding robot MC1a, and the repair welding robot MC1b, a plurality of these pairs may be provided. Although the sensor 4 is shown as a separate body from the welding robot MC1 in FIG. 1, it may be provided integrally with the welding robot MC1 (see FIG. 2).

The host device 1 controls the start and completion of the main welding executed by the welding robot MC1 via the robot control device 2 in an integrated manner. For example, the host device 1 reads out welding-related information input or set in advance by a user (for example, a welding operator or a system administrator; the same applies hereinafter) from the external storage ST, and uses the welding-related information to obtain welding-related information. An execution command for the main welding including the contents is generated and sent to the corresponding robot control device 2. When the main welding by the welding robot MC1 is completed, the host device 1 receives a main welding completion report from the robot control device 2 to the effect that the main welding by the welding robot MC1 is completed, and indicates that the corresponding main welding is completed. Update to status and record in external storage ST. The above-mentioned execution command for the main welding is not limited to being generated by the host device 1, and for example, an operation panel (for example, PLC: Programmable Logical Controller) of equipment in a factory or the like where the main welding is performed, or a robot control device. It may be generated by the operation panel of 2 (for example, TP: Welding Pendant). The teach pendant (TP) is a device for operating the welding robot MC1 connected to the robot control device 2.

Further, the host device 1 controls the start and completion of the bead appearance inspection using the robot control device 2, the inspection control device 3, and the sensor 4 in an integrated manner. For example, when the host device 1 receives the main welding completion report from the robot control device 2, it generates an execution command for bead appearance inspection of the work produced by the welding robot MC1, and the robot control device 2 and the inspection control device 3, respectively. Send to. When the bead appearance inspection is completed, the host device 1 receives an appearance inspection report indicating that the bead appearance inspection is completed from the inspection control device 3, updates the status to the effect that the corresponding bead appearance inspection is completed, and externally. Record in storage ST.

Further, the host device 1 controls the start and completion of repair welding executed by the welding robot MC1 via the robot control device 2 in an integrated manner. For example, when the host device 1 receives the appearance inspection report from the inspection control device 3, it generates an execution command for repair welding of the work produced by the welding robot MC1 and sends it to the robot control device 2. When the repair welding is completed, the host device 1 receives a repair welding completion report indicating that the repair welding is completed from the robot control device 2, updates the status to the corresponding repair welding completion status, and external storage ST. Record in.

Here, the welding-related information is information indicating the content of the main welding executed by the welding robot MC1, and is created in advance for each main welding process and registered in the external storage ST. Weld-related information includes, for example, the number of original workpieces used for main welding, workpiece information including the ID, name and welding location of the original workpiece used for main welding, and the scheduled execution date when main welding is executed. Includes the number of workpieces to be welded and various welding conditions at the time of main welding. The welding-related information does not have to be limited to the data of the above-mentioned items. The robot control device 2 causes the welding robot MC1 to start the execution of the main welding using the original work specified by the execution command based on the execution command of the main welding sent from the higher-level device 1. The welding-related information described above is not limited to being managed by the host device 1 with reference to the external storage ST, and may be managed by, for example, the robot control device 2. In this case, since the robot control device 2 can grasp the state in which the main welding is completed, the actual execution date may be managed instead of the scheduled execution date in which the welding process is scheduled to be executed in the welding-related information. In the present specification, the type of the main welding is not limited, but in order to make the explanation easy to understand, a process of joining a plurality of original works to produce one work will be described as an example.

The host device 1 is connected so that data can be input / output from each of the monitor MN1, the input interface UI1 and the external storage ST, and further, data can be communicated with the robot control device 2. It is connected so that it becomes. The host device 1 may be a terminal device P1 that integrally includes a monitor MN1 and an input interface UI1, and may further include an external storage ST integrally. In this case, the terminal device P1 is a PC (Personal Computer) used by the user prior to the execution of the main welding. The terminal device P1 is not limited to the PC described above, and may be a computer device having a communication function such as a smartphone or a tablet terminal.

The monitor MN1 may be configured by using a display device such as an LCD (Liquid Crystal Display) or an organic EL (Electroluminescence). The monitor MN1 may display, for example, a screen output from the host device 1 indicating that the main welding is completed, that the bead appearance inspection is completed, or that the repair welding is completed. Further, a speaker (not shown) may be connected to the host device 1 instead of the monitor MN1 or together with the monitor MN1, and the host device 1 notifies that the main welding is completed and that the bead appearance inspection is completed. A notification or a voice indicating that the repair welding has been completed may be output via the speaker.

The input interface UI 1 is a user interface that detects a user's input operation and outputs it to the host device 1, and may be configured by using, for example, a mouse, a keyboard, a touch panel, or the like. The input interface UI 1 accepts, for example, an input operation when the user creates welding-related information, or receives an input operation when sending a main welding execution command to the robot control device 2.

The external storage ST is configured by using, for example, a hard disk drive (Hard Disk Drive) or a solid state drive (Solid State Drive). The external storage ST is, for example, data of welding-related information created for each main welding, the status (production status) of the work produced by the main welding or the repair work repaired by the repair welding, and the work information of the work or the repair work. Remember (see above).

The robot control device 2 as an example of the bead appearance inspection device is connected to the higher-level device 1 so as to be able to communicate data, and is also connected to the welding robot MC1 so as to be able to communicate data. When the robot control device 2 receives the execution command of the main welding sent from the higher-level device 1, the robot control device 2 controls the corresponding welding robot MC1 based on the execution command to execute the main welding. When the robot control device 2 detects the completion of the main welding, it generates a main welding completion report indicating that the main welding is completed and notifies the higher-level device 1. As a result, the host device 1 can properly detect the completion of the main welding by the robot control device 2. The method of detecting the completion of the main welding by the robot control device 2 may be, for example, a method of determining based on a signal indicating the completion of the main welding from a sensor (not shown) provided in the wire feeding device 300, or a known method. The method may be used, and the content of the method for detecting the completion of the main welding is not limited.

Further, when the robot control device 2 receives the execution command of the bead visual inspection sent from the higher-level device 1, the robot control device 2 is welded to which the sensor 4 is attached according to the visual inspection program created or prepared in advance by the robot control device 2. The robot MC1 (see FIG. 2) is controlled, and the bead appearance inspection of the corresponding work is executed based on the execution command. The visual inspection report indicating that the bead visual inspection has been completed is sent from the inspection control device 3 to the higher-level device 1, but the robot control device 2 itself or the robot control device 2 that has received an instruction from the inspection control device 3 is higher-level. It may be sent to the device 1. As a result, the host device 1 can appropriately detect the completion of the bead appearance inspection.

Further, when the robot control device 2 receives the repair welding execution command sent from the host device 1, the robot control device 2 controls the corresponding welding robot MC1 based on the repair welding program created by the inspection control device 3. And repair welding is executed. When the robot control device 2 detects the completion of the repair welding, it generates a repair welding completion report indicating that the repair welding is completed and notifies the higher-level device 1. As a result, the host device 1 can properly detect the completion of repair welding based on the robot control device 2. The method of detecting the completion of repair welding by the robot control device 2 may be, for example, a method of determining based on a signal indicating the completion of repair welding from a sensor (not shown) provided in the wire feeding device 300, or a known method. The method may be used, and the content of the method for detecting the completion of repair welding does not have to be limited.

The welding robot MC1 is connected to the robot control device 2 so that data can be communicated with each other. The welding robot MC1 executes main welding or repair welding commanded by the host device 1 under the control of the corresponding robot control device 2. As described above, the welding robot MC1 may be composed of the main welding robot MC1a provided for the main welding and the repair welding robot MC1b provided for the repair welding. When the sensor 4 is integrally attached to the welding robot MC1, the welding robot MC1 drives the sensor 4 according to the appearance inspection program to execute the bead appearance inspection commanded by the host device 1. To support.

