JPH1124724A - Method and device for generating robot program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily add the rule of a work for an object with a new shape without altering the system of the work. SOLUTION: This device is equipped with master data 1 wherein model data 1a, teaching rules 1b and 1c for calculating teaching point data according to respective models and additional models, etc., are registered, an input means 2 which selects one model out of the plural models registered in the master data 1, inputs a feature quantity to the selected model, and also inputs a teaching rule according to another new additional model, and a teaching point calculating means 4 which calculates a teaching point based on the feature quantity inputted from the input means 2 and the teaching rule 1b of the master data 1 and also calculates a teaching point based on the feature quantity and the teaching rule 1c corresponding to the additional model registered in the master data 1.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for generating an operation program for an industrial robot.

[0002]

2. Description of the Related Art In recent years, industrial robots have been actively introduced along with automation of production processes and manufacturing processes.

In such an industrial robot, it is necessary to create an operation program for operating the industrial robot. In particular, when creating a path operation program for a target work, an industrial robot is actually used and the target work is performed in accordance with the target work. A method of teaching by moving the camera has been conventionally employed. That is, a manufactured target work is installed in a work space of an industrial robot on a production line, and teaching is performed between the target work and the industrial robot (this is referred to as conventional technology 1).

Recently, a method of creating a robot program using an off-line device (for example, Japanese Patent Application Laid-Open
Japanese Patent No. 3111613) has also been proposed (this is referred to as prior art 2).

[0005] The program creation method of the prior art 2 is as follows.
At least one sequence pattern for robot operation and teaching point position data in a format including a geometric quantity expression variable are prepared, and the robot operation is performed by an alternative selection from at least one prepared sequence pattern. A sequence included in the program is specified, and a value of a geometric quantity expression variable included in the teaching point position data related to the specified sequence is specified. And
A teaching point position related to a specified sequence is obtained based on a value of a specified geometric quantity expression variable, and a robot operation program is created based on the specified content and the teaching point position. That is, a robot program is generated off-line from the beginning to the end in each case according to the robot language using the input feature amount of the target work.

[0006]

However, the method of the prior art 1 has the following problems. That is, teaching cannot be performed until the target work is manufactured. Since a technician who can perform teaching is required, there is a variation in teaching accuracy due to a difference in technician ability or the like. Since the program is created from the beginning, a large number of teaching steps are required. A large amount of teaching time (days) because the production (manufacturing) line must be stopped and teaching must be performed
Is required.

Further, the method of the prior art 2 has the following problems. That is, since the work order in which the robot can operate is generated from the beginning, a large number of man-hours are required. It is not possible to cope with a target work having a part of the shape or a different mounting part. To change the robot behavior,
You need to generate the program again. An expression for calculating all the teaching point data using the feature amount of the target work needs to be obtained in advance.

The present invention has been devised to solve such problems, and its object is to shorten the program generation time and the teaching time and to make the teaching accuracy uniform, and to solve different works. Another object of the present invention is to provide a method and an apparatus for generating a robot program which can also be applied and which does not require the creation of an equation for obtaining all teaching point data.

[0009]

According to a first aspect of the present invention, there is provided a method for generating a robot program, comprising: a plurality of models in which a target work is classified based on its characteristics; A teaching rule for calculating teaching point data and calculating teaching point data based on another new additional model, a posture master for determining the posture of the machining tool, and a conversion for coordinate conversion Data is provided with master data registered in advance, one model is selected from the plurality of models registered in the master data, and the feature amount input to the selected model and the teaching rule of the master data are selected. Calculating the teaching point based on the characteristic amount and the attribute registered in the master data. A procedure for calculating a teaching point based on a teaching rule corresponding to a model, a procedure for calculating attitude data at each of the calculated teaching points based on the attitude master of the master data, a calculated teaching point data and an attitude It is provided with a procedure for combining the data, a procedure for rearranging the data obtained by the coupling according to the input instruction, and a procedure for converting the rearranged data into data based on the robot origin.

According to a second aspect of the present invention, there is provided a robot program generating method according to the first aspect, wherein the converted data based on the robot origin is converted into a program language defined for the robot. It has a procedure for conversion.

