Movable type double-robot collaborative non-consumable electrode gas shield welding system and welding method

Abstract

The system comprises a welding gun robot, a wire feeding robot and a cooperative control system, and the cooperative control system recognizes three-dimensional point cloud data of a welding seam, the structural shape of a to-be-welded part, the shape of a molten pool and the extension length of a welding wire in real time; calculating the actual moving track, the actual wire feeding path, the actual wire feeding angle and the actual wire feeding speed of the welding gun, comparing the actual moving track of the welding gun with the planned moving track, correcting a deviation value, optimizing the moving track of the welding gun, comparing the actual wire feeding path, the actual wire feeding angle and the actual wire feeding speed with the planned wire feeding path, the planned wire feeding angle and the planned wire feeding speed respectively, correcting the deviation value, and optimizing the welding gun moving track. The wire feeding path, the wire feeding angle and the wire feeding speed are optimized; the cooperative control system is used for cooperatively controlling the welding gun robot and the wire feeding robot to conduct welding operation according to the optimized data. Wire feeding and the welding gun operate independently, the flexibility of the welding system is improved, and precise welding under the complex structure condition can be achieved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of automatic welding, and particularly relates to a mobile double-robot cooperative non-gas shielded welding system and a welding method. BACKGROUND

[0002] Robot welding is a technology for realizing automatic welding by using an industrial robot to carry a welding device. Compared with traditional manual welding, robot welding has high efficiency, high repeatability, and high production quality, and can meet the needs of large-scale production and flexible manufacturing in the manufacturing industry, and is widely used in the fields of automobile manufacturing, aerospace, and electronic assembly.

[0003] Non-gas shielded welding is a highly reliable welding method, and is usually used in precise component welding scenarios such as pressure vessels. The existing non-gas shielded welding system usually has a rigidly fixed wire feeding mechanism and welding torch. The welding wire is fed into the arc from the front of the arc, and is suitable for long straight welds. However, for complex components, the welding torch and wire feeding angle need to be dynamically adjusted according to the weld position. The existing welding system has poor adaptability, and when welding complex welds such as pipe welding and curved welding, the welding wire is always fed at a fixed angle and cannot be dynamically adjusted according to the weld. In addition, in narrow spaces or complex geometric structures, the rigid structure may collide with the workpiece or fixture, causing damage. Therefore, the non-gas shielded welding of large and complex components is mostly completed manually.

[0004] The invention patent with the application number CN202110134833.7 discloses a thin-walled structure laser welding system and method based on multi-robot cooperation. The system uses two parallel guide rails, and an external shaft is connected with a welding robot for welding wall plate parts and an auxiliary robot for pressing the wall plate parts. The welding robot and the auxiliary robot have a stroke along the width direction of the guide rail through the external shaft, and realize the laser welding process. The system improves manufacturing flexibility to a certain extent, but due to the limitation of the guide rail stroke and the requirements of the part structure, it still cannot meet the flexibility required for welding large and complex components. If a single-robot double-robot welding system is used, the manufacturing flexibility can also be improved to a certain extent, but when the double-robot works cooperatively, the robot's center of gravity changes frequently, affecting the stability of the welding and failing to meet the needs of precise welding.

[0005] The existing multi-robot cooperation scheme cooperates multiple heat sources or welding and auxiliary tooling, and there is no double-robot cooperative welding system for flexible and precise welding needs. SUMMARY

[0006] The purpose of the present application is to provide a mobile double-robot cooperative non-gas shielded welding system and a welding method to realize flexible wire feeding and improve welding quality.

[0007] Technical solution: The mobile double-robot cooperative non-gas shield welding system comprises a welding gun robot, a wire feeding robot and a cooperative control system, in the welding process, the cooperative control system identifies the three-dimensional point cloud data of the weld in real time, and calculates the actual running track of the welding gun according to the three-dimensional point cloud data of the weld identified in real time, compares the actual running track of the welding gun with the planned running track, corrects the deviation value, and optimizes the running track of the welding gun; at the same time, the cooperative control system identifies the structure shape of the workpiece to be welded, the molten pool appearance and the wire extension length in real time, and calculates the actual wire feeding path, the wire feeding angle and the wire feeding speed according to the structure shape of the workpiece to be welded, the welding gun pose, the molten pool appearance and the wire extension length identified in real time, compares the actual wire feeding path, the wire feeding angle and the wire feeding speed with the planned wire feeding path, the wire feeding angle and the wire feeding speed respectively, corrects the deviation value, and optimizes the wire feeding path, the wire feeding angle and the wire feeding speed; the cooperative control system is used for cooperatively controlling the welding gun robot and the wire feeding robot to perform welding operation according to the optimized running track of the welding gun, the wire feeding path, the wire feeding angle and the wire feeding speed.

