WELDING PATH AUTONOMOUS OPTIMIZATION DEVICE, WELDING PATH AUTONOMOUS OPTIMIZATION SYSTEM, AND WELDING PATH AUTONOMOUS OPTIMIZATION METHOD

Abstract

To autonomously optimize welding process information of a welding robot.SOLUTION: A welding pass autonomous optimization system 7 comprises: a quality-welding condition correlation database 207 that stores a welding constraint of a welding robot for operating a welding torch; an interference analysis unit 34 that determines a welding path not interfering with an object to be welded on the basis of three-dimensional CAD data 13 of the object and the constraint of the quality-welding condition correlation database 207; and a welding robot operation / welding condition output unit 35 that outputs welding process information on the basis of the welding pass.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The present invention relates to a welding path autonomous optimization device, a welding path autonomous optimization system, and a welding path autonomous optimization method. [Background technology]

[0002] In recent years, automatic calculation methods for welding paths using three-dimensional CAD (computer-aided design) have come to be used when designing welding processes. As typified by Patent Document 1, offline teaching analysis is widely used to avoid interference between the welding torch and the workpiece to be welded. However, detailed welding conditions such as torch operation and welding power source, which are important for quality such as defects in line welds such as arc welding, bead shape, and surface properties, have not been determined, and condition optimization work based on actual equipment is required after the welding process design.

[0003] In addition, in most cases, the results of the simulation of offline teaching performed on a computer during welding process design do not match the positional relationship between the robot and the non-welded workpiece at the actual welding site. Attempts have been made to correct the gap between the positional relationship of the offline teaching and the actual welding by using a three-dimensional measuring machine to correct the coordinates before welding.

[0004] However, a required feature of products that have undergone the arc welding process is a weld bead shape that is free of defects and has good surface properties, so it seems difficult to optimize the quality of the welds using the robot motion coordinate analysis that has been carried out so far in the above offline teaching to avoid interference between the welding torch and non-welded workpieces.

[0005] Meanwhile, in TIG (Tungsten Inert Gas) welding, a method has been described in which a camera is attached to the welding torch and image processing is performed to quantify the movement of the welding torch, the amount of welding material fed, and the welding power source performance, and an attempt has been made to create a database of welding skills. By using the above-mentioned expert skill database, it becomes possible to predict the effects on the quality of the welded joint of the workpiece before and during welding, the movement of the torch during welding, the welding material, and the welding power source. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2012-24867 A Summary of the Invention [Problem to be solved by the invention]

[0007] The method of Patent Document 1 is characterized in that, by simply inputting the relative position information of the gun held by the robot in spot welding and the non-welded workpiece, and the welding point information for each welding point, the idle points including the entry and exit positions of the welding gun in the welding operation are automatically determined. However, the method of Patent Document 1 cannot be applied to arc welding, which performs continuous line welding, or as a pre-evaluation tool for obtaining a weld bead shape with excellent surface quality and no defects.

[0008] As an attempt to correct the gap between the positional relationship of offline teaching and actual welding, there is a teaching method for a welding robot that can correct the welding point position of the teaching data and prevent the accuracy of the teaching data from decreasing, taking into account the variation of the welding gun and the non-welded workpiece at the actual spot welding site. However, this teaching method cannot be applied to arc welding that performs continuous line welding, and cannot be used as a pre-evaluation tool for obtaining a weld bead shape without defects and excellent surface properties.

[0009] Thus, in offline teaching of welding robots for high-quality construction of arc welds, a system is required that combines calculations using numerical coordinate systems that avoid interference in the geometry information obtained from CAD data with a database of factors that affect arc welding quality that have been quantified in attempts to create a database of welding expertise, thereby filling in the gap between offline teaching and the various variations that occur during actual welding.