A robot program generating apparatus according to a third aspect of the present invention calculates a plurality of models obtained by classifying a target work based on the characteristics thereof, calculates teaching point data based on each model, and further develops another new model. A master data in which teaching rules for calculating teaching point data based on the additional model, an attitude master for determining the attitude of the processing tool, and conversion data for coordinate conversion are registered in advance, and the master data One model is selected from the plurality of models registered in, a feature value is input to the selected model, a teaching rule is input based on another new additional model, and a rearrangement order is specified. Input means for inputting the characteristic amount and the tee of the master data. Teaching point calculating means for calculating a teaching point based on a teaching rule and calculating a teaching point based on the feature amount and a teaching rule corresponding to the additional model registered in the master data; and a teaching point calculating means. Posture data calculating means for calculating the posture data at each teaching point calculated based on the posture master of the master data; teaching point data calculated by the teaching point calculating means; and posture data calculated by the posture data calculating means. Data combining means for combining the posture data;
Rearranging means for rearranging the data combined by the data combining means in accordance with an instruction input by the input means; and converting means for converting the data rearranged by the rearranging means into data based on the robot origin. It is provided with.

According to a fourth aspect of the present invention, there is provided a robot program generating apparatus, comprising: a plurality of main body models and component models obtained by classifying a target work based on its characteristics; and teaching point data based on each main body model and each component model. The master data in which the teaching rule for calculating the position, the attitude master for determining the attitude of the processing tool, and the conversion data for coordinate conversion, and the target workpiece are divided into a plurality of body models based on their characteristics. And a part model, and a main body registration file storing a table for registering the characteristic amount for the classified main body model, and a part storing a table for registering the characteristic amount for the classified part model A registration file,
A registration unit for registering the newly registered body model and part model in the master data using the body registration file and the part registration file; and a plurality of body models and part models registered in the master data. Input means for selecting a body model and a part model, inputting a feature amount to the selected body model and part model, and inputting an instruction of a rearrangement order; a feature amount input by the input means; Teaching point calculating means for calculating a teaching point based on a data teaching rule, and calculating a teaching point based on the feature quantity and a teaching rule corresponding to the additional model registered in the master data; and Calculated by point calculation means Attitude data calculating means for calculating attitude data at the teaching point based on the attitude master of the master data, and combining the teaching point data calculated by the teaching point calculating means with the attitude data calculated by the attitude data calculating means. Data combining means, sorting means for sorting the data combined by the data combining means in accordance with an instruction input by the input means, and data sorted by the sorting means with reference to the robot origin. Conversion means for converting the data into data.

According to a fifth aspect of the present invention, there is provided a robot program generating apparatus according to the third or fourth aspect, further comprising: a teaching rule registration file storing a table for registering a teaching rule; and a teaching rule registration file. Registering means for registering the newly registered teaching rule in the master data.

[0014]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows an overall configuration of a robot program generation device for executing the robot program generation method of the present invention.
The example applied to the painting robot is shown. However, it is natural that the present invention can be applied to simple generation of a driving program for other welding robots and the like.

This robot program generating apparatus is a model program which classifies a target work based on its characteristics.
a, Teaching rule 1 for calculating teaching point data based on each classified model data
b, a teaching rule 1c for calculating teaching point data based on another new additional model,
A posture master 1d for determining the posture of the processing tool,
It has master data 1 in which tool data 1e, layout data 1f, robot geometric data 1g, and robot language data 1h are registered in advance.

Then, one model is selected from a plurality of models registered in the master data 1, and the selected model is given a feature amount (length of side, angle, teaching rule based on another new additional model, etc.). ) And inputting the instruction of the sorting order, and based on the model (target work) selected by the input means 2 and the input feature amount, the origin, inclination, Origin / tilt / feature point calculation means 3 for calculating feature points
The teaching point is calculated based on the feature amount input by the input means 2 and the teaching rule 1b of the master data 1, and the teaching point is calculated based on the feature amount and the teaching rule 1c corresponding to the additional model registered in the master data. Teaching point calculating means 4 for calculating a teaching point based on the position data; and attitude data calculating means 5 for calculating attitude data at each teaching point calculated by the teaching point calculating means 4 based on the attitude master 1d of the master data 1. ,
A data combining unit 6 for combining the teaching point data calculated by the teaching point calculating unit 4 with the posture data calculated by the posture data calculating unit 5, and the data combined by the data combining unit 6 are input by the input unit 2. Rearranging means 7 for rearranging according to the input instruction, first converting means 8 for converting the data rearranged by the rearranging means 7 into data based on the robot origin, and first converting means 8 And a second conversion means 9 for converting the data based on the robot origin converted by the above into a program language defined for the robot.

The data calculated by the means 3, 4, 5, 7 and the robot program created by the second conversion means 9 are stored in respective storage areas of the storage means 10 such as a RAM. .

