JPH01188903A - Correcting method for robot teaching data - Google Patents

Abstract

PURPOSE:To reduce the working load of a robot by correcting the teaching data so as to secure the coincidence between three reference points of the old teaching data and a work point of an end factor. CONSTITUTION:A robot controller 20 obtains the positions A of three reference point which are not set on a single straight line from the old teaching data. Then a work point TCP of a welding torch 46 is moved to three reference points which are used in an old teaching job. Thus a position B of the point TCP is obtained at that time point. When these preparatory jobs are through, the controller 20 compares the position A of each reference point of the old teaching data with the point B of each corrected point TCP. Thus a conversion matrix T is calculated for production of the new teaching data.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for modifying teaching data for a robot, and more particularly, the present invention relates to a method for modifying teaching data for a robot. This invention relates to a method for modifying robot teaching data that can be easily modified while taking into consideration the following.

[Background of the Invention] For example, when performing a predetermined work on a work using an industrial robot, it is necessary to move an end effector attached to the tip of the robot along a predetermined path. In this case, the route is generally taught by a teaching box. Here, the teaching box used for teaching the robot is equipped with input means for selecting command values for driving each joint of the robot, a coordinate system as a reference for teaching, or a setting means for setting the operating speed of the robot. The operator teaches the robot by operating these means according to predetermined specifications.

By the way, when an industrial robot that has been taught in this way is used to perform a predetermined work on a workpiece on a line, for example, the rod of the workpiece may be changed during the work, or the robot may be removed from the line for maintenance. If removed, the position of the robot's work point relative to the work may shift. In this case, if the robot were to be taught again, a considerable amount of time would be consumed. Therefore, a method has been proposed in which the teaching data is corrected based on the teaching data before the rod is changed or before the robot is maintained (Japanese Patent Laid-Open No. 60-95605). Of course,
In order to achieve high-speed data modification processing, this method
The posture of the end effector with respect to the workpiece is not taken into consideration. Therefore, if the position of the robot changes significantly, the posture of the end effector with respect to the workpiece becomes inaccurate, resulting in the inconvenience that highly accurate work is impossible.

[Object of the Invention] The present invention has been made in order to overcome the above-mentioned disadvantages, and it is possible to change the position of the three reference points in the old teaching data and to move the end effector to the three reference points in order to correct the data. By correcting the teaching data so that the position of the end effector's work point matches the position of the end effector when the robot is used, the teaching data can be corrected quickly, easily, and accurately, which contributes to improving the operating rate of the robot. The purpose is to provide a method for correcting robot teaching data that is possible.

[Means for Achieving the Object] In order to achieve the above-mentioned object, the present invention, when correcting the teaching data of an articulated robot, corrects three reference points that are not on a straight line and are selected before the teaching data is corrected. After determining the positions of the three working points obtained by moving the working point of the end effector to the three reference points when modifying the teaching data, the positions of the three reference points and The present invention is characterized in that the teaching data is corrected by determining a coordinate transformation matrix that matches the positions of the three working points, and applying the coordinate transformation matrix to the teaching data before modification.

Further, the present invention is characterized in that the coordinate transformation matrix is set so that a triangle formed by the positions of the three reference points and a triangle formed by the positions of the three work points are superimposed.

[Embodiments] Next, preferred embodiments of the robot teaching data correction method according to the present invention will be described in detail with reference to the accompanying drawings.

In FIG. 1, reference numeral 10 indicates a welding system to which the robot teaching data correction method according to the present embodiment is applied, and in this welding system 10, a sequencer 1
Arc welding robot 1 based on sequence control by 2
4 is driven, and a predetermined welding operation is performed on the work 15 positioned on the jig table 17 by the clamp device 16.

In this case, the arc welding robot 14 is taught its operation by the teaching box 18, and is driven and controlled by the hydraulic unit 22 and the welding controller 24 via the robot controller 20.

The arc welding robot 14 is installed on a base 26, and the moving part 28 is moved in the direction of the arrow with respect to the base 26 by a hydraulic motor 30, and the rotating part 32 is moved to the moving part 28 by a hydraulic motor 34. On the other hand, it is configured to be able to turn in the direction of the arrow. Further, one end portion of an arm member 36 is pivotally attached to the rotating portion 32, and this arm member 36 is attached to the rotating portion 32.
It is configured to be movable up and down in the direction of the arrow by a hydraulic cylinder 38 attached to. On the other hand, a mounting member 42 having a hydraulic motor 40 is attached to the other end of the arm member 36, and a welding torch 46 as an end effector is connected to the mounting member 42 via a hydraulic motor 44. In this case, the welding torch 46 is centered around the hydraulic motor 44 and
0, the hydraulic motor 44 rotates in the direction of the arrow.
It is configured to be rotatable in the direction of the arrow. The hydraulic unit 22 described above drives and controls the hydraulic motors 30, 34, 40 and 44 and the hydraulic cylinder 38 that constitute the arc welding robot 14. Further, the welding controller 24 achieves welding current control between the welding torch 46 and the workpiece 15.

