JP4745921B2 - Control method of welding robot - Google Patents

Description

  The present invention relates to a method for controlling a welding robot. In particular, the present invention relates to a control method for causing a 6-axis vertical articulated welding robot to perform a weaving operation of welding while vibrating a tool from side to side.

Referring to FIG. 9, the weaving operation of the arc welding robot is to perform welding while vibrating the torch 1 to the left and right with respect to the weld line, and P _c ( This is generally realized by adding the weaving component W (t) to t). That is, the tip position P (t) of the torch 1 after the addition of the wobbling component is expressed by the following formula (1).

As the whibbing component W (t), a triangular wave or the like is used in addition to the sine wave shown in FIG. 10, and the wobbling translation component (W that changes in time series in accordance with parameters (frequency, vibration, etc.) specified by the teaching data or the like. _x , _Wy , _Wz ) are calculated. In the orthogonal coordinate system (base coordinate system) set for the robot, the position and orientation of the robot tip at the position P _c (t) are expressed by the following equation (2).

Here, P _cx , P _cy , and P _CZ are positions on the orthogonal axis of the base coordinate system, and P _cα , P _cβ , and P _cγ are rotation angles with respect to the orthogonal axis.

  Similarly, the robot tip position P (t) after adding the wobbling component in the base coordinate system is expressed by the following equation (3).

Accordingly, when the wobbling translation component W (t) in the base coordinate system is (W _x , W _y , W _z ), the robot tip position P (t) is expressed by the following equation (4).

  When a general vertical articulated robot as shown in FIG. 11 performs such a whibbing operation, all the joints J1 to J6 swing at the whibbing frequency, and when the whibbing frequency is high (generally 5-10kHz or higher), and the relatively low rigidity of the three axes (joints J1 to J3) that drive the arm is likely to oscillate. As a result, it is difficult to ensure the accuracy of the torch tip position, and accurate welding work can be performed. There was a problem of becoming.

  As a method of solving this problem and enabling a higher frequency of wobbling, the wrist shaft having a higher rigidity can be used without performing a wobbling operation on the basic three axes having low rigidity (the joints J1 to J3 in FIG. 11). It is known that the wobbling operation is executed only by the joints J4 to J6 in FIG. This method is disclosed in Patent Documents 1 and 2, for example.

In the method of Patent Document 1, three axes (joints J4 to J6) are formed so that the extension line in the longitudinal direction of the torch intersects with the rotation center (arm tip position P _arm shown in FIG. 11) of the wrist joints J4 to J6. Wibbing with However, since a specially shaped torch is necessary and the posture of the arm that can be used is limited, the work area becomes extremely narrow and is not practical.

  On the other hand, in the method of Patent Document 2, the shape of the torch is not limited, and the wobbling operation is performed on two axes (joints J4 and J5) excluding the most advanced axis (joint J6) among the wrist axes. . The calculation method is generally as follows.

1) Solve the inverse kinematics of the position Pc (t) before adding the weaving, and obtain the joint angles of all the joints J1 to J6.
2) Find the direction vectors a4 and a5 when the two axes (joints J4 and J5) used for the weaving operation are rotated by a small angle.
3) The weaving component W is decomposed with the direction vectors a4 and a5, and the minute angle changes ΔJ4 and ΔJ5 of the joints J4 and J5 are obtained.
4) The minute angle changes ΔJ4 and ΔJ5 are added to the joint angles of the joints J4 and J5 obtained in 1).

  However, the method of Patent Document 2 requires complicated coordinate transformation and vector calculation when disassembling the weaving component into the corresponding two axes (joint J4, J5) in step 2). An expensive arithmetic unit (CPU) is required. In addition, the wobbling operation pattern is limited to a simple expression such as a sine wave or a triangular wave, and when the operator sets an arbitrary pattern by a program, or the frequency and amplitude are changed in real time according to welding groove and gap fluctuations. It is necessary to prepare a calculation means for each of these various cases.

