JP2002239956A - Teaching data preparation method for articulated robot - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the working efficiency without causing an irrational matter such as calculated action data is unused and without restricting an operator intermittently in teaching data of an articulated robot. SOLUTION: All attitude data corresponding to multiple welding points of a work are found by a programming (step S3-step S7), and after confirming the determined attitude data (step S8), the action data of a midway route connecting between the multiple welding points are found (step S9-step S11).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a method for creating teaching data for an articulated robot, and more particularly to a method for creating teaching data for an articulated robot that efficiently creates teaching data.

[0002]

2. Description of the Related Art Conventionally, when teaching a work posture by directly operating an articulated robot installed on a production line, a robot operator who is familiar with the operation of the articulated robot must perform work at the site of the production line. Must be done, which makes the work inefficient. Further, the operation must be performed in a state where the production line is stopped, so that the operation rate of the production line is reduced.

[0003] In order to improve the efficiency of the teaching or to improve the operation rate of the production line, offline teaching (offline teaching) has recently been performed. That is, a model of an articulated robot, a work to be a work object, and a peripheral structure is constructed on a computer, teaching data is created using the model, and the teaching data is supplied to an articulated robot at a work site. By doing so, there is no need to stop the production line while creating teaching data.

[0004] The teaching data of the articulated robot can be divided into attitude data of the articulated robot corresponding to the welding points (work points) and operation data on an intermediate route connecting the plurality of welding points. it can. The motion data defines a halfway route between the posture data, and is a complicated process because it needs to be set so as not to interfere with the work. In other words, it is not enough to process welding point data alone, it is necessary to consider the overall shape of the work, extract a part of the work that may cause interference, and determine the direction in which the extracted part is evacuated and operated. There is. In addition, it is necessary to determine how many operation data to interpolate between welding points.

Accordingly, the processing for obtaining these operation data is not automated, and the computer operator operates the model of the articulated robot on the screen of the monitor to obtain the operation data while confirming that there is no interference. .

Next, a conventional method of creating teaching data for an articulated robot will be described with reference to FIG.

First, in step S501, undetermined attitude data among welding points is designated, and data of the welding points is input to the apparatus.

In step S502, posture data of the articulated robot is calculated by a program based on the data of the welding points. At this time, when a plurality of postures can be taken, a process of selecting an appropriate one from the plurality of postures is also performed.

Next, in step S503 of branch determination, it is determined whether or not the solution has been obtained in step S502. That is, it is determined whether or not the articulated robot can take a posture with respect to the welding point data. If the solution is not found or if the solution shows a value that indicates an angle outside the range of rotation of the shaft, it is determined that welding is impossible and replanning is performed. Further, the determined posture is confirmed on the monitor screen, and it is determined whether or not the posture is appropriate based on the experience of the operator. If the posture is not appropriate, the condition is changed and recalculation or replanning is performed. If the solution has been normally obtained and the posture is appropriate, the process proceeds to step S504.

Next, in step S504, a non-interference path (operation data) between the welding points is obtained. In this process, as described above, the operator finds and determines an appropriate posture while checking the positions of the model of the articulated robot and the workpiece on the monitor screen. The operation data represents the path between the welding points by some representative postures. In practice, the articulated robot operates by performing linear interpolation or curve interpolation on the representative postures. The number of the representative postures is determined by the distance between the welding points and the shape of the work, and is usually determined by an operator's rule of thumb.

Next, in step S505 of branch determination, the model of the articulated robot is operated on the monitor screen based on the operation data obtained in step S504 to determine whether the route is appropriate (no interference, operation time, etc.). ) Is confirmed, and if the predetermined criterion is satisfied, the next step S50
Move to step 6, if the criterion is not satisfied, step S50
Returning to step 4, the operation data is obtained again.

In the next branch determination step S506, it is determined whether or not processing has been completed for all postures (all paths) of the path table representing teaching data. If there is an unprocessed posture, the unprocessed posture is designated as the next processing target, and the process returns to step S501. If all postures have been completed, the teaching data has been completed, the path table is recorded, and the processing ends.

[0013]

In the above-mentioned prior art, the operator does not need to perform any special operation while the computer is executing the step S502 for calculating the posture data of the articulated robot. It will wait for the output of the result. In addition, in the next step S503, there is a process for judging the quality of the processing result with the intervention of an operator, so that the operator cannot leave the computer terminal.

Further, also in the step S504,
A technique for automating the processing by a computer has been developed. However, even when this technique is used, the processing time is relatively long. Then, in the next step S506, the operator determines whether the result is good or not, so that the operator cannot leave the computer terminal. If automation technology is not used, the operator operates the robot model on the screen of the monitor and confirms that there is no interference, so that the operator cannot leave the computer terminal. The processing time is long.