The inspection control device 3 as an example of the bead appearance inspection device is connected so that data can be communicated with each of the host device 1, the robot control device 2, and the sensor 4. When the inspection control device 3 receives the execution command of the bead appearance inspection sent from the host device 1, the inspection control device 3 inspects the bead appearance of the welded portion of the work produced by the welding robot MC1 (for example, the weld bead formed on the work is preliminarily inspected). (Inspection of whether or not the predetermined welding standard is satisfied) is executed together with the sensor 4. The details of the bead appearance inspection will be described later with reference to FIGS. 4 and 5, but for example, the inspection control device 3 is a sensor based on the welded portion information of the work included in the execution command of the bead appearance inspection. Using the input data related to the shape of the weld bead acquired in step 4 (for example, point cloud data capable of specifying the three-dimensional shape of the weld bead), based on comparison with the master data of the non-defective work preset for each work. Perform bead visual inspection. Hereinafter, such a bead appearance inspection is defined as a "first inspection determination". Further, the inspection control device 3 is equipped with k (k: 1 or more integer) types of artificial intelligence (AI), forms a neural network by the artificial intelligence, and uses the above-mentioned input data as the target AI. Perform a bead appearance inspection to determine the presence or absence of welding defects based on this. Hereinafter, such a bead appearance inspection is defined as a "second inspection determination". In the first embodiment, the inspection control device 3 can execute the first inspection determination and the second inspection determination described above. The inspection control device 3 makes a comprehensive judgment of the bead appearance inspection using the results of executing the first inspection judgment and the second inspection judgment, respectively, and the appearance including the comprehensive judgment result and the notification that the bead appearance inspection is completed. The inspection report is generated, sent to the host device 1, and output to the monitor MN2.

Further, when the inspection control device 3 determines that a welding defect has been detected in the bead appearance inspection of the work, the inspection control device 3 uses the appearance inspection result including the position information of the welding defect location (so-called detection point) to detect the welding defect. Create a repair welding program to make corrections such as repairing parts. The inspection control device 3 associates this repair welding program with the appearance inspection result and sends it to the robot control device 2.

The sensor 4 is connected so that data can be communicated with the inspection control device 3. When the sensor 4 is attached to the welding robot MC1 (see FIG. 2), the sensor 4 three-dimensionally mounts a mounting table on which the work Wk is mounted according to the drive of the manipulator 200 based on the control of the robot control device 2. Can be scanned and is operational. The sensor 4 is data that can identify the three-dimensional shape of the work placed on the mounting table (see FIG. 2) according to the drive of the manipulator 200 based on the control of the robot control device 2 (for example, the point cloud data OD1 described later). Is acquired and sent to the inspection control device 3.

The monitor MN2 as an example of the output device may be configured by using a display device such as an LCD or an organic EL. The monitor MN2 displays, for example, a notification output from the inspection control device 3 to the effect that the bead appearance inspection has been completed, or a screen showing the notification and the result of the bead appearance inspection (for example, the result of the above-mentioned comprehensive determination). Further, a speaker (not shown) may be connected to the inspection control device 3 instead of the monitor MN2 or together with the monitor MN2, and the inspection control device 3 notifies that the bead appearance inspection is completed, or the notification and the bead. An audio indicating the content of the result of the visual inspection (for example, the result of the above-mentioned comprehensive determination) may be output via the speaker.

FIG. 2 is a diagram showing an example of internal configurations of the inspection control device 3, the robot control device 2, and the host device 1 according to the first embodiment. For the sake of clarity, the monitors MN1 and MN2 and the input interface UI1 are not shown in FIG. The work Wk shown in FIG. 2 may be an original work placed before the main welding is performed, or may be a work subject to bead appearance inspection (that is, a work produced by the main welding). The work to be repair-welded may be used.

Under the control of the robot control device 2, the welding robot MC1 executes various processes such as main welding commanded by the host device 1, movement of the sensor 4 during bead appearance inspection, and repair welding. The welding robot MC1 performs, for example, arc welding in the main welding or repair welding process. However, the welding robot MC1 may perform welding other than arc welding (for example, laser welding, gas welding) and the like. In this case, although not shown, the laser head may be connected to the laser oscillator via an optical fiber instead of the welding torch 400. The welding robot MC1 includes at least a manipulator 200, a wire feeding device 300, a welding wire 301, and a welding torch 400.

The manipulator 200 includes articulated arms, and each arm is moved based on a control signal from the robot control unit 25 of the robot control device 2. Thereby, the manipulator 200 can change the positional relationship between the work Wk and the welding torch 400 (for example, the angle of the welding torch 400 with respect to the work Wk) by driving the arm.

The wire feeding device 300 controls the feeding speed of the welding wire 301 based on the control signal from the robot control device 2. The wire feeding device 300 may include a sensor (not shown) capable of detecting the remaining amount of the welding wire 301. The robot control device 2 can detect that the main welding or repair welding process has been completed based on the output of this sensor.

The welding wire 301 is held by the welding torch 400. When electric power is supplied to the welding torch 400 from the power supply device 500, an arc is generated between the tip of the welding wire 301 and the work Wk, and arc welding is performed. The configuration for supplying the shield gas to the welding torch 400 and the like are omitted from the illustration and description for convenience of explanation.

The host device 1 generates execution commands for various processes of main welding, bead appearance inspection, and repair welding by using welding-related information input or set in advance by the user, and sends them to the robot control device 2. As described above, when the sensor 4 is integrally attached to the welding robot MC1, the bead appearance inspection execution command is sent to both the robot control device 2 and the inspection control device 3. The host device 1 has a configuration including at least a communication unit 10, a processor 11, and a memory 12.

The communication unit 10 is connected so that data can be communicated with each of the robot control device 2 and the external storage ST. The communication unit 10 sends an execution command of various processes of main welding, bead appearance inspection, or repair welding generated by the processor 11 to the robot control device 2. The communication unit 10 receives the main welding completion report, the appearance inspection report, and the repair welding completion report sent from the robot control device 2 and outputs them to the processor 11. The execution command of the main welding or the repair welding may include, for example, a control signal for controlling each of the manipulator 200, the wire feeding device 300, and the power supply device 500 included in the welding robot MC1.

The processor 11 is configured by using, for example, a CPU (Central Processing Unit) or an FPGA (Field Programmable Gate Array), and performs various processes and controls in cooperation with the memory 12. Specifically, the processor 11 functionally realizes the cell control unit 13 by referring to the program held in the memory 12 and executing the program.

The memory 12 includes, for example, a RAM (Random Access Memory) as a work memory used when executing the processing of the processor 11, and a ROM (Read Only Memory) for storing a program defining the processing of the processor 11. Data generated or acquired by the processor 11 is temporarily stored in the RAM. A program that defines the processing of the processor 11 is written in the ROM. Further, the memory 12 stores welding-related information data read from the external storage ST, work or repair work status, and work information (see above) of work or repair work sent from the robot control device 2, respectively. Remember.

The cell control unit 13 generates an execution command for executing the main welding, the bead appearance inspection of the work, or the repair welding based on the welding-related information stored in the external storage ST. Further, the cell control unit 13 is for visual inspection regarding the drive of the welding robot MC1 at the time of bead appearance inspection of the work Wk (for example, the work) after the main welding, based on the welding-related information stored in the external storage ST. A program and an execution command of a visual inspection program including this visual inspection program are created. The appearance inspection program may be created in advance and stored in the external storage ST. In this case, the cell control unit 13 simply reads the appearance inspection program from the external storage ST and acquires it. The cell control unit 13 may generate different execution commands for each of the various processes of main welding or repair welding executed by the welding robot MC1. The execution command of the main welding or the repair welding generated by the cell control unit 13, or the execution command of the appearance inspection program including the appearance inspection program is transmitted to the corresponding robot control device 2 or the robot control via the communication unit 10. It is sent to each of the device 2 and the inspection control device 3.

The robot control device 2 is a welding robot MC1 (for example, a sensor 4, a manipulator 200, a wire feeding device 300, etc.) based on an execution command of main welding, bead appearance inspection, or repair welding sent from the host device 1. Controls the processing of the power supply device 500). The robot control device 2 includes at least a communication unit 20, a processor 21, and a memory 22.

