JP2007115011A - Control method of industrial robot - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a control method of industrial robot, in which existing teaching data can be reused as it is without teaching correction even in the event of slippage of relative positional relation between a manipulator and a work. <P>SOLUTION: At the time of teaching, position attitude data of a teaching point is stored as user coordinate value data based on a user coordinate system, and at the time of reproduction, base coordinate value data is calculated based on the user coordinate value data and the user coordinate system, and each joint angle data of a robot is calculated based on the base coordinate value data. When the work is set in a position different from a reference position, positions of a plurality of characteristic points are redesignated to redefine the user coordinate system, and at the time of reproduction, the base coordinate value data is recalculated based on the user coordinate value data and the redefined user coordinate system, and each joint angle data of the robot is calculated based on the recalculated base coordinate value data. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a method for controlling an industrial robot that corrects the position and orientation of created teaching data.

  When processing operations such as arc welding and spot welding are realized by a robot, a method called teaching playback is generally adopted. In the teaching playback method, a work is placed at a predetermined position, the operation of the manipulator with respect to the work is taught in a state where the relative positional relationship between the manipulator and the work is constrained, and this teaching data is reproduced repeatedly. This is a method of continuously processing workpieces of the same type by operating them.

  FIG. 5 is a general configuration diagram of an industrial robot adopting this teaching playback method. In the figure, a manipulator 1 includes a tool T for processing a workpiece W at the tip of an arm 2. The teach pendant 3 is a device for an operator to teach the operation of the manipulator 1 with respect to the workpiece W, and a button group 4 for moving the manipulator 1 is provided on the board surface. As shown in the figure, the button group 4 is generally equipped with a plurality of buttons corresponding to the directions (± X, ± Y, ± Z) of the coordinate system. For example, when the manipulator is moved in the base coordinate system which is a coordinate system unique to the robot individual, when the X + button of the button group 4 is pressed, the manipulator 1 moves in the X + direction of the base coordinate system. As described above, the operator operates the button group 4 to move the manipulator 1 to a desired position and teaches a work path for performing the machining work. Then, the position and orientation coordinate value of the manipulator 1 at the teaching point on the work path is stored in the robot control device 5 as teaching data. The robot control device 5 drives and controls the drive motor of the manipulator 1 based on the teaching data and current position information from an encoder (not shown) to move the tool T to the teaching point.

  FIG. 6 is a diagram for explaining the relationship between the moving direction of the manipulator and the coordinate system during teaching. The manipulator 1 and the workpiece W are the same as those shown in FIG. The work path K is one of the paths necessary for machining the workpiece W, and is formed by the teaching point A and the teaching point B in the case of FIG. The base coordinate system Cb is a coordinate system unique to the manipulator 1. The user coordinate system Cu is a coordinate system that can be arbitrarily defined by the operator. For convenience of explanation, in the base coordinate system Cb and the user coordinate system Cu, X + and Y + in the axial direction are represented by arrows, and the other axial directions are omitted.

  As described above, the operator moves the manipulator 1 to a desired position and teaches the position. However, the button operation for moving is often complicated. For example, assume that the manipulator 1 is moved from the teaching point A that is the current position to the teaching point B that is the target position in order to teach the work route K in FIG. If the direction from the teaching point A to the teaching point B coincides with the X + direction of the base coordinate system Cb, the operator only has to press the X + button of the button group 4. However, the work path K on the workpiece W does not coincide with the axial direction of the base coordinate system Cb. In such a case, the operator needs to move the manipulator 1 by operating the X + button, the Y + button, etc. of the button group 4 in combination. If the workpiece W is large or the work path K is complicated, the operation of moving the manipulator 1 becomes more complicated.

  Therefore, in recent industrial robots, a proposal to eliminate the complexity of the moving operation by defining a user coordinate system Cu that allows the operator to set the origin position and the axial direction of the coordinate system to desired values by himself / herself. (For example, refer to Patent Document 1). As shown in the figure, if the user coordinate system Cu is set so that the direction of the work path K coincides with the X + direction of the coordinate system, it can be moved to the teaching point B by operating only the X + button of the button group 4. it can. That is, since the origin position and direction of the coordinate system can be set to a desired position and direction according to the installation position, shape, etc. of the workpiece W, the complexity of the moving operation can be eliminated and the teaching man-hour can be reduced. is there. The user coordinate system is defined by specifying three feature points on the workpiece as disclosed in Patent Document 1 and the like. Specifically, the user coordinate system Cu is defined by designating arbitrary feature points P1, P2, and P3 on the workpiece W.