As described above, in the teaching data creating method of the articulated robot according to the prior art, the computer operator is restrained by the operation of the computer for a relatively long time, and the time for waiting for the output of the processing result is also increased. Inefficient because there is.

In addition, if a welding point where the posture of the articulated robot cannot be established is detected during the welding process, it is necessary to review the entire welding process. Therefore, if the posture data and the operation data are obtained in the order of the actual operation of the articulated robot, there arises an irrationality that the operation data obtained before the welding point that cannot be established is not used.

Further, since a method is employed in which an operator intervenes in the middle of processing, not only cannot automatic processing be performed during the absence of an operator, for example, at night or on holidays, but also at night or on holidays. In addition, the computer is at rest and the operation efficiency is low from the viewpoint of facilities.

The present invention has been made in view of such a problem, and by calculating all the posture data before calculating the motion data, it is unreasonable that the calculated motion data is not used. It is an object of the present invention to provide a method for creating teaching data of an articulated robot that can eliminate points.

Another object of the present invention is to separate the processing of a computer and an operator in the creation of teaching data of an articulated robot, thereby preventing the operator from being intermittently restricted by the creation of teaching data. And to improve work efficiency.

[0020]

According to the present invention, there is provided a teaching data generating method for an articulated robot, comprising: posture data corresponding to a plurality of working points; and motion data on an intermediate route connecting the plurality of working points. In the method for creating teaching data of an articulated robot to be created, the method further includes a first step of obtaining the posture data corresponding to all of the working points before creating the motion data. Thereby, the processing of the computer and the processing of the operator are separated, so that the operator is not intermittently restrained for the operation of the computer, and the work efficiency is improved.

Further, after the first step, a second step of obtaining the operation data may be provided. Thereby, the unreasonable point that the calculated operation data is not used can be solved.

Still further, in the first step, when the posture of the articulated robot cannot be established with respect to a certain work point, a step of recording that the work point is not workable is provided. You may do so.

[0023]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a method for creating teaching data for an articulated robot according to the present invention will be described below with reference to FIGS.

The teaching data of the articulated robot is represented by a path table 110 shown in FIG. 1, and includes a "gun unit direction" column 110a, a "TCP position" column 110b, and a "axis angle" column. 110c. Each data is a welding point (working point)
Posture data P1 to P4 of the articulated robot corresponding to
The data can be divided into operation data Q1 to Q7 for an intermediate route connecting these plurality of welding points.

FIG. 2 shows these data by the position of the gun unit and shows the positional relationship with the work 80. As can be seen from FIG. 2, the posture data P1 to P4
Is determined from the data of each welding point on the work 80.

On the other hand, regarding the operation data Q1 to Q7,
It defines an intermediate route between the posture data P1 to P4 and needs to be set so as not to interfere with the work 80.

FIG. 3 shows at least an offline teaching device 10 for creating the teaching data described above,
1 shows a robot device 12 that performs a desired operation on a work target based on teaching data created by the offline teaching device 10.

The offline teaching device 10 includes a control unit 14, a monitor 16, a keyboard 18 and a mouse 20 for giving input / output instructions to the control unit 14, and teaches the operation of the articulated robot 50. Things.

The robot apparatus 12 includes an articulated robot 50
And a robot controller 22 for controlling the operation of the articulated robot 50 based on the teaching data.

As shown in FIG. 4, the articulated robot 50
A first base 54 serving as a mounting base and a second base 56
, A first link 58, a second link 60, a third link 62, and a cylindrical gun attaching / detaching portion 64, and the gun unit 52 is connected to the gun attaching / detaching portion 64.

The first base 54 and the second base 56
Are connected by an axis J1 that rotates about a vertical axis L0. The second base 56 and the proximal end of the first link 58 are connected by an axis J2 that rotates in a vertical plane, and the distal end of the first link 58 and the proximal end of the second link 60 are connected by an axis J3 that rotates in a vertical plane. The distal end of the second link 60 and the proximal end of the third link 62 are connected by a shaft J4 that rotates in a vertical plane. The tip of the third link 62 is connected to the gun attaching / detaching portion 64 by a shaft J5.
5 rotates around the central axis L of the third link 62.

The gun unit 52 connected to the gun attaching / detaching portion 64 is a so-called C-type welding gun, and has a pair of electrodes 70 and 72 which open and close along the central axis L. In the closed state, the electrodes 70 and 72 come into contact with the work 80 at a welding operation point (hereinafter, referred to as TCP (Tool Center Point)) on the central axis L.