The communication unit 20 is connected so that data can be communicated with the host device 1, the inspection control device 3, and the welding robot MC1. Although the illustration is simplified in FIG. 2, the robot control unit 25 and the manipulator 200, the robot control unit 25 and the wire feeding device 300, and the power supply control unit 26 and the power supply device 500 During that time, data is transmitted and received via the communication unit 20, respectively. The communication unit 20 receives an execution command for main welding, bead appearance inspection, or repair welding sent from the host device 1. The communication unit 20 sends the work information of the work produced by the main welding or the repair work produced by the correction by the repair welding to the host device 1.

Here, the work information includes not only the ID of the work or repair work, but also the ID of the original work used for the main welding, the name, the welding location, the welding conditions at the time of executing the main welding, and the welding at the time of executing the repair welding. At least the conditions are included. Further, the work information may include information (for example, coordinates) indicating the position of a detection point indicating a defective portion of the work. The welding conditions or repair welding conditions include, for example, the material and thickness of the original work, the material and wire diameter of the welding wire 301, the shield gas type, the flow rate of the shield gas, the set average value of the welding current, and the set average value of the welding voltage. The feeding speed and feeding amount of the welding wire 301, the number of weldings, the welding time, and the like. In addition to these, information indicating the type of main welding or repair welding (for example, TIG welding, MAG welding, pulse welding), the moving speed of the manipulator 200, and the moving time may be included.

The processor 21 is configured by using, for example, a CPU or an FPGA, and cooperates with the memory 22 to perform various processes and controls. Specifically, the processor 21 refers to the program stored in the memory 22 and executes the program to function the welding program creation unit 23, the calculation unit 24, the robot control unit 25, and the power supply control unit 26. Realize.

The memory 22 has, for example, a RAM as a work memory used when executing the processing of the processor 21, and a ROM for storing a program defining the processing of the processor 21. Data generated or acquired by the processor 21 is temporarily stored in the RAM. A program that defines the processing of the processor 21 is written in the ROM. Further, the memory 22 stores the data of the execution command of the main welding, the bead appearance inspection or the repair welding sent from the host device 1, and the data of the work information of the work produced by the main welding or the repair work produced by the repair welding. Remember each one. Further, the memory 22 stores the main welding program of the main welding executed by the welding robot MC1. This welding program is a program that defines specific procedures (processes) for main welding, such as joining a plurality of original workpieces using the welding conditions in main welding.

The main welding program creation unit 23 is based on the main welding execution command sent from the host device 1 via the communication unit 20, and the work information (for example, ID, name, etc.) of each of the plurality of original works included in the execution command. And the welded part of the original work) is used to generate the main weld program of the main weld executed by the welding robot MC1. This welding program includes a welding current, a welding voltage, an offset amount, a welding speed, and a welding torch 400 for controlling a power supply device 500, a manipulator 200, a wire feeding device 300, a welding torch 400, etc. during the execution of the main welding. Various parameters such as the posture of the device may be included. The welding program may be stored in the processor 21 or in the RAM in the memory 22.

The calculation unit 24 performs various calculations. For example, the calculation unit 24 is a welding robot MC1 (specifically, a manipulator 200, a wire feeding device 300, and a wire feeding device 300) controlled by a robot control unit 25 based on the welding program generated by the welding program creation unit 23. Calculation of parameters for controlling each of the power supply devices 500) and the like are performed.

The robot control unit 25 drives the welding robot MC1 (specifically, each of the manipulator 200, the wire feeding device 300, and the power supply device 500) based on the welding program generated by the welding program creation unit 23. To generate a control signal for. The robot control unit 25 sends this generated control signal to the welding robot MC1. Further, the robot control unit 25 welds during the bead appearance inspection so as to cover the operating range of the welding robot MC1 specified in this welding program based on the appearance inspection program sent from the host device 1. The manipulator 200 of the robot MC1 is driven. As a result, the sensor 4 (see FIG. 1) attached to the welding robot MC1 can move along with the operation of the welding robot MC1 and input data regarding the shape of the welding bead of the work Wk (for example, the three-dimensional shape of the welding bead). Specifiable point cloud data) can be acquired.

The power supply control unit 26 drives the power supply device 500 based on the main welding program generated by the main welding program creation unit 23 and the calculation result of the calculation unit 24.

The inspection control device 3 controls the bead appearance inspection process of the work or repair work produced by the main welding by the welding robot MC1 based on the visual inspection execution command sent from the higher-level device 1. The bead appearance inspection is, for example, an inspection of whether or not the weld bead formed on the work or repair work meets a predetermined welding standard (for example, a quality standard), and is composed of the above-mentioned first inspection judgment and second inspection judgment. Will be done. In order to simplify the following description, the inspection control device 3 is based on the input data regarding the shape of the weld bead acquired by the sensor 4 (for example, point cloud data capable of identifying the three-dimensional shape of the weld bead). Whether or not the welding bead formed on the Wk (for example, a work or repair work) satisfies a predetermined welding standard is determined by a comprehensive judgment based on the results of the first inspection judgment and the second inspection judgment described above. The inspection control device 3 has a configuration including at least a communication unit 30, a processor 31, a memory 32, and an inspection result storage unit 33.

The communication unit 30 is connected so that data can be communicated with the host device 1, the robot control device 2, and the sensor 4. Although the illustration is simplified in FIG. 2, data is transmitted and received between the shape detection control unit 35 and the sensor 4 via the communication unit 30, respectively. The communication unit 30 receives the bead appearance inspection execution command sent from the host device 1. The communication unit 30 sends the comprehensive determination result of the bead appearance inspection using the sensor 4 (for example, the presence / absence of welding defect of the weld bead in the work or repair work, the type and position of the welding defect) to the host device 1.

The processor 31 is configured by using, for example, a CPU or an FPGA, and cooperates with the memory 32 to perform various processes and controls. Specifically, the processor 31 refers to the program held in the memory 32 and executes the program to execute the determination threshold storage unit 34, the shape detection control unit 35, the data processing unit 36, and the inspection result determination unit 37. And the repair welding program creation unit 38 is functionally realized.

The memory 32 has, for example, a RAM as a work memory used when executing the processing of the processor 31, and a ROM for storing a program defining the processing of the processor 31. Data generated or acquired by the processor 31 is temporarily stored in the RAM. A program that defines the processing of the processor 31 is written in the ROM. Further, the memory 32 stores data of the execution command of the bead appearance inspection of the work sent from the host device 1, and data of the work information of the work generated by the main welding or the repair work generated by the repair welding. Further, the memory 32 stores the data of the repair welding program created by the repair welding program creation unit 38. The repair welding program uses the welding conditions in repair welding and the position information of the corresponding points (corresponding points) on the operation locus of the welding robot MC1 closest to the detection point (see above) to determine the welding failure of the welding bead. It is a program that defines the specific procedure (process) of repair welding to make corrections such as repairs. This program is created by the repair welding program creation unit 38, and is sent from the inspection control device 3 to the robot control device 2.

The inspection result storage unit 33 is configured by using, for example, a hard disk or a solid state drive. The inspection result storage unit 33 stores data indicating the result of the bead appearance inspection of the welded portion in the work Wk (for example, the work or repair work) as an example of the data generated or acquired by the processor 31. The data showing the result of the bead appearance inspection is, for example, the inspection result determination unit 37 (specifically, the first inspection determination unit 371, the second inspection determination unit 372, ..., Nth inspection included in the inspection result determination unit 37. It is generated by any of the determination units 37N).

