JP3483713B2 - Control method in fillet multi-layer welding robot system - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention uses a robot equipped with a laser sensor and an arc welding torch at its hand (hereinafter referred to as a welding robot or simply a robot) and a system provided with a robot control unit. More specifically, the present invention relates to a technique for performing multi-layer welding of a meat weld portion, and more specifically, to a technique for causing a welding robot to perform a tracking operation in consideration of deformation of a joint during multi-layer welding using a laser sensor.

[0002]

2. Description of the Related Art A laser sensor is mounted on a robot together with a torch for arc welding, and the technique of causing the robot to track the welding line by using the data obtained by sensing the welding line in advance is known as a welding robot. Widely known as real-time tracking. Also, it has been attempted to apply the real-time tracking control to a robot that performs multi-layer welding.

As schematically illustrated in FIG. 1, the multi-layer welding of the fillets of the works W1 and W2 constituting the joint is the first, second, third ... Nth (N: total bead). The sequence of sequentially forming the first layer, the second layer, the third layer ... the Nth beads A1, A2, A3 ... Done in. Generally, the loci of the target position when forming each bead layer are not the same. Here, the "target position" is the welding torch tip position (normally coincides with the tool tip point of the welding robot) required for performing proper welding.

When a multi-layered heap is formed at a corner, the loci B1 to B1 'of the aiming position of the first layer are generally made to substantially coincide with the positions of the fillet lines. As shown by ~ B2 ', B3 ~ B3', a position which is separated from the position of the fillet line along the surface of each of the works W1, W2 by approximately the bead thickness is selected.

The loci B1 to B1 'of the target position of the first layer are
Real-time tracking using a laser sensor can be realized relatively easily by applying it to the first movement cycle of the welding robot. This is because at the time of welding the first layer, the laser sensor may detect the corner line position (a line representative of the position of the corner; the same applies hereinafter) by sensing the corner where the weld bead is not formed. On the other hand, 2
Since the corners are covered with the welding beads formed in advance during the welding of the layer and subsequent layers, it becomes impossible to directly detect the corner line position by the laser sensor.

Therefore, the corner line position is directly detected by the laser sensor, and a shift amount Δ (generally a vector amount having two components of Δx and Δy) preset as a shift amount on the Cartesian coordinate system is added thereto. Trajectories of target positions on the second and subsequent layers B2-B
It has been difficult to adopt the method of realizing 2 ', B3 to B3'.

There is also a method of accumulating the data of the corner line position at the time of the tracking of the first layer and utilizing this for the welding of the second and subsequent layers, but a large amount of data is required to accurately reproduce the position of the corner line. It is necessary to store data, which is not a convenient method.

As a technique capable of solving such a problem,
It has been proposed to indirectly perform the corner line position using a laser sensor (see Japanese Patent Laid-Open No. 16221/1996). If this proposal is applied to the above example, the surfaces G1 and G of both works W1 and W2 that provide the corner line position to the corners.
The target position (the target path of the tool tip point) is determined by obtaining the intersection line position of 2 and adding it to the shift amount Δ (Δx, Δy) preset as the shift amount on the orthogonal coordinate system. Become.

[0009]

The above-mentioned improved technology is excellent in that it makes it easy to apply real-time tracking using a laser sensor even when welding the second and subsequent layers, and does not require storing a large amount of data. ing. However, when this is actually applied to the multi-layer welding of the corners as shown in FIG. 1, it is not uncommon that the expected welding accuracy cannot be obtained due to a slight deviation in the target position during actual welding.

It is considered that the cause of such deviation of the target position is that the works W1 and W2 constituting the joint are deformed by the influence of heat generated by the arc discharge. FIG. 2 is a schematic diagram for explaining the example shown in FIG. As illustrated in the figure, the surface G1 of the work W1 that was at the position of the broken line before the start of welding is
It has the property of being thermally deformed and tilted inward with the progress of multi-layer welding (indicated by G1 → H1). Although illustration is omitted, thermal deformation may occur on the horizontal surface G2 of the work W2.
Generally, the work W determines which work surface is deformed.
It is a case by case, depending on the thickness, material, heating condition, etc. of 1, W2.