The direction which coincides with the axis of the electrode 72 on the main body side from the TCP is defined as an electrode vector Zr, and the direction orthogonal to the electrode vector Zr and opposite to the arm 74 is defined as an arm vector Xr.
r. Further, an arm vector Xr and an electrode vector Z
The direction orthogonal to r is a horizontal vector Yr (see FIG. 5).
And

The driving mechanism for the axes J1, J2, J3, J4 and J5 and the opening / closing mechanism for the electrodes 70 and 72 are driven by actuators (not shown).
1 rotation angle θ1, shaft 2 rotation angle θ2, shaft 3 rotation angle θ
3, the rotation angle θ4 of the shaft 4, the rotation angle θ5 of the shaft J5, and the dimensions of each part of the articulated robot 50.

In this embodiment, the articulated robot 50 has five
Although the shaft type has been described, it is needless to say that six or more axes can be applied, and the axis here includes not only a rotating operation but also a telescopic operation and a moving operation.

The gun unit 52 is not limited to the C-type welding gun, but may be, for example, an X-type welding gun (a welding gun having a pair of opening and closing gun arms supported by a common support shaft) shown in FIG. Is also good.

As shown in FIG. 7, the control unit 14 constituting the offline teaching device 10 is a CPU 26 for controlling the whole of the offline teaching device 10, a ROM 28 as a nonvolatile storage unit, and a volatile storage unit. R
AM 29 and hard disk drive (HDD) 34
A drawing control circuit 30 for controlling drawing on the screen of the monitor 16, an interface circuit 32 to which the keyboard 18 and the mouse 20 are connected, and an external recording medium 36a
, A data creation circuit 38 for creating teaching data, and a simulation circuit 40 for performing a simulation on the screen of the monitor 16 based on the teaching data. The simulation circuit 40 is based on three-dimensional CAD, and has a function of creating the model and investigating interference (such as contact) between the models.

Next, an example of a method of creating teaching data of the articulated robot according to the present embodiment will be described in detail with reference to FIGS.

In this embodiment, the articulated robot 5
The work 0, the work 80, and its surrounding structures are treated as virtual models in the offline teaching device 10, but are described using the same reference numerals as those of the actual device in the following description.

FIGS. 8 to 1 show a method for creating teaching data of the articulated and articulated robot according to the first embodiment.
1 will be described.

In the method for creating teaching data for an articulated robot according to the first embodiment, the method for creating teaching data for an articulated robot is as follows.
Obtain the attitude data for all welding points,
Thereafter, operation data between welding points is obtained.

First, in step S1 of FIG. 8, planning and equipment design are performed. Specifically, the time required for a series of welding processes is examined based on the number of manufactured products, and the required number of articulated robots 50 is determined accordingly. In addition, the type and the like of the gun unit 52 and the articulated robot 50 are selected, and necessary equipment such as a mount on which the articulated robot 50 is installed is designed.

Next, in step S2, the work 80
And the position data of the welding points and the welding order of the welding points are determined.
To enter. Weld point data is stored in gun unit 52
, Ie, TCP, and the direction of the gun unit 52. These are recorded in the path table 120 in advance as values in the “gun unit direction” column 120a and the “TCP position” column 120b in the path table 120 shown in FIG.

In the subsequent steps S3 to S7, the attitude of the articulated robot 50 at all the welding points is determined by the program processing.

That is, in step S3, posture data corresponding to an undetermined posture among the postures of the articulated robot 50 is obtained by calculation. This calculation method uses the position coordinates (X, Y, Z) of the welding point in space and the arm vector X representing the attitude of the gun unit 52.
It can be obtained from a total of six values defined by r, the horizontal vector Yr, and the electrode vector Zr, the dimensions of each part of the articulated robot 50, and the like by a well-known matrix operation method (hereinafter, referred to as an inverse operation).

Next, in step S4 of the branching process,
It is determined whether or not a solution has been obtained in step S3. That is, it is determined whether or not the articulated robot can take a posture with respect to the welding point data. If the solution was not found, and even if the solution was found, the angle values are on the axes J1 to J
5 or the articulated robot 50 interferes with other structures in the determined posture (for example, the first link 58, the second link 60, etc. If so, the process proceeds to step S5. If not, that is, if the solution is normally obtained, the process proceeds to step S6.

This branching process is automatically performed by a computer program, and it is not determined at this time whether or not the operator is in an appropriate posture based on the experience of the operator as in the above-described prior art.

If the solution is not normally obtained, the fact that the posture cannot be established by the articulated robot 50 is recorded in the path table 120 in step S5, and then the process proceeds to step S7. Path table 12 shown in FIG.
0, the welding point P2 in order 2 is the posture of the articulated robot 50 under the conditions [Xr3, Yr3, Zr3, X3, Y3, Z3] of the “gun unit orientation” column 120a and the “TCP position” column 120b. Are not established, and a symbol “*” indicating that the posture cannot be established is recorded in the “Each axis angle” column 120c.

When the solution is obtained, the obtained attitude data is recorded in the path table 120 in step S6, and thereafter, the process proceeds to step S7.