The determination threshold storage unit 34 is composed of, for example, a cache memory provided in the processor 31, is preset by a user operation, and includes a welded portion and a first inspection determination unit 371, ..., The inspection result determination unit 37 included in the inspection result determination unit 37. Information of each threshold value (for example, each threshold value set for each type of welding defect) corresponding to each bead appearance inspection process of the N inspection determination unit 37N is stored. Each threshold value is, for example, an allowable range of displacement of the weld bead, each threshold value of the length, height, and width of the weld bead, and each threshold value of perforation, pit, undercut, and spatter. The determination threshold storage unit 34 has a permissible range (for example, a minimum permissible value, a maximum permissible value, etc.) that satisfies the minimum welding standard (quality) required by a customer or the like as each threshold value at the time of bead appearance inspection after repair welding. May be remembered. It should be noted that these threshold values determine whether or not the inspection results created by each of the first inspection determination unit 371, ..., And the Nth inspection determination unit 37N included in the inspection result determination unit 37 pass the bead appearance inspection. It is used in the judgment process. Further, the determination threshold value storage unit 34 may store the upper limit value of the number of bead appearance inspections for each welded portion. As a result, the inspection control device 3 determines that it is difficult or impossible to correct the defective part by the automatic repair welding by the welding robot MC1 when the upper limit of the predetermined number of times is exceeded when the defective part is corrected by the repair welding. , It is possible to suppress a decrease in the operating rate of the welding system 100.

In the shape detection control unit 35 as an example of the acquisition unit, the robot control device 2 looks at the bead appearance inspection based on the execution command of the bead appearance inspection of the welded portion of the work Wk (for example, the work) sent from the host device 1. While operating the welding robot MC1 to which the sensor 4 is attached based on the inspection program, input data regarding the shape of the welding bead sent from the sensor 4 (for example, point group data capable of specifying the three-dimensional shape of the welding bead). ) Is obtained. When the shape detection control unit 35 reaches a position where the sensor 4 can image the weld bead (in other words, can detect the three-dimensional shape of the welded portion) in response to the drive of the manipulator 200 by the robot control device 2 described above, for example. A laser beam is irradiated from the sensor 4 to acquire input data regarding the shape of the welding bead (for example, point cloud data capable of specifying the three-dimensional shape of the welding bead). When the shape detection control unit 35 receives the input data (see above) acquired by the sensor 4, the shape detection control unit 35 passes the input data to the data processing unit 36.

When the data processing unit 36 as an example of the conversion unit acquires the input data (see above) regarding the shape of the weld bead from the shape detection control unit 35, the data suitable for the first inspection determination by the inspection result determination unit 37. In addition to converting to a format, the data format is converted into a data format suitable for each of the second inspection determination, ..., And the Nth inspection determination in the inspection result determination unit 37. The data format conversion may include a correction process for removing unnecessary point cloud data (for example, noise) included in the input data (that is, point cloud data) as so-called preprocessing, and is used for the first inspection determination. The above-mentioned preprocessing may be omitted. The data processing unit 36 uses a data format suitable for the first inspection determination, and generates image data showing the three-dimensional shape of the weld bead by, for example, executing statistical processing on the input shape data. The data processing unit 36 may perform edge enhancement correction that emphasizes the peripheral portion of the weld bead in order to emphasize the position and shape of the weld bead as the data for the first inspection determination. The data processing unit 36 counts the number of times the bead appearance inspection is executed for each defective welding portion, and the welding inspection result is not good even if the number of bead appearance inspections exceeds the number of times stored in the memory 32 in advance. , It may be determined that it is difficult or impossible to correct the defective portion of welding by automatic repair welding. In this case, the inspection result determination unit 37 generates an alert screen including the position of the welding defect portion and the type of welding defect (for example, perforation, pit, undercut, spatter, protrusion), and displays the generated alert screen. , Is sent to the host device 1 via the communication unit 30. The alert screen sent to the host device 1 is displayed on the monitor MN1. The alert screen may be displayed on the monitor MN2.

The inspection result determination unit 37 can execute a total N (N: an integer of 2 or more) types of bead appearance inspection (for example, each of the first inspection determination and the second inspection determination described above). Specifically, the inspection result determination unit 37 includes a first inspection determination unit 371, a second inspection determination unit 372, ..., And an Nth inspection determination unit 37N. In order to simplify the explanation of FIG. 2 in an easy-to-understand manner, N = 2 will be described, but the same applies even if the integer is N = 3 or more.

The first inspection determination unit 371 reads the combat value for the first inspection result stored in the determination combat value storage unit 34, temporarily stores the combat value, and temporarily stores the first inspection determination (that is, welding acquired by the sensor 4). The bead appearance inspection based on the comparison between the input data regarding the shape of the bead and the master data of the non-defective work predetermined for each work is performed, and the shape reliability of the weld bead (for example, along the linear or curved welding line) is performed. Whether or not), bead chipping, and bead misalignment are inspected (see FIG. 4). FIG. 4 is a table showing appropriate examples of the first inspection determination and the second inspection determination for each of the plurality of inspection items. The first inspection determination unit 371 compares the data converted by the data processing unit 36 for the first inspection determination (for example, image data generated based on the point cloud data) with the master data of the non-defective work (so-called image). Processing). Therefore, as shown in FIG. 4, the first inspection determination unit 371 can inspect the shape reliability of the weld bead, the bead chipping, and the bead misalignment with high accuracy. The first inspection determination unit 371 calculates an inspection score indicating the inspection results of the shape reliability of the weld bead, the bead chipping, and the bead misalignment, and creates the calculated value of the inspection score as the first inspection result.

The second inspection determination unit 372 to the Nth inspection determination unit 37N reads out the fighting value for the second inspection determination stored in the determination fighting value storage unit 34, temporarily stores it, and makes a second inspection determination (that is, that is). Input data related to the shape of the weld bead acquired by the sensor 4 or input data after the input data is preprocessed by the data processing unit 36 by forming a neural network using k = (N-1) types of artificial intelligence. A bead appearance inspection for determining the presence or absence of welding defects based on AI is performed on the data), and the presence or absence of holes, pits, undercuts, spatters, and protrusions of the weld bead is inspected (see FIG. 4). The holes, pits, undercuts, spatters, and protrusions of the weld bead are listed by way of example, and the types of defects inspected by the Nth inspection determination unit 37N are not limited to these. When it is determined that each of the second inspection determination unit 372 to the Nth inspection determination unit 37N has detected a welding defect of the corresponding type, the position of the welding bead in which the welding defect is detected is specified. Each of the second inspection judgment unit 372 to the Nth inspection judgment unit 37N uses a learning model (AI) obtained in advance for each type of welding defect or for each group of welding defect types by learning processing. Determine if there is a welding defect. As a result, each of the second inspection determination unit 372 to the Nth inspection determination unit 37N can inspect, for example, the presence or absence of holes, pits, undercuts, spatters, and protrusions of the weld bead with high accuracy. In addition, each of the second inspection determination unit 372 to the Nth inspection determination unit 37N does not execute the inspection of the shape reliability of the weld bead, the bead chipping, and the bead misalignment executed by the first inspection determination unit 371. The second inspection judgment unit 372 to the Nth inspection judgment unit 37N calculates the inspection results (in other words, the inspection score indicating the probability of occurrence) of holes, pits, undercuts, spatters, and protrusions of the weld bead, and this inspection score. The calculated value of is created as the second inspection judgment result.

Here, any one of the first inspection determination unit 371, ..., the Nth inspection determination unit 37N, or the (N + 1) inspection determination unit (not shown) has a function as a comprehensive determination unit. The inspection judgment unit as a comprehensive judgment unit acquires inspection results from each of the first inspection judgment unit 371, ..., And the Nth inspection judgment unit 37N, and records the acquired inspection results and the judgment threshold storage unit 34. Based on the threshold value, it is determined whether or not each inspection result is passed (that is, whether or not there is a welding defect in the welding bead). The inspection judgment unit as a comprehensive judgment unit creates an appearance inspection report including all the inspection results of the first inspection judgment unit 371, ..., The Nth inspection judgment unit 37N, and information on whether or not each inspection result is acceptable. It is stored in the memory 32 and sent to the host device 1 via the communication unit 30. Here, when the (N + 1) inspection determination unit (not shown) realizes the function as the comprehensive determination unit, the (N + 1) inspection determination unit does not have to have artificial intelligence.

In addition, whether or not the inspection judgment unit as the comprehensive judgment unit is capable of repair welding by the welding robot MC1 based on the inspection result (inspection score) included in the above-mentioned first inspection result or the second inspection result (inspection score). In other words, it may be determined whether repair welding by the welding robot MC1 is preferable or repair welding by hand is preferable), and the determination result may be included in the above-mentioned visual inspection report and output.