[0008] Optionally, the cooperative control system comprises a communication module, a motion planning module and a sensor perception module, the sensor perception module is used for monitoring the actual running data of the welding robot and the wire feeding robot, and sending the actual running data to the motion planning module, the motion planning module compares the actual data with the planned data, corrects the deviation value, and optimizes the running track of the welding gun, the wire feeding path, the wire feeding angle and the wire feeding speed; the motion planning module transmits the optimized data to the welding robot and the wire feeding robot through the communication module.

[0009] The application also provides a welding method of the mobile double-robot cooperative non-gas shield welding system, comprising the following steps:

[0010] Before welding, the welding gun robot and the wire feeding robot are initialized, and the cooperative control system issues a synchronous starting instruction;

[0011] The welding gun robot moves to the starting point of the weld, and the cooperative control system acquires the three-dimensional point cloud data of the weld through visual scanning; the wire feeding robot moves to the working area, and the cooperative control system identifies the structure shape of the workpiece to be welded and the wire extension length;

[0012] The cooperative control system plans the running track and the pose of the welding gun according to the three-dimensional point cloud data of the weld, plans the welding speed of the welding gun through welding requirements, and plans the wire feeding path and the wire feeding angle according to the structure shape of the workpiece to be welded and the pose of the welding gun, and simultaneously plans the wire feeding speed according to the welding speed;

[0013] The cooperative control system sends the planned running track and the welding speed of the welding gun to the welding gun robot, and simultaneously sends the planned wire feeding path, the wire feeding angle and the wire feeding speed to the wire feeding robot, and the two robots cooperatively weld.

[0014] During the welding process, the collaborative control system identifies real-time three-dimensional point cloud data of the weld, real-time structural shape of the workpiece to be welded, real-time molten pool appearance and real-time wire extension length, and obtains the actual movement trajectory and real-time pose of the welding torch according to the real-time three-dimensional point cloud data of the weld; at the same time, the actual wire feeding path and wire feeding angle are obtained according to the real-time pose of the welding torch, the real-time structural shape of the workpiece to be welded, the real-time wire extension length, and the actual wire feeding speed is obtained according to the real-time molten pool appearance and the real-time wire extension length; the collaborative control system compares the actual movement trajectory of the welding torch with the planned movement trajectory, corrects the deviation value, and optimizes the movement trajectory of the welding torch; at the same time, the actual wire feeding path, the wire feeding angle and the wire feeding speed are compared with the planned wire feeding path, the wire feeding angle and the wire feeding speed respectively, the deviation value is corrected, and the wire feeding path, the wire feeding angle and the wire feeding speed are optimized; the welding torch robot and the wire feeding robot perform collaborative work according to the optimized welding torch movement trajectory, wire feeding path, wire feeding angle and wire feeding speed.

[0015] Further, the wire feeding angle optimization method comprises:

[0016] According to the fact that the wire feeding direction is perpendicular to the welding torch movement direction, the wire feeding mechanism and the welding wire cannot interfere with the workpiece to be welded during the welding process, all feasible wire feeding directions are determined, and all feasible wire feeding directions are converted into wire feeding angles relative to the reference reference vector to obtain a wire feeding angle range;

[0017] According to the welding quality requirement, the system stability requirement and the posture flexibility requirement of the wire feeding robot, a welding quality objective function, a system stability objective function and a posture flexibility objective function of the wire feeding robot are determined;

[0018] Discretize the continuous wire feeding angle range, and normalize each objective function;

[0019] Based on the weighted combination of the normalized objective functions, a comprehensive score function is constructed, and finally the wire feeding angle with the maximum function value of the comprehensive score function is selected as the optimal wire feeding angle.

[0020] Further, the wire feeding angle range determination method comprises:

[0021] A local coordinate system is established with the end of the welding torch as the origin, the welding torch movement direction is defined as vector , which represents the direction of the welding torch moving along the weld; assuming that the welding torch is perpendicular to the weld surface, the welding torch posture vector is defined, which represents the direction of the welding torch axis; the wire feeding direction vector is defined, which represents the angle of the wire feeding into the arc; the wire feeding direction is geometrically constrained: , that is, the wire feeding direction vector is located in a plane perpendicular to Within the plane; the wire feeding mechanism and welding wire must not interfere with the workpiece to be welded during the welding process, then the intersection of the spatial volume of the workpiece to be welded and the swept volume of the wire feeding mechanism during its movement is empty, that is... , It is an empty set. The spatial volume of the parts to be welded. The volume swept during the movement of the wire feeding mechanism; while satisfying and In the plane with constraints, Starting from the reference datum, scan to both sides to find all wire feeding directions that satisfy the above constraints, and convert all feasible wire feeding directions to a value relative to the reference datum. The angle is determined by the maximum and minimum values ​​of the wire feed limits at the current welding position, thus defining the wire feed angle. Scope , Minimum wire feeding angle, This is the maximum wire feeding angle.