  FIG. 7 is a flowchart for explaining how teaching data is conventionally stored and played back.

  In step S1 in the figure, the worker defines the user coordinate system by the method described above. In step S2, the manipulator is moved to the teaching point on the work path according to the axial direction of the user coordinate system. In step S3, a teaching point storing operation is performed. At this time, the position and orientation coordinate values of the manipulator 1 stored as teaching data are stored as base coordinate value data. Even if the manipulator 1 is moved with reference to the user coordinate system, the position / orientation coordinate value stored as the teaching data is not the user coordinate value data but the base coordinate value data.

  If all the work route teachings are not completed in step S4, the process returns to step S2 to perform a movement operation and a storage operation to the next teaching point. If the teaching of the work route has been completed, a reproduction operation start operation is performed in step S5. When the reproduction operation is started, the robot controller reads out the position / orientation coordinate value for each teaching point in step S6. Since each joint of the manipulator is controlled by rotation of the motor, the reproduction position needs to be calculated as a joint angle. For this reason, in step S7, an inverse transformation operation is performed to calculate each joint angle of the manipulator based on the base coordinate value data. In step S8, each joint of the manipulator is rotationally driven to move to the position of each calculated joint angle, that is, to the teaching point. Finally, if the reproduction operation for all teaching points is not completed in step S9, the process returns to step S6 to perform the reproduction operation for the next teaching point. If all the teaching points have been reproduced, the flow ends.

JP-A-61-49205 "Robot Manipulator" by Corona, Richard P. Paul

  By the way, when the installation position of the workpiece W is changed every time the reproducing operation is performed, it is necessary for the operator to correct the teaching point using the teach pendant 3. When the teaching point is out of the desired position, the operator moves the manipulator 1 to the desired position using the teach pendant 3 and then corrects (re-teaches) the teaching point. Since it is necessary to repeat as many as the number of teaching points, a large number of work steps are required.

  Usually, a jig for fixing the work W is prepared so that the teaching point is not corrected, and a device for preventing the installation position of the work W from being varied is prepared. However, if the workpiece W becomes large even if a jig is used, it is practically difficult to install the workpiece W without any difference from the teaching. This is because the jig for fixing the work W itself is also large and strong, and the jig becomes expensive, and a large work tends to cause an error in the dimensions of the work itself.

  In this way, even if the user coordinate system is defined to simplify the operation during teaching and reduce the teaching man-hours, if the workpiece installation position changes each time the replay operation is performed, the teaching point must be corrected again. I have a problem that I have to.

  Note that the relative positional relationship between the workpiece W and the manipulator 1 can be allowed to change to some extent during each reproduction operation, and it is possible to automatically correct the teaching point by attaching the sensor device to the robot. . For example, in the case of an arc welding robot, a sensor called a touch sensor is used. A welding wire is fed as a consumable electrode from the tip of an arc welding torch attached to the tip of the manipulator 1. The application of a voltage to the tip of the wire and detecting that the welding wire is in contact with the workpiece W by energization between the welding wire and the workpiece is called a touch detection operation. By detecting the installation position of the workpiece W using the touch sensor and detecting the displacement of the workpiece W position between teaching and machining, the position and orientation coordinate values of the already taught data are corrected appropriately. It becomes possible to do. However, even the method using the touch sensor has a problem that when the number of teaching points is large, the teaching of the touch detection operation increases, that is, a large teaching man-hour is required. Furthermore, since many touch detection operations are performed during the reproduction operation, there is a problem that the tact time is increased.

  Further, in the case of an arc welding robot, it is also possible to automatically correct a teaching point by attaching a sensor device called an arc sensor to the robot. However, the arc sensor can only detect the deviation of the weld line after starting arc welding. In other words, it is not possible to cope with the case where the workpiece W is greatly displaced.

  The present invention teaches the teaching even if the relative positional relationship between the manipulator and the workpiece is deviated, especially when processing a large workpiece where the number of teaching points is large and the workpiece is not accurately set. It is an object of the present invention to provide an industrial robot control method which can be reused as it is without teaching and correcting data.