Therefore, as shown in FIG. 4, the inspection result determination unit 37 executes the first inspection determination and the second inspection determination in combination so as to be suitable for each type of welding defect inspection. It is possible to comprehensively and accurately inspect the shape reliability of weld beads, bead chipping, bead misalignment, perforations, pits, undercuts, spatter, and the presence of protrusions. The above description exemplifies N = 2, but when N = 3, the second inspection determination unit 372 can detect, for example, the presence or absence of a hole in the welding bead or the presence or absence of a pit as a type of welding defect. The Nth inspection determination unit 37N (N = 3) can detect, for example, the presence or absence of undercut, spatter, and protrusions of the welding bead as a type of welding defect by different AIs. That is, in the second inspection judgment, it is arbitrary so that it can be detected by a different AI for each combination of types of welding defects that are inspection items (for example, a combination of (perforated and pit), (undercut, spatter, protrusion)). A plurality of AIs (learning models) may be prepared in.

Here, a process and a processing procedure of the bead appearance inspection executed in the inspection result determination unit 37 will be described with reference to FIG. FIG. 5 is a sequence diagram showing an example of processing procedures for the first inspection determination (point cloud comparison) and the second inspection determination to the Nth inspection determination (AI determination) according to the first embodiment. In the example shown in FIG. 5, the first inspection determination unit 371 has a function as a comprehensive determination unit. Further, in FIG. 5, N = 2 is set in order to make the explanation easy to understand.

The first inspection determination unit 371 determines whether the first inspection result is equal to or higher than the threshold value for the first inspection result (an example of the first determination criterion) (that is, whether the first inspection result passes in the bead appearance inspection). A threshold value for the first inspection result for determining (whether or not) is acquired from the determination threshold storage unit 34 (St31).

The second inspection judgment unit 372 determines whether the second inspection result is equal to or less than the threshold value for the second inspection result (an example of the second judgment criterion) (that is, whether the second inspection result passes in the bead appearance inspection). A threshold value for the second inspection result for determining (whether or not) is acquired from the determination threshold storage unit 34 (St32).

The first inspection determination unit 371 receives input data regarding the shape of the weld bead acquired from the sensor 4 (for example, point cloud data capable of specifying the three-dimensional shape of the weld bead) (St33). Further, the first inspection determination unit 371 acquires the converted input data converted for the second inspection determination executed by the second inspection determination unit 372 from the data processing unit 36 (St34), and the converted input data. The data is output to the second inspection determination unit 372 (St351).

Here, the second inspection determination unit 372 may acquire the converted input data converted for the second inspection determination executed by the second inspection determination unit 372 from the data processing unit 36.

The first inspection determination unit 371 executes image processing using the input data, and inspects the shape reliability of the weld bead, the bead chipping, and the bead misalignment (St361A). The first inspection determination unit 371 calculates the inspection score created for each inspection item (for example, shape reliability, bead chipping, bead misalignment), and creates the calculated value of this inspection score as the first inspection result (St371A). ). The first inspection determination unit 371 compares the created first inspection result with the threshold value for the first inspection result, and whether or not the first inspection result is equal to or less than the threshold value (that is, whether the bead appearance inspection is passed or not). Whether or not) is determined (St381A).

The second inspection determination unit 372 executes an appearance inspection (St361B) for measuring the appearance of the bead using the converted input data acquired from the first inspection determination unit 371, and inspects the inspection items (for example, the shape) included in the appearance inspection. The probability of occurrence of defects, which is the output value of the AI engine (for example, neural network), is calculated for each reliability, bead chipping, bead misalignment, etc., and the calculated value of this inspection score is created as the second inspection result (2nd inspection result). St371B). The second inspection judgment unit 372 compares the created second inspection result with the threshold value for the second inspection result, and whether or not the second inspection result is equal to or less than the threshold value (that is, whether the bead appearance inspection is passed or not). Whether or not) is determined (St381B). The second inspection determination unit 372 outputs the second inspection result and information on whether or not the second inspection result is below the threshold value (that is, information on whether or not the result is acceptable) to the first inspection determination unit 371. (St391).

The first inspection determination unit 371 comprehensively determines whether or not the first inspection result and the second inspection result are acceptable (St401), and includes each inspection result and information on whether or not each inspection result is acceptable. A visual inspection report is created and stored in the memory 32. Further, when the first inspection determination unit 371 determines that both the first inspection result and the second inspection result are acceptable, the first inspection determination unit 371 transmits an appearance inspection report to the higher-level device 1 via the communication unit 30 and first. If it is determined that either the inspection result or the second inspection result is unacceptable, the visual inspection report is sent to the repair welding program creation unit 38 and the higher-level device 1 via the communication unit 30.

In the example shown in FIG. 5, the inspection determination unit having a function as a comprehensive determination unit is not limited to the first inspection determination unit 371 or the second inspection determination unit 372. The inspection determination unit having a function as a comprehensive determination unit may be executed by the third inspection determination unit. In such a case, the third inspection determination unit executes only the function as the comprehensive determination unit. Specifically, the third inspection judgment unit has information on whether or not the first inspection judgment unit 371 has passed the first inspection result, and the second inspection judgment unit 372 has the second inspection result and whether or not it has passed. The information is input, the comprehensive judgment process is executed, and the appearance inspection report is created.

The processing procedures of the first inspection determination (point cloud comparison) and the second inspection determination to the Nth inspection determination (AI determination) shown in FIG. 4 are not limited to the above examples. For example, the processing of steps St33 to St351 (input data conversion and output processing) may be executed after the processing of step St361A (calculating the first inspection result). Further, FIG. 4 shows an example in which the first inspection determination unit 371 executes the comprehensive determination of the first inspection result and the second inspection result, but the execution of the comprehensive determination process in step St39 is performed by the second inspection determination unit 372. It may be executed (that is, it may be executed by any of the first inspection determination unit 371, ..., The Nth inspection determination unit 37N).

Therefore, as shown in FIG. 4, the inspection result determination unit 37 executes the first inspection determination and the second inspection determination in combination so as to be suitable for each type of welding defect inspection. It is possible to comprehensively and accurately inspect the shape reliability of weld beads, bead chipping, bead misalignment, perforations, pits, undercuts, spatter, and the presence of protrusions.

The above description exemplifies N = 2, but when N = 3, the second inspection determination unit 372 can detect, for example, the presence or absence of a hole in the welding bead or the presence or absence of a pit as a type of welding defect. The Nth inspection determination unit 37N (N = 3) can detect, for example, the presence or absence of undercut, spatter, and protrusions of the welding bead as a type of welding defect by different AIs. That is, in the second inspection judgment, it is arbitrary so that it can be detected by a different AI for each combination of types of welding defects that are inspection items (for example, a combination of (perforated and pit), (undercut, spatter, protrusion)). A plurality of AIs (learning models) may be prepared in.

The repair welding program creation unit 38 includes an appearance inspection report of the work Wk (for example, the work or repair work) by the inspection result determination unit 37 and work information (for example, information such as coordinates indicating the position of a welding defect detection point of the work or repair work). ) To create a repair welding program for the work Wk (for example, work or repair work) to be executed by the welding robot MC1. The repair welding program includes welding current, welding voltage, offset amount, welding speed, welding torch 400, etc. for controlling the power supply device 500, the manipulator 200, the wire feeder 300, the welding torch 400, etc. during the execution of repair welding. Various parameters such as the posture of the device may be included. The generated repair welding program may be stored in the processor 31 or in the RAM in the memory 32.