[0010] Therefore, an object of the present invention is to autonomously optimize welding process information for a welding robot. [Means for solving the problem]

[0011] In order to solve the above-mentioned problems, the welding path autonomous optimization device of the present invention includes a database storing constraint conditions for welding by a welding robot that operates a welding torch, an interference analysis unit that determines a welding path that does not interfere with an object to be welded based on three-dimensional CAD data of the object to be welded and the constraint conditions of the database, and an output unit that outputs welding process information based on the welding path. a pre-welding process information correction unit that obtains three-dimensional shape information and a groove shape of the workpiece before welding by the three-dimensional scanner, identifies a difference with the three-dimensional CAD data, and corrects the welding process information; The present invention is characterized by being equipped with:

[0012] The welding path autonomous optimization system of the present invention includes a welding robot that operates a welding torch, a database that stores constraints on welding by the robot, an interference analysis unit that determines a welding path that does not interfere with an object to be welded based on three-dimensional CAD data of the object to be welded and the constraints in the database, and an output unit that outputs welding process information based on the welding path. 、 A three-dimensional scanner that detects the shape of a workpiece, and a pre-welding process information correction unit that acquires three-dimensional shape information and a groove shape of the workpiece before welding by the three-dimensional scanner, identifies a difference with the three-dimensional CAD data, and corrects the welding process information. The present invention is characterized by being equipped with:

[0013] The method for autonomously optimizing a welding path of the present invention includes a step of determining a welding path that does not interfere with an object to be welded based on three-dimensional CAD data of the object to be welded and a constraint condition for welding by a welding robot, and a step of outputting welding process information based on the welding path. acquiring three-dimensional shape information of an object to be welded, the three-dimensional shape information including at least one of three-dimensional shape information of a workpiece before welding and groove shape information; identifying a difference between the three-dimensional CAD data and the three-dimensional shape information of the object to be welded; and correcting the welding process information based on the difference. The present invention is characterized by being equipped with: Other means will be described in the description of the embodiment of the invention. Effect of the Invention

[0014] According to the present invention, it becomes possible to autonomously optimize the welding process information of a welding robot. [Brief description of the drawings]

[0015] [Figure 1] 1 is an overall view of a welding path autonomous optimization system according to an embodiment of the present invention. FIG. [Diagram 2] FIG. 2 is a diagram illustrating a hardware configuration around a control device and an information processing device. [Diagram 3] FIG. 1 is a diagram illustrating a configuration of an information processing device. [Figure 4] FIG. 2 is a diagram showing the logical configuration and operation of a welding path autonomous optimization system. [Diagram 5] FIG. 2 is a diagram for explaining a torch tilt angle θ and a welding target position. [Figure 6] FIG. 4 is a diagram illustrating a torch sweep angle. [Figure 7] FIG. 4 is a diagram illustrating a torch advance angle. [Figure 8] 1 is a graph showing a correlation between welding performance information and welding quality information. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. FIG. 1 is an overall view of an autonomous welding path optimization system 7 according to this embodiment. The welding path autonomous optimization system 7 includes a control device 1 and an information processing device 2. The information processing device 2 can access three-dimensional CAD data 13 and a quality-welding condition correlation database 207, and is further communicatively connected to the control device 1. The information processing device 2 determines a welding path of a welding object related to the three-dimensional CAD data 13 offline and calculates welding process information.

[0017] The control device 1 is connected to a three-dimensional camera 10, a laser displacement meter 11, a molten pool observation camera 12, a digital welding machine 5, a robot control device 6, and an operation device 15. The control device 1 retrieves the welding path determined under the constraint conditions as time-series data of the three-dimensional coordinates of the tip of a welding torch 51, and performs the actual welding.

[0018] The geometric data of the robot 61 and the workpiece to be welded contained in the three-dimensional CAD data 13 and matters related to the movement of the welding torch 51, such as the torch inclination angle and the distance between the torch tip and the workpiece, which are related to the welding quality contained in the quality-welding condition correlation database 207, are constraints on welding.

[0019] The information processing device 2 calculates a welding path including an optimal movement of the welding torch 51 under the welding constraint conditions, and determines the torch tip operation taking into consideration the welding quality in offline teaching. The welding path determined under the welding constraint conditions is called up as time-series data of the three-dimensional coordinates of the tip of the welding torch 51 from the control device 1 by using the operation device 15. The robot 61 operates the welding torch 51 .

[0020] Control device 1 controls digital welding machine 5 and robot control device 6 before and during welding based on measurement values ​​(welding information) measured by various sensor devices exemplified below, in relation to a time-series database of the three-dimensional coordinates of the tip of welding torch 51, and corrects the welding path using welding torch 51 and robot 61 or, if necessary, positioner 62. The various devices below and control device 1 are each connected by a connection cable (not shown).

[0021] Digital welding machine 5 is capable of controlling the current, voltage, arc length, and wire feed speed during welding using the analog output (voltage or current) from control device 1.

