JPS6126111A - Industrial robot - Google Patents

Abstract

PURPOSE:To position an end effector precisely even if the stop position of a moving board is shifted by compensating control shaft command information so that the positional shifts of the end effector at the time of teaching and reproducing are included in an allowable value on the basis of said positional shifts. CONSTITUTION:If it is supposed that the end effector E is fitted to the head of an arm 4, the position of the end effector E on the X coordinate can be calculated as FX on the basis of feedback information obtained from respective encoders. Since the moving board 2 is stopped at a position 2' in case of reproducing the end effector E is moved to a position E' by controlling control shafts beta1, beta2 etc. at the same status as teaching and the position of the end effector E on the X coordinate shaft is located on FX' shifted from that of teaching by DELTAx. Therefore, command information to be applied to the control shafts beta1, beta2, etc. is compensated so that FX' is included in an allowable value on the basis of the difference DELTAx. Thus, the shift of the moving board is compensated by the control shafts beta1, beta2, the position of the end effector E at a reproducing time coincides with that of the teaching time and the work can be precisely processed.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of the Invention) The present invention relates to an industrial robot, and more particularly, to an industrial robot having a first control axis positionable in a predetermined direction, an origin supported by the first control axis, and at least one The present invention relates to an industrial 1-1 pot including one or more second control shafts capable of positioning the end effector in the same direction as the predetermined direction.

(Description of Prior Art) FIG. 1 is a perspective view showing an example of an industrial robot as shown in (a) above. To put it simply, this robot R has 6 degrees of freedom between control axes X and control axes β to β5.
It supports an α-based robot with degrees of freedom. By means of the control axes β1 and β2, the grasping means 7 as an end effector can be positioned at least in the same direction as the control axis X.

Referring to the detailed description of boxes 1 to 1 in FIG. 1, reference numeral 1 denotes a straight horizontal guide, a movable table 2 is supported by this guide 1, and the position in the X direction is forced by a hydraulic motor Mx. 3 is a horizontal first arm that is vertically supported on the movable table 2, and is forced to have a turning angle β1 by a hydraulic motor M. Reference numeral 4 denotes a horizontal second arm that is vertically supported on the third end of the arm, and is forced to have a turning angle β2 by a hydraulic motor M. Reference numeral 5 designates a first rotating body vertically supported at the tip of the arm 4, and the rotating angle β is controlled by a hydraulic motor M3 (not shown).
3 is forced. A vertical arm 6 is supported on the rotating body 5, and is forced to move in the β4 direction by a hydraulic motor M. 7
is horizontal to the lower end of arm 6! 1lllll 1st Yundoev ■
There is a grasping means C as a motor, and the rotation angle β5 is forced by the oil L1 and the motor M5. 1. Also, the claws 7 of the grasping means 7.
can be opened and closed by an oil cylinder (not shown).

These are the 11 beta-based bots.

A second rotating body C 8 is supported coaxially with the rotating body 5, and is forced to have a rotation angle α1 by an electric motor M11 (not shown). 9 is a vertically rotating 11'li which is horizontally supported on the rotating body 8.
l! 1- is present, and the electric motor M12 (this forces the elevation angle α2). 10 is a vertically rotating second arm supported horizontally on the tip of the arm 9, and the electric motor M13 forces the elevation angle α3. A third arm 11 is horizontally supported at the tip of the arm 10, and is forced to have an elevation angle α4 by an electric motor M14.

12 is supported on the arm 11 and on the i-axis at right angles to the elevation angle of the arm 11, and the 21st hand 1-) 1 (welding torch ■)
Holder (with ゛, rotating angle α5 by electric motor M15)
is forced.

The above are α-series robots.

Control axis X, control axis β ・~β5 and J: and control axis α1~
Each movement of α5 is detected by encoders FX, E1 to 13, , F11 to [1, not shown) as detection means.

Each control axis (each degree of freedom) of this robot R is configured to be controlled by a conventionally known control means (for example, a micro combi coater) and by a known playback method.