The sensor 4 is, for example, a three-dimensional shape sensor, which is attached to the tip of the welding robot MC1 and can acquire a plurality of point cloud data capable of specifying the shape of the welded portion on the work Wk (for example, the work). Based on the group data, point cloud data capable of specifying the three-dimensional shape of the welded portion is generated and sent to the inspection control device 3. When the sensor 4 is not attached to the tip of the welding robot MC1 and is arranged separately from the welding robot MC1, the work is based on the position information of the welding portion sent from the inspection control device 3. A laser light source (not shown) configured to be able to scan a welded portion on a Wk (for example, a workpiece or a repair work) and a laser that is arranged so that an imaging region including the periphery of the welded portion can be imaged and irradiated to the welded portion. It may be composed of a camera (not shown) that captures the reflection locus of the reflected laser beam (that is, the shape line of the welded portion) of the light. In this case, the sensor 4 sends the shape data of the welded portion (in other words, the image data of the weld bead) based on the laser beam imaged by the camera to the inspection control device 3. The camera described above includes at least a lens (not shown) and an image sensor (not shown). The image sensor is, for example, a solid-state image sensor such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semi-conductor), and converts an optical image formed on an imaging surface into an electric signal.

(Operation of welding system)
Next, a series of operation procedures of main welding, bead appearance inspection, and repair welding by the welding system 100 according to the first embodiment will be described with reference to FIG. FIG. 3 is a sequence diagram showing a series of processing procedure examples including main welding, bead appearance inspection, and repair welding by the welding system 100 according to the first embodiment. In the explanation of FIG. 3, each of the main welding using a plurality of original workpieces and the repair welding performed based on the failure of the bead appearance inspection of the workpiece (that is, the comprehensive judgment result indicating that there is a welding defect). The operation procedure performed between the host device 1, the robot control device 2, and the inspection control device 3 will be described as an example of the process.

In FIG. 3, the host device 1 acquires the work information (for example, ID, name, and welding location of the original work) of the original work to be the target of the main work, respectively (St1), and the main welding including the work information of the original work. Generate an execution command for. The host device 1 sends a command for executing the main welding including the work information of the original work to the robot control device 2 (St2). The robot control device 2 may execute the processes of steps St1 and St2, respectively, without going through the host device 1. In this case, the memory 22 of the robot control device 2 stores the same data as the data stored in the external storage ST, or the robot control device 2 is connected so that data can be acquired from the external storage ST. Is preferable.

When the robot control device 2 receives the execution command of the main welding sent from the higher-level device 1, the robot control device 2 uses the work information of each of the plurality of original works included in the execution command to execute the main welding by the welding robot MC1. The main welding program is created, and the welding robot MC1 is made to execute the main welding according to the main welding program (St3). When the robot control device 2 determines the completion of the main welding by the welding robot MC1 by various known methods, it generates a main welding completion notification to the effect that the main welding is completed and sends it to the higher-level device 1 (St4). Upon receiving the final welding completion notification, the host device 1 generates an execution command for the appearance inspection program including the work appearance inspection program and sends it to the robot control device 2 (St5), and also executes the work bead appearance inspection. A command is generated and sent to the inspection control device 3 (St6). The robot control device 2 executes a visual inspection program received from the host device 1 with the start of the bead visual inspection, and moves the sensor 4 attached to the welding robot MC1 along the welding line (St7). The sensor 4 acquires point cloud data capable of specifying the three-dimensional shape of the work while the welded portion of the work is scantably moved by the robot control device 2 (St7).

The inspection control device 3 uses the point cloud data that can identify the three-dimensional shape of the weld bead acquired by the sensor 4 as input data, and makes each of the above-mentioned first inspection determination and second inspection determination individually (parallel). Execute (St7). The inspection control device 3 makes a comprehensive determination of the bead appearance inspection of the weld bead of the work based on the results of the individual bead appearance inspections (that is, the first inspection determination and the second inspection determination) in step St7 (that is, the first inspection determination and the second inspection determination). St8).

When the inspection control device 3 determines that repair welding is necessary because the work has a welding defect as a result of the comprehensive determination in step St8 (St9), the inspection control device 3 acquires this welding program from the robot control device 2. A repair welding program is created by modifying a part of this welding program (St9). In addition, a part to be modified is a content indicating, for example, a place (range) where repair welding is performed. Further, although detailed illustration is omitted in FIG. 3, the inspection control device 3 requests the data of the present welding program from the robot control device 2 in step St9, and is sent from the robot control device 2 in response to this request. The data of the main welding program may be acquired, or the data of the main welding program sent from the robot control device 2 after step St3 may be acquired in advance. As a result, the inspection control device 3 can efficiently create the data of the repair welding program by partially modifying the data of the main welding program acquired from the robot control device 2. The inspection control device 3 generates an appearance inspection report including the result of the comprehensive determination in step St8 and the repair welding program and sends it to the robot control device 2 (St10). Further, the inspection control device 3 also sends the similarly generated visual inspection report to the host device 1 (St11).

Upon receiving the visual inspection report in step St11, the host device 1 generates a repair welding execution command for the work and sends it to the robot control device 2 (St12). When the robot control device 2 receives the repair welding execution command sent from the host device 1, the robot control device 2 receives the repair welding based on the repair welding program (received in step St10) for the workpiece specified by the execution command. The welding robot MC1 is made to perform repair welding according to the program (St13). When the robot control device 2 determines the completion of repair welding by the welding robot MC1 by various known methods, the work information of the repair work (for example, the ID of the repair work and each of the plurality of original works used for the main welding). Work information including the ID (for example, the ID and name of the original work, the welding location of the original work, the welding conditions at the time of each execution of the main welding and the repair welding)) is sent to the host device 1 (St14).

When the host device 1 receives the work information including the repair work ID sent from the robot control device 2, it sets a management ID suitable for the user corresponding to the repair work ID and corresponds to the management ID. The data indicating that the repair work welding is completed is saved in the external storage ST (St15).

(Embodiment 2)
In the first embodiment, both the first inspection determination and the second inspection determination are executed in the inspection control device 3. In the second embodiment, an example in which the first inspection determination and the second inspection determination are executed by different devices will be described. Hereinafter, it will be described that the first inspection determination is executed by the inspection control device 3 and the second inspection determination is executed by the higher-level device 1. However, the second inspection determination may be executed by a device other than the host device 1.

(Welding system configuration)
FIG. 6 is a diagram showing an example of internal configurations of the inspection control device 3A, the robot control device 2, and the host device 1A according to the second embodiment. In the description of FIG. 6, the same reference numerals are given to those having the same configuration as each part of FIG. 2, and the description is simplified or omitted, and different contents will be described. Further, the configuration of the welding system 100A according to the second embodiment is the same as that of the welding system 100 according to the first embodiment (see FIG. 1).

The welding system 100A as an example of the bead appearance inspection system includes a higher-level device 1A connected to each of the external storage ST, the input interface UI1 and the monitor MN1, a robot control device 2, an inspection control device 3A, a sensor 4, and the like. The configuration includes the main welding robot MC1a and the repair welding robot MC1b.

In the inspection control device 3A as an example of the bead appearance inspection device, the processor 31A includes a determination threshold storage unit 34, a shape detection control unit 35, a data processing unit 36, an inspection result determination unit 37A, and a repair welding program creation unit. It is a configuration including 38 and. The inspection result determination unit 37A has only the first inspection determination unit 371. Since the configuration of the first inspection determination unit 371 is the same as that of the first embodiment, the description thereof will be omitted.

In the host device 1A as an example of the bead appearance inspection device, the processor 11A has a configuration including a cell control unit 13 and a second inspection determination unit 142 to an Nth inspection determination unit 14N. The memory 12 in the host device 1A stores the fighting value for the second inspection result executed by each of the second inspection determination unit 142 to the Nth inspection determination unit 14N. The second inspection determination unit 142 to the Nth inspection determination unit 14N, like the second inspection determination unit 372 to the Nth inspection determination unit 37N, read out the fighting value for the second inspection result stored in the memory 12 and read the fighting value for the second inspection result. Temporarily memorized, formed a neural network by the second inspection judgment (that is, k = (N-1) type of artificial intelligence, and used the input data regarding the shape of the weld bead acquired by the sensor 4 as the target AI. A bead appearance inspection for determining the presence or absence of defective welding is performed based on the above, and the welding bead is inspected for holes, pits, undercuts, spatters, and protrusions (see FIG. 4).

