Multi-robot cooperative welding method, welding device, equipment and medium

By optimizing the trajectory planning of multi-robot welding using the particle swarm optimization algorithm, and combining inertia weights and learning factors for updating, the problem of local optima in multi-robot welding is solved, and global optimal trajectory planning and welding quality are improved.

CN118893620APending Publication Date: 2024-11-05WUYI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410927690.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-11
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

In existing technologies, multi-robot welding processes are prone to getting stuck in local optima, making it difficult to achieve globally optimal trajectory planning, which leads to unstable welding quality.

Method used

By acquiring the pose information and welding contour of the handling robot and the welding robot, the welding trajectory is optimized using the particle swarm optimization algorithm. The particle positions are updated by combining the inertia weight coefficient, self-learning factor and group learning factor, and the planned trajectory is adjusted to avoid local optima, thus ensuring global search capability and convergence speed.

Benefits of technology

It realizes global optimal trajectory planning in multi-robot collaborative welding process, improves welding quality and stability, avoids getting trapped in local optimal solutions, and has the advantages of fewer parameters and easy calculation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118893620A_ABST
    Figure CN118893620A_ABST
Patent Text Reader

Abstract

The invention provides a multi-robot cooperative welding method, device and equipment and a medium. The method comprises the steps that third pose information of a carrying robot and a welding robot and a welding contour of a to-be-welded workpiece are obtained; a third planning track is obtained according to the third pose information and the welding contour; the welding robot is controlled to weld the to-be-welded workpiece according to the third planning track, and a welded workpiece is obtained; in the welding process, the current welding seam shape of the to-be-welded workpiece is obtained, and the third planning track is adjusted according to the current welding seam shape; the global search capability and the convergence speed can be ensured, and the situation of falling into local optimum is avoided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of image processing, and in particular to a multi-robot collaborative welding method, welding device, equipment and medium. Background Art

[0002] At present, welding robots are widely used in welding workshops. Multi-robot trajectory planning is the basis for completing welding tasks. Various factors such as workpiece posture, welding gun posture, welding position, robot stability, etc. need to be considered during welding. Multi-objective optimization is to consider the actual needs and the optimality of two or more performances at the same time to obtain a comprehensive optimized robot operation trajectory. The weight coefficient method is often used to deal with combined objective optimization problems. This method generally adds corresponding weight coefficients to each optimization objective function, converts the multi-objective problem into a single objective problem for optimization, and the distribution of weight coefficients reflects the degree of individual objectives in the comprehensive optimization objective. At the same time, the order of magnitude or dimension of each optimization objective function must also be considered, which makes it difficult to reasonably give the weight coefficient to each optimization objective function. At the same time, the solution obtained may not be the global optimal solution and fall into the local optimal solution. Summary of the invention

[0003] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0004] The purpose of the present application is to solve one of the technical problems existing in the related art to at least a certain extent. The embodiments of the present application provide a multi-robot collaborative welding method, welding device, equipment and medium, which can ensure the global search capability and convergence speed, and avoid falling into the local optimal situation.

[0005] An embodiment of the first aspect of the present application, a multi-robot collaborative welding method, comprises:

[0006] Obtain the third posture information of the handling robot and the welding robot and the welding contour of the workpiece to be welded;

[0007] Obtaining a third planning trajectory according to the third posture information and the welding contour;

[0008] Controlling the welding robot to weld the workpiece to be welded according to the third planned trajectory to obtain a welded workpiece;

[0009] Wherein, during the welding process, the current weld morphology of the workpiece to be welded is obtained, and the third planned trajectory is adjusted according to the current weld morphology.

[0010] According to certain embodiments of the first aspect of the present application, obtaining a third planned trajectory according to the third posture information and the welding contour includes:

[0011] Determine the global joint changes and average pulses of the handling robot and the welding robot and the total path length and energy consumption of the welding robot according to the third posture information and the welding profile;

[0012] According to the global joint changes and average pulses of the handling robot and the welding robot and the total path length and energy consumption of the welding robot, the optimal initial welding position of the third planning trajectory is obtained;

[0013] A third planning trajectory is obtained according to the optimal initial welding position planning.

[0014] According to certain embodiments of the first aspect of the present application, the optimal initial welding position is obtained by solving a mathematical model of the optimal initial welding position, and the mathematical model of the optimal initial welding position is expressed as: min f(x, y, z) = k gjv ·GJV+k s,e ·(S+E)+k ap ·AP; where f(x, y, z) is the optimal initial welding position, GJV is the global joint variation of the handling robot and the welding robot, AP is the average pulse of the handling robot and the welding robot, S is the total path length of the welding robot, E is the energy consumption of the welding robot, and k gjv , k s,e and k ap They are the influencing factors of global joint variation, the influencing factors of total path length and energy consumption, and the influencing factors of average power consumption.

[0015] According to certain embodiments of the first aspect of the present application, the mathematical model for solving the optimal initial welding position includes:

[0016] Set the initial welding position as the position of the particles in the particle group, and set the movement speed of the particles;

[0017] Obtaining a welding point sequence according to the welding profile;

[0018] Obtaining a joint motion sequence according to the third planning trajectory;

[0019] Calculating a fitness value according to the welding point sequence and the joint motion sequence;

[0020] Obtaining an individual optimal position and a global optimal position according to the fitness value;

[0021] Update the particle position according to the inertia weight coefficient, self-learning factor, group learning factor, individual optimal position and global optimal position;

[0022] The optimal initial welding position is determined according to the updated particle position.

[0023] According to some embodiments of the first aspect of the present application, the updated position of the particle is expressed as: in, is the updated particle position, is the position of the particle before updating, is the updated particle speed, is the velocity of the particle before updating, ω is the inertia weight coefficient, c 1 is the self-learning factor, c 2 is the group learning factor, is the optimal position of an individual, is the global optimal position, r 1 and r 2 Is a random number.

