JPH09204212A - Correcting method for working data for nc robot system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a correcting method for NC data for NC robot system with which working can be performed without fail even if three-dimensional position deviation is generated when a work piece is installed. SOLUTION: When correcting the working data in the NC robot system which has a placer and the main body of robot and inputs coordinates for teaching three points and the working data of the work piece in advance, the really installed object to be worked is sensed by the main body of robot, the actual coordinates for teaching three points corresponding to these coordinates are found, the deviation between the actual coordinates and the teaching coordinates is found and based on the deviation, the three-dimensional position data of placer and the attitude data of the main body of robot are corrected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for correcting machining data in an NC robot device.

[0002]

2. Description of the Related Art Conventionally, in an NC robot apparatus, for example, an NC robot apparatus for welding, NC data, which is welding data (processing data), has been input in advance. Since it is not always correctly installed at the predetermined position on the welding table, the position of the actually-welded object is detected, and the NC data is corrected based on the detected data.

In the correction of the NC data, only the two-dimensional correction, that is, the plane shift amount is taken into consideration.

[0004]

Since the above-described conventional correction method is a two-dimensional correction, it can be applied only when the welding surface is flat or close to it, and the position is three-dimensionally corrected. If there is a gap, there is a problem that it cannot be dealt with.

Therefore, it is an object of the present invention to provide a method of correcting NC data in an NC robot apparatus which can cope with a three-dimensional displacement.

[0006]

In order to solve the above-mentioned problems, a method for correcting NC data in an NC robot apparatus according to the present invention has a placer and a robot body, and has basic data for teaching three points and a workpiece in advance. A method for correcting the above-mentioned processing data in an NC robot device to which processing data is input, wherein the actually installed workpiece is sensed by the robot body to obtain actual data for three-point teaching corresponding to the above basic data. , The deviation amount between this actual data and the basic data is calculated, and based on this deviation amount,
A method for correcting the three-dimensional position data of the placer and the posture data of the robot body.

Further, in the above-mentioned correction method, when the robot body senses three points for teaching by the robot body, three points which are not on a straight line are sensed.
According to the above correction method, even if the installation of the workpiece is three-dimensionally displaced, the displacement can be accurately corrected.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of an NC robot apparatus according to the present invention, for example, a method for correcting NC data in an NC robot apparatus for welding will be described below with reference to FIGS.

In this NC data correction method, a deviation occurs between the work coordinate system {W} defined in the NC data and the work coordinate system {W '} actually set on the welding table. In this case, the amount of deviation is detected and the NC data is corrected.

In this correction, when the target work (workpiece) is a three-dimensional block, the deviation between the coordinate systems {W} and {W '} becomes three-dimensional, and therefore NC data (processing data) ), A three-dimensional correction is needed.

Here, N for welding to which the present correction method is applied
A schematic configuration of the C robot apparatus will be described with reference to FIG.
The NC robot apparatus 1 includes, for example, a placer 5 including a gate frame 4 having traveling wheels 3 that can travel on a guide rail 2 and a horizontal member 4a of the gate frame 4 of the placer 5.
The robot main body 6 is mounted so that it can be raised and lowered and swung freely, and a welding torch 7 attached to the tip of the arm portion 6a of the robot main body 6.

Next, the overall flow of the NC data correcting method in the NC robot apparatus 1 will be described with reference to FIGS. 2 and 3. First, as shown in FIG.
Three points (E, F, G) marked on the work 11 actually set on the welding table are sensed (wire touching sensing) by the welding torch 7, and the coordinates (actual data) of these three points are read. (Step 1 in FIG. 3).

Next, based on the coordinates of these three points, the three points (A, B, C) that have been input (taught) in the system in advance.
The coordinates (basic data) of are corrected (step 2 in FIG. 3). Next, the shift amount of the work 11 is mathematically calculated based on the corrected coordinates of the three points and the coordinates of the three points stored in the NC data (step 3 in FIG. 3).

Then, based on the calculated deviation amount,
The three-dimensional position data (x, y, z) of the placer 5 is corrected, and the posture data (x, y, z) of the robot body 3 is corrected.
z, U, V) are corrected (steps 4 and 5 in FIG. 3).

The contents of steps 2, 4 and 5 will be described in more detail below. First, the correction of the taught three points (step 2) will be described. As shown in FIG. 4A, the triangle formed at the teaching point of the installed work 11 is represented by ΔABC, and as shown in FIG.
The triangle consisting of the teaching points entered in the data is ΔE
It is represented by FG.

If an error occurs during teaching, there should be a difference between the shapes of ΔABC and ΔEFG. Therefore, assuming that no error has occurred at the point A, the points B and C are corrected to become the points B ′ and C ′ based on the following procedure.

That is, the point A is not modified as described above. Next, so that AB ′ = EF,
Take point B'on the extension of line AB. Next, ∠B'A'C '
A point C ′ is set on the plane ABC so that = ∠FEG, AC ′ = EG.

Next, the coordinates of the points B and C actually obtained by the teaching are corrected to the coordinates of the points B'and C '. This modification is performed to reduce the influence of the sensing error on the corrected data, and the 3 sides of the triangle connecting the 3 sensing points input in the NC data are actually sensitized to 3 sides. The coordinates of the actually sensed points are rewritten so that the three sides of the triangle connecting the points are equal. That is, the corrected coordinates are regarded as the coordinates of the actually sensed point.

This prevents an error generated by sensing from propagating and having a great influence on the correction result. Next, each method of correcting the position of the placer and correcting the posture of the robot body based on the deviation amount of the work will be described.

Here, the work coordinate system defined by the NC data creation system is referred to as a basic coordinate system, and the work coordinate system actually set on the welding table is referred to as a work coordinate system. In this correction method, the basic coordinate system and the placer coordinate system match, and the correction is performed by converting the coordinates represented by the work coordinate system into the basic coordinate system.