The model data 1a describes a plurality of model data obtained by classifying the target work based on its characteristics. The teaching rule 1b for calculating the teaching point data based on each model describes a rule for calculating the teaching point data from the feature points of the work. The teaching point data is, for example, data "moves by the spray pattern width in the X-axis direction from the reference point", and "moves by the spray pattern width in the Y-axis direction from the reference point".

A teaching rule 1c corresponding to another new additional model includes a rule for calculating teaching point data according to a formula for calculating teaching point data input from the input means 2 using the feature amount of the target work. Is described.

The posture master 1d for determining the posture of the working tool describes the angle of the robot's wrist (flange surface) with respect to the target work. The tool data 1e describes the dimensions of the tool to be used, the distance from the flange surface of the robot to the point acting on the target work, and the like.

The layout data 1f describes a three-dimensional positional relationship between the robot coordinates and the reference coordinates of the target work. That is, the distance of each axis (X, Y, Z)
The rotation angle (Rx, Ry, Rz) of each axis is described.

The robot geometric data 1g describes the robot geometric data. That is, robot data (robot arm length, robot corner joint data, etc.) necessary to move to a three-dimensional space point is described.

The robot language data 1h describes a robot command sentence. That is, the commands include a command to move the robot, a command to use a tool, a command to operate the transfer device, and a command to transmit and receive signals to and from an external device.

The input means 2 selects one model similar to the target work from a plurality of models registered in the model data 1a of the master data 1, and assigns a feature amount (length of each side or side) to the selected model. This is a block for inputting a teaching rule based on another new additional model, and inputting an instruction of a rearrangement order (a moving order of the robot) based on another new additional model. A method of registering a new teaching rule input from the input means 2 is different from a method of registering a rule in advance, a method of directly inputting and registering data, or a method of registering a combination of a registered variable and a numerical value. There are methods. This registration method does not require knowledge for system development, and is a registration method that can be executed by anyone simply by mastering the necessary point data calculation formula and spraying technique.

The origin calculation section of the origin / inclination / characteristic point calculation means 3 is a block for calculating the center point (coordinate origin) of the target work based on the input data from the input means 2. The inclination calculating unit of the origin / inclination / characteristic point calculating means 3 is a block for calculating the inclination of each surface of the target work. In the case of data of only the length of each side, the angle formed by the two sides is determined by the two sides. Obtain from length.

The feature point calculation section of the origin / inclination / feature point calculation means 3 is a block for calculating three-dimensional position data of a feature point viewed from the center point (coordinate origin) of the target work from the feature amount of the target work. is there.

The teaching point calculating means 4 is a block for calculating teaching point data based on the teaching rule 1b of the master data 1 and the three-dimensional position data of the feature points. Further, the teaching point calculating means 4 is adapted to execute the teaching rule 1 of the master data 1.
This is a block for calculating teaching point data according to c and a formula for calculating each feature point of the additional model input from the input means 2 with reference to one feature point of the target work. A unique point name is assigned to the calculated teaching point so that each point name does not overlap.

The attitude data calculating means 5 is adapted to store the master data 1
The inclination of the flange surface of the robot is calculated by adding the inclination of each surface of the target work calculated by the inclination calculating unit of the origin / inclination / characteristic point calculation means 3 to the posture data described in the posture master 1d of the robot. It is a block.

The data linking means 6 is a block for linking the teaching point data calculated by the teaching point calculating means 4 and the attitude data calculated by the attitude data calculating means 5, and the teaching is performed after the teaching point data. Combine by adding posture data.

The rearranging means 7 is a block for rearranging according to the trajectory of the robot in accordance with the instruction input by the input means 2. The first conversion unit 8 is a center point (origin) of the target work rearranged by the rearrangement unit 7.
This is a block for converting data viewed from the robot origin to data viewed from the robot origin, and is obtained by the following equation (1).

[0031]

## EQU00001 ## Data to be calculated = (data viewed from the work origin) .times. (Layout data) (1) Expression (1) is a calculation expression using a vector because a rotation element is included in the layout data. It is.

Although not shown, the robot program converted by the second conversion means 9 is output from the output means. Next, the operation of the robot program generation device having the above configuration will be described.