FIG. 3 is a diagram showing an example of a workpiece machining state by the robots 1 to 1 described above. It is assumed that the steel plate W1 as a workpiece is positioned in a horizontal state in the moving area of the l''bot l-R by another device (not shown), and the β-based robot grips the channel material W2 with the gripping means 7, and Insert the steel plate W into the channel HA W2 from above as shown in the figure.
Keep it pressed to 1 (). In this state, α series robots 1~
end 1 no [1・-chi T as a lid welding 1
Reply to tack on the missing parts.

FIG. 2 is a diagram showing the coordinate system of the robot in FIG. 1.

The aforementioned control axis X belongs to the first rectangular coordinate system XYZ, and the position of the moving table 2 is based on the position information (X) of the coordinate axis X-hi.
It is expressed as The aforementioned control axes β to β5 belong to the second rectangular coordinate system xyz in which the origin O moves together with the moving table 2, and the position and orientation of the gripping means 7 are (x, y, Z, θ, φ). expressed. The coordinate axis X and the coordinate axis X are parallel.

However, when controlling the position of the robot's movable table 2 as described above to a different position, the total weight of the structure supported by the movable table 2 is large, and the movement range of the movable table 2 is large. Therefore, it is difficult to accurately stop the movable table 2 at the previously taught X-axis position due to the tongue on which inertial force acts. Moreover, the deviation from the teaching point also varies from time to time. In the I-type where the gain of the X-axis board system is set to the first, there are problems such as hunting, etc.
Solving this problem (Ma(paki/, fii.vr-)
When the movable base 2 moves away from the predetermined stop position, when executing the contents of the next slurry, for example, the position of each control axis supported by the movable base 2 is controlled to convey the soak, process, etc. If this occurs, it will no longer be possible to carry out proper transportation, processing, etc.

(Purpose of the Invention) This invention provides an industry in which even if the stop position of the movable base deviates from a predetermined position (for example, a previously taught position), an end effector can be accurately positioned at a predetermined position by making corrections. The purpose is to provide robots for

(Structure of the Invention) As shown in FIG. 9, the present invention includes a command means 21 that provides position command information for controlling the first and second control axes 22 and 23 during teaching, and The position detecting means 24 detects the positions of the first and second control axes during playback and outputs position feedback information, and the position detecting means 24 detects the positions of the first and second control axes and outputs position feedback information, and the position detecting means 24 detects the positions of the first and second control axes and outputs position feedback information.・
The position information (FX) of the first control axis and the second coordinate axis (X
) and the position information (F x ) of the end effector on the first coordinate axis (X) to obtain comprehensive position information (F x ) of the end effector on the first coordinate axis (X).
5. -7- A storage means 26 for storing the position command information and front ranking position information (FX) at the time of eaching as j-evening information, and based on the quality feed pack information inputted at the time of playback. Then, the position information (X') of the first control axis on the first coordinate axis (X) and the position information (Fx
A second operator stage 27 calculates the total position information (FX') of the end effector on the first coordinate axis (X) by calculating (FX) and the comprehensive position information (F
The third trunk means 28 calculates the difference (ΔX) from
ΔX) to obtain the corrected position information (X'), and the end 1 position on the second coordinate axis (X) of the teaching information: [lid position command information ( Cx) and the corrected position information <X'),
An output means 30 is provided for outputting the replaced teaching information to the first and second control axes in order to control them again.

(Explanation of Examples) Hereinafter, the present invention will be explained based on examples.

FIG. 6 is a block diagram showing a control device when the present invention is implemented by a microcomputer. COO
is a control means (microcomputer) for β-based robots
1. which includes CP tJβ and memory MEβ. In the pass line FLUβ of the microcomputer COβ,
Butching box TBβ, X axis 1 novo system sx, β
1-axis υ-bo system Sβ1, β axis υ-bo system Sβ, β ] Novo system Sβ3, β4 axis→Novo system Sβ, β5-axis servo system Sβ5, and gripping means 7 are connected. The teaching box 1-Bβ is equipped with a mode switching switch MS, a speed setting switch SS, a joystick JS, a gripping means switch fGS, and a teaching switch TS.
Further, the servo system 3 is provided with a guide b (not shown), an interpolation switch, etc., and each servo system is provided with a hydraulic motor M-M and encoders Ex--E, respectively.