[0022] The robot control device 6 is communicatively connected to the control device 1 and the information processing device 2 via a LAN (Local Area Network) cable and a LAN port in order to control the coordinates of the robot 61 and the positioner 62, and protocol communication is established.

[0023] The three-dimensional camera 10 photographs the space of 0.01 to 5 m including the welding torch 51, the robot 61, and the workpieces to be welded before welding. This three-dimensional camera 10 functions as a three-dimensional scanner that detects the shape of the workpieces. The control device 1 grasps the positional relationship between these three-dimensionally based on the images captured by the three-dimensional camera 10, and corrects the time series data of the three-dimensional coordinates determined using the information processing device 2. This three-dimensional camera 10 is connected to the control device 1 so as to be able to communicate with each other, and protocol communication has been established.

[0024] The laser displacement meter 11 three-dimensionally measures the positional relationship between the welding torch 51, the workpiece to be welded, and the groove of the workpiece to be welded in a space of 0.01 mm to 500 mm before welding. This control device 1 corrects the time series data of the three-dimensional coordinates determined using the three-dimensional camera 10 and the control device 1 based on the position measured by the laser displacement meter 11. The laser displacement meter 11 is connected to the control device 1 so as to be able to communicate with each other, and protocol communication is established. This laser displacement meter 11 is a displacement meter that identifies the relative position between the welding robot and the groove.

[0025] The molten pool observation camera 12 photographs the tip of the welding torch 51 and the 0.01 to 50 mm space around the groove of the workpiece to be welded during welding. The control device 1 grasps the three-dimensional positional relationship between them from the images photographed by the molten pool observation camera 12, and corrects the time series data of the three-dimensional coordinates determined by the laser displacement meter 11 and the control device 1. The molten pool observation camera 12 is connected to the control device 1 so as to be able to communicate with it, and protocol communication has been established.

[0026] The present invention is not limited to a system in which a plurality of devices operate in cooperation with each other, but may be an autonomous welding path optimization device in which the operation is completed within a single device, and is not limited thereto.

[0027] FIG. 2 is a diagram showing a hardware configuration around the control device 1 and the information processing device 2. As shown in FIG. The control device 1 is configured as, for example, a PLC (Programmable Logic Controller) and includes a CPU unit 101, a protocol communication unit 102 which is an additional unit, a database connection unit 103, an analog input unit 104, and an analog output unit 105. The PLC constituting the control device 1 can expand its functions by attaching additional units to the CPU unit 101.

[0028] The protocol communication unit 102 communicates with the robot control device 6, the three-dimensional camera 10, the laser displacement meter 11, and the molten pool observation camera 12 via Ethernet (registered trademark).

[0029] The database connection unit 103 communicates with a quality-welding condition correlation database 207 and a computer which is an information processing device 2 . The analog input unit 104 converts analog input of the voltage or current of the digital welding machine 5 into digital data.

[0030] The analog output unit 105 converts the digital data of the digital welding machine 5 into a voltage or current and controls the analog output. These digital data can be viewed and monitored via the hub 206 using the monitor display 21 for the information processing device 2 or a tablet terminal which is the operation device 15 for the control device 1 as an HMI (Human Machine Interface).

[0031] The welding information input to the control device 1 is the current, voltage, arc length, and wire feed speed input from the digital welding machine 5 , and is obtained through an analog input unit 104 . In addition, time series data of the three-dimensional coordinates of the tip of the welding torch 51 input from the robot control device 6, and coordinate information of the welding torch 51, the workpiece to be welded, and the welded workpiece groove input from the three-dimensional camera 10 are obtained through a protocol communication unit 102 using Ethernet.

[0032] Furthermore, information on the welding torch 51, the workpiece to be welded, and the groove of the workpiece to be welded input from the laser displacement meter 11 is obtained through an analog input unit 104. Finally, coordinate information on the tip of the welding torch 51 and the groove of the workpiece to be welded input from the molten pool observation camera 12 is obtained through a protocol communication unit 102 using Ethernet (registered trademark).

[0033] Information relating to welding is temporarily stored in the storage in the CPU unit 101 in the control device 1, and is then stored in the quality-welding condition correlation database 207 through the database connection unit 103. By reading any value from the quality-welding condition correlation database 207 using the control device 1 after welding, the control device 1 can also control the operation of the digital welding machine 5 and robot control device 6, and the imaging conditions of the 3D camera 10, laser displacement meter 11, and molten pool observation camera 12.