In order to achieve the above object, according to the first aspect of the present invention, the origin and the axial direction of the orthogonal coordinate system are set to desired values by designating a plurality of feature points on the workpiece when the workpiece is installed at the reference position. A certain user coordinate system is defined, the robot is moved according to the axis direction of this user coordinate system at the time of teaching, the work path is taught, and the position and orientation data of the teaching point on this work path is expressed in an orthogonal coordinate system unique to the robot individual. In a control method for an industrial robot that stores base coordinate value data based on a certain base coordinate system and calculates motion information by calculating each joint angle data of the robot based on the base coordinate value data during reproduction.
The teaching point position and orientation data is stored as user coordinate value data based on the user coordinate system during teaching,
The base coordinate value data is calculated based on the user coordinate value data and the user coordinate system during reproduction, and each joint angle data of the robot is calculated based on the base coordinate value data.
When the workpiece is installed at a position different from the reference position, the user coordinate system is redefined by respecifying the positions of the plurality of feature points,
At the time of reproduction, the base coordinate value data is recalculated based on the user coordinate value data and the redefined user coordinate system, and each joint angle data of the robot is calculated based on the recalculated base coordinate value data. This is a method for controlling an industrial robot.

  According to a second aspect of the invention, there is provided a control method for an industrial robot characterized in that the redesignation of the position of the feature point described in the first aspect is performed by measuring with a position measuring instrument.

  According to the first aspect of the invention, the relative positional relationship between the manipulator and the workpiece has shifted, particularly when machining a large workpiece where the number of teaching points is large and it is difficult to accurately place the workpiece. However, the existing teaching data can be reused as it is without correcting the teaching. For example, even for large teaching data with hundreds of teaching points, as long as the user coordinate system is redefined by re-specifying multiple feature points on the workpiece, one teaching point is corrected. It can be reused as it is.

  According to the second aspect of the invention, re-designation of a plurality of feature points on the workpiece is performed by automatic measurement by the position measuring device, not by the operator, so that the operator himself / herself re-designates the feature points. There is no need to do. That is, in addition to the effects of the first invention, the number of work steps for the operator can be further reduced.

[Embodiment 1]
DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the invention will be described based on examples with reference to the drawings. The details of the mathematical formulas and calculation procedures described in the present embodiment are described in Non-Patent Document 1.

  FIG. 1 is a flowchart for explaining a flow in which teaching data is corrected according to the present embodiment. FIG. 2 is a diagram for explaining that the user coordinate system is defined by designating feature points on the workpiece when the workpiece is placed at the reference position.

  In FIG. 2, the workpiece W is installed at the reference position Wk, and the manipulator 1 includes a tool T for processing the workpiece W at the tip of the arm 2. The base coordinate system Cb is a coordinate system unique to the manipulator 1, and has an origin Ob and an axial direction (Xb, Yb, Zb). The user coordinate system Cu is a coordinate system arbitrarily defined by an operator, and an origin Ou and an axial direction (Xu, Yu, Zu) are defined as illustrated.

  In step S1 of FIG. 1, the operator designates feature points P1, P2, and P3 on the workpiece W when the workpiece W is installed at the reference position Wk. The robot controller 5 calculates and stores a homogeneous transformation matrix Abu representing the position and orientation of the user coordinate system Cu viewed from the base coordinate system Cb based on the positions of the designated feature points P1, P2 and P3. Hereinafter, a method for calculating the homogeneous transformation matrix Abu will be described.

  First, the coordinate value (hereinafter referred to as base coordinate value data) in the base coordinate system of the tip position (hereinafter referred to as TCP) of the tool T at the feature points P1, P2 and P3 is calculated. By the method described in Non-Patent Document 1, the robot control device 5 reads the joint angle of the position (feature point) after the movement of the manipulator 1 as the output of the encoder directly connected to the motor that drives each joint, and is determined in advance. The position of the TCP with reference to the base coordinate system can be calculated by applying a calculation called forward calculation by applying the arm size of the manipulator 1 and the link parameter defining the arrangement of the rotation axis. As a result, the base coordinate value data Bp1, Bp2, and Bp3 of the feature points P1, P2, and P3 are expressed by the following equations.