CO is the control means for the α-based robot i, and CP LJ. and main MF
including. Mike “Don” Pi = 1-Y00゜ bus α Line B LJ ni lma, teaching box d [
3 (x, α welding current WS, α axis) Norvo system Sα1, CX axis 1
No. 1 Bo system Sα, α axis - Bo system Sα, α4 axis -
Bo system Sα4, α5dlII (Pubo system Sα5 is connected.Each Ribo system CJ1, previous J Lmu effect movement 'E
-Enter M11/-M15 and Enter E11~
[It is equipped with 15.

Both microphones 1] computers 00β and C0(x are connected via a known interface).

Each lever system also has an engine that detects the position of each control axis.
1-da is included. Incidentally, since the α system 1" box 1- is not directly related to this invention, its explanation will be omitted below.

FIGS. 4 and 5 are diagrams for explaining the operation of the embodiment of the present invention. The details will be described later, but to briefly explain the dynamics of the embodiment, command information is given to the β-system robot from the deeding means to control each control axis, and the β-system robot is activated at the position indicated by the solid line in FIG. Suppose you are teaching a robot.

To simplify the explanation, assume that the end effector E is at the tip of the arm 4, and the position of the end effector [on the X coordinate axis is determined by each encoder 1. It can be calculated based on the feedback information of s + et al., and this is designated as tX. During playback, as described above, there is a shift in the stop position of the movable base 2, so the movable base stops at the 2' J:4 position, as shown by the dotted line in Figure 4. I'm tired. Therefore, if the C1 control axes β and β2@ are controlled to the same state as during teaching, the end effector will come to position E', and its position on the X coordinate axis will be different from that during teaching by ΔX. c) The result is a shifted FX'. In this case, the J
However, it is not possible to accurately process the workpiece. However, in this embodiment, the command information applied to the control axes β1, β2, etc. is corrected based on the difference ΔX so that it falls within the allowable value. As a result, as shown by the dashed line in FIG. 5, the displacement of the moving platform is corrected by the control axes β1β2, etc.
The position of the end effector E during playback matches that during teaching. Therefore, the workpiece can be machined in a correct manner.

Next, the operation of this embodiment will be explained in detail. FIG. 7 is a flowchart showing the main operations of the control device shown in FIG. 6 during teaching. In step S1, each control axis of the β-system robot 1 is controlled for teaching.
Command information (Cx1,0β11.Cβ
21. cβ31. Cβ4. , Cβ51> are given and each of them is taken into the control device 3. This command information, for example, is sent to 2, i l; I teaching, reproducing steps, which are executed in the control device according to the movement of i +, 2, 3, . . . .

The actual position of each control axis is detected by the two signals assigned to each control axis, and the noise pack information (
FX, ``β11,1β21.1-β31, [-β41
．． [-β51) is taken into the control device within 1".

Information (β11 to Fβ51) in the r-domain information is coordinate-transformed from the β system (multi-joint coordinate system) to the X system (orthogonal coordinate system),
The position/orientation information (X・, Fy・,
FZ・, Fθ1. Fφi) is obtained. Step Sl
, the following equation is performed, and J is x axis F.
The comprehensive position information tX of the end 1 effector at is determined. The first calculation means 25 described above performs this step s4.
Performs the processing content corresponding to .

FX, ==FX H+F x i −
(1) In step S5, the above-mentioned command information (Cx
, to Cβ11-Cβ5i) and the comprehensive position information FX are 1
It is stored in the memory in the control device as teaching information in the step step.In step s6, it is determined whether or not i is the final step. Then, the program returns to step S1. Teaching is performed one after another as described in step 1-.

8Δ FIG. 41 to FIG. 8C are the 6th
This is Noroji X7-1-, which shows the main operations of the control device shown in the figure.

In step S11, the teaching information (Cx,, Cβ, ~Cβ, FX,) of the i step is +
1+ 5i + retrieved from memory.