Japanese Patent Laid-Open No. 1-273374 JP 11-58014 A

  In view of the problem of the conventional control method of the welding robot, the present invention can realize a wobbling operation using only two axes of the fourth and fifth axes without using a specially shaped torch, and the calculation is relatively It is an object of the present invention to provide a welding robot control method that can cope with various wobbling operation patterns by simple calculation.

The present invention is a control method for causing a six-axis vertical articulated welding robot to perform a wobbling operation, the welding robot having first to third axes as basic axes and holding a tool. The fourth to sixth axes are provided as wrist axes, and the fifth axis is positioned at the tip position of the arm including the first to fourth axes, and the following procedure is executed for each control cycle.
The inverse kinematics of the tool tip position before the addition of the wobbling component is solved to calculate at least the angles of the first to third and sixth axes.
A wiping component is added to the tool tip position before adding the wibbing component to obtain a temporary tool tip position after adding the wigbing.
A spherical surface having the radius from the tip position of the arm to the tool tip is obtained with the tip position of the arm as the center.
Using at least the temporary tool tip position, obtain the true tool tip position after adding the wobbling component on the spherical surface.
The joint angles of the fourth and fifth axes are calculated from the true tool tip position after the addition of the weaving component.
The joint angle of the first to sixth axes is output as a command to the manipulator to drive the first to sixth axes.

  Specifically, the true tool tip position after adding the weaving component is an intersection point where a straight line passing through the temporary tool tip position after adding the weaving component and the tip position of the arm intersects the spherical surface.

  As an alternative, the true tool tip position after the addition of the wibbing component is the temporary tool tip after the addition of the wibbing component of the circle of the plane and the spherical surface determined by the direction of the wibbing and the progress of welding. The point closest to the position of.

  As another alternative, the true tool tip position after the addition of the wibbing component is the temporary tool after the addition of the wibbing component among the circles of the plane and the spherical surface determined by the direction of the wibbing and the direction of gravity. The point closest to the tip position.

  In the welding robot control method of the present invention, a spherical surface whose radius is the distance to the tool tip with the tip position of the arm as the center is obtained, and the true tool tip position after adding the weaving component is obtained on this spherical surface. The joint angles of the fourth and fifth axes are calculated from the tool tip position. Therefore, it is not necessary that the extension line of the tool intersects the fifth axis, and a wobbling operation using only two axes of the fourth and fifth axes can be realized without using a specially shaped tool. In addition, the calculation is performed in that the tool tip position calculated in the conventional control for realizing the wobbling operation on the first to sixth axes is calculated as a temporary tool tip position and used for calculating the true tip position. It is comparatively easy and can cope with various wobbling operation patterns, and can be easily switched to the above-described conventional control.

  Next, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

(First embodiment)
Referring to FIG. 1, the welding robot includes a manipulator 11, a control device 12, and a teaching device 13.

The manipulator 11 has a torch 16 as a tool at the tip, and changes its position and posture in a three-dimensional space. The manipulator 11 is a six-axis vertical articulated type, and includes first to third axes (rotary joints RJ _m1 , RJ _m2 , RJ _m3 ) as basic axes, and fourth to fourth wrist axes that hold the torch 16. 6 axes (rotating joints RJ _m4 , RJ _m5 , RJ _m6 ) and a fifth axis (rotating) at the tip position P _arm of the _arm including the first to fourth axes (rotating joints RJ _{m1 to} RJ _m5 ). The joint RJ _m5 ) is located. The second and third axes (rotary joints RJ _m2 and RJ _m3 ) are orthogonal to the first axis (rotary joint RJ _m1 ). The fourth and sixth axes (rotary joints RJ _m4 , RJ _m6 ) are orthogonal to the _fifth axis (rotary joint RJm ₅ ).

The control device 12 includes a processing unit 14 including a CPU, a storage unit 15, an input / output port, and the like. Processor 14, according to a program stored in the storage unit 15, based on input from the angle sensor provided in the rotary joint _RJ m1 _{~RJ m6} teaching device 13 and the manipulator 11 (not shown) or the like, rotary joint _{RJ m1} Drive RJ _m6 .

  Conditions necessary for control are input to the control device 12 using the teaching device 13. This condition includes the positions of the welding start point and end point, the interpolation method between the start point and end point, and the weaving conditions (waveform type, frequency, amplitude, etc.).