[0024] According to certain embodiments of the first aspect of the present application, before obtaining the third posture information of the handling robot and the welding robot and the welding contour, the welding trajectory planning method further includes:

[0025] Obtain the first posture information of the workpiece to be welded;

[0026] Performing trajectory planning according to the first posture information to obtain a first planned trajectory;

[0027] The handling robot is controlled according to the first planned trajectory to transport the workpiece to be welded to a workpiece docking position.

[0028] According to certain embodiments of the first aspect of the present application, before obtaining the third posture information of the handling robot and the welding robot and the welding contour, the welding trajectory planning method further includes:

[0029] Acquire second posture information and a butt joint profile of the workpiece to be welded at the butt joint position of the workpiece;

[0030] Obtaining a second planned trajectory according to the second posture information and the docking contour;

[0031] The transport robot is controlled to dock the workpieces to be welded according to the second planned trajectory.

[0032] An embodiment of the second aspect of the present application, a welding device, includes:

[0033] Vision sensors are used to capture images of handling robots, welding robots, and workpieces to be welded;

[0034] A handling robot, used to hold the workpiece to be welded;

[0035] Welding robot, used for welding the workpiece to be welded;

[0036] The controller is used to obtain third posture information of the handling robot and the welding robot, the welding contour of the workpiece to be welded, and the current weld morphology from the image, obtain a third planning trajectory according to the third posture information and the welding contour, control the welding robot to weld the workpiece to be welded to obtain a welded workpiece according to the third planning trajectory, and adjust the third planning trajectory according to the current weld morphology.

[0037] An embodiment of the third aspect of the present application is an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the multi-robot collaborative welding method as described above is implemented.

[0038] An embodiment of the fourth aspect of the present application is a computer storage medium storing computer executable instructions, wherein the computer executable instructions are used to execute the multi-robot collaborative welding method as described above.

[0039] The above scheme has at least the following beneficial effects: considering the multi-objective performance indicators in the welding process, finding the optimal motion sequence of each robot joint according to the multi-objective indicators, and comprehensively considering the performance indicators in the welding process to ensure the welding quality. This method has the advantages of few parameters and simple calculation. By adjusting the third planning trajectory according to the current weld morphology and improving the weight and learning factor, it can ensure the global search ability and convergence speed and avoid falling into the local optimum. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings are used to provide further understanding of the technical solution of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present application and do not constitute a limitation on the technical solution of the present application.

[0041] Figure 1 is a structural diagram of a welding device provided in an embodiment of the present application;

[0042] Figure 2 is a step diagram of a multi-robot collaborative welding method provided in an embodiment of the present application;

[0043] Figure 3 is a step diagram of a workpiece transport step provided in an embodiment of the present application;

[0044] Figure 4 is a step diagram of the docking steps provided in the embodiments of the present application;

[0045] Figure 5 It is a sub-step diagram of step S320 provided in an embodiment of the present application;

[0046] Figure 6is a step diagram of a mathematical model for solving an optimal initial welding position provided by an embodiment of the present application;

[0047] Figure 7 is the discretization curve diagram of the saddle weld;

[0048] Figure 8 It is a schematic diagram of the collaborative space;

[0049] Fig. 9 yes Figure 1 The motion trajectory curve diagram of the handling robot on the left side of the welding device shown;

[0050] Fig.10 yes Figure 1 The motion trajectory curve diagram of the handling robot on the right side of the welding device shown;

[0051] Fig.11 yes Figure 1 The motion trajectory curve diagram of the welding robot in the welding device shown. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0053] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims or the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0054] The embodiments of the present application are further described below in conjunction with the accompanying drawings.

[0055] An embodiment of the present application provides a welding device.

[0056] Reference Figure 1 , welding device, including: visual sensor, handling robot, welding robot and controller.

[0057] Among them, the visual sensor is used to capture images of the handling robot, the welding robot and the workpiece to be welded; the handling robot is used to clamp the workpiece to be welded; the welding robot is used to weld the workpiece to be welded; the controller is used to obtain the third posture information of the handling robot and the welding robot, the welding contour of the workpiece to be welded and the current weld morphology from the image, obtain the third planned trajectory according to the third posture information and the welding contour, control the welding robot to weld the workpiece to be welded according to the third planned trajectory to obtain the welded workpiece, and adjust the third planned trajectory according to the current weld morphology.

[0058] Furthermore, the welding device also includes a switch, a UR5 electric control cabinet, an I RB 1410 electric control cabinet, a double clamping tool, a large atlantic clamp, an ATI six-dimensional force sensor, a welding table, a welding gun, a welding power supply, a water cooling device and connectors, and the connectors include a visual sensor-connector, a force sensor-connector, a clamp-connector and a robot-connector.

[0059] The handling robot is a UR5 robot, which includes at least two robotic arms, each of which is responsible for holding a workpiece to be welded. The UR5 robot is a six-axis industrial collaborative robot, a robotic arm composed of extruded aluminum tubes and joints, which is light, flexible and safe. It can use programming to move tools and use electrical signals to communicate with other devices.

[0060] The UR5 electrical control cabinet is the control center of the UR5 robot. It integrates various hardware and software to complete the control and monitoring of the robot. It has five electrical interfaces, namely controller I / O, tool I / O, Ethernet, power connection, and robot connection.

[0061] The welding robot is the I RB1410 robot. The I RB1410 robot is a six-axis industrial collaborative robot that can be equipped with various process equipment and has excellent working precision.