上記（１）〜（３）式で、Ｐ１ｘ〜Ｐ３ｘ，Ｐ１ｙ〜Ｐ３ｙ，Ｐ１ｚ〜Ｐ３ｚは、ベース座標値データのＸ，Ｙ，Ｚ成分をそれぞれ表している。

In the above expressions (1) to (3), P1x to P3x, P1y to P3y, and P1z to P3z represent X, Y, and Z components of the base coordinate value data, respectively.

  The following calculation is performed based on the base coordinate value data of the three feature points represented by the above formula, and finally a homogeneous transformation matrix Abu representing the position and orientation of the user coordinate system Cu viewed from the base coordinate system Cb is calculated. To do.

  First, the position of the feature point P1 is set as the origin of the user coordinate system Cu. That is, since the origin Ou of the user coordinate system Cu is the same value as the base coordinate value data Bp1 of the feature point P1, it can be expressed by the following equation.

    Ou = Bp1 = [P1x, P1y, P1z] (4)

  Next, an X-axis vector Xv representing the X-axis of the user coordinate system Cu is calculated. Since the X-axis vector Xv is a vector from the feature point P1 to the feature point P2, the base coordinate value data Bp1 of the feature point P1 expressed by the equation (1) and the base of the feature point P2 expressed by the equation (2) Based on the coordinate value data Bp2, it is expressed by the following equation.

Xv = [Xvx, Xvy, Xvz]
= [((P2x-P1x) / L1,
(P2y-P1y) / L1,
(P2z-P1z) / L1] (5)

  The following equation is substituted for L1 in the above equation (5).

L1 = √ ((P2x−P1x) ² +
^(P2y-P1y) 2 +
(P2z-P1z) ² ) (6)

  Next, the Z-axis vector Zv is calculated from the outer product of the X-axis vector Xv and the vector Qv from the feature point P1 toward the feature point P3. First, the vector Qv is expressed by the following equation.

Qb = [Qvx, Qvy, Qvz]
= [((P3x-P1x) / L2,
(P3y-P1y) / L2,
(P3z-P1z) / L2] (7)

  The following formula is substituted for L2 in the formula (7).

L2 = √ ((P3x−P1x) ² +
^(P3y-P1y) 2 +
(P3z-P1z) ² ) (8)

  Since the Z-axis vector Zv is an outer product of the X-axis vector Xv expressed by the equation (5) and the vector Qv expressed by the equation (7), it is expressed by the following equation.

Zv = [Zvx, Zvy, Zvz]
= [Xvy · Qvz−Xvz · Qvy,
Xvz · Qvx-Xvx · Qvz,
Xvx · Qvy-Xvy · Qvx] (9)

  Next, a Y-axis vector Yv is calculated from the outer product of the Z-axis vector Zv and the X-axis vector Xv. The Y-axis vector Yv is expressed by the following equation.

Yv = [Yvx, Yvy, Yvz]
= [Zvy · Xvz−Zvz · Xvy,
Zvz · Xvx-Zvx · Xvz,
Zvx · Xvy-Zvy · Xvx] (10)

  From the above equations (4), (5), (9) and (10), the homogeneous transformation matrix Abu representing the position and orientation of the user coordinate system Cu viewed from the base coordinate system Cb by the method described in Non-Patent Document 1. Can be expressed by the following equation.

  The homogeneous transformation matrix Abu obtained in this way is stored in the robot controller 5.

  Returning to FIG. 1, in step S <b> 2, the operator determines whether teaching is required for the workpiece W placed at the reference position Wk. In the steps so far, teaching for the workpiece W installed at the reference position Wk is not performed. Therefore, the process proceeds to step S3, and the description is continued assuming that teaching is performed.

  In step S3, the worker moves the TCP of the manipulator 1 to the teaching point R1 (see FIG. 2) on the work path according to the axial direction of the user coordinate system Cu calculated in step S1. In step S4, the operator performs a storing operation for the teaching point R1. At this time, the robot control device 5 calculates and stores a TCP position and orientation coordinate value (hereinafter referred to as user coordinate value data) based on the user coordinate system Cu. Hereinafter, a method for calculating the user coordinate value data Ur1 at the teaching point R1 will be described.