Next, in step S7 of the branching process,
It is determined whether the processing has been completed for all the posture data. If there is an unprocessed posture, the unprocessed posture is designated as the next processing target, and the process returns to step S3. If the processing of all the posture data has been completed, the process moves to the next step S8.

In step S8, the operator displays all the obtained posture data on the screen of the monitor 16.
Alternatively, the operator checks the path table 120, and if there is an inconvenient part in the experience, corrects the part. If the part does not have the required posture, the condition is changed to calculate an appropriate welding posture. .

In the subsequent steps S9 to S11, the motion data between the respective posture data obtained up to the step S7 is obtained. At this stage, since it is confirmed that the posture of the articulated robot 50 is established for each posture data, there is no unreasonable situation that the motion data obtained thereafter is not used.

In step S9, motion data is obtained between the posture data for which motion data has not been obtained.
In this process, the operator finds and determines an appropriate posture while checking the model of the articulated robot 50 and the work 80 on the screen of the monitor 16 as in step S504 in the above-described conventional technique. The number of the operation data is determined by the distance between the posture data and the shape of the work 80, and is usually determined by an operator's rule of thumb.

Next, in step S10, the operation data obtained in step S9 is recorded in the path table 120. At this time, the motion data is recorded so as to be inserted between posture data which are postures at both ends of the path. Figure 1
The motion data Q1 and Q2 recorded in the order 2 and 3 in the path table 130 of FIG.
Shows an example inserted between.

Next, in step S11 of branch determination, it is determined whether or not processing has been completed for all operation data. If there is unprocessed operation data, the unprocessed path is designated as the next processing target, and the process returns to step S9. If the processing has been completed for all the operation data, the process proceeds to step S12 in FIG.

In step S12, additional commands necessary for the operation of the articulated robot 50 are set by program processing. The additional instruction is, for example, an instruction relating to how to move between the posture data and the movement data recorded in the path table 110 (see FIG. 1) obtained at this stage. A linear interpolation motion command, a circular interpolation motion command, and the like are given. In addition, designation of the operation speed and welding timing during that period, and a command for temporarily stopping (waiting) and delaying (delaying) the operation are added as necessary.

At the stage when these instructions are added, the teaching data for operating the articulated robot 50 is formed as a temporary form.

Next, in step S13, the model of the articulated robot 50 is simulated on the screen of the monitor 16 based on the temporary teaching data generated in step S12. The operator confirms that there is no interference with other structures, etc.
Check the cycle time of a series of welding processes.

Next, in step S14, the teaching data is appropriately corrected based on the simulation result in step S13 so as to eliminate interference and complete the teaching data so that the cycle time is within a specified time.

At this stage, the teaching data of the articulated robot 50 as a single unit is completed, and in subsequent steps S15 to S16, all the teaching data of the plurality of articulated robots 50 used in the welding process are obtained. Integrate and check the operation.

That is, in step S15, the teaching data completed up to step 14 is set in a control program called a sequencer. This sequencer has a function of designating the operation timing of each articulated robot 50 in conjunction with the movement of the production line.

Next, in step S16, simulation of a plurality of articulated robots is performed based on the function of the sequencer, and it is confirmed that there is no mutual interference between the articulated robots 50 and the like. Then, if inconvenience is found, it is corrected as appropriate.

Next, in step S17, the created teaching data is supplied to the robot controller 22 to terminate the processing.

As described above, according to the teaching data creating method of the articulated robot according to the first embodiment,
Prior to the creation of motion data, posture data corresponding to all welding points is determined, so it has been confirmed that all posture data are satisfied, and it is unreasonable that the required motion data will not be used thereafter Nothing happens.

Further, since the attitude data is calculated continuously for all the welding points, the operator is not restricted by the processing during that time.

In the first embodiment,
In step S5, if the posture was not found, a record indicating this is recorded and the processing itself is described as continuing, but the processing may be interrupted at this point. That is, if there is enough time before the actual production, the process may be interrupted at this stage, and re-planning such as changing the position of the welding point itself may be performed.

Next, a method for creating teaching data of the articulated and articulated robot according to the second embodiment will be described with reference to FIG.
2 and FIG.

The teaching data creation method for the articulated robot according to the second embodiment performs substantially the same processing as that of the first embodiment. The difference is that the program is adopted and automated. Further, the process of checking and correcting the posture data performed in step S8 is omitted, and the motion data is obtained, and the posture data and the motion data are collectively verified (step S2 in FIG. 13).
12) is performed. As a result, the operation of calculating the posture data and the operation of calculating the motion data can be automatically processed without the intervention of the operator.

More specifically, as shown in FIGS. 12 and 13, posture data is calculated in steps S203 to S207, and thereafter, motion data is calculated in steps S208 to S210. Up to the creation of the command is automated by program processing.