[0062] The I RB 1410 electric control cabinet has five electrical interfaces, namely controller I / O, tool I / O, Ethernet, power connection, and robot connection. It is also equipped with an emergency stop / reset button, a power-on button, a free / manual switch, and a service port RS.

[0063] The visual sensor is the LiDAR Camera L515 RealSense LiDAR Camera. The LiDAR Camera L515 uses LiDAR depth technology, with a depth of field of 70°*55°, an output resolution of 1024*768, and a frame rate of 30fps. The camera has a high level of precision and accuracy, can quickly measure complex objects, and easily handle obstructions or objects behind; it is used in conjunction with robots, allowing robots to pick up workpieces more easily.

[0064] The clamping tool uses two large jaws, which are servo-rotating parallel electric jaws of the RGI series. It has the following characteristics: force, position and speed are controllable, and the jaws can program and adjust the clamping position, clamping force, running speed and rotation angle; there are various communication methods, which can be controlled by Modbus-RTU protocol and IO mode, and USB, EherCAT, CAN, TCP / IP and other protocols can be transferred through protocol converters.

[0065] The model of ATI six-axis force sensor is Mini45 Titanium, which can withstand a maximum force of 240N to 480N and a maximum torque of 12N·m. It can measure all six forces / torques. It includes a sensor, a highly flexible shielded cable, and an intelligent data acquisition system F / T controller. There are 6 M2 threaded holes on one end and 6 M2 threaded holes on the other end.

[0066] The controller is a PC, including a display, keyboard, mouse, and host. It is the control center of the platform and has functions such as receiving environmental data, experimental feedback data, sending robot control instructions, and starting welding commands. Visual Studio 2019, Robot Studio and other software are installed, and various robot control command libraries and communication protocol libraries corresponding to various devices are installed in VS2019. The control program is written in C++.

[0067] The welding table is equipped with M6 threaded holes neatly arranged at a spacing of 100 mm.

[0068] The welding gun is matched with I RB 1410, one of the Esab PSF315 kits.

[0069] Air cooling unit is available with I RB 1410, one of the Esab PSF315 kits.

[0070] Vision-connector, a connector that connects the vision sensor to the end tool position of the robot, with 4 φ6.5mm through holes, 4 φ2.5mm through holes, and 4 M3 threaded holes.

[0071] Pressure-connector, a connector that fixes the ATI six-dimensional pressure sensor to the end tool position of the robot. It has 8 φ2.5mm through holes and 4 M3 threaded holes.

[0072] The gripper-connector is a connector that fixes the gripper to the end tool position of the robot. It has 6 φ2.5mm through holes and 4 M4 threaded holes.

[0073] Robot-Connector, a connector that fixes the handling robot on the welding table, has 4 M8 threaded holes with a center distance of 132mm, 4 M8 threaded holes with a center distance of 140mm, and 4 φ7mm through holes.

[0074] A switch is a hardware connection that connects different devices together to achieve data transmission and communication.

[0075] The platform layout of the welding device is as follows.

[0076] The UR5 robot is installed through the 4 φ9mm through holes on the UR5 robot base. Four M8 studs are used to pass through the 4 φ9mm through holes of the UR5 base and fixed to the M8 threaded holes with a center distance of 132mm between the robot-connector to complete the fixation of the UR5 robot to the robot-connector. Four M6 studs are used to pass through the 4 φ7mm through holes of the robot-connector and fixed to the M6 ​​threaded holes on the welding table to complete the fixation of the robot-connector to the welding table. Both handling robots are fixed to the welding table through the robot-connector, and are fixed symmetrically on the left and right.

[0077] Installation of the I RB 1410 robot. The I RB 1410 robot is placed on one side of the welding table, in the middle of the double-carrying robot and facing the double-carrying robot. Cork is used as the base of the I RB 1410 robot to prevent the jitter generated during startup from changing the position of the I RB 1410 robot in the world coordinate system.

[0078] To install the visual sensor, use 4 M2 studs to fix the visual sensor on the visual connector, and use 4 M6 studs to fix the visual connector on the threaded hole of the end flange of the handling robot. A visual sensor is installed at each end of the handling robot and fixed by the visual connector.

[0079] Installation of the ATI six-dimensional force sensor: fix the sensor to the pressure-connector with 6 M2 studs, and fix the pressure-connector to the vision-connector with 4 M3 studs to complete the installation of the ATI six-dimensional sensor.

[0080] The atlas jaw is installed by fixing it to the jaw-connector through 4 M4 studs, and the jaw-connector is fixed to the ATI six-dimensional pressure sensor with 6 M2 studs to complete the installation of the atlas jaw.

[0081] Installation of arc welding gun, fix the arc welding gun by the flange of the sixth axis of I RB 1410.

[0082] Installation of the air cooling device: The spray pipe at the end of the air cooling device is installed next to the arc welding gun, and the spray pipe is aimed at the position to be welded.

[0083] The platform communication of the welding device is as follows.

[0084] Communication connection of UR5 robot; at the physical connection level, it communicates through its own control cabinet, and its electric control cabinet communicates with the switch through the Ethernet communication protocol, and the switch communicates with the controller through the TCP / IP communication protocol to transmit data. At the software communication level, the IPv4 addresses of the UR5 robot and the controller are in the same network segment to establish a local area network. The dual-handling robot establishes a Socket communication connection with the controller, and the software VS2019 on the controller uses the Socket code. The server side is the controller, and the dual-handling robot is the client. The client and the server communicate directly through the TCP protocol. When the three are connected successfully, the controller can transmit the command to control the robot's movement to the corresponding robot electric control cabinet. The electric control cabinet controls the robot's movement, and the robot can also transmit robot-related data to the controller.