[0022] Furthermore, the objective function for welding quality Represented as: , The optimal wire feeding angle is determined empirically.

[0023] System stability objective function Represented as: ;

[0024] Objective function for the posture flexibility of the wire feeding robot Represented as: , For the first wire feeding robot The angle of each joint This refers to the joint number of the wire feeding robot. This refers to the number of joints in the wire-feeding robot. For the first Median range of motion of each joint , and The first The upper and lower limits of the range of motion of each joint.

[0025] Furthermore, the continuous wire feeding angle range is discretized, and each objective function is normalized; including:

[0026] First, set the continuous wire feeding angle range Discretized One candidate wire feeding angle , The objective function is then normalized, and the normalization function is as follows:

[0027] ;

[0028] ;

[0029] ;

[0030] in, for The normalization function, for The normalization function, for The normalization function, The wire feeding angle is The objective function for welding quality, The wire feeding angle is The system stability objective function, The wire feeding angle is The objective function for the posture flexibility of the wire feeding robot.

[0031] Furthermore, a comprehensive scoring function was constructed. Represented as: ,in These are the weighting coefficients. for The normalization function, for The normalization function, for The normalization function, The wire feeding angle is The objective function for welding quality, The wire feeding angle is The system stability objective function, The wire feeding angle is The objective function for the posture flexibility of the wire feeding robot.

[0032] The present invention also provides an electronic device, comprising:

[0033] Memory, used to store computer programs;

[0034] A processor for executing the computer program to implement the method.

[0035] The present invention also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the method described.

[0036] Beneficial effects: Compared with the prior art, the significant technical effects of this invention are as follows: It adopts a dual-robot collaborative architecture, equipped with a welding torch robot and a wire feeding robot. Both robots have independent motion control capabilities. The welding torch robot is mainly responsible for identifying the weld seam and planning the welding torch movement trajectory, while the wire feeding robot adjusts the wire feeding angle in real time according to parameters such as the part structure, welding position information and arc shape during the welding process. It has higher flexibility and collaborative capabilities and can realize operations under complex working conditions. Through path decoupling design, the end of the wire feeding mechanism can form a flexible working space around the welding torch path, realizing flexible wire feeding in multiple directions and angles. It overcomes the defects of traditional fixed wire feeding mechanisms that are prone to interference and poor adaptability in complex welding scenarios such as inside pipes and curved surfaces. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of a dual-robot working scenario in an embodiment of the present invention;

[0038] Figure 2 This is a partial schematic diagram of the operation of the two robots in an embodiment of the present invention;

[0039] Figure 3 This is a schematic diagram of the welding gun robot in an embodiment of the present invention;

[0040] Figure 4 This is a schematic diagram of the wire feeding robot in an embodiment of the present invention;

[0041] Figure 5 This is a schematic diagram of the welding method flow of the welding system in an embodiment of the present invention;

[0042] Figure 6 A schematic diagram of the interference condition of the wire feeding mechanism in a traditional welding system where the welding torch and the wire feeding mechanism are rigidly fixed.

[0043] Figure 7 This is a schematic diagram illustrating how interference is avoided by adjusting the position of the wire feeding mechanism in this invention;

[0044] Figure 8 A schematic diagram of interference conditions when performing internal pipe welding using a conventional welding system;

[0045] Figure 9 This is a schematic diagram illustrating how interference is avoided by adjusting the position of the wire feeding mechanism in this invention;

[0046] The diagram is labeled as follows: 1-Welding torch robot, 11-Welding torch, 12-Welding torch robotic arm, 13-Welding torch hydraulic lifting mechanism, 131-Welding torch main hydraulic lifting mechanism, 132-Welding torch auxiliary hydraulic lifting mechanism, 14-Welding torch integrated housing, 15-Welding torch mobile chassis, 2-Wire feeding robot, 21-Wire feeding mechanism, 22-Wire feeding robotic arm, 23-Wire feeding hydraulic lifting mechanism, 231-Wire feeding main hydraulic lifting mechanism, 232-Wire feeding auxiliary hydraulic lifting mechanism, 24-Wire feeding integrated housing, 25-Wire feeding mobile chassis. Detailed Implementation

[0047] The present invention will now be described in detail with reference to the accompanying drawings.

[0048] The present invention and its embodiments are described below. This description is not restrictive, and actual embodiments are not limited thereto. In short, if those skilled in the art are inspired by this description and, without departing from the spirit of the invention, design similar structures and embodiments to this technical solution, such designs should fall within the protection scope of the present invention.