Thereafter, the operator verifies (confirms and corrects) the posture data and the operation data in step S2.
12 is performed.

Here, as for the step S208 for obtaining the operation data by the program processing, since a method and a program for automating this processing have recently been developed, the technology may be adopted.

Hereinafter, an example of a technique for automating step S208 will be described in detail with reference to FIGS.

First, in step S301 of FIG. 14, the gun unit 52 of the articulated robot 50 sets the welding point P0 of the work 80 to a position to be welded. The welding point P0 corresponds to any one of P1 to P4 in the path table 120.

Since the welding point P0 is the initial position, the welding point P0 is recorded in the temporary path table 140 for the operation data and is initialized (see order 1 in FIG. 17).

As shown in FIG. 17, the path table 140
Is composed of a “gun unit orientation” column 140a, a “TCP position” column 140b, and a “each axis angle” column 140c, and the “each axis angle” column 140c includes rotation angles θ1 to θ.
5 is comprised.

Next, in step S302 of FIG. 14, the gun unit 52 located at the welding point P0
Is set as the investigation start position Ps.

Next, in step S303, the arm 74 and the electrode
A center point C (see FIG. 18A) is defined at a position where 72 can be seen. Then, a radial straight line 1090 is set at a predetermined angular width from the center point C, and the arm 74 and the electrodes 70 and 72 are set.
Intersection 1092 with is obtained.

Although the intersection 1092 is determined on the plane to simplify the description, the intersection in the three-dimensional shape is actually determined by using the data in the depth direction. 1096 and the solid (block) 1094 described below are also treated as three-dimensional shapes instead of surfaces.

Next, in step S304, FIG.
As shown in B, a plurality of intersections 1092 are connected by a line segment to set an annular line 1092b forming a closed section 1092a. Then, a grid-like line having a predetermined interval is set on the closed section 1092a, and an intersection 1092c existing in the closed section 1092a is extracted from the intersection of the grid-like line.

Next, in step S305, FIG.
As shown in A, a solid 1094, which is a square surface with the intersection 1092c extracted as the center, is filled so as to have no gap, and is set as a gun internal space.

Next, in step S306, FIG.
As shown in B, the work 80 is arranged so that the relative position with respect to the gun unit 52 matches. And work 80
And the part where the solid 1094 overlaps the work model 10
96 (a target work portion, see FIG. 19C). At this time, a portion 80a of the work 80 that does not overlap with the solid 1094 becomes irrelevant when investigating interference, and is therefore excluded. Each solid 1094 constituting the work model 1096 is distinguished as a work solid 1098. Even when the gun unit 52 moves, the work model 1096 and each work solid
98 has an initial position fixed.

As described above, since the work 80 is modeled and handled by a plurality of blocks, processing becomes easy, and unnecessary portions (for example, non-overlapping portions 8
Since 0a) is automatically excluded, no useless processing is performed.

Next, in step S307, a principal component line (reference line) M1 of the work model 1096 is calculated by a principal component analysis technique.

The method of calculating the principal component line M1 will be described in detail. First, as shown in FIG. 20A, the center point coordinates 1098a (Xs, Ys, Zs) of each work solid 1098 are obtained.
Is specified.

Next, as shown in FIG. 20B, the sum of the squares of the distance s between each center point coordinate 1098a and the principal component line M1 is minimized. That is, the principal component line M1 that satisfies Σ | s | ² = min is defined. More specifically, a variance, an eigenvalue of a covariance matrix, and an eigenvector are calculated from each center point coordinate 1098a, and Xs, Ys, Zs are used to calculate X,
A center-of-gravity position G1, which is an average value of Y and Z coordinates, is obtained, and the eigenvector passing through the center-of-gravity position G1 becomes a principal component line M1.

Steps S308 to S31 thereafter
In FIG. 2, as shown in FIG.
It is checked whether or not there is any interference when the operation is performed linearly from the position to the extraction position (investigation end position) Qe.

That is, first, in step S308, the extraction position Qe is determined. This withdrawal position Qe
As shown in FIG. 21, an arm vector X based on the TCP of the gun unit 52 at a point on the main component line M1 is
r is moved while being coincident with the main component line M1, and a position where the gun unit 52 and the electrodes 70 and 72 do not interfere with each other is set as the extraction position Qe.

Next, in step S309, the posture of the articulated robot 50, that is, the rotation angles θ1 to θ5, is determined based on the position and posture of the gun unit 52 specified by the extraction position Qe. This calculation method is based on the position coordinates (X, Y, Z) of the extraction position Qe in the space and the arm vector Xr, the lateral vector Yr, and the electrode vector Zr representing the attitude of the gun unit 52. It may be obtained by calculation.

