Control device and program

Abstract

The present invention appropriately controls an articulated robot for friction stir welding without capturing images of a weld line. Disclosed is a control device for an articulated robot that comprises a rotary tool that makes it possible to friction stir weld (FSW) two members to be welded. The control device comprises a first acquisition unit that acquires control instruction data for moving the rotating rotary tool along a target trajectory while the rotary tool is pressed into an intended welding site between the two members to be welded, a robot control unit that outputs command values based on the control instruction data to rotate the rotary tool and move the rotary tool along the target trajectory, and a second acquisition unit that acquires actual measurement data for the movement trajectory of the rotary tool when the rotary tool is actually working on the two members to be welded under control by the robot control unit. The robot control unit performs feedback control such that the difference between the actual measurement data and the target trajectory is 0.

Description

Control Device and Program

[0001] The present disclosure relates to a control device for controlling an articulated robot for friction stir welding (FSW: Friction Stir Welding).

[0002] Regarding a control device for controlling an articulated robot for friction stir welding, a technique is known in which the relationship between the rotating tool and the joint line (the mating surface of two members to be joined) is imaged during actual operation, and feedback control is performed according to the joint deviation derived from the imaging information (for example, Patent Document 1). A technique is known in which the position and orientation of the joining tool corresponding to the marker position information obtained by detecting the positions of a plurality of markers provided on the tip shaft portion of the manipulator are calculated, the calculated position and orientation of the joining tool are compared with the estimated position and estimated orientation output from the manipulator, a correction amount for correcting the position and orientation of the joining tool is obtained, and the manipulator is driven based on a command signal corrected using this correction amount to correct the position and orientation of the joining tool (for example, Patent Document 2).

[0003] Japanese Patent Application Laid-Open No. 2016-198800 Japanese Patent Application Laid-Open No. 2024-145832

[0004] However, in the above Patent Document 1, since imaging the joint line is a prerequisite, for example, it can only be applied to butt joints with a clear joint line, and there is a problem that it cannot be applied when there is no joint line such as an overlap joint. Further, in the above Patent Document 2, there is a problem that a correction map is required and the control becomes complicated.

[0005] Therefore, on one side, an object of the present disclosure is to appropriately control an articulated robot for friction stir welding without imaging the joint line.

[0006] In one aspect, a control device for an articulated robot equipped with a rotary tool that enables friction stir welding (FSW) of two members to be joined is provided, comprising: a first acquisition unit that acquires control training data for moving the rotating rotary tool along a target trajectory while pushing it into the planned joining location between the two members to be joined; a robot control unit that outputs command values ​​based on the control training data to rotate the rotary tool and move it along the target trajectory; and a second acquisition unit that acquires measured data of the movement trajectory of the rotary tool when it is actually working on the two members to be joined under the control of the robot control unit, wherein the robot control unit performs feedback control so that the difference between the measured data and the target trajectory becomes zero.

[0007] In one aspect, this disclosure makes it possible to appropriately control an articulated robot for friction stir welding without imaging the joint line, and also enables high-precision control without corrective control for the position and angle of the rotating tool.

[0008] This is a schematic diagram showing the overall system including the control device of this embodiment. This is a functional block diagram schematically showing an example of the control device's functions. This is an explanatory diagram of feedback control. This is an explanatory diagram schematically showing the relationship between the target trajectory and the rotating tool. This is a schematic explanatory diagram (1) of the mating surface between two members to be joined. This is a schematic explanatory diagram (2) of the mating surface between two members to be joined. This is a schematic flowchart showing one of the characteristics of the operation of the control device of this embodiment. This is an explanatory diagram of Figure 5, and is a conceptual diagram of time-series position control.

[0009] The following describes each embodiment in detail with reference to the attached drawings. Note that the dimensional ratios in the drawings are merely examples and are not exhaustive. Furthermore, shapes and other details in the drawings may be partially exaggerated for illustrative purposes. Also, for clarity, in some cases, only a portion of parts with the same attribute are assigned reference numerals in the drawings.

[0010] Figure 1 is a schematic diagram showing the overall system 1 including the control device 100 of this embodiment.

[0011] The control device 100 in this embodiment controls an articulated robot 9 equipped with a rotary tool 90 (see Figure 3) that enables friction stir welding (FSW). Friction stir welding is a technique for joining two members to be joined, and is mainly used for joining dissimilar materials. However, it may also be used for joining aluminum alloys. The combination of dissimilar materials is basically arbitrary, and may be, for example, aluminum and stainless steel, or aluminum and copper. The two members to be joined may be members that make up a vehicle body.