Next, coordinates used for control will be described. Two coordinate systems are set for the manipulator 11. First, a rotating joint coordinate system (manipulator coordinate system) given by six joint angles J1 to J6 is set. The position and posture of the manipulator 11 in the manipulator coordinate system (the position and posture of the torch tip 16a) are expressed by these joint angles J1 to J6. Also, an orthogonal coordinate system XYZ (base coordinate system Σbase) fixed with respect to the three-dimensional space is set. The position and posture of the manipulator 11 (the position and posture of the torch tip 16a) in the base coordinate system Σbase are the translation component (P _x , P _y , P _z ) and the roll pitch pitch yaw posture angle (P _α , P _β , P _γ ). Further, an orthogonal coordinate system X _line -Y _line -Z _line (welding line coordinate system Σlime) fixed with respect to the three-dimensional space is set for the trajectory of the torch tip 16a on the workpiece. The calculation of the welding coordinate system Σline will be described in detail later.

Next, the principle of the present invention common to this embodiment and the second and third embodiments described later will be described. Referring to FIG. 2, since the fourth axis (rotary joint RJ _m4 ) and the fifth axis (rotary joint RJ _m5 ) are orthogonal to each other as described above, the torch tip that can be moved by changing these axes The position of 16a is a spherical surface Q (corresponding to "latitude" and "longitude" in terms of the globe). The center of the spherical surface Q is the arm tip position P _arm , and the radius is the distance from the arm tip position _Parm to the torch tip 16a (symbol Lw in FIG. 1). This distance Lw is determined by the link length between the fifth and sixth axes (rotating joints RJ _m5 , RJ _m6 ) and the shape of the tool 16. In the conventional method in which all of the first to sixth axes are used for the whibbing operation, the weaving component W in the base coordinate system with respect to P _c (t) of the central locus of the weld line in the base coordinate system for each time t. The position of the torch tip 16a is obtained by adding (t). In this case, the position P _c (t) of the torch tip 16a before the addition of the weaving component always exists on the spherical surface Q, but the position of the torch tip 16a after the addition of the weaving component (P (t) = Pc (t) + W (t) )) Generally does not exist on the spherical surface Q. Therefore, in the present invention this position P (t), that is, the position P _Q of a point on the sphere Q from the position of the temporary torch tip _16a, i.e., the position of the true torch tip 16a obtains, fourth and from this position P _Q The angle of the fifth axis (rotating joints RJ _m4 , RJ _m5 ) is calculated. For the first to third axes and the sixth axis (rotary joints RJ _{m1 to} RJ _m3 , RJ _m6 ), the joint angles J1 to J3 and J6 obtained from the position P (t) are directly used as command values. , fourth and fifth axes (rotary joint _RJ m4, _{RJ m5)} uses a joint angle J4, J5 determined from the position _{P Q} as a command value. As a result, high-frequency wobbling is enabled only by the fourth and fifth axes (rotary joints RJ _m4 , RJ _m5 ). The method for determining the position on the spherical surface Q is different for each embodiment.

In the method described in Patent Document 2, the approximate solution of the joint angles of the fourth and fifth axes is obtained using the direction vector when the fourth and fifth axes are rotated by a minute angle. On the other hand, in the present invention, an exact solution of the joint angles of the fourth and fifth axes is obtained in a forward geometric manner based on the spherical surface Q with the arm tip position _Parm as the center. In this respect, the calculation principle of the present invention is completely different from that described in Patent Document 2.

  Next, with reference to FIGS. 3A and 3B, the control method of this embodiment (the operation of the processing unit 14 of the control device 12) will be described in detail according to the flowcharts of FIGS. 4A and 4B.

First, the start and end positions P1 and P2 taught from the storage unit 15 in step S1, the interpolation conditions between these teaching positions P1 and P2, and the wobbling conditions (for example, operating methods such as sine waves and triangular waves, amplitude, frequency, etc.) ). Teaching positions P1, P2 in this embodiment the translation component in the base coordinate system _{Σbase (P x, P y,} P z) with the roll-pitch-yaw attitude angle _{(P α, P β, P} γ) is given by. As an interpolation condition, the position P _c (t) of the torch tip 16a before the addition of the wobbling component is linearly moved at a constant speed. Further, as a wobbling condition, the wobbling operation is performed with a sine wave having a translation component amplitude of ΔW and a frequency f.