In the present embodiment, a new shape in which an unregistered attachment part (additional model) shown in FIG. 3 is attached to a target work shown in FIG. 4 (ie, a target work having a registered shape as shown in FIG. 2). The following describes a case in which painting is performed on a target work). Needless to say, many shapes of products (workpieces to be painted) flow on the painting line of the factory, but the shapes are often similar to some extent or can be modeled in several patterns. Therefore, in the robot program generation method and the generation apparatus of the present embodiment, as shown in FIG. 2, a model capturing the characteristics of the target work is classified and created in advance, and this is registered as a part of the master data 1. Shall be kept. The data of the attachment parts shown in FIG. Incidentally, FIG. 14 is an input screen of a formula for calculating a corner point of a target work, and FIG.
Reference numeral 5 denotes an input screen for calculating a formula for calculating a corner point of a mounting part.

As a result, when painting a target work having a shape shown in FIG. 2, for example, the on-site operator selects a model similar to the work to be painted for which a painting program is to be created, and designates a model on the screen. It is only necessary to read the data to be input as the characteristic amount from the drawing or the like and input the numerical value.

That is, the operator at the site uses the input means 2 to select a model similar to the target work from the plurality of models registered in the model data 1a of the master data 1 (FIG. 2).
Is selected, a feature amount is input to the selected model, and an instruction of a sorting order is input. That is, dimensions of H1, H2, W1, W2, D1,
And the angle between the bottom surface and each side surface. The angle between the bottom surface and each side surface does not necessarily need to be 90 °, and may be, for example, 88 ° or 89 °. Also, as the painting order (order of rearrangement), for example, front, right side, back,
Specify as left side, top side, etc.

Here, in the present embodiment, the work to be painted is a new work in which the unregistered mounting parts shown in FIG. 3 are attached to the registered work shown in FIG. 2 (hereinafter referred to as a basic work) as described above. Since the work has a shape, the calculation formula of the teaching point data for the attachment part shown in FIG. 3 is input using one feature point of the basic target work shown in FIG. 2 as an attachment reference.

For example, assuming that the corner (point 9) shown in FIG. 4 is a mounting reference, the coordinate calculation formula of the corner (point 13) is as follows.

[0038]

## EQU2 ## Coordinate of corner (point 9) Coordinate of corner (point 13) x coordinate: -D1 / 2 x coordinate: (x coordinate of corner (point 9) y coordinate: W1 / 2-W2 y coordinate: ( X coordinate of corner (point 9) + W3 z coordinate: H2 z coordinate: (z coordinate of corner (point 9))

The origin calculation unit of the origin / inclination / feature point calculation means 3 calculates the center point (coordinate origin) of the target work based on the input data from the input means 2. This corresponds to the teaching rule 1b registered in the master data 1,
For example, if a rule such as “the center of gravity of the bottom surface is the coordinate origin” is described,
The three-dimensional data of the center point (coordinate origin) of the target work is (0, 0, 0).

When the data of the target work is data of only the length of each side, the inclination calculating section of the origin / tilt / characteristic point calculation means 3 calculates the angle formed by the two sides from the length of the two sides. That is, the inclinations from the bottom surface of all the side surfaces required for preparing a later painting program are calculated.

The feature point calculation unit of the origin / inclination / feature point calculation means 3 calculates three-dimensional position data of a feature point viewed from the center point (coordinate origin) of the target work from the feature amount of the target work. That is, based on the coordinate origin, the coordinates of each corner are calculated based on the previously input dimensions of the target work, the calculation formula of the teaching point data for the attached parts, and the like. In the example shown in FIG. 4, points 1 to 20
These are the three-dimensional coordinates up to. In the following processing, the coordinates of each corner (points 13 to 20) calculated on the basis of the calculation formula of the teaching point data for the attachment part are the same as those of the respective corners (points 1 to 12) of the reference target work. It shall be treated as the same as coordinates.

The teaching point calculating means 4 calculates the teaching points required for the painting work according to the teaching rule 1b of the master data 1, and further calculates their coordinate data. Teaching rule 1b
Varies completely depending on the performance of the painting robot, the pattern of spraying paint from the spray nozzle, and the like. For example, as shown in FIGS. 5A and 5B, the spray pattern has a narrow width, and the height direction is 90 mm. Assuming the case of, "Once painted while driving horizontally, the next is the pattern width 2
It moves upward by 45 mm, which is one-half, and moves in the horizontal direction and sprays while overlapping half. Is described. Then, from point 4 shown in FIG.
Next to the horizontal moving painting up to the point, it moves to a point (supposed to be a point 1B) raised by 45 mm from the point 1 (in the direction of the point 5) (supposed to be a point 1B), and while performing the horizontal moving painting again, 45 mm from the point 4
Only the point (in the direction of the point 12) is returned to the point (supposedly the point 4B). By such a procedure, teaching points (that is, points 1B, 4B, etc.) other than the actual corners (points 1 to 20) are calculated, and their coordinates (three-dimensional data based on the origin) are calculated. You.