Considering now the target position B2 at the time of welding the second layer illustrated in FIG. 1 as an example, in the conventional technique (including the above-mentioned improved technique), this target position B2 is the position before the start of welding. On the premise of the position G1, the position shift amount Δ2 is set with reference to the corner line position B1 (detected by the laser sensor). For example, when a coordinate system in which the directions of G1 and G2 are respectively the Y axis and the X axis is set, the position shift amount Δ2 designating the target position B2 is designated by the vector amount Δ2 (0, d). Here, d is the distance between the corner line position B1 and the aim position B2.

When the target position Bi of the i-th layer (i = 1, 2, 3 ... N) is determined for the N-layer multi-layer welding by the same designation method, it is expressed by the following equation (1). The respective components Δix and Δiy of the position shift amount Δi are set in advance. In the example of FIG. 2, naturally Δ2x = 0 and Δiy = d.

Δi = (Δix, Δiy) (1) Now, when the thermal deformation as shown in the figure occurs, the target position (torch tip position) that should be realized is indicated by the symbol C2 instead of B2. Move to a position like. However, with the above-described method of determining the target position, the target position remains B2 even if thermal deformation occurs, and it is not possible to realize a robot path in which the tip of the torch passes through the position C2. That is,
Since the conventional technology does not have a technical means to appropriately change the target position according to thermal deformation, it is a difficult problem to solve in cases where the thermal deformation is large or where the deviation of the target position greatly affects the welding quality. It was.

Therefore, an object of the present invention is to enable real-time tracking using a laser sensor even during fillet multi-layer welding of the second and subsequent layers, while depending on the thermal deformation of the work constituting the joint. It is to provide a control method for determining a movement path of a robot while changing a target position. In addition, the quality of fillet multi-layer welding is to be improved through this.

[0015]

SUMMARY OF THE INVENTION The present invention relates to a system control method for controlling a robot equipped with an arc welding torch and a laser sensor by means of a robot controller, and tracking a route movement during the execution of welding by the robot. By performing movement and shift adjustment of the moving path in parallel considering the thermal deformation of the work or the work surface,
This is a solution to the above technical problem.

The tracking operation is performed on the basis of the position of the corner line. To detect the position of the corner line for that purpose, the positions of the two work surfaces providing the corners are detected by using a laser sensor, and This is accomplished by determining the position of the corner line based on

On the other hand, the detection results of the positions of the two work surfaces are used to control the shift amount of the moving path according to the thermal deformation of the work surfaces. In a preferred embodiment, the shift amount of the movement path with respect to the work surface without thermal deformation is preset in the system as the reference shift amount. The set reference shift adjustment amount is converted into a new shift adjustment amount based on the detection result of the positions of the two work surfaces, and the shift amount of the movement path is controlled according to the thermal deformation of the work surfaces.

In the exemplary embodiment, the reference shift adjustment amount is set as a two-dimensional reference shift adjustment amount, and is converted into a new two-dimensional shift amount based on the detection result of the positions of the two work surfaces. Thereby, the shift amount of the moving path can be controlled according to the thermal deformation of the work surface.

In one preferable conversion method, a new two-dimensional shift is performed by rereading two straight lines representing the positions of two work surfaces as values expressed on an oblique coordinate system constituted by coordinate axes. Conversion to quantity takes place. Further, after the replacement, the correction may be performed in a form including the distance adjustment from the detected corner line position.

With the above-mentioned structure of the present invention, the target position is changed according to the thermal deformation of the work constituting the joint while applying the real-time tracking using the laser sensor even in the fillet multi-layer welding after the second layer. While moving, you can decide the movement route of the robot. Therefore, even if thermal deformation occurs along with the progress of corner multi-layer welding, deviation of the target position can be avoided.

[0021]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT FIG. 3 is a main block for explaining the overall configuration of a system including a robot controller which can be used when the present invention is carried out. As shown in the figure, the robot controller 30 that controls the entire system is
The central processing unit (hereinafter, CPU) 31 has
In U31, a memory 32 including a ROM, a memory 33 including a RAM, a non-volatile memory 34, a robot axis controller 35, a teaching operation panel 38 including an LCD 37, and a general-purpose interface 39 for connection with an external device are connected to a bus. Has been done.