[0034] FIG. 3 is a diagram showing the configuration of the information processing device 2. As shown in FIG. The information processing device 2 is configured as a computer having a CPU (Central Processing Unit) 201, a RAM (Random Access Memory) 202, a ROM (Read Only Memory) 203, a HDD (Hard Disk Drive) 204, and a communication interface 205.

[0035] The communication interface 205 is connected to a hub 206, which is an external communication device, and exchanges information with a quality-welding condition correlation database 207. The three-dimensional CAD data 13 is stored in the HDD 204, and motion analysis software 209 is installed therein. Data exchange and software execution are performed within the information processing device 2 as necessary.

[0036] FIG. 4 is a diagram showing the logical configuration and operation of the welding path autonomous optimization system 7. The welding path autonomous optimization system 7 has a welding process information creation unit 3, a pre-welding process information correction unit 8, and an in-welding process information correction unit 9. The welding path autonomous optimization system 7 executes a welding path autonomous optimization method. The welding path autonomous optimization method includes a step of determining a welding path that does not interfere with a welding object based on three-dimensional CAD data 13 of the welding object and constraint conditions for welding by a welding robot, and a step of outputting welding process information based on the welding path.

[0037] Welding process information creation unit 3 has a quality-welding condition correlation database 207, a constraint condition addition unit 33 such as a torch tilt angle / target position, a weld shape input unit 31, a welding condition input unit 32, three-dimensional CAD data 13, motion analysis software 209, an interference analysis unit 34, and a welding robot operation / welding condition output unit 35. Quality-welding condition correlation database 207 must be constructed before executing welding path autonomous optimization system 7. Moreover, interference analysis unit 34 is embodied by CPU 201 executing motion analysis software 209.

[0038] Constraint condition adding unit 33 outputs information such as the torch tilt angle θ and the distance between the torch tip and the workpiece from the viewpoint of the quality of the weld for the quality-welding condition correlation database 207 and the weld shape and welding conditions input by weld shape input unit 31 and welding condition input unit 32. The welding constraint conditions output by constraint condition adding unit 33 are combined with three-dimensional CAD data 13 to form constraint conditions for interference analysis unit 34. Interference analysis unit 34 performs motion interference analysis under these constraint conditions. This enables interference analysis unit 34 to find a robot operation that does not interfere with the workpiece while satisfying constraint conditions such as the torch quality.

[0039] The inventors have found that welding performance information, such as the movement and tilt angle of welding torch 51 in robotic welding or manual welding, forward / backward angles, welding target position, distance between the torch tip and the workpiece, and time history of current / voltage of the welding power source, correlates with welding quality information, such as the flank angle, toe radius, bead width, bead height, leg length, and weld defects of the weld bead.

[0040] The inventors have found that the torch tilt angle of welding torch 51 and the welding target position have a large effect on the welding quality, that is, the degree of bead sagging.

[0041] FIG. 5 is a diagram for explaining the torch tilt angle θ and the welding target position. 5, the base materials 41 and 42 are planar, and the base material 41 is supported vertically on the base material 42 which is placed substantially horizontally. The welding torch 51 has a wire 52 of length L W Only the distance L is sent out. D refers to the distance from the tip of the torch to the point where the wire 52 comes into contact with the base material 41. The "distance between the tip of the torch and the workpiece" refers to the distance L D This is what I mean.

[0042] This welding torch 51 is inclined by a predetermined torch inclination angle θ, and is moved, for example, toward the back side of the paper surface. The torch inclination angle θ is the inclination angle of the welding torch 51, and refers to the inclination angle on a plane perpendicular to the welding proceeding direction.

[0043] Target position L T indicates the distance from the apex of the L-shape formed by the flat plates of the base materials 41 and 42 to the intersection point of the axis of the welding torch 51 and the base material 42. F is the bead width, which is the distance from the side formed by the flat plates of the base materials 41 and 42 to the end of the molten bead 40. F is one of the indicators of welding quality.

[0044] FIG. 6 is a diagram for explaining the torch sweep angle. 6, base materials 41 and 42 are planar, and base material 41 is supported vertically on base material 42 which is placed substantially horizontally. Welding torch 51 moves to the left side of the drawing, and molten bead 40 is formed between base materials 41 and 42. Here, the torch sweep angle refers to a rearward inclination angle α1 of welding torch 51 which is parallel to the welding advance direction.