[0012] Figure 1 schematically illustrates various externally connected elements 130 in relation to the hardware configuration of the control device 100.

[0013] The various externally connected elements 130 include an articulated robot 9 and a motion capture device 24.

[0014] As described above, the articulated robot 9 is an articulated robot equipped with a rotary tool 90 (see Figure 3) for friction stir welding. Typically, the articulated robot 9 is fixed to the site. The articulated robot 9 may be in the form of a multi-joint robot. The articulated robot 9 has an end portion (hand portion 93) that holds the rotary tool 90 via a joining head 92 and incorporates an electric motor (not shown) for high-speed rotation of the rotary tool 90. The articulated robot 9 also incorporates electric motors (not shown) for each joint (one joint 99 is schematically shown in Figure 3) to realize movement at each joint.

[0015] The motion capture system 24 measures the position of markers 80 (see Figure 3) attached to the articulated robot 9 in three dimensions. Preferably, the markers 80 are attached to three or more locations on the articulated robot 9 so that the orientation (angle) of the rotation axis of the rotating tool 90 can be calculated (only two markers 80 are shown in Figure 3). The markers 80 may also be placed on non-rotating parts other than the rotating tool 90 (e.g., the joining head 92). In modified examples, other distance measuring devices such as laser trackers and depth cameras may be used instead of or in addition to the motion capture system 24. The position of the distance measuring device is determined by various data (training data).

[0016] The control device 100 is formed by a computer. The control device 100 includes a CPU (Central Processing Unit) 111, RAM (Random Access Memory) 112, ROM (Read Only Memory) 113, auxiliary storage device 114, drive device 115, and communication interface 117 connected by a bus 119, as well as a wired transceiver 125 and a wireless transceiver 126 connected to the communication interface 117.

[0017] The auxiliary storage device 114 is, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and is a storage device that stores data related to application software, etc.

[0018] The wired transceiver unit 125 includes a transceiver unit capable of communicating using a wired network 128 based on protocols such as CAN (Controller Area Network) or LIN (Local Interconnect Network). Various elements 130 are connected to the wired transceiver unit 125. However, some or all of the various elements 130 may be connected to the bus 119 or to the wireless transceiver unit 126.

[0019] The wireless transceiver 126 is a transceiver capable of communicating using a wireless network. The wireless network may include a mobile phone wireless communication network, the internet, VPN (Virtual Private Network), WAN (Wide Area Network), etc. The wireless transceiver 126 may also include a Near Field Communication (NFC) unit, a Bluetooth (registered trademark) communication unit, a Wi-Fi (Wireless Fidelity) transceiver unit, an infrared transceiver unit, etc.

[0020] The control device 100 may also be connectable to the recording medium 116. The recording medium 116 stores a predetermined program. The program stored on the recording medium 116 is installed on the auxiliary storage device 114 of the control device 100 via the drive device 115. The installed predetermined program becomes executable by the CPU 111 of the control device 100. For example, the recording medium 116 may be a recording medium that records information optically, electrically, or magnetically, such as a CD (Compact Disc)-ROM, flexible disk, magneto-optical disk, etc., or a semiconductor memory that records information electrically, such as ROM, flash memory, etc.

[0021] Figure 2 is a functional block diagram schematically showing an example of the functions of the control device 100. Figure 3 is an explanatory diagram schematically showing the relationship between the target trajectory and the rotating tool 90. Figures 4A and 4B are schematic explanatory diagrams of the mating surface 34 between the members to be joined 31 and 32.

[0022] Furthermore, some or all of the functions of the control device 100 described below may be implemented by other control devices (for example, a control device that may be built into the articulated robot 9, and / or an external server computer).

[0023] As shown in Figure 2, the control device 100 includes a target trajectory generation unit 150, a training data generation unit 152, a training data acquisition unit 154, a robot control unit 156, a motion capture data acquisition unit 158, a measured data acquisition unit 160, a position training data storage unit 170, and an axis orientation training data storage unit 172.

[0024] Furthermore, each function from the target trajectory generation unit 150 to the motion capture data acquisition unit 158 ​​can be realized by the CPU 111 shown in Figure 2 executing one or more programs in a storage device (for example, the ROM 113 or auxiliary storage device 114 shown in Figure 2). In addition, the position training data storage unit 170 and the axis orientation training data storage unit 172 can be realized by, for example, the ROM 113 or auxiliary storage device 114 shown in Figure 2.