[0085] Communication connection of I RB 1410 robot; at the physical connection level, it communicates through its own control cabinet, and its electric control cabinet communicates with the switch through the Ethernet communication protocol, and the switch communicates with the PC through the TCP / IP communication protocol to transmit data. At the software communication level, I RB 1410 robot can establish a local area network with the PC. I RB 1410 robot uses Robot Studio to control it. The software is placed in the controller and uses Socket communication to establish communication between Robot Studio and VS2019. VS2019 is the server and Robot Studio is the client. When the connection is successful, the I RB 1410 robot can be controlled by VS2019, and vice versa, motion data can be transmitted to the controller.

[0086] Communication connection of LiDAR Camera L515 visual sensor; At the physical connection level, the camera is connected to the PC host via USB3.1. At the software communication level, use the Intel SDK2.0 function package, download and install it to the PC, reference the SDK in the software VS2019, and access and control the LiDAR Camera L515 visual sensor by writing code to access the Intel SDK2.0.

[0087] Communication connection of ATI six-dimensional pressure sensor; At the physical connection level, ATI six-dimensional pressure sensor communicates with ATINetBox, ATI NetBox communicates with the switch through Ethernet communication protocol, and the switch communicates with the PC through TCP / IP communication protocol to transmit data. The default IPv4 address of ATI six-dimensional sensor is 192.168.1.1, which can establish a local area network with the PC. ATI six-dimensional sensor uses Socket communication to establish a connection, and the PC can receive data from ATI six-dimensional pressure sensor.

[0088] Communication connection and power supply of the Dahuan gripper; fix the two grippers on a handling robot respectively, and provide 24V DC voltage from the controller I / O interface of the corresponding handling robot electric control cabinet. Connect the red line of the gripper communication line to the positive pole of the 24V DC voltage regulator of the electric control cabinet, and the black line to the negative pole of the 24V DC voltage regulator of the electric control cabinet. At the physical connection level, the blue line is connected to the RS-485 to USB module T / R-, and the green line is connected to the RS-485 to USB module T / R+. The RS-485 to USB data cable transmits data with the PC via USB. At the software communication level, the Dahuan gripper communicates with the PC through the Modbus-RTU protocol. The software VS2019 calls the C++ library of Modbus-RTU to control the Dahuan gripper.

[0089] Communication connection of welding gun; At the physical connection level, the welding gun is connected to the I RB 1410 electric control cabinet through a cable. At the software communication level, I RB 1410 communicates with the welding gun and welding power supply through the DSQC651 communication module; DSQC651 is a mixed type signal module, including 8 digital input interfaces, 8 digital output interfaces and two analog output interfaces; in the official software RobotStud io, configure the digital input and output signals of I RB 1410 to control the start and stop of the welding power supply and the analog output signals of the welding power supply voltage and current to realize the control of welding by the software RobotStud io; finally, through the control of the software RobotStud io by the software VS2019, the welding gun and welding power supply are indirectly controlled.

[0090] The welding device can be divided into multiple modules, including an image recognition module, a trajectory planning module, a communication module, a robot control module, a welding module and a welding recognition module.

[0091] For the image recognition module, the image recognition module includes two LiDAR Camera L515 visual sensors, which are equipped with an image recognition algorithm and a hand-eye calibration algorithm. The two LiDAR Camera L515 visual sensors are each fixed on a flange at the end of a handling robot. The hand-eye calibration algorithm is set on the PC side, and its function is: when the welding task is started, the robot hand-eye calibration is used to obtain the position of the workpiece to be welded, and the data is processed and sent to the trajectory planning module; when the workpiece to be welded is in the docking preparation position, the position of the workpiece to be welded at this time is transmitted to the trajectory planning module; before welding starts, the weld image is collected, and the image is processed and sent to the trajectory planning module. The image recognition algorithm is as follows: the image of the environment collected by the camera is used to recognize the workpiece to be welded based on the Opencv visual library programming. The steps of hand-eye calibration are as follows: the collected image is corrected according to the preset camera parameters to obtain a corrected image; the image is subjected to median filtering, histogram equalization, grayscale processing, adaptive threshold segmentation, morphological opening and closing operations, edge extraction, feature point stereo matching and template matching to obtain the two-dimensional information and distance information of the contour of the workpiece to be welded; based on the two-dimensional information and distance information of the contour of the workpiece to be welded, a workpiece coordinate system with the camera coordinate system as the parent coordinate system is established; based on the conversion between the workpiece coordinate system and the world coordinate system, the spatial position information of the workpiece to be welded is obtained. The image recognition module works as follows: when the welding task is started, the image recognition algorithm will identify the workpiece to be welded. Through hand-eye calibration, the position and posture of the workpiece to be welded relative to the camera coordinates can be obtained. After the rigid body matrix transformation, the position and posture of the workpiece to be welded in the world coordinate system can be obtained, and the position and posture information is sent to the trajectory planning module; when the workpiece to be welded is transported to the docking preparation position by the handling robot, the two LiDAR Camera L515 visual sensors will scan the position and posture of the docking part of the workpiece of the handling robot on the other side, and send the position and posture to the trajectory planning module, so that the handling robot can make minor adjustments to ensure the accuracy of the docking; when the workpiece to be welded is successfully docked, the weld image scanned by the LiDARCamera L515 visual sensor will be analyzed and processed, and then uploaded to the trajectory planning module.

[0092] For the trajectory planning module, the trajectory planning module is set in the PC, and the trajectory planning module applies the trajectory planning algorithm. The trajectory planning module plans the motion trajectory and posture of the dual handling robot during the clamping, handling, workpiece docking, and welding process; before welding, the motion trajectory and posture of the handling robot and the welding robot end welding gun are planned; the planned path is simulated to prevent collision and interference during the task. The working method of the trajectory planning module is as follows: receiving the posture of the workpiece to be welded in the world coordinate system sent by the image recognition module, the trajectory planning algorithm uses the Moveit motion planner to automatically plan the path of clamping, docking preparation position, and docking position; receiving the weld processing analysis sent by the image recognition module, planning the collaborative welding trajectory of the three robots and the welding gun to ensure the completion of the saddle weld task.