(Operation of welding system)
Next, a series of processing procedures including main welding, bead appearance inspection, and repair welding by the welding system 100A according to the second embodiment will be described with reference to FIG. 7. FIG. 7 is a sequence diagram showing a series of processing procedure examples including main welding, bead appearance inspection, and repair welding by the welding system 100A according to the second embodiment. In the description of FIG. 7, the upper device 1A and the robot control device 2 are inspected for each process of the main welding using a plurality of original workpieces and the repair welding performed based on the failure of the bead appearance inspection of the workpieces. The operation procedure performed with the control device 3A will be illustrated and described. Further, in the description of FIG. 7, the same step number is assigned to the process that overlaps with the process of FIG. 3, and the description is simplified or omitted, and different contents will be described.

In FIG. 7, after step St6, the robot control device 2 executes the visual inspection program received from the host device 1A with the start of the bead visual inspection, and puts the sensor 4 attached to the welding robot MC1 on the welding line. Move along (St7A). The sensor 4 acquires point cloud data capable of specifying the three-dimensional shape of the work while the welded portion of the work is scantably moved by the robot control device 2 (St7A). The inspection control device 3A uses the point cloud data capable of identifying the three-dimensional shape of the weld bead acquired by the sensor 4 as input data, and executes the above-mentioned first inspection determination (St7A). Further, the inspection control device 3A generates the execution command of the second inspection determination described above by the processor 31A and sends it to the higher-level device 1A (St21).

When the host device 1A receives the execution command for the second inspection determination sent from the inspection control device 3A in step St21, the host device 1A makes the second inspection determination based on the execution command, the second inspection determination unit 142 to the Nth inspection determination unit. Execute at each of 14N (St22). Since the details of the second inspection determination executed in step St22 are the same as those described in the first embodiment, the description thereof will be omitted. The host device 1A generates a processing result of the second inspection determination (that is, detection of the presence or absence of welding defects for each inspection item by AI processing) and sends it to the inspection control device 3A (St23). The inspection control device 3A makes a comprehensive judgment of the bead appearance inspection of the work based on the results of the first inspection judgment by the inspection control device 3A in step St7A and the second inspection judgment by the higher-level device 1A in step St22. (St8A). Since the details of the comprehensive determination executed in step St8A are the same as those described in the first embodiment, the description thereof will be omitted. Since the processing after step St8A is the same as that in FIG. 3, the description thereof will be omitted.

As described above, the welding system 100A as an example of the bead appearance inspection system according to the second embodiment inputs input data (for example, point group data OD1) regarding the welding bead of the work produced by welding in the inspection control device 3A. The welding system 100A uses the input data and the master data MD1 of the non-defective work, and makes the first inspection determination of the welding bead by the inspection control device 3A based on the comparison between the input data and the master data MD1, and k (k: 1). The above integers) types of artificial intelligence are installed, and the second inspection judgment of the weld bead is performed based on the processing of k types of artificial intelligence that targets the input data. This is performed by each of the determination units 14N. The inspection control device 3A inspects the bead appearance of the weld bead based on the respective judgment results of the first inspection judgment unit 371 of the inspection control device 3A and the second inspection judgment unit 142 to the Nth inspection judgment unit 14N of the higher-level device 1A. Is generated by the inspection result determination unit 37 and output to an output device (for example, monitor MN2).

The processing procedure of the bead appearance inspection executed in the inspection result determination unit 37A in the inspection control device 3A and in the processor 11A in the higher-level device 1A will be described with reference to FIG. FIG. 8 is a sequence diagram showing an example of processing procedures for the first inspection determination (point cloud comparison) and the second inspection determination to the Nth inspection determination (AI determination) according to the second embodiment. In the description of FIG. 8, the processing procedure of the bead appearance inspection performed between the inspection control device 3A and the higher-level device 1A, in which the higher-level device 1A is equipped with k types of artificial intelligence, will be described. In the description of FIG. 8, the same step number is assigned to the process that overlaps with the process of FIG. 5, and the description is simplified or omitted, and different contents will be described.

The first inspection determination unit 371 receives input data regarding the shape of the weld bead acquired from the sensor 4 (for example, point cloud data capable of specifying the three-dimensional shape of the weld bead) (St33). Further, the first inspection determination unit 371 acquires the converted input data converted for the second inspection determination executed by the second inspection determination unit 142 from the data processing unit 36 (St34), and the converted input data. The data is sent to the second inspection determination unit 142 in the host device 1A (St352).

Here, the second inspection determination unit 142 receives the converted input data converted for the second inspection determination executed by the second inspection determination unit 142 from the data processing unit 36 in the inspection control device 3A. May be good.

The first inspection determination unit 371 executes image processing using the input data, and inspects the shape reliability of the weld bead, the bead chipping, and the bead misalignment (St361A). The first inspection determination unit 371 calculates the inspection score created for each inspection item (for example, shape reliability, bead chipping, bead misalignment), and creates the calculated value of this inspection score as the first inspection result (St371A). ). The first inspection determination unit 371 compares the created first inspection result with the threshold value for the first inspection result, and whether or not the first inspection result is equal to or less than the threshold value (that is, whether the bead appearance inspection is passed or not). Whether or not) is determined (St381A).

The second inspection determination unit 142 executes an appearance inspection (St362B) for measuring the appearance of the bead using the converted input data acquired from the first inspection determination unit 371, and inspects the inspection items (for example, the shape) included in the appearance inspection. The probability of occurrence of defects, which is the output value of the AI engine (for example, neural network), is calculated for each reliability, bead chipping, bead misalignment, etc., and the calculated value of this inspection score is created as the second inspection result (2nd inspection result). St372B). The second inspection determination unit 142 compares the created second inspection result with the threshold value for the second inspection result, and whether or not the second inspection result is equal to or less than the threshold value (that is, whether the bead appearance inspection is passed or not). Whether or not) is determined (St382B). The second inspection determination unit 142 makes the first inspection in the inspection control device 3A the second inspection result and the information on whether or not the second inspection result is equal to or less than the threshold value (that is, the information on whether or not the second inspection result is passed). It is sent to the determination unit 371 (St392).

The first inspection determination unit 371 comprehensively determines whether or not the first inspection result and the second inspection result are acceptable (St402), and includes each inspection result and information on whether or not each inspection result is acceptable. A visual inspection report is created and stored in the memory 32. Further, when the first inspection determination unit 371 determines that both the first inspection result and the second inspection result are acceptable, the first inspection determination unit 371 transmits an appearance inspection report to the higher-level device 1A via the communication unit 30 and first. If it is determined that either the inspection result or the second inspection result is unacceptable, the visual inspection report is sent to the repair welding program creation unit 38 and the host device 1A via the communication unit 30.

As described above, in the welding system 100 according to the first embodiment and the welding system 100A according to the second embodiment, the shape detection control unit 35 that acquires the input data regarding the welding bead and the defect of the welding bead set by the user are defective. The first inspection judgment unit 371 (an example of the first judgment unit) that stores the threshold value for the first inspection result (an example of the first judgment criterion) used for the inspection and executes the first inspection judgment for the weld bead, and the user. K (k (an integer of k: 1 or more) that stores the threshold value for the second inspection result (an example of the second judgment criterion) used for the inspection of the set defect of the weld bead and executes the second inspection judgment for the weld bead. 2nd inspection judgment unit 372 to Nth inspection judgment unit 37N, or 2nd inspection judgment unit 142 to Nth inspection judgment unit 14N (an example of the second judgment unit), and whether or not the welding bead has a defect is determined. The first inspection determination unit 371 is provided with a comprehensive determination unit for determining whether or not the first inspection result acquired by the first inspection determination satisfies the threshold value for the first inspection result (an example of the first determination criterion). The first judgment result of ) Is satisfied or not, and the second judgment result is output to the comprehensive judgment unit. The comprehensive determination unit creates and outputs the appearance inspection result of the weld bead based on the first determination result and the second determination result.

