JPH0754446B2 - Robot controller - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention is used when a work is arranged in a work container (hereinafter referred to as a pallet) by a robot and a work is taken out from the pallet (hereinafter referred to as a palletizing work). The present invention relates to a robot coordinate system correction device that corrects a robot coordinate system.

[Prior Art] In assembly work of machines and the like, as shown in FIG. 4, a palletizing work in which a robot (not shown) sequentially takes out works from a pallet (7) on which a plurality of works (6) are arranged is performed. Accompany. In this case, when a so-called teaching playback type robot having no function of coordinate calculation is to perform palletizing work, the 6 × 4 = 24 grid point positions, that is, the work placed on the pallet (7). The position of (6) must be taught, and when the pallet (7) has many grid points, it is not practical.

Therefore, recently, using a robot with a coordinate calculation function,
This robot has 3 points (D1) and (D
It is becoming common to teach only the coordinates 2) and (D3) and to calculate all other grid points (DX) by coordinate calculation. That is, the three points (D
By teaching the coordinates of 1), (D2), and (D3), the plane of the pallet (7) is uniquely defined, and the coordinates of all other grid points (DX) can be calculated.

FIG. 5 is a block diagram of a robot coordinate system correction device that calculates the coordinates of other grid points by teaching the coordinates of three points. In FIG. 5, (1) is a reference point storage unit that describes reference points (D1), (D2) and (D3) that determine the working coordinate system of the plane of the pallet (7), and (2) is a reference point. A work coordinate calculation unit that calculates the position of a work point (DX) having a predetermined coordinate value as a combination of vectors based on the reference points (D1) to (D3) stored in the point storage unit (1). .

Next, FIG. 6 shows the relationship between the reference points (D1), (D2) and (D3) by the correction device for the conventional robot coordinate system and the diagonal calculation points calculated by the work coordinate calculation section (2) and the amount of deviation. It is an explanatory view showing the occurrence of. In FIG. 6, reference points (D1), (D2) and (D3) are the origin of the work coordinate system, and reference points (D2) and (D3) are the vertical coordinate axes Y, respectively. And a point that determines the coordinate axis X in the horizontal direction. As described above, the position of the work point (D4) having a certain coordinate value is calculated by the work coordinate calculation unit (2) by combining the vectors.

By the way, when a robot having a multi-joint mechanism is positioned at a calculation work point (D4) calculated by the work coordinate calculation unit (2), the robot may be affected by a reference error of each joint angle of the robot and an error of an arm length. Will be positioned at a working point (D4 ') that does not necessarily match the calculated working point (D4). That is, a diagonal calculation point (work point (D4) for calculation of a rectangle determined by the reference points (D1), (D2) and (D3)
If the robot is to be positioned at, the robot will be positioned at the work point (D4 ') separated by the deviation amount Δ. It should be noted that the maximum error will appear at the diagonal point that is most distant from the generally taught reference point (calculation reference point).

[Problems to be Solved by the Invention] As described above, the conventional correction device for the robot coordinate system is
Even if the work coordinate calculation unit (2) calculates the work point (D4) on the work coordinate system and attempts to position the work point (D4) on the calculation using the robot, the accuracy error of the robot itself is Due to the influence, there is a problem that the positioning cannot be performed accurately.

The present invention has been made to solve the above problems,
Suppresses the influence of the accuracy error of the robot itself,
An object of the present invention is to provide a robot coordinate system correction device capable of accurately positioning a robot at a work point calculated in the work coordinate system.

[Means for Solving the Problems] In the robot controller according to the present invention, the predetermined positions of each of the three corners of the four corners of the work array in which the works are arranged on the pallet in a grid pattern at predetermined intervals. Is stored based on the teaching, a reference point storage unit for correction in which a predetermined position of the remaining one corner of the four corners of the work array is stored based on the teaching, and the stored contents of the reference point storage section Based on the above, a correction vector calculation unit that calculates a predetermined position of the work of the remaining one corner of the four corners, and a correction vector indicating the difference between the calculation result and the stored contents of the correction reference point storage unit A correction work coordinate calculation unit that corrects a command position based on the stored contents of the point storage unit by the correction vector calculated by the correction vector calculation unit and outputs the corrected command position to the machine drive unit. It

Further, the corrected work coordinate calculation unit calculates the position of the work based on the stored contents of the reference point storage unit, the number of arrangements of the work arrangement, and the work arrangement position, and the calculation result is based on the arrangement position of the work. The correction vector is proportionally corrected, and the correction result is output to the mechanical drive unit.

[Operation] In the present invention, the predetermined positions of the work at the three corners of the work array on the pallet are taught, and the respective taught positions are stored.

Further, when a predetermined position of the work in the remaining one corner of the four corners of the work array is taught, the taught position is stored.

Next, the predetermined position of the work of the remaining one corner of the four corners is calculated based on the predetermined positions of the works of the three corners of the work arrangement, and the calculation result and the remaining one of the four corners according to the teaching are calculated. A correction vector indicating the difference between the corner and the predetermined position of the work is calculated.

Then, the respective work positions calculated based on the respective predetermined positions of the works at the three corners of the work arrangement are corrected based on the correction vector, and the corrected command positions are output to the machine drive unit, so that the hand is accurately controlled. Position on each work.

Also, the position of the work is calculated based on each predetermined position of the work on the three corners of the work array, the number of the work array and the work array position, and the calculation result is proportionally distributed based on the array position of the work. The correction is performed using the corrected vector and the correction result is output to the machine drive unit.

[Embodiment] An embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a block diagram of a robot coordinate system correction apparatus according to the present invention. In FIG. 1, (1) is a reference point storage unit that stores reference points (D1), (D2) and (D3) that determine a plane work coordinate system, and (3) stores correction reference points. Correction reference point storage unit, (4) correction vector calculation unit,
(5) is a correction work coordinate calculation unit.