[0093] As for the communication module, the communication module is set in the PC to realize the communication between each module and the PC, and realize the information exchange between each module through the PC. It has the functions of command transmission, data transmission, monitoring and control, collaboration and coordination, etc.

[0094] As for the robot control module, the robot control module includes: UR5 and its electric control cabinet, I RB 1410 and its electric control cabinet, Dahuan gripper, gripper control algorithm, ATI six-dimensional sensor, and force / position control algorithm. The robot control module receives the trajectory operation information transmitted by the trajectory planning module and controls the robot to complete the corresponding trajectory action; during the robot's tight coordination operation, the robot's movement is adaptively adjusted due to the robot's motion accuracy or external interference to avoid robot damage and ensure the stability of welding quality. The working mode of the robot control module is as follows: the trajectory planning module will transmit the planned trajectory information to the PC, and the PC will transmit it to the corresponding robot electric control cabinet, and finally the electric control cabinet will control the robot's joint movement based on the information; after completing the docking of the workpieces to be welded, the handling robot will enter a tight coordination state. If the two robots do not move in unison, huge internal forces will be generated, resulting in the failure of the welding task, and in severe cases, the robot will be damaged. The state of the internal force is monitored in real time through the ATI six-dimensional pressure sensor. The force / position coordination algorithm will process the data of the pressure sensor and follow the position of the robot to ensure that the double handling robots are still in a tight coordination state without causing damage to the robots.

[0095] For the welding module, the welding module includes welding power supply, welding gun, gas cooling protection device, and welding configuration data in the PC. The welding module receives the control command of the PC, turns on the welding protection gas device, turns on the welding power supply according to the control command, and controls the welding gun to weld; the welding current and voltage can be changed on the PC to meet the task requirements of different workpieces.

[0096] The weld seam recognition module includes two LiDAR Camera L515 visual sensors, image recognition algorithms, and weld seam morphology database. During the welding process, the LiDAR Camera L515 visual sensor will identify the weld seam morphology in real time, monitor the quality of the welding process, and adjust the robot's posture at any time to ensure the smooth completion of welding.

[0097] The welding device performs welding using the following multi-robot collaborative welding method.

[0098] Reference Figure 2 , a multi-robot collaborative welding method comprises the following steps:

[0099] Step S310, obtaining the third posture information of the handling robot and the welding robot and the welding contour of the workpiece to be welded;

[0100] Step S320, obtaining a third planning trajectory according to the third posture information and the welding contour;

[0101] Step S330, controlling the welding robot to weld the workpiece to be welded according to the third planned trajectory to obtain a welded workpiece;

[0102] During the welding process, the current weld morphology of the workpiece to be welded is obtained, and the third planning trajectory is adjusted according to the current weld morphology.

[0103] Before welding, the multi-robot collaborative welding method also includes: a workpiece transportation step and a docking step.

[0104] Reference Figure 3 , the workpiece transport step includes the following steps:

[0105] Step S110, obtaining the first pose information of the workpiece to be welded;

[0106] Step S120, performing trajectory planning according to the first pose information to obtain a first planned trajectory;

[0107] Step S130: Controlling the transport robot to transport the workpiece to be welded to a workpiece docking position according to the first planned trajectory.

[0108] Reference Figure 4 , the docking step comprises the following steps;

[0109] Step S210, obtaining second posture information and a butt joint profile of the workpiece to be welded at the butt joint position of the workpiece;

[0110] Step S220, obtaining a second planning trajectory according to the second posture information and the docking profile;

[0111] Step S230: Controlling the transport robot to dock the workpieces to be welded according to the second planned trajectory.

[0112] In general, the steps of the multi-robot collaborative welding method are as follows: initialize the camera and calibrate the camera. The image recognition module controls the camera to scan the environment, find and identify the workpiece to be welded, obtain the posture information of the workpiece to be welded through hand-eye calibration, and calculate the first posture information of the workpiece to be welded in the world coordinate system through rigid body matrix transformation. The image recognition module sends the converted first posture information to the trajectory planning module. The trajectory planning module will plan the motion trajectory of the handling robot's clamping position and the workpiece docking preparation position according to the current posture information to obtain the first planned trajectory. The trajectory planning module sends the first planned trajectory to the robot control module. The robot control module controls the handling robot to move to a posture suitable for clamping the workpiece to be welded based on the first planned trajectory, and controls the large jaws to clamp the workpiece to be welded, and then the handling robot will move to the workpiece docking position. The image recognition module will determine the second posture information and docking contour of the workpiece to be welded held by the handling robot at this time through hand-eye calibration, and send the second posture information and docking contour to the trajectory planning module; the trajectory planning module will adjust the posture of the handling robot according to the second posture information of the two workpieces to be welded to ensure that the workpieces can be docked, and plan the motion trajectory of the docking position to obtain the second planning trajectory. The robot control module will plan according to the received motion instructions, control the handling robot to follow the adjustment, and start docking. If the docking fails, the robot will follow the adjustment of the posture according to the force / position control algorithm to avoid damage to the robot. After the docking is successful, the robot control module will feedback the corresponding third posture information of the handling robot and the welding robot back to the PC. The trajectory planning module plans the third planning path of the three-robot collaborative operation of welding the saddle-shaped weld based on the third posture information of the handling robot and the welding robot and the welding contour of the workpiece to be welded, and simulates the third planning path to avoid collision or interference between the three robots during the tight / loose coordination process. The robot control module and welding module start, start welding, and complete the saddle weld task; the weld recognition module monitors the current weld morphology of the workpiece during the welding process, and compares the current weld morphology with the morphology library, and transmits the comparison results to the trajectory planning module and the robot control module, adjusts the third planning trajectory according to the current weld morphology, and controls the three robots to complete the task with the adjusted third planning trajectory to ensure the stability of welding quality. After welding is completed, one handling robot will release the clamp, and another handling robot will clamp the workpiece and place it in the finished product area.