[0025] The target trajectory generation unit 150 generates a target trajectory related to the movement trajectory of the rotating tool 90 of the articulated robot 9. Figure 3 conceptually shows the target trajectory 300 together with the rotating tool 90 of the articulated robot 9. The target trajectory 300 may be managed in a three-dimensional coordinate system. The target trajectory 300 can be derived based on the fixed position of the workpiece (fixed position at the site) and the shape data of the workpiece (design data) when the two members to be joined are considered as workpieces. The workpiece is fixed to the site by a jig or the like.

[0026] The training data generation unit 152 generates training data for control. The training data for control is control data for moving the rotating tool 90 along a target trajectory while pushing it into the planned joining locations between two members to be joined. Figures 4A and 4B schematically illustrate two members to be joined, 31 and 32. In Figure 4A, the two members to be joined, 31 and 32, are plate-shaped members and are overlapped in the thickness direction. The planned joining locations between the two members to be joined, 31 and 32, are the overlapping portions between the members to be joined, 31 and 32. The target trajectory is a trajectory that reflects the depth of the rotating tool 90 (the amount it is pushed into the workpiece), and is the position (depth) where it penetrates the member to be joined 32 in the thickness direction. In Figure 4B, the planned joining locations between the two members to be joined, 31 and 32, are the locations along the mating surface 34 (see Figure 4B) between the members to be joined, 31 and 32. Figure 4B shows a view of the two joining members 31 and 32 along the direction of insertion of the rotary tool 90. In this case as well, the target trajectory is the trajectory that reflects the depth of the rotary tool 90 (the amount it is inserted into the workpiece).

[0027] In this embodiment, the control training data associates command values ​​with each position on the target trajectory in a time series from a reference point. The command value is a command value for moving the rotation tool 90 to the next position, and may be a value indicating the target position of the rotation tool 90 (a position on the target trajectory that is the target position in the next processing cycle), or a value indicating the target movement vector to that target position, etc.

[0028] Furthermore, in this embodiment, the control training data may include training data regarding the orientation (angle) of the rotation axis of the rotation tool 90, along with training data regarding the position of the rotation tool 90. In this case, the training data regarding the angle of the rotation axis of the rotation tool 90 may be data to which the target angle of the rotation axis of the rotation tool 90 is associated with each position on the target trajectory. The target angle of the rotation axis of the rotation tool 90 at each position on the target trajectory may be generated in such a way that it is possible to join the joining surface at a right angle or a constant angle. Hereinafter, when distinguishing between them, the training data regarding the position of the rotation tool 90 will also be referred to as "position training data," and the training data regarding the angle of the rotation axis of the rotation tool 90 will also be referred to as "axis orientation training data."

[0029] In this embodiment, control training data is generated by moving the articulated robot 9 along a target trajectory in an unloaded state. In this case, the unloaded state may be achieved by removing the rotating tool 90, or by treating a soft material different from the actual workpiece as the workpiece. By using a soft material, an essentially unloaded state can be created. Alternatively, it may be achieved by moving the rotating tool 90 in air without using any soft material. In either case, the acquisition (generation) of control training data may be achieved by a motion capture data acquisition unit 158, described later, which acquires motion information (time-series data) when the rotating tool 90 is moved (i.e., motion capture 24 may be used). In the modified example, the control training data may be generated based on design information.

[0030] When the training data generation unit 152 generates control training data, it stores the generated control training data in the storage unit. Specifically, the training data generation unit 152 stores the position training data in the position training data storage unit 170 and the axis orientation training data in the axis orientation training data storage unit 172.

[0031] The target trajectory and control training data may be generated separately for each type of workpiece. Unless otherwise specified, the following explanation will focus on a specific type of workpiece.

[0032] The training data acquisition unit 154 acquires (reads) control training data from the position training data storage unit 170 and the axis orientation training data storage unit 172.

[0033] The robot control unit 156 controls the articulated robot 9 by outputting control command values ​​to the articulated robot 9. The command values ​​are derived based on the control training data, as described above. In this case, the robot control unit 156 may also derive command values ​​related to the amount of rotation and rotation speed around each joint of the articulated robot 9, as further lower-level command values ​​to realize the movement of the rotation tool 90 to the target position, based on the control training data.

[0034] In this embodiment, the robot control unit 156 basically drives the rotary tool 90 to rotate and moves it along a target trajectory by outputting command values ​​based on control training data. In other words, position sensors and angle sensors that may be built into the articulated robot (rotary tool) are not referenced during control.