Next, in step S310 of the branch determination, it is determined whether the solution in the inverse operation of step S309 has been normally obtained, that is, whether the TCP can reach the extraction position Qe. The solution is not obtained, or even if the solution is obtained, the angle value is out of the rotation operation range of the axes J1 to J5, and the articulated robot 50 interferes with another structure in the obtained posture (for example, other If there is any interference with the first link 58, the second link 60, or the like of the work or the column in the factory, the process proceeds to step S311. Otherwise, that is, if the solution is normally obtained, the process proceeds to step S312.

In this interference investigation, when the X-type welding gun 52a is employed as the gun unit, the inspection is particularly made on both the open state and the closed state of the gun unit.

If the solution has not been obtained normally, in step S311, the angle α ° is set around the horizontal vector Yr.
A rotating operation for rotating is performed. This rotation operation is performed when the gun unit 52 is in the withdrawn position Q as shown by the two-dot line in FIG.
This means that the work model 1096 is rotated around e in a range that does not interfere with the work model 1096. In this state, the arm vector Xr, the horizontal vector Yr, and the electrode vector Zr
And returns to step S309. Also, the angle α °
May be investigated as both positive and negative angles.

Note that steps S309 to S31
If the loop formed by 1 is executed several times in succession, the extraction position Qe is reset to a suitable position on the principal component line M1 where the posture of the articulated robot 50 is established at a more distant position, Then, control goes to the next step S312.

Further, the process of rotating α ° is as follows.
The rotation may be performed not only about the horizontal vector Yr but also about an axis such as the arm vector Xr or the electrode vector Zr. The same applies to the following rotation processing.

Next, moving to FIG. 15, in step S312, the gun unit 52 is moved linearly from the investigation start position Ps to the extraction position Qe as shown by the path V1 in FIG. , 72 and the work model 1096 do not interfere with each other. Investigation of this interference is automatically conducted by the function of the simulation circuit 40. When the simulation circuit 40 is used, a three-dimensional investigation that cannot be understood on the screen of the monitor 16 which is a two-dimensional expression can be reliably performed.

If it is determined in step S313 of the branch determination that there is interference as a result of the examination in step S312, the flow proceeds to step S314. If it is determined that there is no interference, the direct pull-out operation is performed in this one operation. Then, the process moves to step S331, which is the end process.

As described above, when the shape of the work 80 is simple, the extraction path can be determined by one operation, so that the processing time can be reduced.

In the example shown in FIG. 21, it is clear that the electrode 70 interferes with the projection 1096a of the work model 1096 during the movement of the route V1, and in this case, the process proceeds to step S314.

Steps S314 to S31
8, the work model 10 starts from the survey start position Ps.
Investigation is made as to whether or not there is interference when the device is operated linearly up to the position of the center of gravity G1 of 96.

That is, in step S314,
As shown in FIG. 22, a path V2 connecting the investigation start position Ps and the position of the center of gravity G1 is defined, and the posture of the gun unit 52 in which the arm vector Xr matches the path V2 based on the position of the center of gravity G1. Assume.

Then, in step S315, the attitude of the articulated robot 50 is determined by the above-described inverse operation using the assumed attitude.

Next, in step S316 of the branch determination, it is checked whether the solution in the inverse operation has been normally obtained as in step S310. At this time, in addition to the inverse operation processing, it may be checked whether or not the gun unit 52 interferes with the work model 1096.

If the solution is not normally obtained, a rotation operation for rotating the horizontal vector Yr by α ° is performed as in step S311 (step S317).
Then, the arm vector Xr and the horizontal vector Y in this state
After determining r and the electrode vector Zr, the process returns to step S315.

If a solution is obtained, the process proceeds to step S318, as in step S312.
2 is operated linearly along the path V2 from the investigation start position Ps to the center of gravity position G1 to investigate the interference.

Steps S315 to S31
If the loop formed in step 7 is executed several times in succession, it is determined that the gun unit 52 cannot be arranged at the position of the center of gravity G1, and the process proceeds to step S324, which is a discontinuation mask process.

If it is determined in step S318 and step S330 to be described later that there is interference, the process proceeds to step S324 through step S319 of branch determination, and if it is determined that there is no interference, the process proceeds to this center of gravity. In the next step S32,
Move to 0.

In step S320, the posture of the articulated robot 50 at that time is additionally recorded in the path table 140.

Then, in step S321, as in step S312, the gun unit 52 is operated linearly from the position of the gun unit 52 at that time to the withdrawal position Qe to check for interference. In the example shown in FIG. 22, the investigation is performed along the principal component line M1.

If it is determined in step S322 of the branch determination that there is interference as a result of the investigation in step S321, the flow proceeds to step S323. It moves to step S331 which is processing.