  In step S2, a path equation (the following formula (5)) between the teaching positions P1 and P2 in the base coordinate system is obtained.

  In Expression (5), ΔP is a movement amount per unit time. The position Pc (t) given by this equation (5) is the position of the torch tip 16a before adding the weaving component at each time t.

Next, a welding line coordinate system Σline is obtained in step S3. In the present invention, a unit vector _{X line,} a unit obtained the direction of the Y coordinates as the outer product of the unit vector _{X line} and the gravity direction unit vector g vector _{Y line,} the unit in the direction of Z-coordinate of the direction of the X coordinates to welding direction A unit vector Z _line obtained as an outer product of the vector X _line and the unit vector Y _line is assumed. These unit vectors X _line , Y _line , and Z _line are expressed by the following equation (6).

In step S4, a 3 × 3 rotation matrix baseR _line of the weld line coordinate system Σline viewed from the base coordinate system Σbase is obtained.

  Next, in step S5, time t is initialized (t = 0).

  Steps S6 to S17 are repeated every control cycle Δt (usually 10 to 20 ms) of the welding robot. First, in step S6, time t is updated to time t + Δt.

Calculating the Wibingu addition previous position P _{c (t)} by substituting the time t of updating the path equation P _{c (t)} in step S7.

In step S8, a wobbling translation component line W (t) (= (w _x , w _y , w _z )) in the weld line coordinate system Σline is calculated. In the present embodiment, since the above-described sinusoidal wobbling pattern is employed, the wobbling translation component linew (t) is expressed by the following equation (7).

Next, in step S9, the wigbing translation component line W (t) in the weld line coordinate system Σline is rotationally transformed using the rotation matrix baseR _line, and the wobbling translation component W (t) in the base coordinate system Σbase (= (W _x , _Wy , _Wz )). The wobbling translation component W (t) in the base coordinate system Σbase is expressed by the following equation (8).

Subsequently, in step S10, as shown in the following equation (9), the towing tip 16a before adding the weaving component is added to the position P _c (t) of the torch tip 16a, and the torch after adding the weaving component is added. The position P (t) of the tip 16a is obtained. The position P (t) of the torch tip 16a corresponds to the “temporary tool tip position” in the present invention.

In step S11, the inverse kinematics of the position P _c (t) before the weaving addition is calculated, and the joint angles J _c (J _c1 , J _c2 , J _c3 , J _c4 , J) at the position P _c (t) are calculated. _Jc5 , _Jc6 ) is calculated.

In step S12, forward kinematics of the rotation angles J _c1 , J _c2 , J _c3 of the basic three axes (rotary joints RJ _{m1 to} RJ _m3 ) are calculated, and a rotation matrix baseR _arm indicating the arm tip position P _arm and the arm direction is calculated. Ask. When J _c23 = J _c2 + J _c3 , the link length between the rotary joints RJ _m2 and RJ _m3 is L2, and the link length between the rotary joints RJ _m3 and RJ _m5 is L3, the arm tip position P _arm is expressed by the following equation (10): The rotation matrix baseR _arm indicating the arm direction is expressed by the following equation (11).

In step S13, an equation of a spherical surface Q is obtained with a distance Lw from the arm tip position P _arm to the torch tip 16a as a radius centered on the arm tip position P _arm (= (X _arm , Y _arm , Z _arm )). The equation of the spherical surface Q is as the following equation (12).

Next, in step S14, it obtains an intersection _{P Q} of the straight line and the spherical Q passing through the arm tip position _{P arm} and Wibingu position after adding P (t). The intersection point P _Q corresponds to the "true tool tip position after Wibingu component summing" in the present invention.