[0035] However, in this embodiment, the robot control unit 156 may perform feedback control based on the relationship between the measured data and the target trajectory, which will be described later. This feedback control will be described later.

[0036] The motion capture data acquisition unit 158 ​​acquires movement information (time-series data) of the articulated robot 9 from the motion capture 24. The movement information may be time-series data of the position information of each marker 80, or time-series data of the position information of a single reference point that can be derived based on the position information of each marker 80. In this case, the reference point may be the center position of each marker 80, etc. The movement information of the articulated robot 9 to be acquired should be information that allows the actual measurement data acquisition unit 160 to accurately derive (acquire) the position of the rotating tool 90 by calculation.

[0037] The measured data acquisition unit 160 acquires measured data of the movement trajectory of the rotation tool 90 based on the movement information from the motion capture 24. That is, the measured data acquisition unit 160 acquires time-series position data (measured data of the movement trajectory) by calculating the position of the rotation tool 90 at each point in time based on the movement information from the motion capture 24. When the movement information from the motion capture 24 is used to acquire (generate) control training data, the same motion capture 24 is used. That is, the positional relationship between the articulated robot 9 and the motion capture 24 is the same when acquiring measured data and when acquiring control training data, and the motion capture 24 is basically always fixed. In this case, the relative position between the motion capture 24 and the maximum range of motion of the rotation tool 90 of the articulated robot 9 can be kept constant.

[0038] In actual operation, control based solely on training data may not be sufficient to control the position of the rotating tool 90 along the target trajectory due to bending and warping of various parts of the articulated robot 9 caused by reaction forces from the workpiece.

[0039] In this embodiment, the robot control unit 156 generates command values ​​based on the relationship between the measured data acquired by the measured data acquisition unit 160 and the target trajectory. Specifically, the robot control unit 156 generates command values ​​in a direction that eliminates the difference (positional deviation) between the measured data and the target trajectory. The robot control unit 156 also generates command values ​​in a direction that eliminates the difference (angle-related deviation) between the angle of the rotation axis of the rotating tool 90 based on the measured data and the target angle of the rotation axis of the rotating tool 90 at each position on the target trajectory. Figure 2A shows a block diagram of this feedback control. In the example shown in Figure 2A, a final command value is calculated such that the deviation becomes zero based on the deviation between the command value (target position and target angle) based on the target trajectory and the corresponding measured values ​​from the imaging device (e.g., motion capture 24) (hereinafter also referred to as "deviation reduction processing based on feedback control"). The final command value is then given to the articulated robot 9 via the control device (motor driver).

[0040] Thus, according to this embodiment, the articulated robot 9 for friction stir welding can be appropriately controlled without imaging the bonding line on the mating surface 34 between the members 31 and 32 to be joined with a camera. Therefore, according to this embodiment, it can be applied not only to butt joints but also to cases where there is no bonding line such as lap joints. In this embodiment, although the feedback control is performed by collating the relationship between the measured data acquired by the measured data acquisition unit 160 and the target trajectory in time series, the correction to the nearest position on the target trajectory (correction by the nearest neighbor search method) may be realized. Also, the command value may be corrected according to temperature conditions, the bending angle of the processing path, etc.

[0041] Next, referring to FIGS. 5 and 6, an operation example of the control device 100 of this embodiment will be described.

[0042] FIG. 5 is a schematic flowchart showing a feature of the operation of the control device 100 of this embodiment. FIG. 6 is an explanatory diagram of FIG. 5 and is a conceptual diagram of time series position control.

[0043] The processing routine shown in FIG. 5 may be repeatedly executed, for example, every predetermined processing cycle (for example, 10 ms).

[0044] In step S400, the control device 100 determines whether or not the control flag is off. The control flag is in an "off" state in the initial state and becomes on during the execution of control for one workpiece. If the determination result is "YES", the process proceeds to step S402, and otherwise, the process proceeds to step S404.

[0045] In step S402, when the control start condition is satisfied, the control device 100 starts the control and sets the current value of the counter to the initial value 0. The counter functions as a timer that measures the elapsed time from the start of control.

[0046] In step S404, the control device 100 increments the current value of the counter by 1.