If there is interference, in step S323, an update process is performed in which the current position of the gun unit 52 is set as a new investigation start position and replaced with the previous investigation start position Ps. That is, in the example of FIG. 22, since the gun unit 52 could be pulled out to the position of the center of gravity G1, it is determined that there is no longer any need to consider the part deviating from the gun internal space. The survey start position Ps is also updated to recreate the model 1096.

The direction of the gun unit 52 is set so that the arm vector Xr matches the direction of the principal component line at that time.

Then, the work solid 1098 is extracted and updated in the same manner as in step S306, and a new principal component line M1 and a new center of gravity position G1 are obtained and updated in the same manner as in step S307. It returns to step S314. After returning to step S314, the new work solid 109 obtained in step S323 is obtained.
8. The processing is continued for the principal component line M1 and the center of gravity position G1.

In this way, the parts that do not enter the gun internal space are sequentially removed from the processing target, so that a path through which the gun unit 52 is pulled out even for a work 80 having a complicated shape can be obtained.

However, if the loop formed from step S314 to step S323 has been executed a predetermined number of times or more, it is very difficult to pull out the gun unit 52 from the workpiece 80. Since it is determined that there is, the process is terminated and re-planning is performed.

Next, in step S319, it is determined that there is interference due to the operation along the route Vn (n = 1, 2, 3,...).
Step S330 will be described. In this case, only the portion of the work model 1096 that is close to the opening of the gun unit 52 is extracted (masked), and the extracted portion is preferentially determined for the extraction path.

First, in step S324 of FIG. 16, as shown in FIG. 23, the center of gravity position G1 is used as a reference.
A portion 1096b of the work model 1096 on the opening side of the gun unit 52 and a portion 1096 on the side opposite to the opening.
c. In this classification process, a process is devised so that the gun unit 52 is pulled out only in a portion near the opening, and a portion 1096 on the opposite side to the opening.
“c” suspends the processing and extracts the opening side portion (new target work portion) 1096b. Then, in steps S325 to S330, which are subsequent processes, the new target work portion 1096b is handled as a substitute for the work model 1096.

Next, in step S325, a new target work part 1 is set in the same manner as in the processing in step S307.
The main component line M2 and the center of gravity position G2 are obtained for 096b.

In step S326, similarly to step S314, a path V3 connecting the investigation start position Ps and the center of gravity G2 is defined, and the arm vector Xr is set to a path V3 based on the center of gravity G2. It is assumed that the posture of the gun unit 52 is matched.

Next, in step S327, the attitude of the articulated robot 50 is determined by the above-described inverse operation in the assumed attitude, as in step S315.

Next, in step S328 of the branch determination, it is checked whether the solution in the inverse operation has been normally obtained as in step S316.

If the solution is not normally obtained, a rotation operation for rotating the horizontal vector Yr by α ° is performed as in step S317 (step S329).
Then, the process returns to step S327.

If a solution is found, in step S330, as in step S318, the gun unit 5
2 is operated linearly from the investigation start position Ps to the center of gravity position G2 along the path V3 to investigate the interference. And
Returning to step S319, the determination of this interference investigation is performed.

As described above, even when an appropriate route is not found even when a route search is performed on the entire work model 1096, mask processing is performed on the work model 1096, so that the opening of the gun unit 52 is formed. The extraction route can be obtained by giving priority only to the near portion.
Further, in the subsequent processing, by combining with the updating processing of the work model 1096 in the above-mentioned step S323, the work model 1096 sequentially becomes a simple shape, and it becomes easy to obtain a drawing path.

Steps S327 to S32
If the loop formed at step 9 is executed several times in succession, it is determined that the gun unit 52 cannot be arranged at the position of the center of gravity G2, and the process returns to step S324 to perform further mask processing. However, if the number of times of the mask processing is equal to or more than a predetermined number of times, it is determined that the mask processing is not effective for the shape of the workpiece 80, and the process returns to step S320, which is a drawing processing in which the mask processing is not performed. Is calculated.

In step S331, which is the end processing, the coordinates and vector data of the investigation end position Qe are stored in the path table 140 (FIG. 17) as the operation data.
) Is inserted between the welding points of the path table 120.

Thereafter, the flow returns to FIG. 12, and the loop end determination in step S210 may be performed.

As described above, in the method for creating teaching data of the articulated robot according to the second embodiment,
The processing for obtaining the posture data and the motion data can be executed continuously, and further, since there is no need for the operator to intervene between them, the computer can be continuously operated for a relatively long time. Therefore, for example, if this processing is performed at night, the operation efficiency of the computer is improved, and the operator looks at the processing result, and
Since the posture data and the motion data can be verified collectively, work efficiency is good.

In the above description, the welding robot has been described as an example, but it is needless to say that the present invention can be applied to other robots, for example, a painting robot.

Further, the teaching data creating method for the articulated robot according to the present invention is not limited to the above-described embodiment, but may take various configurations without departing from the gist of the present invention.