  First, base coordinate value data Br1 of the teaching point R1 is calculated. Since the base coordinate value data Br1 can be calculated by a method similar to the method for obtaining the equations (1) to (3), it is expressed by the following equation.

    Br1 = [R1x, R1y, R1z, R1α, R1β, R1γ] (12)

  From the above equation (12), the homogeneous transformation matrix Abr to TCP at the teaching point R1 viewed from the base coordinate system can be calculated by the method described in Non-Patent Document 1 as the following equation. In the following formula, ◯ represents a numerical value representing a rotation matrix.

  From the equations (11) and (13), a homogeneous transformation matrix Aur representing the position and orientation of the teaching point R1 viewed from the user coordinate system can be calculated by the following equation. In the following formula, ◯ represents a numerical value representing a rotation matrix.

  Furthermore, the user coordinate value data Ur1 of the teaching point R1 can be calculated from the above equation (14) as follows.

Ur1 = [Ur1x, Ur1y, Ur1z, Ur1α, Ur1β, Ur1γ]
= [Sx, Sy, Sz, Ur1α, Ur1β, Ur1γ] (15)

  In the above equation (15), Ur1x, Ur1y, and Ur1z represent the position coordinate values of the X axis, the Y axis, and the Z axis on the user coordinate system, respectively, and Sx, Sy, and The same value as Sz. Ur1α, Ur1β, and Ur1γ are posture coordinate values (rotation angles) of α, β, and γ on the user coordinate system that are obtained from the rotation matrix portion in equation (14).

  The user coordinate value data Ur1 of the teaching point R1 obtained in this way is stored in the robot control device 5. Since there are almost tens to hundreds of teaching points, not just R1, the operator stores all the teaching points by repeating steps S3 and S4 in FIG.

  In step S5 of FIG. 1, when all the work route teachings are completed, the process proceeds to step S6. In step S6, it is determined whether or not the installation position of the workpiece W has been changed. If the installation position of the workpiece W has not been changed, the process proceeds to step S7, and the reproduction operation is started. Here, assuming that the installation position of the workpiece W has been changed, the description returns to step S1 and the description is continued assuming that the user coordinate system is redefined. The description after step S7 will be described later.

  When the installation position of the workpiece W is changed, the process returns to step S1 to redefine the user coordinate system. FIG. 3 is a diagram for explaining that a new user coordinate system is redefined by re-designating feature points on the workpiece when the workpiece is placed at a position different from the reference position.

  In FIG. 3, the workpiece W is installed at a position Wn different from the reference position Wk. The teaching point R1n is a position corresponding to the teaching point R1 taught when the workpiece W is set at the reference position Wk. The manipulator 1 and the base coordinate system Cb are the same as those in FIG. The user coordinate system Cun is a coordinate system redefined by the operator, and an origin Oun and an axial direction (Xun, Yun, Zun) are defined as illustrated.

  The operator re-specifies the feature points P1n, P2n, and P3n on the workpiece W installed at a position Wn different from the reference position. The robot controller 5 recalculates the homogeneous transformation matrix Abun of the user coordinate system Cun representing the position of the user coordinate system Cun viewed from the base coordinate system Cb, based on the positions of the redesignated feature points P1n, P2n, and P3n. And remember again. The calculation method is the same as the method for calculating equation (11) described above. Therefore, the homogeneous transformation matrix Abun of the user coordinate system Cun after re-designation with the base coordinate system Cb as a reference can be expressed by the following equation.

  The homogeneous transformation matrix Abun obtained in this way is stored in the robot controller 5 as the position of the user coordinate system Cun viewed from the base coordinate system Cb.

  Next, in step S2, teaching for the workpiece W has already been completed. That is, there is no need to re-teach the work path for the workpiece W installed at a different position Wn, and the process proceeds to step S7.

  In step S7, a reproduction operation start operation is performed. Specifically, the teaching data is reproduced by inputting a start button provided in the robot controller 5, a start signal from a host controller that controls the production line, and the like.