Next, the operation of the robot coordinate system correction apparatus according to the present invention will be described with reference to FIG. 2 showing generation of a correction vector and correction of work point coordinates using the correction vector. In addition, in FIG. 2, reference points (D1), (D2)
And (D3) are points where the reference point (D1) determines the origin of the work coordinate system, and the reference points (D2) and (D3) determine the vertical coordinate axis Y and the horizontal coordinate axis X, respectively.

First, the correction vector calculation unit (4) has two coordinate axes X and Y.
With respect to, the coordinates of the diagonal reference point (D4 ') of the reference point 1 are calculated by combining the vectors based on the reference points (D1) to (D3) stored in the reference point storage unit (1).

Then the correction vector calculator (4) is actually numbered,
The reference point (D) stored in the correction reference point storage unit (3)
The shift amount between 4) and the calculated diagonal reference point (D4 ′), that is, the coordinate correction vector ΔP is calculated.

Then, the correction work coordinate calculation unit (5) calculates the correction vector ΔP.
The calculation points (lattice points) of the working coordinate system are corrected based on the above. The method of correcting each calculation point of the work coordinate system performed by the correction work coordinate calculation unit (5) is the proportional distribution method. That is, the reference point (D1), the coordinate vector P _1, P ₂ and (D2) and (D3)
If P ₃ and the ratio coefficients of the working point (DX) to be calculated to the reference points (D1) and (D2) are u and v, the coordinate vector P _x of the desired point is P _x = P ₁ + (P ₂ −P _{1 ) u + (P 3 -P 1} ) v + ΔP · u · V (1) to become. Here, ΔP is a correction vector, and ΔP = (P ₄ −P ₁ ) − {(P ₂ −P ₁ ) + (P ₃ −P ₁ )} (2). As described above, the first formula has the correction term (ΔP · u · V) which is not the conventional calculation formula, and the correction work coordinate production unit (5) does not calculate the diagonal reference point (D4 ′). In addition to the correction calculation, the correction calculation is also performed for other coordinate points.

In this way, it is possible to calculate the working point in which the error is corrected in advance, in consideration of the occurrence of an error when positioning the robot or the like. When this is applied to an actual pallet (7), it becomes as shown in FIG. That is, the robot can be positioned without displacement from the reference point (D4) and each working point (DX) with respect to the reference points (D1), (D2), and (D3).

Although the correction calculation using the correction vector is performed using the proportional distribution method in this embodiment, the correction calculation may be performed using other methods.

Further, although the working coordinate system is a plane in this embodiment, it may be a three-dimensional working coordinate system, and in this case, both the reference point and the correction reference point are expanded to three-dimensional coordinates.

[Effects of the Invention] As described above, according to the present invention, the predetermined positions of each of the three corners of the work array on the actual pallet according to the teaching are stored, and based on the stored contents, the four positions of the work array are stored.
A correction vector that indicates the difference between the predetermined position of the work of the remaining one corner of the corners and the difference between this calculation result and the predetermined position of the work of the remaining one corner of the four corners of the work array obtained by the teaching. Then, the position of each workpiece calculated based on the above-mentioned stored contents is corrected by the correction vector, and the corrected command position is output to the machine drive unit, thereby the error of the robot itself or the error due to the pallet distortion. Is corrected, so that the hand can be accurately positioned on each work.

Also, the position of the work is calculated based on each predetermined position of the work on the three corners of the work array, the number of the work array and the work array position, and the calculation result is proportionally distributed based on the array position of the work. Since the correction result is corrected by the corrected vector and the correction result is output to the machine driving unit, even if there are a plurality of works, the hand can be accurately positioned on each work by simple calculation. There is.

[Brief description of drawings]

FIG. 1 is a block diagram of a robot coordinate system correction device according to the present invention, and FIG. 2 is an explanatory diagram of a coordinate system correction using a correction vector by the robot coordinate system correction device shown in FIG. The figure is an explanatory view when the correction of the coordinate system is applied to the pallet, Fig. 4 is an explanatory view of the pallet, Fig. 5 is a configuration diagram of a conventional robot coordinate system correction device, and Fig. 6 is a conventional robot coordinate system. FIG. 6 is an explanatory diagram of the generation of the positional deviation amount of the diagonal calculation points by the correction device of FIG. In each figure, 1 is a reference point storage unit, 3 is a correction reference point storage unit,
Reference numeral 4 is a correction vector calculation unit, and 5 is a correction work coordinate calculation unit. In the drawings, the same reference numerals indicate the same or corresponding parts.

Claims (2)

[Claims] 1. A reference point storage unit in which predetermined positions of each of the three corners of the four corners of the work array in which the works are arranged on the pallet at predetermined intervals in a grid pattern are stored based on teaching. A correction reference point storage unit in which a predetermined position of a work piece in the remaining one corner of the four corners of the array is stored based on a teaching, and the remaining one corner of the four corners based on the stored contents of the reference point storage unit. A correction vector calculation unit that calculates a predetermined position of the workpiece and calculates a correction vector indicating the difference between the calculation result and the storage content of the correction reference point storage unit, and a command based on the storage content of the reference point storage unit. A robot control apparatus comprising: a correction work coordinate calculation unit that corrects a position by the correction vector calculated by the correction vector calculation unit and outputs the corrected command position to a machine drive unit. 2. The corrected work coordinate calculation unit calculates the position of the work based on the stored contents of the reference point storage unit, the number of work arrangements and the work arrangement position, and the calculation result is calculated for the work. The robot controller according to claim 1, wherein the controller is corrected by a correction vector that is proportionally distributed based on the array position, and the correction result is output to a mechanical drive unit.