[0045] FIG. 7 is a diagram for explaining the torch advance angle. 7, base materials 41 and 42 are planar, and base material 41 is supported vertically on base material 42 which is placed substantially horizontally. Welding torch 51 moves to the left side of the drawing, and molten bead 40 is formed between base materials 41 and 42. Here, the torch advance angle refers to a forward inclination angle α2 of welding torch 51 which is parallel to the welding advance direction.

[0046] The inventors have also found that the power output of a welding machine has a large effect on the welding quality, particularly on undercut. Here, the power output of a welding machine refers to the product of the current flowing through the welding machine and the voltage applied to the welding machine.

[0047] FIG. 8 is a graph showing the correlation between welding performance information and welding quality information. The horizontal axis of the graph indicates the leg length predicted from the quality-welding condition correlation database 207. The vertical axis indicates the actually measured leg length. Here, the leg length is one piece of welding quality information. Each plot shows welding based on each welding performance information, and has a predetermined correlation. In this way, when the welding performance information is known, the welding quality information can be predicted.

[0048] Therefore, welding path autonomous optimization system 7 clarifies the relationship between welding record information and welding quality information by multivariate analysis or the like, and stores the degree of influence of each welding record on the welding quality as a numerical value such as a correlation coefficient in quality-welding condition correlation database 207. By determining a welding path using the degree of influence of the welding record on the welding quality as a constraint condition, welding path autonomous optimization system 7 can calculate a welding path that ensures a predetermined welding quality.

[0049] In addition, when it is difficult to apply multivariate analysis to the relationship between the welding performance information and the welding quality information, the correlation coefficient may be obtained by a machine learning method using artificial intelligence (AI).

[0050] Returning to Fig. 4, the explanation will be continued. The material of the workpiece to be welded and welding groove shape information are mainly input in the weld part shape input unit 31. Information on the welding groove shape includes mill sheet information such as steel type, type of welding wire, welding wire diameter, welding method, groove pattern (I-shape, V-shape, L-shape, Y-shape, K-shape, J-shape, X-shape, U-shape, H-shape, etc.), groove angle, root length, root gap length, roughness during groove preparation, etc.

[0051] Welding conditions are mainly input in the welding condition input unit 32. Information regarding the welding conditions includes the set voltage of the welding power source, current, welding speed, weaving frequency, weaving width, wire feed speed, wire extension length, shielding gas type and flow rate, and in the case of multi-layer welding, the interpass time and the presence or absence of chipping.

[0052] The three-dimensional CAD data 13 includes three-dimensional model information of the workpiece to be welded having groove shape information, the robot 61, the jig for welding restraint, the positioner 62, and the weld beads of each layer. Regarding the information in the weld shape input unit 31 and the welding condition input unit 32, it is preferable that the information be input to annotations in the three-dimensional CAD.

[0053] 3 executes motion analysis software 209 to realize interference analysis unit 34 (see FIG. 4), which can calculate all quantitative information related to kinematics including the position, speed, and acceleration of all components included in 3D CAD data 13, and dynamics including joint reaction force, inertial force, and power requirement. This function is often provided as one of the functions of commercially available 3D CAD software, and welding process information creation unit 3 determines the welding path by offline teaching regarding the motion of the welding robot so as not to interfere with the workpiece to be welded, the jig, and positioner 62.

[0054] Interference analysis unit 34 determines a welding path by changing the geometry on software for three-dimensional CAD data 13 using an API (Application Programming Interface) of the three-dimensional CAD based on constraint conditions such as the tilt angle of welding torch 51 and the distance between the torch tip and the work (a value substantially equivalent to the welding wire target position) obtained from quality-welding condition correlation database 207. In other words, interference analysis unit 34 determines a welding path that does not interfere with the object to be welded based on three-dimensional CAD data 13 of the object to be welded and the constraint conditions of quality-welding condition correlation database 207.

[0055] In changing the geometry, interference analysis unit 34 executes, for example, the angle and rotation of welding torch 51. The absolute amount of the geometry change is determined by a multivariate analysis based on the output information of weld shape input unit 31 and welding condition input unit 32 stored in the annotation information in three-dimensional CAD data 13 and the quality-welding condition correlation database 207.