The posture data calculating means 5 is provided with a posture master 1d
The inclination of the flange surface of the robot is calculated by adding the inclination of each surface of the target work calculated by the inclination calculation unit to the posture data described in (1). That is, as the teaching data of the painting robot, in addition to the above teaching point data, another attitude data of the spray nozzle (painting gun) at the tip of the robot arm is required.
This attitude data can be generally defined as rotation angle data (Rx, Ry, Rz) about three axes of XYZ. This definition is described in the posture master 1d. For example, a rule such as "the nozzle tip always keeps a right angle to the spray surface during spraying on the side surface (spray at a right angle)" is described. Then, the posture data of the nozzle tip at the time of painting on the surface is calculated with reference to the previously calculated inclination from the side surface.

The data combining means 6 combines the calculated teaching point data with the posture data. In other words, the teaching position data is added after the teaching point data.

However, this data is a set of the coordinate data and the posture data of each painting point, and is merely data as a "point". Therefore, it is necessary to teach painting order data connecting these data as "points". When calculating the teaching point, although the rough painting order is shown, it is only a tentative rule for calculating drive data in a general purpose, and does not include the know-how of painting for each target work.

Therefore, at this stage, the on-site operator rearranges the above set of point data in a desired painting order, thereby completing the driving data in consideration of the painting know-how. That is, the rearranging unit 7 performs rearrangement according to the trajectory of the robot in accordance with the instruction input from the input unit 2 by the site operator. In the order in which the robot moves, the names of all or some of the teaching point data and the order in which the point data are rearranged are manually input.

By the above operation, teaching data required by the painting robot is completed. Since these data are data of the teaching point based on the coordinate origin of the workpiece to be painted, the painting robot is actually used. That is, it is necessary to convert the data into data based on the tip of the coating gun.

Therefore, the first converting means 8 sets the center point (origin) of the target work rearranged by the rearranging means 7.
Convert the data as viewed from the robot origin to the data as viewed from the robot origin. This work is merely a matter of coordinate conversion, and includes the shape and dimensions of the paint gun (tool data 1e), the relative position between the coordinate origin of the target work and the robot reference origin (layout data 1f), and the robot shape data (robot data). It can be easily converted from the geometric data 1g).

After such operations, the second conversion means 9 converts the obtained data into a format and converts the obtained data into a robot program in a programming language corresponding to the robot necessary for actually driving the painting robot. Get.

In this way, the on-site operator can easily complete a complicated driving program of the painting robot simply by instructing a very small part of the data.
Next, the above operation description will be applied to a more specific target work.

The target work is a FRP washing place shown in FIG. 4, and the target work is resin spraying on the target work. Here, an example of the robot program is shown in FIG.
In FIG. 7, "step" indicates a line number, and "instruction"
Indicates robot operation, tools, etc., “x coordinate” indicates robot position x coordinate, “y coordinate” indicates robot position y coordinate, “z coordinate” indicates position data of robot position z coordinate, and “Rx” indicates robot position. The robot posture x coordinate, “Ry” indicates the robot posture y coordinate, and “Rz” indicates each posture data of the robot posture z coordinate.

The teaching rule 1b for calculating the teaching point data includes a point (X, Y, Z) that is raised from the four corners of the bottom surface of the target work in three directions (X, Y, Z directions) by the pattern width. Then, the point to be obtained is represented by the following equation (the directions of X, Y, and Z are shown in FIG. 2).

[0053]

[Expression 3] Move in X direction: (X + pattern width, Y, Z) Move in Y direction: (X, Y + pattern width, Z) Move in Z direction: (X, Y, Z + pattern width) 1d designates the angle (posture) of the robot (tool) with respect to the target work. FIG. 8 shows an example of the attitude master data.

The tool data 1e is shown in FIG.
As shown in (b), the pattern width is 100 mm, and the tool dimensions are D × W × H = 100 × 50 × 50. Also,
The layout data 1f is shown as three-dimensional data from the center of the robot to the center of the target work, as shown in FIG. In FIG. 9, Rx, Ry, and Rz are X, Y,
The Z-axis rotation angle (deg), X, Y, and Z are lengths (mm).

The robot geometric data 1g is data such as arm length and angle of the robot. The robot language data 1h describes the grammar of the robot language, how to assign line numbers (step numbers), and the like.