[0047] In step S406, the control device 100 calculates a target position (a position on the target trajectory) corresponding to the current value of the counter based on the control teacher data. The target position corresponding to the current value of the counter corresponds to the position (a position on the target trajectory) of the rotary tool 90 that should be located when the time corresponding to the current value of the counter has elapsed. Similarly, the control device 100 calculates a target angle (the target angle of the rotation axis of the rotary tool 90) corresponding to the current value of the counter based on the control teacher data.

[0048] In step S408, the control device 100 acquires the position and angle (the angle of the rotation axis of the rotary tool 90) of the rotary tool 90 at the current time based on the measured data at the current time.

[0049] In step S410, the control device 100 calculates the difference between each value obtained in step S406 and each value obtained in step S408. That is, it calculates the difference between the target position and the position based on the measured data at the current time (hereinafter, also referred to as "position deviation" for the sake of distinction) and the difference between the target angle and the angle based on the measured data at the current time (hereinafter, also referred to as "angle deviation" for the sake of distinction).

[0050] In step S412, the control device 100 determines whether each value (position deviation and angle deviation) obtained in step S410 is within the allowable range. Whether it is within the allowable range may be determined by setting a predetermined threshold value. In this case, the predetermined threshold value may be adapted according to requirements such as the required joining accuracy. The predetermined threshold value may be set separately for each of the position deviation and the angle deviation. When both the position deviation and the angle deviation are within the allowable range, the process proceeds to step S414, and in other cases (that is, when at least one of the position deviation and the angle deviation is not within the allowable range), the process proceeds to step S416.

[0051] In addition, in a modified example, the derivation of the angle deviation may be omitted. In this case, similar processing may be executed using only the position deviation.

[0052] In step S414, the control device 100 calculates a command value corresponding to the current value of the counter based on the control training data, and outputs the calculated command value to the articulated robot 9. In this case, the command value is used as is.

[0053] In step S416, the control device 100 calculates a command value corresponding to the current value of the counter based on the control training data, and after performing a deviation reduction process on the calculated command value, outputs it to the articulated robot 9. At this time, as described above, the control device 100 performs a deviation reduction process on the command value based on feedback control so that the position deviation and / or angle deviation are eliminated. For example, the control device 100 may derive the final command value (command value after deviation reduction processing) by adding or subtracting a value obtained by multiplying the position deviation and / or angle deviation by a coefficient to the command value.

[0054] In step S418, the control device 100 determines whether the control termination condition has been met. The control termination condition may be met, for example, when the current value of the counter reaches the upper limit. If the determination result is "YES", the process terminates; otherwise, the next processing cycle starts from step S400.

[0055] In this way, according to the process shown in Figure 5, the deviation (position deviation and / or angular deviation) is calculated at each elapsed time from the start of control (for example, every 10 ms corresponding to the processing cycle), and a command value is generated. In the example shown in Figure 6, a schematic example of a deviation reduction method based on feedback control at four time points is shown. In this example, the deviation is relatively large at 10 ms from the start of control, but through the deviation reduction process based on feedback control, the deviation becomes small by 40 ms from the start of control, and the control is completed. Note that in the example shown in Figure 6, the deviation is shown as each position in a two-dimensional coordinate system of XY, but in reality, the deviation may be calculated in three-dimensional position as described above.

[0056] Although each embodiment has been described in detail above, the invention is not limited to any particular embodiment, and various modifications and changes are possible within the scope described in the claims. Furthermore, it is possible to combine all or more of the components of the embodiments described above.

[0057] The following additional information is disclosed regarding each of the above embodiments.

[0058] [Note 1] A control device for an articulated robot equipped with a rotary tool that enables friction stir welding (FSW) of two members to be joined, comprising: a first acquisition unit that acquires control training data for moving the rotating rotary tool along a target trajectory while pushing it into the planned joining location between the two members to be joined; a robot control unit that outputs command values ​​based on the control training data to rotate the rotary tool and move it along the target trajectory; and a second acquisition unit that acquires measured data of the movement trajectory of the rotary tool when it is actually working on the two members to be joined under the control of the robot control unit, wherein the robot control unit performs feedback control so that the difference between the measured data and the target trajectory becomes zero.

[0059] [Note 2] The control device according to Note 1, further comprising a third acquisition unit that acquires motion information of the articulated robot via a distance measuring device such as motion capture, wherein the second acquisition unit acquires the measured data based on the measured positions of at least three locations of the articulated robot based on the motion information.

[0060] [Note 3] The control device according to Note 1 or 2, wherein the control training data has the command value associated with each position on the target trajectory in a time series from a reference point, and the second acquisition unit acquires the measured data in the same time series from the same reference point as the control training data.