  In step S8, the teaching data is read from the robot controller 5. In the teaching data, the TCP reproduction position of the manipulator 1 is stored as user coordinate value data. For example, the user coordinate value data Ur1 represented by the equation (15) is stored as the reproduction position of the teaching point R1n. Based on this user coordinate value data Ur1, a homogeneous transformation matrix Aurn representing the position and orientation of the teaching point R1 viewed from the user coordinate system is calculated. That is, the homogeneous transformation matrix Aurn is an unknown value at the time of the reproduction operation after the work W is placed at a different position Wn, and thus a process of calculating based on the user coordinate value data Ur1 is performed. The homogenous transformation matrix Aurn can be easily derived by performing an operation opposite to the operation for obtaining the user coordinate value data Ur1 of the equation (15) based on the homogeneous transformation matrix Aur of the equation (14). The homogeneous transformation matrix Aurn of the teaching point R1n calculated here is the same as the expression (14). This means that if the position and orientation coordinate values of TCP are stored as user coordinate value data based on the user coordinate system, the same homogeneous transformation matrix Aur can be used even if the installation position of the workpiece W changes. That's what it means.

  Next, in step S9, base coordinate value data Br1n of the teaching point R1n is calculated. For this purpose, a homogeneous transformation matrix Abrn representing the TCP position of the teaching point R1n viewed from the base coordinate system is calculated. The homogeneous transformation matrix Abun representing the position of the user coordinate system as viewed from the base coordinate system has already been calculated by the equation (16). Further, the homogeneous transformation matrix Aurn representing the position of the TCP viewed from the user coordinate system at the teaching point R1n is the same as the equation (14), and thus has already been calculated. Therefore, the homogeneous transformation matrix Abrn representing the TCP position of the teaching point R1n viewed from the base coordinate system can be calculated by the following formula based on the formulas (14) and (16). In the following formula, ◯ represents a numerical value representing a rotation matrix.

  Further, base coordinate value data Br1n of the teaching point R1n is calculated from the above equation (17).

  Br1n = [Pbx, Pby, Pbz, Pbα, Pbβ, Pbγ] (18)

  Pbx, Pby, and Pbz respectively represent the position coordinate values of the X axis, the Y axis, and the Z axis on the base coordinate system. Pbα, Pbβ, and Pbγ are posture coordinate values (rotation angles) of α, β, and γ on the base coordinate system that are obtained by the rotation matrix portion in Expression (17).

  In step S10, based on the base coordinate value data Br1n obtained by the equation (18), an inverse transformation operation for calculating each joint angle of the manipulator is performed. In step S11, each joint of the manipulator is rotationally driven to move to the calculated position of each joint angle, that is, to the teaching point R1n. Finally, if the reproduction operation for all teaching points is not completed in step S12, the process returns to step S8 to perform the reproduction operation for the next teaching point. If all the teaching points have been reproduced, the flow ends.

As described above, even if the relative positional relationship between the manipulator and the work is deviated, the existing teaching data can be reused as it is without correcting the teaching. Especially when processing large workpieces where the number of teaching points is large and it is difficult to accurately place the workpiece, even if there is a large amount of teaching data with several hundred teaching points, As long as the feature point is redesignated and the user coordinate system is redefined, the teaching point can be reused without modification.

[Embodiment 2]

  Next, Embodiment 2 of the present invention will be described. In the second embodiment, by re-designating the position of the feature point by measuring with a position measuring device such as a touch sensor, the robot does not redefine the user coordinate system, but automatically by the robot. It is.

  FIG. 4 is a flowchart for explaining the flow of automatically correcting teaching data according to the second embodiment of the present invention. In the figure, steps S1, S3 to S5, and S8 to S12 to which broken lines are given are substantially the same as the steps having the same reference numerals in FIG. Hereinafter, only steps S7a and S7b different from those of the first embodiment indicated by the solid line will be described.

  When a reproduction operation start operation is performed in step S7, the amount of deviation (ΔXb, ΔYb, ΔZb) from the reference position in the base coordinate system is detected for each of the three feature points by the touch detection operation by the touch sensor in step S7a. To detect.

  In step S7b, the shift amount is added to the base coordinate value data of the three feature points. Based on the base coordinate value data after the addition, the calculations of the equations (1) to (11) are performed to calculate the homogeneous transformation matrix Abu of the user coordinate system with the base coordinate system as a reference, and the robot control device Just remember.

  In the second embodiment, an example in which correction is automatically performed using a touch sensor as a position measuring device has been described, but the present invention is not limited to this. For example, the present invention includes a case where measurement is performed using an image processing apparatus such as a laser sensor and correction is automatically performed.