The ROM 32 stores a program for controlling the entire system including the robot controller 30 itself. This includes a program necessary for screen-inputting various data for implementing the present invention in a mode described later. A part of the RAM 33 is used for temporary storage of data for the processing performed by the CPU 31. Further, in order to carry out the present invention, the RAM 33 is provided with some buffer registers for storing sensing data, path correction data, etc., which will be described later. Non-volatile memory 3
A register for storing position data and a parameter for determining a path shift amount at the time of multiple layers is set in 4.

Further, to the robot axis controller 35, each servo motor provided in each axis mechanism portion of the robot 40 is connected via a servo circuit 36. The external device connected to the general-purpose interface 39 includes a power supply device 50 for the arc welding torch 2 mounted on the hand of the robot 40 and a laser sensor 3 for performing real-time tracking.

The structure, function, measurement principle, etc. of the laser sensor are well known, but the outline will be described with reference to FIGS. 4 and 5. FIG. 4 exemplifies the main configuration of the laser sensor used in this embodiment.
It includes a projection unit 10 and a light detection unit 13. The projection unit 10 includes a laser oscillator 11 and an oscillating mirror (galvanometer) 12 for beam scanning, and the light detection unit 13 includes an optical system 14 for image formation and a light receiving element 15.

On the other hand, the laser sensor control section 20 includes a mirror drive section 21 for swinging the swing mirror 12 and a laser drive section 2 for driving the laser oscillator 11 to generate a laser beam.
2. A signal detection unit 23 that detects the incident position (reflection point) S of the laser beam 5 on the object based on the light receiving position of the light receiving element 15 is connected. These are general-purpose interfaces 39 of the robot controller 30 via the line 24.
(See FIG. 3).

When the laser sensor 3 receives an operation command from the robot controller 30 via the line 24, the laser drive unit 22 drives the laser oscillator 11 to generate the laser beam 5. In parallel with this, the swinging of the swing mirror 12 is started by the mirror driving unit 21. As a result, the laser beam generated from the laser oscillator 11 scans the object surface G.

The laser beam diffused and reflected at the reflection point S on the object surface G forms an image on the light receiving element 15 according to the position of the reflection point S by the optical system 14. The light receiving element 15 includes a split type element and a one-dimensional CCD array which is a light receiving element, or a position detection detector (PS which is a non-split type integral type element).
D: Position Sensitive Detector), or a CCD camera having a two-dimensional CCD array is used.

Here, a one-dimensional CC is used as the light receiving element 15.
It is assumed that a D array is used. The light (image of the reflected light) that strikes the light receiving surface of the light receiving element 15 is converted into an electric charge,
It is stored in that cell. The charge accumulated in the cell is 1 every predetermined period according to the CCD scanning signal from the signal detection unit 23.
The signals are sequentially output from the leading edge and sent to the robot controller 30 via the signal detection unit 23 and the line 24.

The scanning cycle of the CCD is set sufficiently shorter than the scanning cycle of the swing mirror 12 (for example, several hundredths).
The change of the swing angle of the swing mirror 12 and the CCD
The transition of the element output state can be grasped at any time. C
CD element output status is the maximum output cell position (cell number)
The cell position where reflected light hits is detected.
From this position, the position of the reflection point S on the sensor coordinate system is obtained.

FIG. 5 shows the detection position s in the light receiving element 15.
Position of the reflection point S on the sensor coordinate system (Xs, Ys
FIG. 6 is a diagram illustrating a method of obtaining). It is assumed that the origin (0, 0) of the sensor coordinate system is on the line connecting the center of the optical system 14 and the center point of the light receiving element 15, and this line is the Y axis, and the axis orthogonal to this Y axis is the X axis.

Then, the distance from the origin of the sensor coordinate system to the center of the optical system is L1, and the light receiving element 15 from the center of the optical system.
Is the distance from the sensor origin to the swing center of the swing mirror 12 in the X-axis direction, and the Y-axis distance from the sensor origin to the swing center of the swing mirror is L0.
The angle of the reflected light of the laser beam by the oscillating mirror 12 with respect to the Y-axis direction is θ, and the light receiving position on the light receiving surface of the light receiving element 15 is s.