[0049] To further enhance manufacturing flexibility, this invention employs a dual-robot collaborative welding system that decouples the wire feeding mechanism from the welding torch, allowing for independent path and posture planning. One robotic arm holds the argon arc welding torch at its end, while the other holds the wire feeding mechanism. The welding torch robot positions the weld seam and determines the torch's trajectory. The wire feeding robot plans the wire feeding path, angle, and speed based on the workpiece's shape, the torch's position, and the welding speed, achieving flexible wire feeding, improving welding quality, and holding significant importance for welding complex components.

[0050] Example 1:

[0051] like Figures 1 to 4 As shown, this embodiment discloses a mobile dual-robot collaborative non-consumable electrode gas shielded welding system, including a welding torch robot 1, a wire feeding robot 2, and a collaborative control system. Figure 1 This demonstrates a collaborative working scenario of two robots welding a workpiece. Figure 2 This is a partial schematic diagram of the end effector of the dual-robot operation, illustrating the working positions of the welding torch 11 and the wire feeding mechanism 21 at the welding point during welding. Figure 3 and Figure 4 The specific structures of the welding gun robot and the wire feeding robot were shown separately.

[0052] like Figure 3As shown, the welding torch robot 1 includes: a welding torch 11, a welding torch robotic arm 12, a welding torch hydraulic lifting mechanism 13, a welding torch integrated housing 14, and a welding torch mobile chassis 15. The welding torch hydraulic lifting mechanism 13 includes a welding torch main hydraulic lifting mechanism 131 and a welding torch auxiliary hydraulic lifting mechanism 132. The welding torch 11 is mounted at the end of the welding torch robotic arm 12 and is responsible for performing welding. The welding torch robotic arm 12 is mounted at the end of the welding torch auxiliary hydraulic lifting mechanism 132. It is a 6-axis robotic arm responsible for adjusting the posture of the welding torch and positioning it in a precise space, ensuring that the welding torch can be aligned with the weld at the optimal angle. The welding torch main hydraulic lifting mechanism 131 is fixedly mounted on the welding torch moving chassis 15, which can realize a wide range of vertical position adjustment of the welding torch to adapt to welds of different heights. A U-shaped connector is fixed at the top of the welding torch main hydraulic lifting mechanism 131. The welding torch auxiliary hydraulic lifting mechanism 132 is fixedly mounted on the connector by wall-mounting, realizing small-range fine adjustment. The welding torch integrated housing 14 is fixed on the welding torch moving chassis 15. It integrates the drive controller, servo motor, gas cylinder and other components of the welding torch robot 1 to realize power supply and local control. The welding torch moving chassis 15 can move flexibly around large and complex workpieces, expanding the robot's workspace.

[0053] like Figure 4 As shown, the wire feeding robot 2 includes: a wire feeding mechanism 21, a wire feeding robotic arm 22, a wire feeding hydraulic lifting mechanism 23, a wire feeding integrated box 24, and a wire feeding mobile chassis 25. The wire feeding hydraulic lifting mechanism 23 includes a main wire feeding hydraulic lifting mechanism 231 and a secondary wire feeding hydraulic lifting mechanism 232. The wire feeding mechanism 21 is installed at the end of the wire feeding robotic arm 22 and is responsible for feeding the wire. The wire feeding robotic arm 22 is installed at the end of the auxiliary wire feeding hydraulic lifting mechanism 232. It is a 6-axis robotic arm and is responsible for adjusting the posture of the wire feeding mechanism and positioning it in a fine space to ensure that the welding wire can be fed into the welding space at the optimal angle. The main wire feeding hydraulic lifting mechanism 231 is fixedly installed on the wire feeding moving chassis 25 and can realize a large range of vertical position adjustment of the wire feeding mechanism 21 to adapt to welds of different heights following the welding torch 11. A U-shaped connector is fixed at the top of the main wire feeding hydraulic lifting mechanism 231. The auxiliary wire feeding hydraulic lifting mechanism 232 is installed on the connector by hanging it on the wall to realize small-range fine adjustment. The wire feeding integrated box 24 is fixed on the wire feeding moving chassis 25 and integrates the control system, drive motor and wire feeding speed regulation system of the wire feeding robot 2. The wire feeding moving chassis 25 has the same structure and function as the welding torch moving chassis 15 and can move flexibly around large and complex workpieces, expanding the robot's workspace.

[0054] The wire feeding robot is independent of the welding torch robot and can dynamically adjust the wire feeding angle and speed. A vision sensor is installed on the wire feeding mechanism to collect data such as weldment parameters, molten pool morphology, and wire extension length in real time, transmitting this data to the collaborative control system. The collaborative control system can adjust the wire feeding parameters based on feedback from the vision sensor.

[0055] In this embodiment, welding torch robot 1 is responsible for holding the welding torch 11, and wire feeding robot 2 holds the wire feeding mechanism 21. The two mobile robotic arms (welding torch robotic arm and wire feeding robotic arm) operate independently, improving the flexibility of the mobile robot's operation. The welding torch used in this embodiment is a tungsten inert gas (TIG) welding torch.