First, the operator at the site inputs the characteristic amount of the basic target work (the target work shown in FIG. 2) and the characteristic amount of the mounting part (the additional model shown in FIG. 3) by using the input means 2. The data to be input is the length of each side as described above,
The angle between the sides. Here, the feature amount of the model of the basic target work is set as follows (see FIG. 2). "D1 = 1500 mm, W1 = 2000 mm, W
2 = 1200 mm, H1 = 400 mm, H2 = 100 m
m, the angle formed by each side = 90 ° ”Further, the characteristic amount of the attachment part is set as follows (FIG. 3).
reference). "D1 = 50 mm, w1 = 100 mm, h1 =
With this setting, the coordinates of each point as viewed from the origin (point 13) of the mounting part are:

[0057]

(Equation 4) Becomes

Further, the operator at the site uses the point 9 of the basic target work as an attachment reference, and inputs an amount of deviation from the point 9 of the origin (point 14) of the attached part. The shift amount is “x = 0
mm, y = 100 mm, z = 0 mm ". That is,
W3 = 100 mm.

The operator at the site inputs the moving order by the input means 2. That is, the point name of the teaching point data and the magnitude of each coordinate are entered as the moving order of the robot. Point names dictate that points do not overlap. For example, as shown in FIG. XY 5] indicates that the lower side corner 4 of FIG. 4 is designated five times in the Z-axis direction.

The following equation

[0061]

(Equation 5) Indicates that the point data is to be rearranged so that the point data X coordinates with the letter “Z” added to the point name are in descending order. Also, "1 XY FIG. 4
Indicates that the point is designated as point 1.

The origin calculating section of the origin / inclination / characteristic point calculating means 3 calculates the center point (coordinate origin) of the basic target work based on the input data from the input means 2. The center point is the center of the bottom surface of the basic target work, and this is the center point (0,
0,0). The reference coordinate axes are shown in FIG.

With this setting, the coordinates of each point viewed from the center point (coordinate origin) of the basic target work are:

[0064]

(Equation 6) Becomes

The inclination calculating section of the origin / inclination / characteristic point calculation means 3 calculates the inclination of each surface. For example, hypotenuse a and base b
And the angle θ

[0066]

[Mathematical formula-see original document] θ = cos ^-1 (a / b) However, when the inclination of each side is input, such calculation is unnecessary.

The feature point calculation unit of the origin / inclination / feature point calculation means 3 calculates the three-dimensional position data of the feature point viewed from the center point (coordinate origin) of the basic target work from the feature amount of the basic target work by the following equation. calculate.

[0068]

(Equation 8)

The feature point calculation unit of the origin / inclination / feature point calculation means 3 uses the three-dimensional position data of the feature point 9 calculated by the above equation as the attachment reference position, and calculates the coordinates of the point 14 according to the above [Equation 2]. Based on the calculation formula and the respective coordinates from the origin (point 14) of the mounting part shown in [Equation 4] above, the characteristic points (points 13 to points) of the mounting part viewed from the center point (coordinate origin) of the basic target work 20) The three-dimensional position data is calculated.

The teaching point calculation means 4 calculates the pattern width from the characteristic points of the basic target work calculated by the above equation [Equation 8] and the characteristic points of the attachment parts calculated by the above [Equation 2] and [Equation 4]. Find the point moved only in the required direction. However, the point to be calculated is calculated within the range of the surface of the target work (see the example of the basic target work shown in FIGS. 6A and 6B). “XY” and the like described in the middle of the point name indicate the moving direction. That is, “XY” is a point moved in the Z direction, and “XZ” is a point moved in the Y direction.
The point moved in the direction, “YZ” indicates the point moved in the X direction. FIG. 11 shows the value of point 4 shown in FIG.

As described above, necessary points are obtained from the feature points according to the rules. Further, the names of the points are set as unique names as described above, and are added as subscripts. This point name is used for sorting the teaching posture data and the point data.

The attitude data calculating means 5 is provided with the attitude master 1d
The inclination of the flange surface of the robot is calculated by adding the inclination of each surface of the target work calculated by the inclination calculation unit to the posture data described in (1). That is, the same posture is selected by the alphabet and the posture master data key described in the point name, and the teaching posture data is calculated (multiplied by the vector) by the following equation using the inclination of the target work and the posture master 1d.

[0073]

## EQU9 ## Inclination of target work + posture master = teaching posture data As shown in FIG. 12, the data combining means 6 combines the calculated teaching point data and teaching posture data to create teaching data.