Then, the coordinate position (Xs, Ys) of the reflection point S of the laser beam 5 is calculated by the following equations (2) and (3). Xs = s · [(L1−L0) · tan θ + D] / (xa + L2 · tan θ) (2) Ys = [L1 · s + L2 · (L0 · tan θ−D)] / (s + L2 · tan θ) ... (3) Position (Xs, Ys) of the reflection point S obtained in the sensor coordinate system
) Is the coordinate system recognized by the robot using the posture data of the robot and the calibration data (data representing the relationship between the sensor coordinate system and the robot coordinate system) (here, work coordinates shown in FIG. 6 described later). System Σw O
-XYZ) converted into three-dimensional data. A series of data groups obtained for each scanning cycle of the laser beam 5 is analyzed by software processing in the robot controller 30 and, for example, the position of the laser beam scanning locus on the surface G (calculated from the positions of the plurality of reflection points S). Is required.

Next, FIG. 6 is a sketch showing a general arrangement when the present invention is applied to the case shown in FIG.
As shown in FIG. 1, a welding torch 2 and a laser sensor 3 are mounted on the hand of the robot designated by reference numeral 1. Reference numeral 5 represents a tool tip point of the robot set at the tip position (torch tip) of the welding wire 4 fed from the welding torch 2.

This drawing shows a state in which the welding of the second layer is in progress, and the bead A2 of the second layer is formed halfway. Laser beam LB projected from the laser sensor 3
Is a surface H1 (G1) of one workpiece W1 that provides a corner in order to sense a region preceding the torch tip 5.
1), the bead A1 of the first layer already formed along the corner line 7, the surface H2 (G of the other work W2 that provides the corner)
2) Light spot trajectories 6a, 6 that repeatedly pass through the three
Make b and 6c. H1 (G1) and H2 (G2)
The notation indicates that the surfaces G1 and G2 before the start of welding may be moved by thermal deformation after the start of welding (the same applies hereinafter).

For the welding robot, as indicated by the arrow,
It is assumed that a moving path is taught which moves from the corner starting point P to the corner ending point P ′ along the corner line 7 while the welding torch 2 is ignited. The robot is set to P
The movement operation with P'as the teaching path is repeated N times, which is equal to the total number of layers, but the actual movement path each time is different depending on the target position as described above. What is important here is that the influence of thermal deformation is taken into consideration when changing the route (changing the direction of shifting from the corner line 7 and the distance) depending on the difference in the target position.

In this embodiment, the present invention is implemented by the following method. (1) Maximum expected total number N in the non-volatile memory 34
The number of registers RG1, RG2 ... RG corresponding to max
Set Nmax and an addressing counter that specifies a register number. (2) The reference shift amount Δi for welding the i-th layer is stored in advance in the i-th register RGi. Here, the reference shift amount Δi is referred to as "the target positions B1 to B4 which are considered to be optimal when thermal deformation does not occur" (hereinafter, referred to as "reference target position") at the corner line 7 (laser sensor 3). It is expressed by a two-dimensional shift amount based on the position of (detected by 1.). (3) The reference shift amount Δi is the corner line 7 on the Cartesian coordinate system Σw (work coordinate system O-XYZ) already set in the robot.
Shift amount Δ in the X-axis direction based on the position of (teaching path)
x and a value representing the shift amount Δy in the Y-axis direction are, for example, mm
The unit is set in advance by operating the teaching operation panel 38. Here, it is assumed that the normal directions of the work surfaces G1 and G2 before welding are parallel to the Y axis and the X axis of the coordinate system Σw, respectively.

(4) As an example, the total number of layers N = 4, and the reference shift amount Δ1 representing the reference target positions B1 to B4.
It is assumed that .about..DELTA.4 is set as shown in FIG. Register R designated by each addressing counter value 1, 2, 3, 4
The data stored in G1 to RG4 is as follows. RG1; (0,0) RG2; (0, d) RG3; (d, 0) RG4; (d, d) (5) In any of the first to fourth layers, the laser sensor 3 is used to make a corner. The position of the part line 7 is detected, and real-time tracking with position correction based on it is executed.
This tracking correction is the teaching route (planned corner line position)
This is for compensating for the positional deviation of the corner line 7 actually detected by the laser sensor 3, and has a different meaning from the position shift for changing the target position for each layer welding.
Both are in a relationship of being added when determining the torch tip position.