Here, with respect to the arc welding robot 14, the base coordinate system 0° fixed to the base 26 and the workpiece 1 of the welding torch 46 are
Hand coordinate system 0.5 fixed at work point TCP. At the same time, coordinate systems 0+ to 06 are defined for each joint of the arc welding robot 14.

The base coordinate system O0 consists of three orthogonal axes x, , Y, , and zo, and the rotation directions around each axis are A0, Bo, and Co, respectively. Hand coordinate system 0. is the orthogonal 3 of X, , y, , z
It consists of axes, and the rotation directions around each axis are A, , B, respectively.
,C,. The coordinate system 0 is a coordinate system that is displaced with respect to the base coordinate system 0° by the hydraulic motor 30, and the coordinate system 0
□ is a coordinate system rotated with respect to the coordinate system O3 by the hydraulic motor 34, and the coordinate system O8 is a coordinate system rotated with respect to the coordinate system 02 by the hydraulic cylinder 38.

Further, the coordinate system 04 is a coordinate system fixed to the coordinate system 03, and is added to explain the arc welding robot 14 as a 6-axis robot. Furthermore, the coordinate system 0. is a coordinate system rotated with respect to the coordinate system 04 by the hydraulic motor 40, and the coordinate system 06 is rotated with respect to the coordinate system 0. by the hydraulic motor 44. It is a coordinate system that rotates with respect to

On the other hand, the teaching box 18 is configured as shown in FIG. That is, the teaching box 18 basically includes a main body section 52 that performs teaching etc. to the arc welding robot 14, and a display section 54 that displays the operating method of the teaching box 18 and the like.

The main body 52 is connected to the robot controller 20 via a connection cable 55, and on the top thereof there is a mode changeover switch 56 for switching between a teach mode, a play mode, etc., which will be described later, and a mode switch 56 for manually operating the arc welding robot 14. A joystick 58 for controlling the arc welding robot 14, a numeric keypad 60 for selecting functions, inputting data, etc., and an emergency stop button 62 for stopping the operation of the arc welding robot 14 in an emergency are provided. In this case, joystick 58
is configured to be able to tilt in the directions of arrows α and β and to rotate in the direction of arrow γ, and the correspondence between the tilting direction or rotation direction and the driving direction of the arc welding robot 14 is set by the numeric keypad 60. Ru. Further, the operating speed of the arc welding robot 14 is set by the tilt angle or rotation angle of the joystick 58. In addition,
An operation switch 64 is provided at the tip of the joystick 58, and by pressing the operation switch 64, the arc welding robot 14 becomes operable.

The main body portion 52 includes a pair of arm members 6 that protrude diagonally upward.
6a, 66b, and these arm members 66a
, 66b through mounting screws 68a, 68b, the display section 54 is rotatably mounted in the direction of arrow θ. In this case, the display section 54 has an LCD display 70, on which the operating method of the teaching box 18 selected by the mode changeover switch 56 and the numeric keypad 60 is displayed.

The welding system to which the robot teaching data correction method according to the present embodiment is applied is basically configured as described above. Next, a teaching method using this system will be described.

First, the power of the teaching pendant 18 is turned on. In this case, a main menu screen 80 shown in FIG. 3 is displayed on the LCD display 70 of the teaching box 18. In this main menu screen 80, rTE
ACH is a teaching mode in which the arc welding robot 14 is taught using the joystick 58, and [PLAYJ is a play mode in which desired teaching data is called up using a keyboard (not shown) of the robot controller 20 and the arc welding robot 14 is operated. It is. rAUTOJ is an auto mode that automatically executes the play mode described above based on a request from the sequencer I2, etc.
The actual welding work is done in this auto mode. Further, rEDITJ is an editing mode for editing the teaching data stored in the teaching box 18, and in this mode, a three-dimensional shift, which is a teaching data modification work according to this embodiment, is performed. rPAR
AJ is a welding torch 46 in the arc welding robot 14
This is a parameter setting mode for setting parameters such as dimensions.