In step 812, the command information CX1 for the control axis
It is determined whether it is the same as that of step 3(i-1). The same means that the control axis X has not moved from the previous step, and the fact that they are not the same means that the control axis X has moved from the previous step. In step 813, the correction amount ΔX1-1 determined in the previous step is cleared. Step 8
14, command information (Cβ
11 to Cβ5H) are coordinate-transformed from the β system to the X system, and command information < c x H, Cy, Cz·, Cθ, Cφ) is obtained. In step S15, the following expression mlN
Then, the corrected position information x1' is obtained. 1
) The start correction means 29 performs processing corresponding to step 815.

x・' =Cx・→−Δxi−1 −(
2) In step 816, the original command information Cx in the teaching information and this corrected position information x,
l is replaced, and the command information after replacement (x, l.

Cy, Cz,, Co1. Cφi) is coordinate-transformed from the X system to the β system, thereby obtaining command information (Cβ·' to Cβ51') for the control axes β1 to β5. In step 817, this command information and the original command information CX1 for the control axis X are output to each control axis, thereby controlling each control axis. As mentioned above
The output means 30 performs processing corresponding to step S17.

In step S18, it is determined whether the robot (β-based robot) has stopped. This judgment can be made, for example, by β
System [This is performed depending on whether the transmission from one hot encoder is constant or not. The robot is f? Stop LI I= <
'c la LC, s'i IS 19 G;-, 43 i'
(, paJ Sugo axis
~ "β51') or f+ ("
In the case of J43"
1 Due to the poor graininess, the control axis X (1) stops.
16 Due to poor lees quality, the control axis of the teaching machine
The one node back information 1-x1 about the control axis X and the f-back information LX,' about the control axis

In step S20, control axes β to β5. Feedback information about (Fβ1i' ”−”−” 5i
') is coordinate transformed from the β system to the X system, and the information ([-Xi'
, I-y', 'z', F
θ ・ ′, φ 1 ′) is obtained. In step 821, calculation of the a3-C linear equation is performed, thereby obtaining the comprehensive position information [x1' of the end 1-noecta of the X-axis h-c.

The second calculation means 27 described in previous step S21
Perform the processing content corresponding to .

FX ' = FX, ' +FX ' ・(
3) In step S22, the -C linear equation is calculated,
As a result, the previous FiS general position information FX during teaching, and the above general position information FX during playback.
'The difference ΔX1 is determined (see FIG. 4). As mentioned above
This third calculation means 28 corresponds to this step 322.
Perform the processing contents.

In step S23, it is determined whether or not the difference ΔX1 is within the allowable value. The program does not enter step S24 and proceeds to step 824. In step S24, the following equation is calculated, and the previous correction amount ΔX1-1 is corrected using the difference ΔX, and the program returns to step S15. The operations of steps 815 to S24 are repeated until the difference ΔX1 falls within the allowable value.
The purpose is to correct the deviation of the control axis X using the control axes β, β2, etc., so that the position of the T-end effector of the reproduction IS coincides with the teaching position.

When the difference ΔX falls within the allowable value, the correction amount Δx1-1 is stored in the memory as the correction amount ΔX of the current step 1 in step S25. In step S26, it is determined whether or not i is the final step. In this case, in the next playback step (for example, step 111), the control axis
), the correction amount ΔX stored in the memory is cleared in step 813.Fl is used in step S15, so the difference ΔXi+1' mentioned above can be ascertained (easily tolerated). Clearing the compensation amount ΔX at step 1ε313 is the stop position 1 of the control axis
This is because when the control axis X moves, the correction h1ΔX is calculated based on the movement of the control axis X.

According to this embodiment, the loop from step 815 to step 82/I in FIG. 8 corrects the difference ΔX so that it always falls within the allowable value. Therefore, the positional accuracy of the end effector during reproduction is extremely high.

Also, in this example, the previous)! In order to find the difference ΔX of (see step S22 in the figure), not only the position error is corrected when the control axis
1. Even if there is a stop position error in β2, etc., it is corrected together with these 6 errors. Therefore, from this point of view as well, the positional accuracy of the first effector during reproduction is extremely high.