[0056] Welding robot operation / welding condition output unit 35 outputs point data of time-series trajectory changes focusing on the coordinate system of the tip of welding torch 51 of robot 61 to a file in CSV (Comma Separated Values) format or to quality-welding condition correlation database 207. This output data is taken into control device 1. Welding robot operation / welding condition output unit 35 is an output unit that outputs welding process information based on the welding path. The welding process information output by welding robot operation / welding condition output unit 35 includes at least information on the torch tilt angle and welding target position of welding torch 51.

[0057] 1, for example, and includes a three-dimensional shape information acquisition unit 81, a difference information acquisition unit 82, and a robot operation / welding condition correction unit 83. Pre-welding process information correction unit 8 acquires three-dimensional shape information and groove shape information of the work before welding by a three-dimensional camera 10, identifies differences with three-dimensional CAD data 13, and corrects the welding process information.

[0058] The three-dimensional shape information acquisition unit 81 acquires three-dimensional shape information of the workpiece before welding using the three-dimensional camera 10, the laser displacement meter 11, etc., and compiles the three-dimensional shape information of the workpiece. The three-dimensional shape information acquisition unit 81 acquires three-dimensional shape information of the object to be welded, including at least one of the three-dimensional shape information of the workpiece before welding and groove shape information. The difference information acquisition unit 82 compares the three-dimensional shape information of the workpiece acquired by the three-dimensional shape information acquisition unit 81 with the output data of the welding robot operation / welding condition output unit 35, and evaluates the deviation of the positional relationship. The difference information acquisition unit 82 functions as a first difference information acquisition unit that identifies the difference between the three-dimensional shape information and groove shape information of the workpiece, and the three-dimensional CAD data 13. In other words, the difference information acquisition unit 82 identifies the difference between the three-dimensional CAD data 13 and the three-dimensional shape information of the object to be welded.

[0059] Robot operation / welding condition correction unit 83 issues a command to robot 61 based on the positional relationship deviation information output from difference information acquisition unit 82. In other words, robot operation / welding condition correction unit 83 corrects the welding process information based on the difference output from difference information acquisition unit 82.

[0060] The pre-welding process information correction unit 8 automatically performs high-quality robot welding using the welding robot operation / welding conditions obtained from the welding robot operation / welding condition output unit 35, automatically recognizes gaps in the positional relationship between the computer and reality caused by workpiece placement, groove processing accuracy, etc., and applies corrections to the robot operation.

[0061] The welding process information correction unit 9 is composed of a welding fluctuation evaluation unit 91 , a difference information acquisition unit 92 , and a robot operation / welding condition correction unit 93 . The welding fluctuation evaluation unit 91 acquires three-dimensional shape information of the workpiece during welding using the molten pool observation camera 12 and the like, and compiles the three-dimensional shape information of the workpiece. Furthermore, the welding fluctuation evaluation unit 91 evaluates the state of the molten pool during welding and the fluctuation of its relative position with respect to the groove from the images taken by the molten pool observation camera 12.

[0062] The difference information acquisition unit 92 compares the three-dimensional shape information of the workpiece acquired by the during-welding variation evaluation unit 91 with the output data of the welding robot operation / welding condition output unit 35, and evaluates the deviation of the positional relationship. The difference information acquisition unit 92 functions as a second difference information acquisition unit that identifies the difference between the three-dimensional shape information of the workpiece acquired by the during-welding variation evaluation unit 91 and the three-dimensional CAD data 13. Robot operation / welding condition correction unit 93 corrects the welding process information based on the positional relationship deviation information output by difference information acquisition unit 92, and issues a command to the robot motion based on the corrected welding process information. At this time, robot operation / welding condition correction unit 93 corrects either the position of welding torch 51 or the welding conditions based on the positional relationship deviation.

[0063] The welding process information correction unit 9 automatically performs high-quality robot welding using the welding robot operation / welding conditions obtained from the welding robot operation / welding condition output unit 35, automatically recognizes the gap in the positional relationship between the computer and the reality, such as in the groove, and applies corrections to the robot operation.

[0064] The control device 1 shown in Fig. 1 passes information to the robot control device 6 shown in Fig. 1, which becomes an input for the motion control of the robot 61 and the positioner 62. The information calculated by the control device 1 is input to a difference information acquisition unit 82 before welding, and data is input to a difference information acquisition unit 92 during welding.