[0129]

As described above, according to the teaching data creating method for an articulated robot according to the present invention, all the posture data are obtained prior to the calculation of the movement data, so that the calculated movement data can be used. There is no unreasonable occurrence of not being performed, and by separating the processing of the computer and the operator, it is possible to prevent the operator from being intermittently restrained by the computer operation by creating the teaching data, thereby improving the work efficiency. Is achieved.

[Brief description of the drawings]

FIG. 1 is a diagram showing an example of a path table obtained in a teaching data creation method according to the present embodiment.

FIG. 2 is an explanatory diagram showing a relationship among a workpiece, a welding point, and a movement path.

FIG. 3 is a configuration diagram showing an offline teaching device and a robot device according to the present embodiment.

FIG. 4 is an explanatory diagram illustrating a configuration of an articulated robot.

FIG. 5 is an explanatory view showing a gun unit in FIG. 4;

FIG. 6 is an explanatory view showing an X-type welding gun.

FIG. 7 is a block diagram illustrating a circuit configuration of a control unit in the offline teaching device.

FIG. 8 is a process block diagram (part 1) illustrating a method for creating teaching data of the articulated robot according to the first embodiment;

FIG. 9 is a process block diagram (part 2) illustrating a method for creating teaching data of the articulated robot according to the first embodiment;

FIG. 10 is a diagram showing a path table in a process of obtaining posture data.

FIG. 11 is a diagram showing a path table in a state where motion data is inserted in the middle of posture data.

FIG. 12 is a process block diagram (part 1) illustrating a method for creating teaching data of an articulated robot according to a second embodiment;

FIG. 13 is a process block diagram (part 2) illustrating a method for creating teaching data of the articulated robot according to the second embodiment;

FIG. 14 is a process block diagram (part 1) illustrating a method for obtaining posture data in the method for creating teaching data of the articulated robot according to the second embodiment.

FIG. 15 is a process block diagram (part 2) illustrating a method of obtaining posture data in the method of generating teaching data of the articulated robot according to the second embodiment.

FIG. 16 is a process block diagram (part 3) illustrating a method for obtaining posture data in the method for creating teaching data of the articulated robot according to the second embodiment.

FIG. 17 is a diagram illustrating a path table in a process of obtaining posture data in the method for generating teaching data of the articulated robot according to the second embodiment.

FIG. 18A is a diagram illustrating a procedure for setting a line radially from a center point, and FIG. 18B is a diagram illustrating a procedure for extracting an intersection in a closed space by drawing a line in a grid pattern. .

19A is a diagram illustrating a procedure for setting a solid around an intersection, FIG. 19B is a diagram illustrating a procedure for extracting an overlapping portion between a solid and a work, and FIG. 19C is a diagram illustrating an extraction process. FIG. 9 is a diagram showing a work model that has been selected.

FIG. 20A is a diagram illustrating a center point of each solid, and FIG. 20B is a diagram illustrating a procedure for obtaining a principal component line.

FIG. 21 is a diagram showing a procedure for obtaining a drawing point and a drawing path (V1).

FIG. 22 is a diagram showing a drawing path (V2).

FIG. 23 is a diagram illustrating a mask process.

FIG. 24 is a process block diagram showing a method for creating teaching data of an articulated robot according to the related art.

[Explanation of symbols]

10 Offline teaching device 12 Robot device 16 Monitor 50 Articulated robot 52 Gun unit 70, 72
Electrode 74 Arm 80 Work 110, 120, 130, 140 Path table 1094 Solid 1096 Work model 1096b Opening side part G1, G2
Center of gravity position M1, M2 Principal component line Ps Study start position P1, P2, P3, P4 Attitude data Q1, Q2, Q3, Q4, Q5, Q6, Q7 Operation data Qe Study end position

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Ryo Nakajima 1-10-1 Shinsayama, Sayama City, Saitama Prefecture Honda Engineering Co., Ltd. F term (reference) 3C007 AS11 BS11 BT05 CT05 CV08 CW08 LS10 LS11 LV05 MT01 5H269 AB12 AB33 BB09 BB14 CC09 DD06 EE14 NN16 PP17 SA08 SA10

Claims (3)

[Claims] 1. A teaching data creating method for an articulated robot for creating posture data corresponding to a plurality of working points and movement data on an intermediate route connecting the plurality of working points, A method for creating teaching data for an articulated robot, comprising a first step of obtaining the posture data corresponding to all of the work points beforehand. 2. The teaching data creation method for an articulated robot according to claim 1, further comprising a second step of obtaining the motion data after the first step. How to make. 3. The teaching data creating method for an articulated robot according to claim 1 or 2, wherein in the first step, when a posture of the articulated robot cannot be established with respect to a certain work point, A method for creating teaching data for an articulated robot, comprising the step of recording that the work point is not workable.