Next, when the operator selects the teach mode from the main menu screen 80 and sets the mode selector switch 56 to the [1] position, 1. A teach mode menu screen 82 shown in FIG. 4 is displayed on the LCD display 70.

In this case, the rJOINT 123J is connected to the hydraulic motors 30 and 34 of the arc welding robot 14 for the movement of the joystick 58 in the arrow α, β and T directions.
This is the mode in which the hydraulic cylinder 38 is driven. In addition, "rJOINT 456" is a joystick 58
Hydraulic motor 40 for movement in the directions of arrows α and β
and 44 are the driving modes. rBASE
XYZJ is a mode in which the working point TCP of the welding torch 46 moves in the Xo, Yo, and Z0 directions of the base coordinate system 0° with the base 26 as a reference in response to the operation of the joystick 58. Furthermore, rBAsE ABCJ is the working point T of the welding torch 46 with respect to the movement of the joystick 58.
CP is the three orthogonal axes x, , y, , of the base coordinate system 0°,
This is a mode in which it rotates around z in the Ao, Bo, and Co directions. On the other hand, rHAND XYZ is a hand coordinate system O in which the work point TCP of the welding torch 46 is based on the work point TCP with respect to the operation of the joystick 58.

This mode moves in the three orthogonal axes x, , y, , and z. Similarly, rHAND ABC, the working point TCP of the welding torch 46 is the orthogonal three axes X of the hand coordinate system 0,
, Y, , Z, in the Ao, B, , C, directions. “BASE MDIJ and rHAN
D MDIJ indicates the position of the work point TCP of the welding torch 46 in the base coordinate system O0 or in the hand coordinate system 0. This is a mode in which corrections are made based on correction data that is input as a reference. "rMEMORI ZE" indicates a mode for importing the current state of the arc welding robot 14 into a storage means (not shown) of the robot controller 20 as teaching data.

Therefore, for example, the mode "rBASE XYZ from the teach mode when 3 keys are input from the numeric keypad 60" is selected. Next, while pressing the operation switch 64, the operator tilts or rotates the joystick 58 by a predetermined amount in the directions of arrows α, β, and T. in this case,
Data regarding the tilt direction, rotation direction, tilt angle, and rotation angle of the joystick 58 are transferred to the robot controller 20 via the connection cable 55. The robot controller 20 converts each coordinate system 0. A pulse signal corresponding to the movement amount from 0 to 06 is generated and output to the hydraulic unit 22. The hydraulic unit 22 moves the working point TCP of the welding torch 46 in the arc welding robot 14 to a desired welding position on the workpiece 15 according to the base coordinate system O0 based on the pulse signal. In this case, the work point TCP is moved in the 70 direction by tilting the joystick 58 in the arrow α direction, moved in the X0 direction by tilting the joystick 58 in the arrow β direction, and moved in the X0 direction by rotating it in the arrow γ direction. move in each direction.

On the other hand, when the operator confirms that the working point TCP of the welding torch 46 has moved to the desired welding position on the workpiece 15 and that the welding torch 46 has assumed the desired attitude relative to the workpiece 15 by the operation described below, the teaching box A teaching data storage command is input using the 18 numeric keys 60. In this case, when O and I on the numeric keypad 60 are pressed in accordance with the teach mode menu screen 82 shown in FIG. be remembered. By repeating the above operations for each desired welding position, the teaching operation of the arc welding robot 14 is completed.

Note that during this teaching work, the welding torch 4
3 which is not on a straight line with the work point TCP of 6 set as the line
By positioning at two reference points, pulse data of each axis is sampled and stored. In this case, arbitrary teaching points that are not on a straight line may be selected as the three reference points.

When the teaching work of the arc welding robot 14 is completed, the operator returns the screen of the LCD display 70 of the teaching box 18 to the main menu screen 80 shown in FIG. 3. Next, set the mode selector switch 56 to [21
The arc welding robot 14 is set to play mode, and the arc welding robot 14 is made to perform a playback operation to confirm the teaching data.

Next, arc welding robot 1 based on the teaching data
When the operation check in step 4 is completed, the operator returns the screen of the LCD display 70 to the main menu screen 80, and then selects [3] on the numeric keypad 60 to enter the auto mode. In this case, the arc welding robot 14 is connected to the sequencer 1
A desired welding operation is performed on the workpiece 15 based on the teaching data output from the robot controller 20 under the control of the robot controller 20.