[0065] In the three-dimensional shape information acquisition unit 81, sensor data from scanned laser displacement meters 11 mounted on a three-dimensional camera 10, a robot 61, etc. is input to the control device 1, and the data taken into the control device 1 is retained in the control device 1 as three-dimensional coordinate point cloud data in a form comparable to the point cloud data of the three-dimensional CAD, and is also transferred to the information processing device 2.

[0066] The difference information acquisition unit 82 sequentially calculates and outputs the positional relationship deviation between the point cloud data obtained via the welding robot operation / welding condition output unit 35 and the point cloud data of the workpiece obtained by the three-dimensional shape information acquisition unit 81 or the like, by pattern matching or the like. The difference information acquisition unit 82 is embodied on the control device 1, but may be embodied on the information processing device 2.

[0067] 1 and the voltage / current setting values ​​of the welding power source based on successive values ​​of the deviation in positional relationship obtained from difference information acquisition unit 82. Robot operation / welding condition correction unit 83 further determines changes in the welding speed, addition of weaving, and change in weaving width based on welding records in quality-welding condition correlation database 207 that provide similar weld quality, and defines new welding conditions by tuning the conditions in constraint condition addition unit 33 so that the quality under the welding conditions defined in welding robot operation / welding condition output unit 35 can be reproduced in three-dimensional shape information acquisition unit 81 as well.

[0068] 1 captures video of the molten pool and the front of the welding torch 51 while avoiding the arc light, and detects state information of the molten pool during welding. The welding fluctuation evaluation unit 91 quantitatively grasps values ​​such as the fluctuation in the groove width in front of the welding torch 51 and the width of the molten pool by using technology such as deep learning in the video with image fluctuations under the welding phenomenon.

[0069] The difference information acquisition unit 92 sequentially compares the point cloud data obtained via the welding robot operation / welding condition output unit 35 with the fluctuation values ​​of the work groove width and the molten pool width obtained by the welding fluctuation evaluation unit 91, and outputs the deviation in positional relationship by pattern matching, etc.

[0070] Robot operation / welding condition correction unit 93 changes the height of welding torch 51 and the voltage / current setting values ​​of the welding power source based on the sequential values ​​of the deviation in positional relationship obtained from difference information acquisition unit 92. Robot operation / welding condition correction unit 93 further determines a change in the welding speed, addition of weaving, or change in weaving width based on welding records that provide similar weld quality in quality-welding condition correlation database 207, and defines new welding conditions by tuning the welding conditions determined by welding robot operation / welding condition output unit 35.

[0071] In the present embodiment described above, it is possible to perform welding with a welding robot trajectory that reflects welding deformation and molten pool fluctuations, and it is possible to provide an offline teaching system for a welding robot for high-quality construction of arc welds, and a system that autonomously optimizes a taught robot welding path under the variability of actual welding.

[0072] In this embodiment, the following effects can be further obtained. (1) It is possible to provide offline teaching not only for spot welding but also for arc welding having line welds. (2) It is possible to determine offline teaching based not only on interference between the welding robot's welding torch and the workpiece, but also on offline teaching that focuses on improving the quality of welds. (3) The system recognizes the difference between the positional relationship between offline teaching in arc welding and actual welding, and achieves stable welding by correcting the robot operation and the welding power source based on the difference.

[0073] (Modification) The present invention is not limited to the above-described embodiment, and includes various modified examples. For example, the above-described embodiment has been described in detail to clearly explain the present invention, and is not necessarily limited to those having all of the configurations described. It is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. In addition, it is also possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.

[0074] The above configurations, functions, processing units, processing means, etc. may be realized in part or in whole by hardware such as an integrated circuit. The above configurations, functions, etc. may be realized by software by a processor interpreting and executing a program that realizes each function. Information such as the program, table, file, etc. that realizes each function can be stored in a recording device such as a memory, a hard disk, or an SSD (Solid State Drive), or a recording medium such as a flash memory card or a DVD (Digital Versatile Disk).