[0056] The collaborative control system includes a communication module, a motion planning module, and a sensor perception module. The sensor perception module includes a first vision sensor mounted on the welding torch robot and a second vision sensor mounted on the wire feeding robot. The operator issues welding tasks on the host computer, the motion planning module performs collaborative path planning and welding parameter settings, and transmits the data to welding robot 1 and wire feeding robot 2 through the communication module. During welding, the vision sensors on the two robots monitor the welding status and the surrounding environment in real time and transmit the data back to the motion planning module. The motion planning module compares the actual values ​​with the ideal values, corrects the deviation values, and optimizes the motion trajectory of the welding torch and the wire feeding mechanism.

[0057] Specifically: The collaborative control system plans the welding speed, welding torch trajectory, and pose based on welding requirements and identified weld seam data, ensuring that welding torch robot 1 moves according to a strict empty motion trajectory and speed. Simultaneously, based on the identified workpiece shape, welding torch pose, and welding speed, the collaborative control system plans the wire feeding path, wire feeding angle, and wire feeding speed, ensuring the wire feeding nozzle is always maintained at the optimal wire feeding direction and distance relative to the welding torch. Welding torch robot 1 and wire feeding robot 2 do not communicate directly; their collaborative relationship is established by the collaborative control system.

[0058] Before welding, the motion planning module plans the welding speed according to welding requirements, the welding torch trajectory and posture based on the 3D point cloud data of the weld seam, and the wire feeding path and angle based on the structural shape of the workpiece and the welding torch posture. Simultaneously, it plans the wire feeding speed based on the welding speed. During welding, the first vision sensor on the welding torch robot 1 identifies the 3D point cloud data of the weld seam in real time and calculates the relative position between the weld seam centerline and the end of the welding torch, sending this information to the motion planning module. The motion planning module calculates the actual motion trajectory of the welding torch using the relative position information, compares it with the planned trajectory, corrects deviations, and optimizes the welding torch trajectory. Simultaneously, it... The second vision sensor on the wire feeding robot identifies the real-time structural shape of the workpiece to be welded, the real-time pose of the welding torch, the real-time morphology of the molten pool, and the real-time extension length of the welding wire. It obtains the actual wire feeding path, wire feeding angle, and wire feeding speed, and transmits them to the motion planning module through the communication module. The motion planning module compares the actual wire feeding path, wire feeding angle, and wire feeding speed with the planned wire feeding path, wire feeding angle, and wire feeding speed, corrects the deviation values, and optimizes the wire feeding path, wire feeding angle, and wire feeding speed. The optimized welding torch motion trajectory, wire feeding path, wire feeding angle, and wire feeding speed are sent to the welding torch robot and the wire feeding robot, respectively. The two robots work together to ensure welding quality.

[0059] Example 2:

[0060] like Figure 5 As shown, the welding method using the aforementioned welding system includes the following steps:

[0061] First, initialize both robots (welding gun robot and wire feeding robot), returning them to zero and ensuring the system is in normal working condition. After the collaborative control system issues a welding command, the two robots start synchronously.

[0062] The welding torch robot moves to the starting point of the weld, uses visual scanning to acquire the three-dimensional point cloud data of the weld, and uploads it to the collaborative control system; the wire feeding robot moves to the working area, identifies the structural shape of the workpiece to be welded, and uploads it to the collaborative control system.

[0063] The collaborative control system plans the welding torch movement trajectory that strictly follows the center line of the weld seam based on the three-dimensional point cloud data of the weld seam, and calculates the matching welding torch posture and welding current, voltage and other parameters. It plans the welding speed of the welding torch robot according to the welding requirements, and plans the wire feeding path and wire feeding angle according to the structural shape of the workpiece and the welding torch posture. At the same time, it plans the wire feeding speed according to the welding speed.

[0064] The collaborative control system sends the planned welding torch trajectory and welding speed to the welding torch robot, and at the same time sends the planned wire feeding path, wire feeding angle and wire feeding speed to the wire feeding robot, and the two robots cooperate in welding.