Furthermore, in this embodiment, the teaching information regarding the control axis X is the command information Cx itself given to the control axis X at the time of teaching (step S in FIG.
5), feedback information from the encoder (there is no ゛. Now, after the taming, play it back on a trial basis and check the control axis
Let us consider a case where one of several points with the same position is in a bad position and only point A is to be re-taught. If the feedback information from 2[n=+-da is used as the teaching information for the control axis Then, feedback information different from the previous one becomes teaching information regarding the control tall axis x. That is, the operator [yo, control 11
Even though I moved only to qlβ and re-taught a certain point that I made before), the actual control l1llllX
This is the same as moving 6. In this case, the certain point mentioned above will have different teaching information from the other points, and the correction operation during playback will not be performed properly.
9. Accurate positioning is not possible. However, in this embodiment, the teaching information regarding the control axis , the teaching information of -C regarding the control axis X does not change at all. Therefore, the correction operation during playback is successfully performed, and the positional viscosity of the end effector remains extremely high.

J: If the β-system robot is accurately controlled to the target position in this manner, the α-system]] robot supported by it can naturally perform accurate temporary (-IIJ welding, etc.).

Note that the robot R described above supports the control axes β to β5 that can be positioned three-dimensionally on the control axis X, but for example, the control axis It supports a control axis that can be positioned two-dimensionally in the X direction and the direction perpendicular to this.

Furthermore, 1 that can be positioned in the same direction as the control axis
The b-sized structure supports two control axes. Also, even if the control axis
It is sufficient if it can be expressed as Z.

(Effect of the Invention) According to the present invention, ], control axis XO') stop 1-i☆ error prevention is corrected, and control axis β1. β2 etc. stop J1st place I
I KGj difference is also corrected, and J: d, [nd soil]
“Positioning the lid in place extremely accurately is the key to
゛Saru.

[Brief explanation of drawings]

1) FIG. 1 is a perspective view showing an example of an industrial robot. FIG. 2 is a diagram of the 86 standard system 1-: Bot 1~ in FIG. 1. FIG. 3 is a small-sized diagram showing an example of a workpiece processing state by the robot shown in FIG. 1. Figure 4 J3 J, σ 5th
The figure is a diagram for explaining the first embodiment of the present invention. FIG. 6 is a diagram of the control device when the present invention is implemented in the microcomputer 1. Figure 7 shows one of the control devices in Figure 6 during teaching.
Figures 88 to 8C show the rough behavior of the control device of Figure 6 during playback. 9 is a diagram for explaining the present invention in detail.
... Microcomputer for β-system robot 1, FX and FX' ... Position feedback information of the control axis X on the coordinate axis X during teaching and playback, respectively, FX and F Position feedback information of end 1 effector on coordinate axis X during playback, FX and FX'
...Comprehensive position information of the end effector of coordinate axis X10 during teaching and playback respectively

Claims (1)

[Claims] (1) A first control axis capable of positioning in a predetermined direction; and one or more second controls whose origin is supported by the first control axis and capable of positioning the end effector in at least the same direction as the predetermined direction. and one of the position information of the first control axis is a first coordinate axis (
X) one of the positional information of the end effector is parallel to the first coordinate axis (X) in a second Cartesian coordinate system whose origin moves together with the first control axis; an industrial robot expressed on a second coordinate axis (x), comprising: command means for giving position command information to the first and second control axes for controlling them during teaching; and during teaching and playback. position detecting means for detecting the positions of the first and second control axes and outputting position feedback information at the time of teaching; The position information (FX) of the control axis and the position information (Fx) of the end effector on the second coordinate axis (x) are calculated to calculate the overall position information of the end effector on the first coordinate axis (X). (@FX
a first calculation means for calculating the position command information (@FX@) at the time of teaching; a storage means for storing the position command information and the comprehensive position information (@FX@) at the time of teaching as teaching information; Position information of the first control axis on the first coordinate axis (X) (
FX') and the position information (Fx') of the end effector on the second coordinate axis (x),
) on the end effector's comprehensive position information (@FX'@
) for calculating the comprehensive position information (@F@@X
@) and the general position information at the time of playback (@FX'@)
and a third calculating means for calculating the difference (Δ@X@) from the position command information (Cx) of the end effector on the second coordinate axis (x) of the teaching information to the difference (Δ@X @
) to obtain the corrected position information (x'); and the position command information (Cx) of the end effector on the second coordinate axis (x) in the teaching information and the corrected position information. and output means for replacing the information (x') with the teaching information after the replacement and providing the replaced teaching information to the first and second control axes in order to control them during reproduction.