[0075] In each embodiment, the control lines and information lines are those that are considered necessary for the explanation, and not all control lines and information lines in the product are necessarily shown. In reality, it may be considered that almost all components are connected to each other. Furthermore, the communication means connecting the devices is not limited to wireless LAN, but may be changed to wired LAN or other communication means. [Explanation of symbols]

[0076] 1 Control device 10 3D Camera 11 Laser displacement meter (displacement meter) 12 Molten pool observation camera 13 3D CAD data 101 CPU Unit 102 Protocol communication unit 103 Database Connection Unit 104 Analog Input Unit 105 Analog output unit 15 Operating equipment 2. Information processing equipment 21 Monitor Display 200 Computers 201 CPU 202 RAM 203 ROM 204 HDD 205 Communication Interface 207 Quality-Welding Condition Correlation Database (Database) 209 Motion Analysis Software 5 Digital Welding Machine 51 Welding Torch 6 Robot control device 61 Robot 62 Positioner 7. Welding Path Autonomous Optimization System 206 Hub 3. Welding Process Information Creation Department 33 Constraint Condition Addition Section 31 Weld shape input section 32 Welding condition input section 34 Interference Analysis Department 35 Welding robot operation / welding condition output section (output section) 8. Pre-welding process information correction section 81 3D shape information acquisition unit 82 difference information acquisition unit (first difference information acquisition unit) 83 Robot operation / welding condition correction section 9. Welding process information correction section 91 Welding Fluctuation Evaluation Section 92 Difference information acquisition unit (second difference information acquisition unit) 93 Robot operation / welding condition correction section

Claims

1. A database storing constraints on welding by a welding robot that operates a welding torch; an interference analysis unit that determines a welding path that does not interfere with the object to be welded based on three-dimensional CAD data of the object to be welded and constraints in the database; an output unit that outputs welding process information based on the welding path; A three-dimensional scanner for detecting the shape of the workpiece; A pre-welding process information correction unit that acquires three-dimensional shape information and a groove shape of a workpiece before welding by the three-dimensional scanner, identifies a difference from the three-dimensional CAD data, and corrects the welding process information; A welding path autonomous optimization device comprising:

2. The welding process information includes at least information on the tilt angle and the welding target position of the welding torch. The welding path autonomous optimization device according to claim 1 .

3. The constraint conditions include at least constraint conditions of the inclination angle and the welding target position of the welding torch. The welding path autonomous optimization device according to claim 1 .

4. The pre-welding process information correction unit A three-dimensional shape information acquisition unit that acquires three-dimensional shape information and groove shape of a workpiece before welding; A first difference information acquisition unit that identifies a difference between three-dimensional shape information and a groove shape of the workpiece and the three-dimensional CAD data; The welding path autonomous optimization device according to claim 1, further comprising:

5. A displacement meter for identifying the relative position of the welding robot and the groove; A camera for observing the molten pool, a welding process information correction unit that evaluates the state of the molten pool during welding and the fluctuation of the relative position with respect to the groove from the image captured by the camera, identifies a difference from the three-dimensional CAD data, and corrects the welding process information; The welding path autonomous optimization device according to claim 1, further comprising:

6. The welding in-progress process information correction unit is A welding fluctuation evaluation unit that evaluates the state of the molten pool during welding and the fluctuation of the relative position with respect to the groove from the image taken by the camera; A second difference information acquisition unit that identifies a difference between the three-dimensional shape information of the workpiece acquired by the welding fluctuation evaluation unit and the three-dimensional CAD data; The welding path autonomous optimization device according to claim 5, further comprising:

7. A welding robot that operates a welding torch; A database storing constraints for welding by a robot; an interference analysis unit that determines a welding path that does not interfere with the object to be welded based on three-dimensional CAD data of the object to be welded and constraints in the database; an output unit that outputs welding process information based on the welding path; A three-dimensional scanner for detecting the shape of the workpiece; A pre-welding process information correction unit that acquires three-dimensional shape information and a groove shape of a workpiece before welding by the three-dimensional scanner, identifies a difference from the three-dimensional CAD data, and corrects the welding process information; A welding path autonomous optimization system comprising:

8. determining a welding path that does not interfere with the object to be welded based on three-dimensional CAD data of the object to be welded and a constraint condition for welding by a welding robot; outputting welding process information based on the welding path; Acquiring three-dimensional shape information of an object to be welded, the three-dimensional shape information including at least one of three-dimensional shape information of a workpiece before welding and groove shape information; Identifying a difference between the three-dimensional CAD data and three-dimensional shape information of the object to be welded; correcting the welding process information by the difference; A welding path autonomous optimization method comprising:

9. A step of inputting groove shape information during welding; Detecting state information of a molten pool during welding; Evaluating a positional relationship deviation based on the groove shape information and the state information of the molten pool; correcting either the position of the welding torch or the welding conditions based on the deviation of the positional relationship; The method of claim 8, further comprising:

Images