Next, in step S15, joint angles J ₄ ′ and J ₅ ′ of the joints RJ _m4 and RJ _m5 are calculated from the intersection point P _Q , the arm tip position P _arm , and the rotation matrix baseR _arm indicating the arm direction. A method for calculating the joint angles J ₄ ′ and J ₅ ′ will be described in detail with respect to a second embodiment described later.

The joint angle output as a command value to the step S16 all axes (rotary joint _RJ m1 _{~RJ m6),} to operate the manipulator 11. Specifically, for the basic three axes (rotary joints RJ _{m1 to} RJ _m3 ) and the most advanced axis of the wrist (rotary joint RJ _m6 ), the position P _c (t) before the weaving addition calculated in step S11. The joint angles J _{c1 to} J _c3 and J _c6 calculated by the inverse kinematics are output. For the fourth and fifth axis contrast (rotary joint _RJ m4, _{RJ m5),} joint angle _{J 4} calculated from the intersection _{P Q} in step S15 ', _{J 5'} outputs a.

  The processing from step S6 to step S16 is repeated until the end position P2 taught in step S17 is reached.

Thus, in the present embodiment, the fourth and fifth axes (rotary joint RJ _m4 ) of the wrist without swinging the basic three axes (rotary joints RJ _{m1 to} RJ _m3 ) having relatively low rigidity at a high weaving frequency. , RJ _m5 ) alone can be performed, and a higher frequency whibbing operation is possible. Further, it is not necessary that the extension line of the torch 16 intersects the fifth axis, and a wobbling operation using only two axes of the fourth and fifth axes can be realized without using the torch 16 having a special shape. Further, the tool tip position calculated in the conventional control for realizing the wobbling operation on the first to sixth axes is calculated as the position of the temporary torch tip 16a (step S10), and the tool tip position is the fourth target for the wobbling operation. And the true tip position for obtaining the joint angle of the fifth axis (step S14). In this respect, the calculation is comparatively easy and can cope with various wobbling operation patterns. It is also possible to easily switch to the control for realizing the weaving operation using all of the first to sixth axes. For example, the conventional method can be used when the frequency of the whibbing operation is low, and the method of the present embodiment can be employed when the frequency of the whibbing operation is high.

In the present embodiment, the joint angles J ₄ ′ and J ₅ ′ of the fourth and fifth axes for performing the weaving operation are determined by the intersection point P _Q which is the closest position to the position of the temporary torch tip 16a on the spherical surface Q. Therefore, the error compared with the conventional case where the first to sixth axes are all used to realize the wobbling operation is minimized.

(Second Embodiment)
Second embodiment of the present invention shown in FIGS. 5A, 5B, the step of the fourth and fifth joint angles J ₄ ', J _5' method of determining the intersection point P _Q on a spherical surface Q to be used in the calculation of (Figure 4B Only S14) is different from the first embodiment. That is, in the second embodiment, a plane Pw including the Y coordinate direction Y _line (wibbing width direction) of the weld line coordinate system Σline and the Z coordinate direction Zline is considered. Then, the point determined at the intersection P _Q the closest to the position P of the temporary torch tip 16a (t) on交円C of this plane Pw and spherical Q, using the intersection point P _Q joint angle J ₄ ' , J ₅ '.

In the method of the present embodiment, the error compared with the case of realizing the weaving operation using all of the first to sixth axes is larger than that of the first embodiment, but the welding progress direction (welding The direction of the wobbling operation is always perpendicular to the X _line direction of the line coordinate system Σline.

Next, it will be described in detail the calculation of the intersection point _{P Q} with reference to FIG 6A-6D. In the following description, the base coordinate system Σbase is used as a reference unless otherwise specified.

As described above, when the wobbling operation is executed only with the fourth and fifth axes (rotary joints RJ _m4 , RJ _m5 ), the trajectory drawn by the torch tip 16a is referred to as an arm tip position P _arm (hereinafter referred to as point G). ) At the center, and is located on a spherical surface Q having a radius of a line segment connecting the point G and the torch tip 16a. Here, the plane formed by the X coordinate direction (welding progress direction) X _line of the weld line coordinate system Σline and the Z coordinate direction Z _line is denoted by S, and the intersection of the spherical surface Q and the plane S is denoted by C.