The rearranging means 7 rearranges the teaching data according to the moving order inputted from the input means 2. As shown in FIG. 13, the order of rearrangement is a part or all of the point names, the magnitude of each axis of each teaching point data, and the like.

All the teaching point data is three-dimensional data as viewed from the origin of the target work. For this reason, this data is converted by the first conversion means 8 into three-dimensional coordinates as viewed from the robot origin, and then, if necessary, into data as viewed from the flange surface of the robot and data as viewed from the tool tip.

The coordinates of each feature point as viewed from the robot origin are shown below.

[0077]

Xyz 1: 1250-1000-5002: 2750-1000-5003: 27501000-5004: 12501000-5005: 1250-1000-4006: 2750-1000-4007: 1250 -200 -400 8: 2750 -200 -400 9: 1250 -200-100 10: 2750 -200-100 11: 2750 1000-100 12: 1250 1000-100 13: 1250-100-100 14: 1300-100- 100 15: 13000-100 16: 1250-100 17: 1250-100-50 18: 1300-100-50 19: 1300 0-50 20: 1250 0-50 When coming riding may perform conversion in consideration of the conveying speed of the conveying device.
This conversion is a rotation translation conversion, and converts the teaching point data into data viewed from the center point (see Equation 9).

[0078]

[Equation 11]

After such work, the second conversion means 9 converts the obtained data into a format (converted into a format recognizable by the robot), adds a step (line number), and sets the Obtain the necessary robot program to actually drive.

Thereafter, the robot program is output in a format that can be read by the painting robot. This output is
Use a conversion tool provided by each manufacturer.
FIG. 16 shows another embodiment of the robot program generation device of the present invention.

That is, the robot program generation device of the present embodiment classifies the target work into a plurality of main body models and component models based on the features in the circuit configuration shown in FIG. A main body registration file 11a storing a table for registering the feature amount, and a component registration file 11 storing a table for registering the feature amount for the classified component model
b, a file storage unit 11 for storing a teaching rule registration file 11c storing a table for registering teaching rules, and a main body registration file 11a
The newly registered body model and part model are registered in the model data 1a of the master data 1 using the part registration file 11b, and the newly registered teaching rule is registered in the master data 1 using the teaching rule registration file 11c. Teaching rules 1b, 1
c and a registering means 12 for registering the information in c. The other block configuration is the same as that shown in FIG. 1, and thus the same reference numerals are given to the same blocks, and detailed description is omitted.

However, in this embodiment, since the model is classified into the main body model and the part model, the model data 1a
, The data of the body model and the part model are classified and registered. Also, teaching rule 1b
The teaching rule of the main body model is registered in, and the teaching rule of the part model is registered in the teaching rule 1c. That is, the component model in the present embodiment is the same as that described as the attachment component in the above embodiment.

That is, the robot program generation device of the present embodiment makes it possible to easily register a work at the time of generating a robot program, thereby enabling creation of a robot program intended by a robot program creator.

FIG. 17 shows the shape of a target work used in the description of this embodiment. In the target work shown in FIG. 17, ID1 = 0, IW1 = W3 (= 100 m
m) is the target work shown in FIG. 4 described in the above embodiment. However, the target work shown in FIG. 17 and the target work shown in FIG. 4 are different in the way of giving the names (numbers) of the corner points.

The body registration file 11a is a file for registering the feature amount of the body model, the added part model, and the mounting position of this part model. FIG. 18 shows a state where the main body registration file 11a is called on the screen. Using this screen, a feature amount of the main body model is input (a shape table is created). That is, the dimensions H1, H2, W1, W2, and D1 are input as the feature quantities of the main body model. In addition, although not shown, the angles formed by the respective side surfaces, the four corners of the component name, the corner name of the mounting reference position of the main body model, and the position data of ID1, IW1, and IH1, which are the component mounting positions, are registered. FIG. 19 shows data examples of four types of body models having a common basic shape registered using the body registration file 11a shown in FIG.

The part registration file 11b is a file for registering the feature amount of the part model. FIG. 20 shows a state where the component registration file 11b is called on the screen. Using this screen, the feature amount of the part model and the like are input (a shape table is created). That is, as the part model, the four corners of the part name and the feature amounts BD1, BW1, B
Register H1. This registration is performed by inputting the number of each feature amount from the shape drawing of the target model. In addition, by inputting the mounting direction (rotation angle) of the component model, it is not necessary to re-register a component model having the same shape only with a difference in direction, as in the case of two component models shown in FIG. FIG. 21 shows the component registration file 1 shown in FIG.
1B shows an example of data of three types of part models registered using 1b and having a common basic shape.