That is, in this embodiment, "torch tip position" = "position on teaching path" + "tracking correction amount"
The torch tip position when the robot moves is determined based on the relationship of + “shift amount for determining target position”. And
Regarding the third item “shift amount for determining the target position” on the right side, the technical idea of the present invention is applied, and thermal deformation of the workpiece is considered.

(6) Regarding the position of the corner line 7 described in (5) above, the position of the line of intersection of the surfaces H1 (G1), H2 (G2) is determined by laser beam scanning and the surfaces H1 (G1), H2 ( G2)
It is determined as the intersection point position of the two straight lines represented by the light spot trajectories 6a and 6b (see FIG. 6) formed in. There are two reasons for adopting such a method. One is to make it possible to estimate the position of the corner line 7 even if the corner line 7 is hidden by the bead during welding of the second and subsequent layers.
This is to know the thermal deformation of 1) and H2 (G2). The latter detection result is reflected in the above-mentioned (5) "shift amount for determining aiming position" in real time.

(7) The planes H1 (G1) and H2 (G2) are defined as the intersection points of the two straight lines represented by the light spot loci 6a and 6b.
In order to obtain the intersection line position of the plane H, as shown in FIG.
1 (G1) and H2 (G2) are set in advance with parameters capable of designating ranges L1 and L2 that are apart from the position of the corner line 7 by an appropriate distance (there is no possibility of being covered by the beads A1 and the like). .

As the parameter, for example, as shown in the figure, φs for determining a constant angle range at both ends of the full scanning angle range φfull of the laser beam can be adopted. It is also possible to specify the time measured from the time when both ends of the scanning angle range are reached. Although FIG. 8 shows the state where only the surface G1 is thermally deformed to H1, even if the surface G2 is thermally deformed, no particular problem occurs in designating the ranges L1 and L2 in such a manner.

(8) There are various conceivable ways of changing the target position depending on the thermal deformation of the surfaces H1 (G1) and H2 (G2), but the point is that the target positions are equivalent before and after the thermal deformation. Can be changed by an appropriate method so that the position is maintained. Although various methods can be adopted as the changing method, here, a method using the oblique coordinate system will be described with reference to FIG.

In this method, the surfaces H1 (G1), H2 (G
A straight line (corresponding to 6a, 6b) detected in 2) is regarded as a coordinate axis of a two-dimensional oblique coordinate system that represents thermal deformation, and the set value Δi of the reference software amount is reinterpreted to determine the target position. , Find the target position by the following geometric calculation. 1. As shown in FIG. 9, when Q is a reference target position that represents an arbitrary set value Δi (Δx, Δy) of the reference software amount, QV is a perpendicular line drawn from the point Q to the surface G1, and QU is a point Q.
It is a vertical line that has been dropped on the surface G2. Surfaces G1 and G2 are Y
Considering the coordinate system o-xy corresponding to the axes, the y coordinate value of V is Δy and the x coordinate value of U is Δx.

2. As shown (exaggerated), surface G
When G1 and G2 are thermally deformed into H1 and H2, y
An oblique coordinate system Oob-x in which the axes x and x correspond to the planes H1 and H2
ask for obyob. The position and direction of the xob axis and the yob axis are detected by the laser sensor 3, and the position of Oob is obtained from this. The displacement at the corner position Oob due to thermal deformation is so small that it can be ignored (however, the present invention can be applied even if there is some displacement).

3. xob such that oobUob = oU (= Δx)
Find the point Uob on the axis. 4. Point Vob on the yob axis where oobVob = oV (= Δy)
Ask for. 5.3 Find the remaining vertices Qob of the parallelogram having vertices oob, Vob, and Uob, and set this point as the target position. The X and Y coordinate values of the origin (corner line position detected by the laser sensor) oob of the oblique coordinate system represented on the work coordinate system O-XYZ and the point Qob obtained above are respectively (pxob, pyob).
), (Qxob, qyob), the shift amount δi (δix, δiy) considering thermal deformation is given by the following equations (4) and (5). δix = qxob-pxob (4) δiy = qyob-pyob (5) 6. The diagonal oobQ of the parallelogram oobUobQobVob
A point Qob (γ) on ob where oobQob (γ) = γoobQob
May be set and this may be the target position. Here, γ is a coefficient for adjusting the distance from the corner line position oob to the target position measured, and is generally set to be close to 1. Alternatively, a point Qob 'at which oobQob' = | Δi | is set on the diagonal line oobQob, and this point may be set as a target position.