By the way, if the rod of the workpiece 15 is changed while the arc welding robot 14 is performing welding work, or if the arc welding robot 14 is removed from the line for maintenance and reinstalled, the welding torch 46
A case may occur in which the work point TCP of the workpiece 15 is not set to the desired digit 1. In such a case, in this embodiment, the teaching data is corrected by selecting the edit mode on the main menu screen 80.

Therefore, a method for correcting teaching data will be explained based on the flowcharts shown in FIGS. 5a and 5b.

After displaying the main menu screen 80 shown in FIG. 3 on the LCD display 70 of the teaching box 18, the operator enters [4] from the numeric keypad 60 to convert the old teaching data to new teaching data in the edit mode. Perform work (3D shift).

Therefore, first, the robot controller 20 determines the positions of the three reference points that are not on a straight line from the old teaching data (STPI).

In this case, since the data of the reference point is held as pulse data, the position is calculated as follows. That is, the robot controller 20 generates pulses P for each axis related to each reference point.

(i=1.2,...6) and this pulse P8
The current angle r1 of each joint with respect to the adjacent coordinate systems OI to 06 is determined from (i=1.2, . . . 6). Here, kli is a parameter related to the resolution of the potentiometer and the reduction ratio in the coordinate system 01. k2! is an offset amount for expressing the pulse P obtained by the potentiometer as an increase or decrease with respect to the reference midpoint when a reference midpoint is set for the range of motion of the joint in the coordinate system O. Also, k3
i is a parameter for converting an angle (deg) into radians. Note that in this embodiment, since the coordinate system 04 is fixed with respect to the coordinate system 03, the pulse P4 is set to 0.

Next, a transformation matrix T is calculated from the current angle r calculated based on equation (1). In this case, the transformation matrix T is the coordinate system 0.degree. with respect to the base coordinate system 0.degree. The position and orientation of the transformation matrix T are set as tffi7(
m=1.2.3, n=1.2.3.4), and (i=1.2, . . . 6) are set. Here, the transformation matrix T is Tmi(a'
, b) ...If defined as (3),
The position A of the reference point in the old teaching data is determined using the transformation matrix T, A = 'T' -h E a -h
It is calculated as o + b (4). Note that a is the left 4 rows 3 of the transformation matrix T.
Indicates the column component, and b indicates the component in the 4th row and 1st column on the right side of the transformation matrix T. In addition, h is a vector indicating the offset amount of the work point TCP with respect to the coordinate system 06, and the Xs, Yo, and Zo force direction components with respect to the base coordinate system 0° are expressed as h+,
hz, h: +, defined as follows.

Next, the working point TCP of the welding torch 46 is moved to the three reference points used in the old teaching 5TP2), and the position B of each working point TCP at that time is determined (SrF2). In this case, the position B of each work point TCP can be determined in the same manner as in step 1 described above.

When the above preparation work is completed, the robot controller 2
0 calculates a transformation matrix T, 4 for creating new teaching data by comparing the position A of each reference point in the old teaching data and the position B of each work point TCP after correction. Below, according to FIG. 6 a to C, the transformation matrix T,
The calculation procedure of step 4 will be explained. In this case, the base coordinate system O0
Let the three reference point positions A for , AI, A2, and A
3, and the positions B of the three work points TCP after correction are respectively B
0, B2, and B3.

First, a transformation matrix T1 for translating B to the origin of the base coordinate system 0° is determined (SrF2, see FIG. 6a). Next, A8,kt,k,,B3,B2,B, are translated in parallel using the transformation matrix TI obtained in step 4 (
SrF2). In this case, the positions after parallel translation are A
I*, A21. A3'l, Bl", B 2", B
3", then A j"= T' , , A=
j,
(6) Bj”=T’, B, ・
...(7) (j=1.2.3). Next, a transformation matrix T2 for translating AI'k obtained in step 5 in parallel to B- is obtained (SrF6, see FIG. 6a). Using this transformation matrix T2, A-1A? , A?
, when (SrF2) is translated in parallel, the positions after translation are +"% kz", A, respectively. Close, A.”
m= 7" 2Aj*...(
8) (j=1.2.3). As a result, the triangle δ consisting of A1, A2, and A3
The triangle ε consisting of B8, B2, and B3 is AI. and B-
is set at the origin of the base coordinate system O0 in common with (
(See Figure 6).