Let the foot of the perpendicular drawn from the point G on the plane S be the point H. At this time, the length h of the line segment GH is a vector (vector GE) from the point G to the point E (the intersection of the straight line extending from the point H to the Z coordinate direction Z _line on the plane S and the intersection circle C). It is an inner product of unit vectors X _line (= (v _x , v _y , v _z )) in the weld line direction. Therefore, when the coordinates of the point G are (x ₅ , y ₅ , z ₅ ) and the coordinates of the point E are (x _e , y _e , z _e ), the length h is expressed by the following equation (13). .

Using this length h, the coordinates (x _h , y _h , z _h ) of the point H are expressed by the following equation (14).

  The point H is the center coordinate of the intersection circle C, and the radius of the intersection circle C is the line segment HE. Here, when the length of the line segment GE is 1 (constant), the length of the line segment HE is expressed by the following formula (15) from the three-square theorem.

The tip position P (t) of the temporary torch tip 16a is given as a target value in the weaving width direction, and the length of the line segment EP (t) is u. The coordinates (x _r , y _r , z _r ) of the position P (t) are expressed by the following equation (16) using the unit vector Y _line (= (w _x , w _y , w _z )) in the weaving width direction. expressed.

The intersection of the straight line GP (t) and spherical Q comes out the intersection _{P Q} to be obtained. Since the intersection point P _Q is in the direction of the vector HP (t) as viewed from the point H, the coordinate of the intersection point PQ is represented as unknowns d (d> 0) using the following equation (17).

  From the equations (16) and (17), the square of the length of the line segment HW is expressed as the following equation (18).

  On the other hand, the square of the length of the line segment HE (equal to the length of the line segment HW) is expressed by the following expression (19) from the expression (15).

  Therefore, using the fact that the right sides of the equations (18) and (19) are equal, the variable d is expressed as the following equation (20).

On the other hand, substituting the equation (14) Equation (16) into equation (17), the intersection point _{P Q} coordinates _{(x w, y w, z} w) is expressed as the following equation (21).

By substituting equation (20) into equation (21), the intersection point P _Q , that is, the position of the true torch tip 16 a can be obtained.

  The above calculation is not limited to the case where the position P (t) of the temporary torch tip 16a is located outside the intersection circle C as shown in FIG. 6C, and the position P (t) intersects as shown in FIG. 6D. It is also established when it is located inside the circle C.

Next, a method of calculating joint angles J ₄ ′ and J ₅ ′ of the fourth and fifth axes (rotating joints RJ _m4 and RJ _m5 ) from the intersection point P _Q obtained by Expression (21) (step S15). This will be described in detail with reference to FIGS. FIG. 7B is a view of FIG. 7A viewed from the direction of the rotation axis of the fifth axis (rotary joint RJ _m5 ) as indicated by an arrow A. Here is the third coordinate system axes (rotary joint _{RJ m3)} a (orthogonal coordinate system) and sigma _3, direction cosines _{X 3 (e x, e y} , e z), Y 3 (f x, f y, f _z ), Z ₃ (g _x , g _y , g _z ) are set as shown in FIG. 7B.

Extending a straight line from the point G to the X-axis direction of the sigma ₃ coordinate system, the perpendicular foot dropping off from the intersection P _Q to this straight line and the point N. Since the joint angle J ₅ ′ of the fifth axis (rotary joint RJ _m5 ) is an angle formed by the line segment GN and the line segment GP _Q , the following equation (22) is established.

'But are determined, joint angles _{J 4} of the fourth axis (rotary joint _{RJ m4)'} This equation (22) joint angle _{J 5} from the joint angle _{J 5} by 'may be positive or negative. However, assuming that the torch tip 16a does not pass through a singular point (a point at which the joint angle J ₅ ′ becomes 0) and the joint angle of the rotary joint RJ _m5 detected during the previous control cycle is positive, the current control is performed. If the joint angle J ₅ ′ in the cycle is positive, and if the joint angle of the rotary joint RJ _m5 detected in the previous control cycle is negative, the joint angle J ₅ ′ in the current control cycle is also negative. Can be determined.