[0065] During welding, the first vision sensor on the welding torch robot identifies the real-time 3D point cloud data of the weld seam, while the second vision sensor on the wire feeding robot simultaneously identifies the real-time structural shape of the workpiece, the real-time molten pool morphology, and the real-time extension length of the welding wire. This data is then transmitted to the motion planning module. Upon receiving the data, the motion planning module calculates the relative position of the weld seam centerline and the end of the welding torch based on the real-time 3D point cloud data, thus obtaining the actual motion trajectory of the welding torch. Simultaneously, the motion planning module determines the actual wire feeding path based on the real-time pose of the welding torch, the real-time structural shape of the workpiece, and the real-time extension length of the welding wire. The wire feeding angle is calculated, and the actual wire feeding speed is obtained based on the real-time molten pool morphology and the real-time extension length of the welding wire. The collaborative control system compares the actual movement trajectory of the welding torch with the planned movement trajectory, corrects the deviation value, and optimizes the movement trajectory. At the same time, the actual wire feeding path, actual wire feeding angle, and actual wire feeding speed are compared with the planned wire feeding path, wire feeding angle, and wire feeding speed, respectively, correct the deviation value, and optimize the wire feeding path, wire feeding angle, and wire feeding speed. The welding torch robot and the wire feeding robot work collaboratively based on the optimized welding torch movement trajectory, wire feeding path, wire feeding angle, and wire feeding speed.

[0066] The welding torch robot 1 can move freely along the trajectory of the welding torch to perform welding. All data during the welding process are transmitted to the collaborative control system, which adjusts the welding parameters of the welding torch robot 1 based on the feedback information. Specifically, the collaborative control system makes dynamic decisions by comparing the preset welding process parameters with the real-time welding status fed back by the welding torch robot. For example, when the width of the molten pool is detected to be less than the preset value, the system determines that the heat input is insufficient and then dynamically adjusts the welding parameters, appropriately increasing the welding current or decreasing the welding speed to ensure that the welding quality meets the requirements.

[0067] The wire feeding robot 2 can move along the wire feeding path to accurately feed the welding wire into the welding area. After receiving the welding task, the collaborative control system will dynamically calculate the feasible region of the wire feeding angle and determine the optimal wire feeding direction within the feasible region. The wire feeding mechanism 21 is equipped with a second vision sensor to monitor the wire feeding process and can collect parameters such as wire feeding angle and speed in real time. If an abnormality is found in the wire feeding process or a complex part structure is encountered, the collaborative control system will recalculate the wire feeding angle, optimize the wire feeding parameters and the wire feeding path, and ensure the stability and accuracy of wire feeding.

[0068] Methods for optimizing wire feeding angle include:

[0069] First, based on the principle that the wire feeding direction is perpendicular to the welding torch movement direction, and that the wire feeding mechanism and welding wire must not interfere with the workpiece during welding, all feasible wire feeding directions are determined. These feasible wire feeding directions are then converted into wire feeding angles relative to a reference vector, thus obtaining the range of wire feeding angles. Specifically:

[0070] Establish a local coordinate system with the tip of the welding torch as the origin, and define the direction of the welding torch's movement as a vector. This indicates the direction in which the welding torch moves along the weld; assuming the welding torch is perpendicular to the weld surface, the welding torch attitude vector is defined. , indicates the direction of the welding torch axis; defines the wire feed direction vector. This indicates the angle at which the welding wire is fed into the electric arc. To ensure that the welding wire is stably fed into the molten pool, the wire feeding direction must be geometrically constrained. The wire feeding direction should be perpendicular to the direction of the welding torch movement, i.e., satisfying... Therefore, the wire feeding direction vector It should be located in a perpendicular to Within the plane. The wire feeding mechanism and welding wire must not interfere with the workpiece during welding; therefore, the intersection of the spatial volume of the workpiece and the swept volume of the wire feeding mechanism during its movement must be empty. Let the spatial volume of the workpiece be... The volume swept during the movement of the wire feeding mechanism is Then there is , Let be an empty set, and in a plane satisfying the above conditions, let Starting from the reference datum, scan to both sides to find all wire feeding directions that satisfy the above constraints, and convert all feasible wire feeding directions to a value relative to the reference datum. The angle is given by the maximum and minimum values, which represent the wire feed limits at the current welding position. The wire feed angle can then be determined from these values. Scope , Minimum wire feeding angle, This is the maximum wire feeding angle.

[0071] Secondly, based on the welding quality requirements, system stability requirements, and wire feeding robot posture flexibility requirements, the objective functions for welding quality, system stability, and wire feeding robot posture flexibility are determined; specifically:

[0072] Within the determined range of wire feeding angles, the collaborative control system can select an optimal wire feeding angle based on constraints. First, the collaborative control system should prioritize ensuring welding quality. For specific weld types and welding processes, there exists an empirically optimal wire feed angle. The welding process is usually provided directly from the welding process library. During the optimization process, the selected process should be considered. as close as possible Therefore, a welding quality objective function is proposed. The smaller the function value, the higher the theoretical welding quality. Secondly, the location least likely to experience collisions or process fluctuations should be selected, i.e., the point farthest from the two limit boundaries. This provides the maximum safety margin for the robot and wire feeding mechanism in the event of minor vibrations or workpiece positioning errors. Based on this, the system stability objective function is proposed. The larger this function value, the more stable the system. Finally, the posture of the wire-feeding robot should be within its flexible working range, which can be calculated based on the robot's inverse kinematics model. Angle of each joint This paper proposes an objective function for the posture flexibility of a wire feeding robot. The function is defined as the sum of the squares of the deviations of all joint angles from their respective joint space medians, where This refers to the joint number of the wire feeding robot. This refers to the number of joints in the wire-feeding robot. For the first Median range of motion of each joint , and The first The upper and lower limits of the range of motion of each joint. The smaller the value, the closer all joint angles are to the midpoint of their own range of motion, resulting in greater flexibility and rigidity.