Therefore, it is necessary to obtain a conversion matrix in advance. This conversion matrix has 4 rows and 4 columns,
The rotation matrix has 3 rows and 3 columns, and the translation matrix has 3 rows and 1 column.

First, a method of correcting the placer will be described.
You. When considering coordinate conversion, set the workpiece coordinate system {W}.
Think fixed. Therefore, it may be
As shown in FIG. 5, the robot coordinate system {R} is changed to {R
_m }, The placer coordinate system {B} moves to {B '}.
Fig. 5 shows the seat of each part when the work coordinate system is fixed.
The standard system is shown, and {T} shows the torch coordinate system.

By the way, placer data is (x, y,
z). This shows the coordinates of the origin of the robot coordinate system {R} seen in the placer coordinate system {B},
When this is corrected, it becomes the coordinates of the origin of the robot coordinate system {R _m } seen in the placer coordinate system {B '}.

Here, the origins of the coordinate systems {R} and {R _m } coincide with each other, and the following formula is established. Therefore, the corrected placer data (X, Y, Z) is represented by the following formula. To be done.

[0025]

[Equation 1]

Therefore, by performing the coordinate conversion as shown in this equation, the three-dimensional position of the placer 2 can be corrected. Next, the attitude data correction of the robot body 3 will be described.

The posture data of the robot body 3 is (x,
y, z, U, V), which is (x, y, z,
Change the data in the form of u, v, w). However, u = cos U × cos V v = sin U × cos V w = sin V.

The posture data modified as described above is represented by the robot coordinate system {R}, and is corrected and converted into the robot coordinate system {R _m }. When the corrected robot data is represented by (x ', y', z ', u', v ', w'), each data is represented by the following equations and equations.

[0029]

[Equation 2]

Therefore, the posture data of the robot body 3 can be corrected by performing coordinate conversion as shown in the above equations and equations. In this way, the basic coordinates and the actual coordinates of the three teaching points are detected, the shift amount of the work 11 is obtained, and the above-described calculation is performed based on this shift amount, whereby the three-dimensional position of the placer 2 is calculated. The position data and the posture data of the robot body 3 can be corrected.

The correction is performed by a computer, the corrected NC data is input to the NC robot apparatus 1 side, and the welding operation is performed based on the corrected NC data.

In the above embodiment, when sensing three points for teaching, it was described that three points are not on a straight line, but a plurality of planes, for example, three planes exist independently. In this case, three points that are not on a straight line are sensed on each surface.

Further, as shown in FIG. 6, there are three surfaces (a).
~ (C) are perpendicular to each other, the first
3 points not specified on surface (a), 2 points not specified on surface 2 (b), surface 3 (c)
It can be corrected by one-point sensing which is not particularly determined above.

For example, as a concrete work, the case of painting the inside of the three-dimensional block 21 as shown in FIG. 7 can be considered. This solid block 21 is, for example, a flat plate 22.
And the horizontal material 23 attached to the inner surface of the flat plate 22.
And a longitudinal member 24 attached at a right angle to the horizontal member 23.

Therefore, when sensing the coating position of the three-dimensional block 21, the NC robot apparatus 3 for coating is used.
An ultrasonic sensor instead of the coating gun 35 is attached to the tip of the arm 34 of the robot body 33 arranged on the first placer 32, and sensing as shown in FIG. 6 is performed.

Further, in the above embodiment, NC
As the robot apparatus, the welding robot apparatus and the painting robot apparatus have been described as an example, but other NCs for processing are used.
This method can also be applied to a robot device.

[0037]

As described above, according to the correction method of the present invention, the actually installed workpiece is sensed by the robot body to obtain the actual data for three-point teaching corresponding to the basic data, and this actual data is obtained. The amount of deviation between the data and the basic data is obtained, and the three-dimensional position data of the placer and the attitude data of the robot body are corrected based on this amount of deviation, so that even if the workpiece is three-dimensionally displaced. Therefore, the deviation can be accurately corrected.

[Brief description of drawings]

FIG. 1 is a front view showing a schematic configuration of an NC robot device according to an embodiment of the present invention.

FIG. 2 is a perspective view illustrating a sensing operation by the NC robot device according to the same embodiment.

FIG. 3 is a flowchart showing a method of correcting processing data by the NC robot apparatus according to the same embodiment.

FIG. 4 is a diagram illustrating a method of correcting machining data by the NC robot apparatus according to the same embodiment.

FIG. 5 is a diagram showing a coordinate system of the NC robot apparatus in the same embodiment.

FIG. 6 is a perspective view illustrating a sensing operation in an NC robot device according to another embodiment of the present invention.

FIG. 7 is a front view showing a schematic configuration of an NC robot apparatus according to another embodiment of the present invention.

[Explanation of symbols]

 1 NC Robot Device 5 Placer 6 Robot Main Body 7 Welding Torch 11 Work 21 Solid Block 31 NC Robot Device 32 Placer 33 Robot Main Body

Claims (2)

[Claims] 1. A method of correcting the above-mentioned machining data in an NC robot apparatus having a placer and a robot main body, in which basic data for teaching three points and machining data of a workpiece are inputted in advance, which is actually installed. The workpiece is sensed by the robot body to obtain the actual data for three-point teaching corresponding to the basic data, the deviation amount between the actual data and the basic data is obtained, and the tertiary of the placer is calculated based on the deviation amount. A method for correcting machining data in an NC robot apparatus, characterized by correcting original position data and posture data of a robot body. 2. A device for teaching a work piece by a robot body.
The method for correcting machining data in an NC robot apparatus according to claim 1, wherein when sensing points, three points that are not on a straight line are sensed.