[0113] In some embodiments, the above welding method can be used to weld saddle-shaped welds. In actual production, the connecting welding between cylindrical tubes is one of the more common forms of pipe connection. From a geometric point of view, when a circular tube with a larger diameter (main tube) and a circular tube with a smaller diameter (branch tube) intersect each other vertically, the shape of the workpiece interface weld is like a saddle, and the weld is called a saddle-shaped weld. Saddle-shaped welds mainly have several forms such as orthogonal, oblique, and eccentric intersections.

[0114] Reference Figure 5 In some embodiments, obtaining a third planning trajectory according to the third posture information and the welding contour includes the following steps:

[0115] Step S321, determining the global joint changes and average pulses of the handling robot and the welding robot and the total path length and energy consumption of the welding robot according to the third posture information and the welding profile;

[0116] Step S322, obtaining the optimal initial welding position of the third planning trajectory according to the global joint changes and average pulses of the handling robot and the welding robot and the total path length and energy consumption of the welding robot;

[0117] Step S323, obtaining a third planning trajectory according to the optimal initial welding position planning.

[0118] In the whole multi-robot collaborative welding process, after determining the motion trajectory of each robot, it is expected that the global joint variation of the robot joint angle within the unit control cycle is as small as possible, and the robot motion is more stable. Different initial welding positions correspond to different global joint variations, so the global joint variation is also an important indicator for selecting the optimal initial welding position. The objective function corresponding to the global joint variation is: m θ i (j) is the joint coordinate at the current moment, m θ i (j-1) is the joint coordinate at the previous moment, m θ i (j)- m θ i (j-1) is the joint change at the current moment, m is the number of robots, N is the joint number, and n is the number of weld discretizations.

[0119] In order to improve the stability of the robot, pulsation shock is also an indicator that needs to be considered. The average pulse represents the speed of torque change. Smaller shocks can make the torque change more stable. The average pulse is expressed as: is the joint jerk, and J is the joint jerk.

[0120] When planning the trajectory of multiple robots, the energy consumption and path length of the welding robot are indicators that need to be considered. When analyzing the indicators, the robot dynamics and kinematics are involved.

[0121] The shorter the path length of the end motion of the welding robot, the easier it is for the welding gun to ensure stable welding. The distance between weld point j and point j+1 is the distance between the two weld points, so the total path length S of the welding robot is: d j,j+1 Represents the distance between weld point j and point j+1.

[0122] During the welding process, industrial robots consume a lot of energy, so it is very necessary to consider the energy consumption of industrial robots. The trajectory planning of the welding robot can obtain information such as the robot's joint angle, speed, acceleration, etc. Based on the joint angle, speed, and acceleration, the required joint torque can be calculated by using inverse dynamics. The energy consumption of the welding robot is expressed as follows through the Lagrangian method: Where E is energy consumption, q i is the joint velocity, τ is the joint torque, t j is the time step.

[0123] According to the above performance indicators, the mathematical modeling of the optimal initial welding position layout is established: Based on this formula, the optimal initial welding position can be determined.

[0124] The optimal initial welding position is obtained by solving the mathematical model of the optimal initial welding position. The mathematical model of the optimal initial welding position is expressed as: min f(x, y, z) = k gjv ·GJV+k s,e ·(S+E)+k ap ·AP; where f(x, y, z) is the optimal initial welding position, GJV is the global joint variation of the handling robot and the welding robot, AP is the average pulse of the handling robot and the welding robot, S is the total path length of the welding robot, E is the energy consumption of the welding robot, and k gjv , k s,e and k ap They are the influencing factors of global joint variation, total path length and energy consumption, and average power consumption. Specifically, according to the empirical weight values, the following coefficient values ​​can be taken: k gjv =0.3, k s,e =0.3, k ap =0.4.

[0125] Reference Figure 6 , solving the mathematical model for the optimal initial welding position includes the following steps:

[0126] Step S410, setting an initial welding position as the position of a particle in the particle group, and setting a moving speed of the particle;

[0127] Step S420, obtaining a welding point sequence according to the welding profile;

[0128] Step S430, obtaining a joint motion sequence according to the third planning trajectory;

[0129] Step S440, calculating the fitness value according to the welding point sequence and the joint motion sequence;

[0130] Step S450, obtaining the individual optimal position and the global optimal position according to the fitness value;

[0131] Step S460, updating the position of the particle according to the inertia weight coefficient, the self-learning factor, the group learning factor, the individual optimal position and the global optimal position;

[0132] Step S470, determining the optimal initial welding position according to the updated particle position.

[0133] The updated particle position is expressed as: in, is the updated particle position, is the position of the particle before updating, is the updated particle speed, is the velocity of the particle before updating, ω is the inertia weight coefficient, c 1 is the self-learning factor, c 2 is the group learning factor, is the optimal position of an individual, is the global optimal position, r 1 and r 2 is a random number between 0 and 1.

[0134] Specifically, in the particle swarm algorithm, multiple particles coexist and cooperate to find the best solution. Each particle flies to a better position in the problem space based on its own experience and the experience of other particles in the group. The best position that an individual particle has experienced during the flight is called the individual optimal position pBest, and the best position that the entire group has experienced so far is called the global optimal position gBest.