Under the assumptions described so far, FIG.
With reference to the flowchart of No. 0, the outline of the processing contents at the time of executing the multi-layer welding will be described. Only the processing related to the detection operation by the laser sensor and the path movement of the robot will be described, and the processing related to the control of the welding torch and the like will be omitted.

The process is started with the initial value of the addressing counter value i designating the register RGi as 1. First, the laser sensor 3 is activated (step M1), the operation command sentence of the operation program is read, and the processing for moving the robot toward the teaching point position P is started (step M2). With the laser sensor 3, the works W1, W2
Is detected (Yes output in step M3), the detection data of the laser sensor 3 obtained in the above-described scanning angle range φs is immediately used, and appropriate points included in the ranges L1 and L2 (see FIG. 8) (at least 2 positions each),
The position of the straight line represented by L1 and L2 (including direction information) is obtained and stored in the buffer memory (step M4).

The straight line obtained in step M4 corresponds to the yob axis and the xob axis shown in FIG. The processing procedure for converting the sensor data obtained by the laser sensor 3 into the coordinate system set in the robot (the current position data of the robot and the coordinate system conversion data are used) and the data of the positions of a plurality of points are used. Since a processing procedure for obtaining a straight line represented by the above points is well known, detailed description thereof will be omitted.

Next, the position oob of the corner line 7 is set to step M4.
It is obtained as the position of the intersection of the two straight lines (yob axis and xob axis) obtained in (step M5). Then, a tracking correction amount for compensating the deviation amount from the teaching route is obtained from the obtained data of the position oob of the corner line 7 and stored in the buffer memory (step M6).

Then, the data of the reference position shift amount Δi stored in the register RGi designated by the addressing counter value i is read (step M7). And
A position for determining the target position using the read data of the reference position shift amount Δi, the xob axis obtained in steps M4 and M5, the data indicating the position / direction of yob, and the corner position oob. The shift amount δi is calculated (step M8). The calculation contents of the position shift amount Δi are as described above.

When the position shift amount δi is obtained, the movement target position of the robot is calculated by the teaching path position + tracking correction amount + position shift amount δi. The movement command created based on the calculation result is output to the servo system (step M9).

If the process for the robot to reach the teaching point P'has not been completed (NO output at step M10), the process returns to step M4 and steps M4 to M10 are repeated. If the process for the robot to reach the teaching point P'has been completed (Yes output in step M10), the process proceeds to step M11 to increment the addressing counter value i by 1.

Next, it is checked whether or not the total number of layers of the multi-layer welding (here, four layers) has been completed (step M12), and if there is no output (multi-layer welding is not completed), step Returning to M2, the processing cycle described above is repeated. When the multi-layer welding for 4 layers is completed,
Since the Yes output is produced in step M12, the process is terminated after the process of stopping the laser sensor 3 and the robot.

Through the above processing, regarding the robot movement at the time of welding the first to fourth layers, the robot path is determined in real time while determining the target position in consideration of the thermal deformation of the work in parallel with the real time tracking of the robot. You can decide.

[0055]

According to the present invention, while applying the real-time tracking using the laser sensor even in the fillet multi-layer welding of the second and subsequent layers, while changing the target position according to the thermal deformation of the work constituting the joint. It has become possible to determine the movement route of the robot. Further, through this, it becomes possible to improve the willingness of fillet multi-layer welding, which is performed under conditions where thermal deformation is likely to occur.

[Brief description of drawings]

FIG. 1 is a diagram illustrating multi-layer welding on a fillet portion.

FIG. 2 is a schematic diagram for explaining deviation of a target position due to thermal deformation in the example shown in FIG.

FIG. 3 is a main block for explaining an overall configuration of a system including a robot control device that can be used when implementing the present invention.