Next, a vector vA perpendicular to the triangle δ and a vector ■ perpendicular to the triangle ε are determined (SrF8). in this case,
Let the vector from AI$11 to A2*11 be A, 2°,
The vector from A1° to A3** is AI3**, the vector from B to B2″ is B, ?, and the vector from Bl″ to B,″
Let the vector leading to B I3'' be va = AI
z ” X Also −(9)V B =
B Iz' X B 13”...0
becomes ω. Also, find the cross product of vectors ■1 and ■, (
SrF9), and if this cross product is the vector ■9, then ■v = ■A X 7m
...(becomes IQ.Furthermore, find the angle θ1 formed by the vector V and ■8 (STPIO). Then, from this angle θ, find the rotation matrix R7 for rotating 01 around the vector ■9. (STPII).

Therefore, using the rotation matrix R0 obtained in step 11, A
2° and A,. is rotated by θ1 around the vector vv (STP12).

In this case, triangles α and β are set on the same plane (
(See Figure 6C). In addition, rotationally converted A2゜ and A
3°! ” ” and A3*0, A2”” = R, A, “” ...02) A
3"'l = R, A3"...G3).

Next, A. The vector from to A2"" is A+ z
"", the vector from B to 82" is B10
'', and the angle θ2 formed by these vectors A1□-9 and B12'' is determined (STP13). From this angle θ2, a rotation matrix R2 for rotating by θ22 around the vector ■3 is determined (STP14). Here, using the rotation matrix R2 obtained in step 14, A? If O and A3°0 are rotated by θ2 around vector VB, triangles α and β will be superimposed. Therefore, the transformation matrix T is calculated using each matrix obtained in the above transformation process (STP15
). In this case, the transformation matrix T,4 is set in step 4.6.11
And from each matrix obtained in step 14, 'r, = T', -'R
2R, T2 "17°...04). New teaching data is obtained by applying the transformation matrices T and 4 obtained in the above manner to the old teaching data (S T PI3).

[Effects of the Invention] As described above, according to the present invention, the positions of the three reference points before the teaching data is corrected and the working point of the end effector when the end effector is moved to the reference point during data correction. A coordinate transformation matrix is obtained from the position of , and new teaching data is created using this coordinate transformation matrix. In this case, there is no need to perform teaching again to correct the data, so the work is reduced and data correction becomes extremely easy. In addition, since the coordinates are translated and rotated based on the work point by the end effector, not only the position but also the posture of the work point relative to the workpiece is accurately corrected, and the corrected teaching data can be used to perform high-precision work. can be continued. Furthermore, since it does not take a long time to create the data, the impact on the manufacturing line to which the robot is applied is extremely small, and it can also contribute to improving the operating rate.

Although the present invention has been described above with reference to preferred embodiments, the present invention is not limited to these embodiments.
Of course, various improvements and changes in design are possible without departing from the gist of the present invention.

[Brief explanation of the drawing]

FIG. 1 is a schematic configuration diagram of a welding system to which the robot teaching data correction method according to the present invention is applied; FIG. 2 is a perspective view of the configuration of a teaching box in the welding system shown in FIG. 1; FIGS. The figure is an explanatory diagram of each menu screen displayed on the teaching box shown in Figure 2. Figures 5a and b are flowcharts showing the procedure of the robot teaching data correction method according to the present invention. Figures 6a to C are FIG. 3 is a procedure explanatory diagram of a robot teaching data correction method according to the present invention. 10... Welding system 12... Sequencer 1
4... Arcn receptacle robot 18... Teaching box 20... Robot controller 22... Hydraulic unit 24... Welding controller 52... Main body part 54... Display part 5
6...Mode selector switch 58...Joystick 60...Numeric keypad 62
...Emergency stop button 64...Operation switch 70
...LCD display patent applicant Honda Motor Co., Ltd. FIG, 3 FIG, 4

Claims (2)

[Claims] (1) When modifying the teaching data of an articulated robot, find the positions of the three selected reference points that are not on a straight line before modifying the teaching data, and then calculate the end effector work point when modifying the teaching data. After determining the positions of the three work points obtained by moving the three reference points to the three reference points, a coordinate transformation matrix for matching the positions of the three reference points and the three work points, and A method for correcting teaching data for a robot, characterized in that teaching data is corrected by applying a coordinate transformation matrix to teaching data before correction. (2) In the method according to claim 1, the coordinate transformation matrix is 3
A method for modifying teaching data for a robot, characterized in that a triangle formed by the positions of one reference point and a triangle formed by the positions of three work points are set to overlap.