[0061] [Appendix 4] The control device according to any one of Appendix 1 to 3, wherein the measured data further includes measured data of the angle of the rotation axis of the rotating tool, and further comprises a fourth acquisition unit that acquires posture-maintaining data to which the target angle of the rotation axis of the rotating tool is associated with each position on the target trajectory, and based on the control training data, the command value includes a position command value and an angle command value, and the robot control unit further performs feedback control so that the difference between the position command value and the measured position and the difference between the angle command value and the measured angle become zero.

[0062] [Note 5] The control device according to any one of Notes 1 to 4, wherein the control training data is generated by moving the articulated robot in an unloaded state in such a manner that the rotating tool moves along the target trajectory. [Note 6] The control device according to Note 4 or 5, wherein the first acquisition unit, the second acquisition unit, the third acquisition unit, and the fourth acquisition unit each acquire corresponding data or information via an optical distance measuring device, and the relative position between the optical distance measuring device and the maximum movable range of the rotating tool of the articulated robot remains unchanged during the generation of the control training data and during actual operation.

[0063] [Note 7] A control program for an articulated robot equipped with a rotary tool that enables friction stir welding (FSW) of two members to be joined, the program causing a computer to execute: a first acquisition process for acquiring control training data for moving the rotating rotary tool along a target trajectory while pushing it into the planned joining location between the two members to be joined; a robot control process for moving the rotary tool along the target trajectory while rotating it by outputting command values ​​based on the control training data; and a second acquisition process for acquiring measured data of the movement trajectory of the rotary tool when it is actually working on the two members to be joined under the control of the robot control process, wherein the robot control process includes performing feedback control so that the difference between the measured data and the target trajectory becomes zero.

[0064] 31, 32 Joint target members, 9 Joint robot, 90 Rotation tool, 100 Control device, 154 Training data acquisition unit (1st acquisition unit, 4th acquisition unit), 156 Robot control unit, 158 Motion capture data acquisition unit (3rd acquisition unit), 160 Actual measurement data acquisition unit (2nd acquisition unit)

Claims

1. A control device for an articulated robot equipped with a rotary tool that enables friction stir welding (FSW) of two members to be joined, comprising: a first acquisition unit that acquires control training data for moving the rotating rotary tool along a target trajectory while pushing it into the planned joining location between the two members to be joined; a robot control unit that outputs command values ​​based on the control training data to rotate the rotary tool and move it along the target trajectory; and a second acquisition unit that acquires measured data of the movement trajectory of the rotary tool when it is actually working on the two members to be joined under the control of the robot control unit, wherein the robot control unit performs feedback control so that the difference between the measured data and the target trajectory becomes zero.

2. The control device according to claim 1, further comprising a third acquisition unit that acquires motion information of the articulated robot via a distance measuring device, wherein the second acquisition unit acquires the measured data based on the measured positions of at least three locations of the articulated robot based on the motion information.

3. The control device according to claim 2, wherein the measured data further includes measured data of the angle of the rotation axis of the rotating tool, and further comprises a fourth acquisition unit that acquires attitude-maintaining data to which the target angle of the rotation axis of the rotating tool is associated with each position on the target trajectory, and based on the control training data, the command value includes a position command value and an angle command value, and the robot control unit further performs feedback control so that the difference between the position command value and the measured position and the difference between the angle command value and the measured angle become zero.

4. The control device according to any one of claims 1 to 3, wherein the control training data is generated by moving the articulated robot in an unloaded state in such a manner that the rotating tool moves along the target trajectory.

5. The control device according to claim 4, wherein the first acquisition unit, the second acquisition unit, the third acquisition unit, and the fourth acquisition unit each acquire corresponding data or information via an optical distance measuring device, and the relative position between the optical distance measuring device and the maximum movable range of the rotating tool of the articulated robot remains unchanged during the generation of the control training data and during actual operation.

6. A control program for an articulated robot equipped with a rotary tool that enables friction stir welding (FSW) of two members to be joined, the program comprising: a first acquisition process for acquiring control training data for moving the rotating rotary tool along a target trajectory while pushing it into the planned joining location between the two members to be joined; a robot control process for moving the rotary tool along the target trajectory while rotating it by outputting command values ​​based on the control training data; and a second acquisition process for acquiring measured data of the movement trajectory of the rotary tool when it is actually working on the two members to be joined under the control of the robot control process, wherein the robot control process includes performing feedback control so that the difference between the measured data and the target trajectory becomes zero.

Images