[0135] The particle state is expressed by the D-dimensional velocity v i =(v i1 ,v i2 ,···,v iD ) and position x i =(x i1 ,x i2 ,···,xiD ) represents the movement of particles. The movement of particles is determined by the current position and speed as well as the optimal position and global optimal position of the particles. Through multiple iterations, the particle swarm will gradually approach the optimal solution of the problem.

[0136] The inertia weight coefficient ω determines the exploration ability of the particle. The larger the ω, the stronger the exploration ability of the global space, and the smaller the ω, the stronger the exploration ability of the local area. Therefore, in the process of particle iteration, the global exploration ability of the particle is guaranteed in the early stage of iteration, and the particle can explore a larger range in order to find the global optimal position area. The local exploration ability of the particle is guaranteed in the later stage of iteration to find the optimal position. Among them, the inertia weight coefficient ω is expressed as: In the formula, k is the current number of iterations, T is the maximum number of iterations; ω end =0.4;ω start =1.1.

[0137] Self-learning factor c 1 represents the tendency of particles to approach the individual optimal position pBest, and the group learning factor c 2 represents the tendency of the particle to approach the global optimal position gBest, and adjust c appropriately 1 、c 2 In order to facilitate particle optimization. The larger learning factor of the initial particles is conducive to improving the global search ability of the particles, and the convergence of the particles is also strengthened. The smaller learning factor of the particles in the later stage is conducive to finding the global optimal individual. Among them, k is the current number of iterations, T is the maximum number of iterations; c max =2.1; c min =0.8.

[0138] Through the mathematical model of the optimal initial welding position min f(x,y,z)=k gjv ·GJV+k s,e ·(S+E)+k ap AP is used to solve the problem, with the goal of solving the optimal initial welding position f(x, y, z).

[0139] Specifically, the solution steps are as follows:

[0140] Initialize the particle swarm parameters, initialize the particle position and particle velocity by randomly generating, and the position of the particles in the particle swarm is equivalent to the initial welding position.

[0141] Discretize the weld seam of the workpiece to be welded to obtain a sequence of weld points; and calculate the motion trajectory relative to the end of the robot based on kinematic transformation. Figure 7 , Figure 7 is the discretization curve of the saddle weld.

[0142] The motion trajectory of the robot end relative to the robot base is calculated according to the third planning trajectory to obtain the joint motion sequence.

[0143] The fitness value is calculated according to the welding point sequence and the joint motion sequence, and the fitness value is a multi-objective function.

[0144] The individual optimal position and the global optimal position are obtained according to the fitness value.

[0145] Determine whether the maximum number of iterations has been reached.

[0146] When the maximum number of iterations is not reached, the inertia weight coefficient, self-learning factor, and group learning factor are adjusted, and the position of the particle is updated according to the inertia weight coefficient, self-learning factor, group learning factor, individual optimal position, and global optimal position. Return to the step of discretizing the weld seam of the workpiece to be welded to obtain a sequence of weld points.

[0147] When the maximum number of beatings is reached, the updated particle position is used as the optimal initial welding position.

[0148] By adjusting the weights and learning factors, updating the position and velocity of the particles, combining the individual's historical optimal position and the group's historical optimal position, and combining the particle's own inertia to jointly determine the particle's motion state, until the iteration ends and the optimal initial welding position is output, a convenient multi-objective function solving algorithm with fast convergence speed and strong global exploration capability is realized.

[0149] In this embodiment, the multi-objective performance indicators in the welding process are taken into account, and the optimal motion sequence of each robot joint is found according to the multi-objective indicators, and the performance indicators in the welding process are comprehensively considered to ensure the welding quality. This method has the advantages of few parameters and simple calculation. By improving the weights and learning factors, it can ensure the global search ability and convergence speed, and avoid falling into the local optimum.

[0150] The simulation modeling was performed using simulation software. In the world coordinate system, the base of the left handling robot UR5 is [0,700mm,0], the base of the right handling robot UR5 is [0,-700,0], and the base of the welding robot IRB 1410 is [1300,0,0]. Whether multiple robots can collaborate to complete the saddle-shaped task, the collaborative space (the workspace that the robots can reach together) is an important consideration. The Monte Carlo method is used to solve the collaborative space of multiple robots, and the range of the collaborative space is approximately: X[-270mm,270mm], Y[-300mm,300mm], Z[-200mm,800mm]; refer to Figure 8 , Figure 8The schematic diagram of the collaborative space. In the collaborative space, the multi-robot collaborative welding method can obtain the end motion trajectory of each robot, and the multi-objective performance index is reasonable, which verifies the feasibility. In the collaborative space, through simulation verification, refer to Fig. 9 , Fig. 9 This is the motion trajectory curve of the handling robot on the left; refer to Fig.10 , Fig.10 This is the motion trajectory curve of the transport robot on the right; refer to Fig.11 , Fig.11 The figure is the motion trajectory curve of the welding robot; the motion trajectory of each robot is very smooth in the whole process, and there is no major mutation, which proves the rationality and reliability of the trajectory of each robot planned by this method. In the collaborative space, the motion trajectory of the end of the double handling robot and the motion trajectory of the end of the welding robot are obtained according to the motion trajectory of each robot. The simulation results show that multi-robot collaboration can successfully complete the welding task, and the path of each robot is smooth and reasonable, which proves the feasibility of realizing the multi-robot collaborative welding task through the multi-robot collaborative welding method.

[0151] An embodiment of the present application provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the multi-robot collaborative welding method as described above when executing the computer program.

[0152] The electronic device may be any intelligent terminal including a computer.

[0153] In general, for the hardware structure of the electronic device, the processor can be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, etc., to execute related programs to implement the technical solutions provided in the embodiments of the present application.