As a result, the inspection control devices 3 and 3A detect the presence or absence of defects in the weld bead based on the first inspection determination based on the comparison between the input data indicating the three-dimensional shape of the weld bead and the master data MD1 and the AI process. Since it can be executed in combination with the two inspection determinations, the appearance inspection of the weld bead of the work produced by the main welding can be performed more efficiently. Therefore, the inspection control devices 3 and 3A can enhance the convenience for the user of the visual inspection of the weld bead.

Further, the first inspection determination unit 371 according to the first embodiment executes the first inspection determination based on the comparison between the input data and the master data MD1 on the weld bead. Further, the second inspection determination unit 372 to the Nth inspection determination unit 37N according to the first embodiment is equipped with k types of artificial intelligence, and a second inspection based on processing by k types of artificial intelligence using input data. Perform the determination on the weld bead. As a result, the inspection control devices 3 and 3A according to the first embodiment are defective in the weld bead based on the first inspection determination based on the comparison between the input data indicating the three-dimensional shape of the weld bead and the master data MD1 and the AI process. Since it can be executed in combination with the second inspection determination for detecting the presence or absence of the weld bead, the appearance inspection of the weld bead of the work produced by the main welding can be performed more efficiently.

Further, the inspection control devices 3 and 3A further include a data processing unit 36 that converts the input data into a format of input data in which the second inspection determination by the k second inspection determination units can be executed. The k second inspection determination units execute the second inspection determination using the converted input data converted by the data processing unit 36. As a result, the inspection control devices 3 and 3A convert one input data into the data format required by each of the k second inspection determination units, thereby converting the second inspection determination unit 372 to the Nth inspection determination unit. The accuracy of the AI processing executed in each of the 37Ns can be improved, and the detection accuracy of the presence or absence of defective welding beads (for example, perforation, pit, undercut, spatter) can be improved. Therefore, the inspection control devices 3 and 3A cover the appearance inspection items of the weld bead detected with high accuracy by the first inspection judgment and the appearance inspection items of the weld bead detected with high accuracy by the second inspection judgment. Can be inspected.

The appearance inspection items of the weld bead to be the target of the first inspection determination are the shape of the weld bead, the chipping of the weld bead, and the misalignment of the weld bead. The appearance inspection items of the weld bead to be subject to the second inspection determination are holes, pits, undercuts, spatters, and protrusions of the weld bead. As a result, the inspection control devices 3 and 3A have the appearance inspection items of the weld bead (for example, the shape of the weld bead, the chipping of the weld bead, the misalignment of the weld bead) and the second inspection, which are detected with high accuracy by the first inspection judgment. It is possible to comprehensively inspect the appearance inspection items of the weld bead (for example, holes in the weld bead, pits, undercuts, spatters, protrusions) that are detected with high accuracy by the judgment.

Further, any one of the first inspection determination unit 371, the k second inspection determination units 372 to the N inspection determination unit 37N, or the second inspection determination unit 142 to the Nth inspection determination unit 14N may be used. It functions as a comprehensive judgment unit, and creates and outputs the appearance inspection results of weld beads. As a result, the inspection control devices 3 and 3A have the appearance inspection items of the weld bead (for example, the shape of the weld bead, the chipping of the weld bead, the misalignment of the weld bead) and the second inspection, which are detected with high accuracy by the first inspection judgment. Efficiently determine whether or not the inspection result is acceptable, including the appearance inspection items of the weld bead (for example, holes in the weld bead, pits, undercuts, spatter, protrusions) that are detected with high accuracy by the judgment. And the appearance inspection result of the weld bead can be created. The comprehensive judgment unit is one of the inspections of one of the first inspection judgment unit 371, the k second inspection judgment units 372 to the Nth inspection judgment unit 37N, or the second inspection judgment unit 142 to the Nth inspection judgment unit 14N. The function is not limited to the determination unit, and the function may be realized by the (N + 1) inspection determination unit (not shown).

Although various embodiments have been described above with reference to the drawings, it goes without saying that the present disclosure is not limited to such examples. It is clear that a person skilled in the art can come up with various modifications, modifications, substitutions, additions, deletions, and equality within the scope of the claims. It is understood that it naturally belongs to the technical scope of the present disclosure. Further, each component in the various embodiments described above may be arbitrarily combined as long as the gist of the invention is not deviated.

The present disclosure is useful as a bead appearance inspection device and a bead appearance inspection system that make the setting operation by the user more efficient with respect to the bead appearance inspection of the workpiece produced by the present welding.

1,1A Upper device 2 Robot control device 3,3A Inspection control device 4 Sensor 10, 20, 30 Communication unit 11, 11A, 21, 31, 31A Processor 12, 22, 32 Memory 13 Cell control unit 23 Welding program creation unit 24 Calculation unit 25 Robot control unit 26 Power supply control unit 33 Inspection result storage unit 34 Judgment threshold storage unit 35 Shape detection control unit 36 Data processing unit 37, 37A Inspection result judgment unit 371 First inspection judgment unit 142, 372 Second inspection judgment Part 14N, 37N Nth inspection judgment part 100, 100A Welding system 200 Manipulator 300 Wire feeder 301 Welding wire 400 Welding torch 500 Power supply device MC1 Welding robot MC1a Main welding robot MC1b Repair welding robot MN1, MN2 Monitor ST External storage

Claims (6)

An acquisition unit that acquires input data related to welding beads,
A first determination unit that stores a first determination criterion set by the user and is used for inspection of defects in the weld bead and executes a first inspection determination for the weld bead.
The k (k: 1 or more integer) second determination unit that stores the second determination criterion used for the inspection of the defect of the welding bead set by the user and executes the second inspection determination for the welding bead. When,
A comprehensive determination unit for determining whether or not the welding bead has the defect is provided.
The first determination unit outputs the first determination result of whether or not the first inspection result acquired by the first inspection determination satisfies the first determination criterion to the comprehensive determination unit.
The k second determination units output to the comprehensive determination unit the second determination result of whether or not the second inspection result acquired by the second inspection determination satisfies the second determination criterion.
The comprehensive determination unit creates and outputs the appearance inspection result of the weld bead based on the first determination result and the second determination result.
Bead visual inspection equipment. The first determination unit executes the first inspection determination based on the comparison between the input data and the master data on the weld bead.
The second determination unit is equipped with k types of artificial intelligence, and executes the second inspection determination based on the processing by the k types of artificial intelligence using the input data on the weld bead.
The bead visual inspection apparatus according to claim 1. A conversion unit for converting the input data into an executable input data format for the second inspection determination by the k second determination units is further provided.
The k second determination units execute the second inspection determination using the converted input data converted by the conversion unit.
The bead visual inspection apparatus according to claim 1. The appearance inspection items relating to the shape of the weld bead to be subject to the first inspection determination are the shape of the weld bead, the chipping of the weld bead, and the misalignment of the weld bead.
The appearance inspection items relating to the defect of the weld bead to be subject to the second inspection determination are perforations, pits, undercuts, spatters, and protrusions of the weld bead.
The bead visual inspection apparatus according to claim 1. Either the first determination unit or the k second determination unit creates and outputs the appearance inspection result of the weld bead as the comprehensive determination unit.
The bead visual inspection apparatus according to claim 1. An acquisition unit that acquires input data related to welding beads,
A first determination unit that stores a first determination criterion set by the user and is used for inspection of defects in the weld bead and executes a first inspection determination for the weld bead.
The k (k: 1 or more integer) second determination unit that stores the second determination criterion used for the inspection of the defect of the welding bead set by the user and executes the second inspection determination for the welding bead. When,
A comprehensive determination unit for determining whether or not the welding bead has the defect is provided.
The first determination unit outputs the first determination result of whether or not the first inspection result acquired by the first inspection determination satisfies the first determination criterion to the comprehensive determination unit.
The k second determination units output to the comprehensive determination unit the second determination result of whether or not the second inspection result acquired by the second inspection determination satisfies the second determination criterion.
The comprehensive determination unit creates and outputs the appearance inspection result of the weld bead based on the first determination result and the second determination result.
Bead visual inspection system.

Images