Next, a joint angle J ₄ ′ of the fourth axis (rotary joint RJ _m4 ) is obtained. When the length of the line segment GN is n, the following formula (23) is established.

The coordinates (x _N , y _N , z _N ) of the point N are expressed by the following formula (24) based on the length n.

  Moreover, the length of the line segment NW is represented by the following formula (25) from the three-square theorem.

The view of the point G along the direction cosines X ₃ Figure 7B from point N shown in FIG. 7C.

  When the angle α and the angle β are determined as shown in FIG. 7C, the following equations (26) and (27) are established.

Since the parameters on the right side of the equations (26) and (27) are all known, the angles α and β can be calculated from these equations. Further, the following expressions (28) to (31) are established with respect to the sign of the joint angle J ₅ ′ of the _fifth axis, and the joint angle J ₄ ′ of the fourth axis is obtained from these.

  Other configurations and operations of the second embodiment are the same as those of the first embodiment (see FIGS. 1, 4A, and 4B).

(Third embodiment)
Figure 8A, even the third implementation of the invention shown in 8B, the steps of the fourth and fifth joint angles J ₄ ', J _5' method of determining the intersection point P _Q on a spherical surface Q to be used in the calculation of (Figure 4B Only S14) is different from the first embodiment. That is, in the third embodiment, a plane Pw ′ determined by the Y coordinate direction Y _line (wibbing width direction) of the weld line coordinate system Σline and the vector G _inv in the opposite direction of gravity g is considered. Then, to determine a point close to the position P of the most tentative torch head 16a on交円C of this plane Pw 'and spherical Q (t) at the intersection P _Q, joint angle J ₄ using the intersection point P _Q ', J ₅ ' is calculated.

  In the method of the present embodiment, the error compared to the case of realizing the whibbing operation using all of the conventional first to sixth axes is larger than in the case of the first embodiment. Is always perpendicular to.

  Other configurations and operations of the third embodiment are the same as those of the first embodiment (see FIGS. 1, 4A, and 4B).

The schematic diagram which shows the robot which concerns on 1st Embodiment of this invention. The typical perspective view of the wrist part of the manipulator for demonstrating the principle of this invention. The typical perspective view of the wrist part of the manipulator for explaining the control method of a 1st embodiment of the present invention. The schematic diagram which looked at the wrist part of the manipulator for demonstrating the control method of 1st Embodiment of this invention from the rotation direction of the 5th axis | shaft. The flowchart for demonstrating the control method of 1st Embodiment of this invention. The flowchart for demonstrating the control method of 1st Embodiment of this invention. The typical perspective view of the wrist part of the manipulator for demonstrating the control method of 2nd Embodiment of this invention. The schematic diagram which looked at the wrist part of the manipulator for demonstrating the control method of 2nd Embodiment of this invention from the rotation direction of the 5th axis | shaft. Schematic perspective view of the manipulator wrist portion for explaining a method of calculating the coordinates of the intersection point P _Q in the second embodiment. Schematic view of the manipulator wrist portion for explaining a method of calculating the coordinates of the intersection point P _Q in the second embodiment the rotating direction of the fifth axis. Schematic view of the wrist portion as viewed from the line segment HG direction of the manipulator for explaining a method of calculating the coordinates of the intersection point P _Q in the second embodiment (when the position P (t) is located outside of交円C). The intersection P _Q schematic view the manipulator wrist portion for explaining a method of calculating the coordinates as seen from the line segment HG direction (when the position P (t) is located inside the交円C). Schematic perspective view of the manipulator wrist portion for explaining a method of calculating the intersection P _Q joint angles of the fourth and fifth axes. The schematic diagram which looked at the wrist part of the manipulator for demonstrating the method of calculating the joint angle of the 4th and 5th axis | shaft from intersection _PQ from the rotation direction of the 5th axis | shaft. Schematic view of the manipulator wrist portion for explaining the intersecting point P _Q a method of calculating the joint angle of the fourth and fifth axis line segment HG direction. The typical perspective view of the wrist part of the manipulator for demonstrating the control method of 3rd Embodiment of this invention. The schematic diagram which looked at the wrist part of the manipulator for demonstrating the control method of 3rd Embodiment of this invention from the rotation direction of the 5th joint. The schematic diagram for demonstrating the control method of the conventional whibbing operation | movement. The graph which shows an example of a wobbling operation | movement pattern. Schematic diagram showing a general vertical articulated robot.