FIG. 4 is a diagram illustrating a configuration of a main part of a laser sensor used in this embodiment.

FIG. 5 is a diagram illustrating a measurement principle of a laser sensor.

FIG. 6 is a sketch diagram showing an outline of the overall arrangement in the embodiment.

FIG. 7 is a diagram illustrating setting contents of reference shift amounts Δ1 to Δ4 representing reference target positions B1 to B4 in the embodiment.

FIG. 8 shows surfaces H1 (G1) and H2 using a laser sensor.
It is a figure explaining the parameter setting for calculating | requiring the intersection line position of (G2).

FIG. 9 is a diagram illustrating a calculation method for determining a shift amount δi in consideration of thermal deformation.

FIG. 10 is a flowchart illustrating an outline of processing contents when executing multi-layer welding in the embodiment.

[Explanation of symbols]

1 Robot hand part 2 welding torch 3 Laser sensor 4 welding wire 5 Torch tip position (tool tip point) 6a, 6b, 6c Light spot locus 10 Projector 11 Laser oscillator 12 Swing mirror (galvanometer) 13 Detector 14 Optical system for imaging 15 Light receiving element 20 Laser sensor controller 21 Mirror drive 22 Laser drive 23 Signal detector 24 lines 30 Robot controller 31 CPU 32 ROM 33 RAM 34 Non-volatile memory 35 axis controller 36 Servo circuit 37 LCD 38 Teaching operation panel 39 General-purpose interface 40 Robot (servo motor of main body mechanism) 50 power supply A1 to A3 welding beads B1 to B4 (reference) Target position C2 Target position (considering thermal deformation) G1, G2 Work surface (before welding starts) H1, H2 Work surface (after starting welding) S incident position (reflection point) W1, W2 work

Front page continuation (58) Fields surveyed (Int.Cl. ⁷ , DB name) B23K 9/127 508 B23K 9/095

Claims (5)

(57) [Claims] 1. An arc welding torch, a laser sensor,
A control method in a fillet multi-layer welding robot system for performing fillet multi-layer welding using a system including a robot equipped with the arc welding torch and laser sensor, and a system including a robot controller, wherein the fillet multi-layer deposition of the robot is performed. The path movement during welding is
In parallel with the tracking operation based on the detection of the position of the corner line representing the corner, the movement path of the robot is realized so that the target position of the arc welding torch corresponding to the welding of each layer of the fillet multi-layer is realized. The position of the corner line is detected by using the laser sensor to detect the position of the two work surfaces that provide the corner, and the position of the two work surfaces is detected. Based on the detection result of the positions of the two work surfaces, the thermal deformation of the work surfaces is determined based on the detection result of the positions of the two work surfaces. The control method in the fillet multi-layer welding robot system, which is controlled according to the above. 2. A shift amount of the movement path with respect to the work surface without thermal deformation is preset in the system as a reference shift amount, and the set reference shift adjustment amount is set to a position of the two work faces. Convert to a new shift adjustment amount based on the detection result
It, the shift amount of the moving path in response to thermal deformation of the workpiece surface is to be controlled, the control method in the fillet multi-layer welding robot system of claim 1,. 3. The reference shift amount is preset in the system as a two-dimensional reference shift amount based on the corner line position, and the set two-dimensional reference shift adjustment amount is used for the two workpieces. The shift amount of the movement path is controlled according to the thermal deformation of the work surface by converting into a new two-dimensional shift amount based on the detection result of the surface position. Control method for fillet multi-layer welding robot system. 4. The two-dimensional reference shift adjustment amount is set to the two
The two straight lines representing the positions of one work surface are read as values expressed on the oblique coordinate system configured as the coordinate axes, so that the conversion into the new two-dimensional shift amount is performed. A control method in a fillet multi-layer welding robot system according to claim 3. 5. The two-dimensional reference shift adjustment amount is set to the two
Two straight lines representing the position of one work surface are read as values expressed on an oblique coordinate system configured as a coordinate axis, and further corrected to include the distance adjustment from the detected corner line position. The control method in the fillet multi-layer welding robot system according to claim 3, wherein the conversion into the new two-dimensional shift amount is thereby performed.