In this way, re-designation of a plurality of feature points on a workpiece is performed by automatic measurement by a position measuring device, not by an operator, so it is necessary for the operator himself to re-designate feature points. Absent. That is, in addition to the effects of the first embodiment, it is possible to further reduce the work man-hours of the worker.

It is a flowchart figure for demonstrating the flow in which correction | amendment of teaching data is performed by Embodiment 1 of this invention. It is a figure for demonstrating that a user coordinate system is defined by designating the feature point on a workpiece | work when the workpiece | work is installed in the reference position. It is a figure for demonstrating that the new user coordinate system is redefined by re-specifying the feature point on a workpiece | work when a workpiece | work is installed in the position different from a reference position. It is a flowchart for demonstrating the flow in which correction | amendment of teaching data is automatically performed by Embodiment 2 of this invention. It is a general block diagram of the industrial robot which employ | adopted the teaching playback system. It is a figure for demonstrating the relationship between the moving direction of the manipulator at the time of teaching, and a coordinate system. It is a flowchart for demonstrating how teaching data is conventionally memorize | stored and reproduced | regenerated.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Manipulator 2 Arm 3 Teach pendant 4 Button group 5 Robot controller Abr (representing the position and orientation of the teaching point R1 viewed from the base coordinate system) Homogeneous transformation matrix Abrn (the position and orientation of the teaching point R1n viewed from the base coordinate system) Homogeneous transformation matrix Abu (represents the position and orientation of the user coordinate system viewed from the base coordinate system) Homogeneous transformation matrix Abun (represents the position and orientation of the user coordinate system viewed from the base coordinate system) Homogeneous transformation matrix Aur ( Homogenous transformation matrix Aurn (representing the position and orientation of the teaching point R1 viewed from the user coordinate system) Homogeneous transformation matrix Bp1 representing the position and orientation of the teaching point R1n viewed from the user coordinate system Base coordinate value data (feature point P1)
Bp2 base coordinate value data (feature point P2)
Bp3 base coordinate value data (feature point P3)
Br1 Base coordinate value data (teaching point R1)
Br1n base coordinate value data (teaching point R1n)
Cr Base coordinate system Cu User coordinate system Cun User coordinate system (after redefinition)
K work path Ob origin Ou (in the base coordinate system) origin Ou (in the user coordinate system) origin Un (in the user coordinate system after redefinition) origin P1 feature point P2 feature point P3 feature point P1n feature point P2n feature point P3n feature point Qv Vector R1 Teaching point R1n Teaching point T Tool Ur1 User coordinate value data W Work Wk Reference position Wn Position different from the reference position Xb Axial direction (base coordinate system X)
Xu axis direction (user coordinate system X)
Xv X-axis vector Yb Axis direction (base coordinate system Y)
Yu axis direction (user coordinate system Y)
Yv Y-axis vector Zb Axis direction (base coordinate system Z)
Zu axis direction (user coordinate system Z)
Zv Z-axis vector

Claims (2)

When the workpiece is set at the reference position, a user coordinate system is defined in which the origin and the axial direction of the Cartesian coordinate system are the desired values by specifying multiple feature points on the workpiece, and this user coordinate system is used during teaching The robot is moved according to the axis direction of the robot, the work path is taught, and the position and orientation data of the teaching point on this work path is stored as base coordinate value data based on the base coordinate system that is an orthogonal coordinate system unique to the robot individual. In a control method for an industrial robot that performs motion control by calculating each joint angle data of the robot based on this base coordinate value data during reproduction,
The teaching point position and orientation data is stored as user coordinate value data based on the user coordinate system during teaching,
The base coordinate value data is calculated based on the user coordinate value data and the user coordinate system during reproduction, and each joint angle data of the robot is calculated based on the base coordinate value data.
When the workpiece is installed at a position different from the reference position, the user coordinate system is redefined by respecifying the positions of the plurality of feature points,
At the time of reproduction, the base coordinate value data is recalculated based on the user coordinate value data and the redefined user coordinate system, and each joint angle data of the robot is calculated based on the recalculated base coordinate value data. A method for controlling an industrial robot. The industrial robot control method according to claim 1, wherein the re-designation of the position of the feature point is performed by measuring with a position measuring device.

Images