[0154] The memory can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory can store an operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory, and the processor is called to execute the methods of the embodiments of this application.

[0155] The input / output interface is used to realize information input and output.

[0156] The communication interface is used to realize the communication interaction between this device and other devices. Communication can be achieved through wired methods (such as USB, network cable, etc.) or wireless methods (such as mobile network, WIFI, Bluetooth, etc.).

[0157] The bus transmits information between the various components of the device (such as the processor, memory, input / output interface, and communication interface). The processor, memory, input / output interface, and communication interface communicate with each other within the device through the bus.

[0158] An embodiment of the present application provides a computer storage medium storing computer executable instructions for executing the multi-robot collaborative welding method as described above.

[0159] It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium. In the above description of this specification, the description of reference terms "one embodiment / example", "another embodiment / example" or "certain embodiments / examples" and the like means that the specific features, structures, materials or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0160] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0161] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0162] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0163] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0164] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above-mentioned units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. Although the embodiments of the present application have been shown and described, it can be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and purposes of the present application, and the scope of the present application is defined by the claims and their equivalents.

[0165] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the embodiments. Technical personnel familiar with the field can make various equivalent modifications or substitutions without violating the spirit of the present application. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present application.

Claims

1. A multi-robot collaborative welding method, It is characterized in that include: Obtain the third posture information of the handling robot and the welding robot and the welding contour of the workpiece to be welded; Obtaining a third planning trajectory according to the third posture information and the welding contour; Controlling the welding robot to weld the workpiece to be welded according to the third planned trajectory to obtain a welded workpiece; Wherein, during the welding process, the current weld morphology of the workpiece to be welded is obtained, and the third planned trajectory is adjusted according to the current weld morphology.

2. The multi-robot collaborative welding method according to claim 1, It is characterized in that The obtaining a third planning trajectory according to the third posture information and the welding contour includes: Determine the global joint changes and average pulses of the handling robot and the welding robot and the total path length and energy consumption of the welding robot according to the third posture information and the welding profile; According to the global joint changes and average pulses of the handling robot and the welding robot and the total path length and energy consumption of the welding robot, the optimal initial welding position of the third planning trajectory is obtained; A third planning trajectory is obtained according to the optimal initial welding position planning.

3. The multi-robot collaborative welding method according to claim 2, It is characterized in that The optimal initial welding position is obtained by solving the mathematical model of the optimal initial welding position. The mathematical model of the optimal initial welding position is expressed as: min f(x, y, z) = k gjv ·GJV+k s,e ·(S+E)+k ap ·AP; where f(x, y, z) is the optimal initial welding position, GJV is the global joint variation of the handling robot and the welding robot, AP is the average pulse of the handling robot and the welding robot, S is the total path length of the welding robot, E is the energy consumption of the welding robot, and k gjv , k s,e and k ap They are the influencing factors of global joint variation, the influencing factors of total path length and energy consumption, and the influencing factors of average power consumption.

4. The multi-robot collaborative welding method according to claim 3, It is characterized in that The mathematical model for solving the optimal initial welding position includes: Set the initial welding position as the position of the particles in the particle group, and set the movement speed of the particles; Obtaining a welding point sequence according to the welding profile; Obtaining a joint motion sequence according to the third planning trajectory; Calculating a fitness value according to the welding point sequence and the joint motion sequence; Obtaining an individual optimal position and a global optimal position according to the fitness value; Update the particle position according to the inertia weight coefficient, self-learning factor, group learning factor, individual optimal position and global optimal position; The optimal initial welding position is determined according to the updated particle position.

5. The multi-robot collaborative welding method according to claim 4, It is characterized in that The updated particle position is expressed as: in, is the updated particle position, is the position of the particle before updating, is the updated particle speed, is the velocity of the particle before updating, ω is the inertia weight coefficient, c 1 is the self-learning factor, c 2 is the group learning factor, is the optimal position of an individual, is the global optimal position, r 1 and r 2 Is a random number.

6. The multi-robot collaborative welding method according to claim 1, It is characterized in that Before obtaining the third posture information of the handling robot and the welding robot and the welding contour, the welding trajectory planning method further includes: Obtain the first posture information of the workpiece to be welded; Performing trajectory planning according to the first posture information to obtain a first planned trajectory; The handling robot is controlled according to the first planned trajectory to transport the workpiece to be welded to a workpiece docking position.

7. The multi-robot collaborative welding method according to claim 6, It is characterized in that Before obtaining the third posture information of the handling robot and the welding robot and the welding contour, the welding trajectory planning method further includes: Acquire second posture information and a butt joint profile of the workpiece to be welded at the butt joint position of the workpiece; Obtaining a second planned trajectory according to the second posture information and the docking contour; The transport robot is controlled to dock the workpieces to be welded according to the second planned trajectory.

8. A welding device, It is characterized in that include: Vision sensors are used to capture images of handling robots, welding robots, and workpieces to be welded; A handling robot, used to hold the workpiece to be welded; Welding robot, used for welding the workpiece to be welded; The controller is used to obtain third posture information of the handling robot and the welding robot, the welding contour of the workpiece to be welded, and the current weld morphology from the image, obtain a third planning trajectory according to the third posture information and the welding contour, control the welding robot to weld the workpiece to be welded to obtain a welded workpiece according to the third planning trajectory, and adjust the third planning trajectory according to the current weld morphology.

9. An electronic device, It is characterized in that include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the multi-robot collaborative welding method according to any one of claims 1 to 7 is implemented.

10. A computer storage medium, It is characterized in that Computer executable instructions are stored, and the computer executable instructions are used to execute the multi-robot collaborative welding method according to any one of claims 1 to 7.