[0073] Then, the continuous wire feeding angle range is discretized, and each objective function is normalized; specifically:

[0074] In the decision-making process, first define the continuous range Discretized One candidate wire feeding angle , The objective function is then normalized for comparison purposes. The normalization function is as follows:

[0075] ;

[0076] ;

[0077] ;

[0078] in, for The normalization function, for The normalization function, for The normalization function, The wire feeding angle is The objective function for welding quality, The wire feeding angle is The system stability objective function, The wire feeding angle is The objective function for the posture flexibility of the wire feeding robot.

[0079] Function Transform into , The larger the function value, the better.

[0080] Finally, based on the weighted combination of the normalized objective functions, a comprehensive scoring function is constructed, and the wire feeding angle with the largest comprehensive scoring function value is selected as the optimal wire feeding angle; specifically:

[0081] Construct a comprehensive scoring function ,in As a weighting coefficient, in high-quality welding, it is set Final choice The function with the largest value As the optimal wire feeding angle .

[0082] like Figures 6 to 9 As shown, when the workpiece to be welded has a complex structure or is inside a welding pipe, the traditional fixed wire feeding mechanism is difficult to adjust its posture and is prone to interference with the workpiece. However, the dual-robot collaborative welding system designed in this invention can flexibly adjust the direction and angle of the wire feeding mechanism by decoupling the welding torch from the wire feeding mechanism, thus avoiding interference.

[0083] The present invention also provides an electronic device, comprising:

[0084] Memory, used to store computer programs;

[0085] A processor for executing the computer program to implement the method.

[0086] The present invention also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the method described.

[0087] The above description is merely a preferred embodiment of the present invention and is not intended to further limit the present invention. All equivalent changes made based on the description and drawings of the present invention are within the protection scope of the present invention.

Claims

1. A mobile dual-robot collaborative non-consumable electrode gas shielded welding system, characterized in that, The system includes a welding torch robot, a wire feeding robot, and a collaborative control system. During the welding process, the collaborative control system identifies the 3D point cloud data of the weld seam in real time and calculates the actual trajectory of the welding torch based on this data. It then compares the actual trajectory with the planned trajectory, corrects deviations, and optimizes the welding torch's trajectory. Simultaneously, the collaborative control system identifies the structural shape of the workpiece, the molten pool morphology, and the wire extension length in real time. Based on these parameters, it calculates the actual wire feeding path, angle, and speed. The actual wire feeding path, angle, and speed are then compared with the planned paths, angles, and speeds to correct deviations and optimize these parameters. The collaborative control system coordinates the welding torch robot and the wire feeding robot to perform welding operations based on the optimized welding torch trajectory, wire feeding path, angle, and speed.

2. The mobile dual-robot collaborative non-consumable electrode gas shielded welding system according to claim 1, characterized in that, The collaborative control system includes a communication module, a motion planning module, and a sensor perception module. The sensor perception module is used to monitor the actual operating data of the welding robot and the wire feeding robot, and sends the actual operating data to the motion planning module. The motion planning module compares the actual data with the planned data, corrects the deviation value, and optimizes the welding torch trajectory, wire feeding path, wire feeding angle, and wire feeding speed. The motion planning module transmits the optimized data to the welding robot and the wire feeding robot via the communication module.