Explanation of symbols

11 the manipulator 12 the controller 13 teaching apparatus 14 processor 15 memory 16 torch 16a torch tip _{RJ m1, RJ m2, RJ m3} , RJ m4, RJ m5, RJ m6 rotating joint

Claims (3)

A control method for causing a 6-axis vertical articulated welding robot to perform a wobbling operation,
The welding robot includes first to third axes as basic axes, fourth to sixth axes as wrist axes for holding a tool, and an arm tip position including the first to fourth axes. The fifth axis is located at
For each control cycle,
Solving the inverse kinematics of the tool tip position before adding the wobbling component to calculate at least the angles of the first to third and sixth axes;
Add a wibbing component to the tool tip position before adding the wibbing component to obtain a temporary tool tip position after adding the wigbing,
Finding a spherical surface centered on the tip position of the arm and having a radius from the tip position of the arm to the tool tip,
Using at least the temporary tool tip position, find the true tool tip position after adding the wobbling component on the spherical surface,
Calculating the joint angles of the fourth and fifth axes from the true tool tip position after the addition of the wobbling component;
Outputting the joint angles of the first to sixth axes as commands to the manipulator to drive the first to sixth axes;
The Wibingu component summing the true tool tip position after shall be the wherein the straight line passing through the tip positions of said the tentative tool tip position after the Wibingu component summing arm is an intersection which intersects with the spherical soluble Control method for contact robot. A control method for causing a 6-axis vertical articulated welding robot to perform a wobbling operation,
The welding robot includes first to third axes as basic axes, fourth to sixth axes as wrist axes for holding a tool, and an arm tip position including the first to fourth axes. The fifth axis is located at
For each control cycle,
Solving the inverse kinematics of the tool tip position before adding the wobbling component to calculate at least the angles of the first to third and sixth axes;
Add a wibbing component to the tool tip position before adding the wibbing component to obtain a temporary tool tip position after adding the wigbing,
Finding a spherical surface centered on the tip position of the arm and having a radius from the tip position of the arm to the tool tip,
Using at least the temporary tool tip position, find the true tool tip position after adding the wobbling component on the spherical surface,
Calculating the joint angles of the fourth and fifth axes from the true tool tip position after the addition of the wobbling component;
Outputting the joint angles of the first to sixth axes as commands to the manipulator to drive the first to sixth axes;
The true tool tip position after the addition of the weaving component is the largest at the position of the temporary tool tip after the addition of the weaving component among the circles of the spherical surface and the plane determined by the direction of the weaving and the progress of welding. the method of welding robot characterized in that a point close. A control method for causing a 6-axis vertical articulated welding robot to perform a wobbling operation,
The welding robot includes first to third axes as basic axes, fourth to sixth axes as wrist axes for holding a tool, and an arm tip position including the first to fourth axes. The fifth axis is located at
For each control cycle,
Solving the inverse kinematics of the tool tip position before adding the wobbling component to calculate at least the angles of the first to third and sixth axes;
Add a wibbing component to the tool tip position before adding the wibbing component to obtain a temporary tool tip position after adding the wigbing,
Finding a spherical surface centered on the tip position of the arm and having a radius from the tip position of the arm to the tool tip,
Using at least the temporary tool tip position, find the true tool tip position after adding the wobbling component on the spherical surface,
Calculating the joint angles of the fourth and fifth axes from the true tool tip position after the addition of the wobbling component;
Outputting the joint angles of the first to sixth axes as commands to the manipulator to drive the first to sixth axes;
The true tool tip position after the addition of the wibbing component is closest to the position of the temporary tool tip after the addition of the wibbing component among the circles of the spherical surface and the plane determined by the direction of the wibbing and the direction of gravity. the method of welding robot you characterized by a point.

Images