3. A welding method for a mobile dual-robot collaborative non-consumable electrode gas shielded welding system as described in any one of claims 1-2, characterized in that, Includes the following steps: Before welding, the welding gun robot and wire feeding robot are initialized, and the collaborative control system issues a synchronous start command. The welding torch robot moves to the starting point of the weld, and the collaborative control system uses visual scanning to acquire the three-dimensional point cloud data of the weld; the wire feeding robot moves to the working area, and the collaborative control system identifies the structural shape of the workpiece to be welded and the extension length of the welding wire. The collaborative control system plans the welding torch motion trajectory and welding torch posture based on the three-dimensional point cloud data of the weld seam, plans the welding speed of the welding torch robot according to the welding requirements, and plans the wire feeding path and wire feeding angle according to the structural shape of the workpiece to be welded and the welding torch posture. At the same time, it plans the wire feeding speed according to the welding speed. The collaborative control system sends the planned welding torch trajectory and welding speed to the welding torch robot, and at the same time sends the planned wire feeding path, wire feeding angle and wire feeding speed to the wire feeding robot, and the two robots cooperate in welding. During the welding process, the collaborative control system identifies the real-time 3D point cloud data of the weld seam, as well as the real-time structural shape of the workpiece, the real-time molten pool morphology, and the real-time extension length of the welding wire. Based on the real-time 3D point cloud data of the weld seam, it calculates the actual motion trajectory and real-time pose of the welding torch. Simultaneously, based on the real-time pose of the welding torch, the real-time structural shape of the workpiece, and the real-time extension length of the welding wire, it calculates the actual wire feeding path and angle, and the actual wire feeding speed based on the real-time molten pool morphology and the real-time extension length of the welding wire. The collaborative control system compares the actual motion trajectory of the welding torch with the planned trajectory, corrects deviations, and optimizes the welding torch motion trajectory. Simultaneously, it compares the actual wire feeding path, angle, and speed with the planned wire feeding path, angle, and speed, corrects deviations, and optimizes the wire feeding path, angle, and speed. The welding torch robot and the wire feeding robot then perform collaborative operations based on the optimized welding torch motion trajectory, wire feeding path, angle, and speed.

4. The welding method according to claim 3, characterized in that, Methods for optimizing wire feeding angle include: Based on the fact that the wire feeding direction is perpendicular to the welding gun movement direction, and that the wire feeding mechanism and welding wire cannot interfere with the parts to be welded during the welding process, all feasible wire feeding directions are determined, and all feasible wire feeding directions are converted into wire feeding angles relative to the reference vector to obtain the range of wire feeding angles. Based on the welding quality requirements, system stability requirements, and wire feeding robot posture flexibility requirements, the welding quality objective function, system stability objective function, and wire feeding robot posture flexibility objective function are determined. Discretize the continuous range of wire feeding angles and normalize each objective function; Based on the weighted combination of the normalized objective functions, a comprehensive scoring function is constructed, and the wire feeding angle with the largest comprehensive scoring function value is finally selected as the optimal wire feeding angle.

5. The welding method according to claim 3, characterized in that, Methods for determining the wire feeding angle range include: Establish a local coordinate system with the tip of the welding torch as the origin, and define the direction of the welding torch's movement as a vector. This indicates the direction in which the welding torch moves along the weld; assuming the welding torch is perpendicular to the weld surface, the welding torch attitude vector is defined. , indicates the direction of the welding torch axis; defines the wire feed direction vector. This indicates the angle at which the welding wire is fed into the electric arc; geometric constraints are applied to the wire feeding direction: That is, the wire feeding direction vector Located in a perpendicular to Within the plane; the wire feeding mechanism and welding wire must not interfere with the workpiece to be welded during the welding process, then the intersection of the spatial volume of the workpiece to be welded and the swept volume of the wire feeding mechanism during its movement is empty, that is... , It is an empty set. The spatial volume of the parts to be welded. The volume swept during the movement of the wire feeding mechanism; while satisfying and In the plane of constraints, with Starting from the reference datum, scan to both sides to find all wire feeding directions that satisfy the above constraints, and convert all feasible wire feeding directions to a value relative to the reference datum. The angle is determined by the maximum and minimum values ​​of the wire feed limits at the current welding position, thus defining the wire feed angle. Scope , Minimum wire feed angle, This is the maximum wire feeding angle.

6. The welding method according to claim 3, characterized in that, Welding quality objective function Represented as: , The optimal wire feeding angle is determined empirically. System stability objective function Represented as: ; Objective function for the posture flexibility of the wire feeding robot Represented as: , For the first wire feeding robot The angle of each joint This refers to the joint number of the wire feeding robot. This refers to the number of joints in the wire-feeding robot. For the first Median range of motion of each joint , and The first The upper and lower limits of the range of motion of each joint.

7. The welding method according to claim 3, characterized in that, Discretize the continuous wire feeding angle range and normalize each objective function; including: First, set the continuous wire feeding angle range Discretized One candidate wire feeding angle , The objective function is then normalized, and the normalization function is as follows: ； ； ； in, for The normalization function, for The normalization function, for The normalization function, The wire feeding angle is The objective function for welding quality, The wire feeding angle is The system stability objective function, The wire feeding angle is The objective function for the posture flexibility of the wire feeding robot.

8. The welding method according to claim 3, characterized in that, Constructed comprehensive scoring function Represented as: ,in These are the weighting coefficients. for The normalization function, for The normalization function, for The normalization function, The wire feeding angle is The objective function for welding quality, The wire feeding angle is The system stability objective function, The wire feeding angle is The objective function for the posture flexibility of the wire feeding robot.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the method as described in any one of claims 3-8.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the method described in any one of claims 3-8.

Images