That is, the robot program generation device of the present embodiment calls the tables stored in the main body registration file 11a and the parts registration file 11b,
It is characterized in that a new target work consisting of a body model and a part model can be registered only by inputting necessary data into the table.

The data of the body model and the part model registered using the body registration file 11a and the part registration file 11b as described above are stored in the model data 1a of the master data 1.

The origin / inclination / feature point calculation unit 3 calculates the origin, the inclination, and the feature point based on these data. However, since these calculation methods are as described in the above embodiment, Here, the description is omitted. Further, the subsequent processing from the teaching point calculation means 4 to the second conversion means 9 is the same as that described in the above-described embodiment, and therefore, the description is omitted here.

On the other hand, in the teaching rule 1b, for example, assuming that the spray pattern of the spray nozzle has a wide width and the height direction is 240 mm in FIGS. After painting while moving, move up by 120 mm, which is half the pattern width,
And it moves in the horizontal direction and sprays while overlapping half. Is described. Then, next to the horizontal moving paint from point 7 to point 12 shown in FIG.
Move to a point that is 120 mm above point 7 (in the direction of point 11), and perform horizontal movement painting again,
It returns to the point raised 20 mm upward (direction of point 8). In such a procedure, teaching points other than the actual corners (points 1 to 20) are calculated, and their coordinates (three-dimensional data with reference to the origin) are calculated.

A table for registering such teaching rules is stored in the teaching rule registration file 11c in the form of a list shown in FIG. In FIG. 24, the number of the corner which is the starting point of each work is written in the vertical column of each axis of X, Y, Z, and beside the corner number,
A numerical field for writing the moving distance is provided. That is, the numeral "45" is entered in the numerical value column indicated by the reference numeral 51, and this numeral is shifted from the point raised by 45 mm from the point 7 to the point 12 as shown in FIG. This indicates horizontal movement. The numeral "165" is written in the column indicated by the reference numeral 52, and this numeral is changed from the point 12 at a point 120 mm further up from that point (that is, a point 165 mm up from the point 7) to the point 7
This indicates horizontal movement to the side. Eventually, this FIG.
Are rules for moving the spray nozzles as indicated by arrows in FIG. Such rules are registered by the teaching rule registration file 11c and stored in the teaching rules 1b and 1c.

In the present embodiment, when it is desired to change the moving distance of the spray nozzle in such a rule,
The teaching rules stored in the teaching rules 1b and 1c are called up on the screen by the teaching rule registration file 11c, and the teaching in the corresponding numeric value column in the teaching rule displayed on the screen is directly changed by directly changing the numeric value. Rules can be changed.

For example, taking the case where the spray nozzle passes through the corner of point 8, as shown in FIG. 27, the spray nozzle moves in an arc at the corner of point 8. Therefore, the corner portion of the point 8 may be insufficiently coated.
Therefore, in this case, the teaching rule of the portion passing near point 8 is read out on the screen, and the numerical value in the numerical value column of the moving distance of the rule passing point 8 is changed (for example, the moving distance in the Z direction is 45 mm. If this is the case,
0 mm, which is changed so as to be closer to the point 8). As a result, the movement trajectory of the spray nozzle is
Move upward as shown by the two-dot chain line and approach point 8.

FIG. 25 shows another movement pattern different from the movement pattern of the spray nozzle shown in FIG. 23. FIG. 26 shows a rule for moving the spray nozzle according to the movement pattern shown in FIG. I have.

Note that such a body registration file 11a
Process after registering or changing a new body model or component model by using the teaching rule registration file 11c, or by registering or changing the teaching rule by using the teaching rule registration file 11c (ie, from the teaching point calculation means 4 to the second storage means). 9 are the same as in the above-described embodiment, and a description thereof will not be repeated.

[0096]

According to the robot program generating method and the generating apparatus according to the first to third aspects of the present invention, teaching can be performed only with the drawing, so that the actual target work is unnecessary, and the apparatus itself is a program. The generation time of the program can be shortened, and the generation of the program can be performed off-line, so that the teaching period can be shortened. Therefore, the number of teaching steps can be significantly reduced. Further, since it is only necessary to input a feature amount from a work drawing, teaching technology is not required, and teaching accuracy can be made uniform. In addition, since the teaching point generation rule is the same, it can be applied to different works. Further, since there is no need to create an expression for obtaining all teaching point data (an expression for obtaining teaching point data using the feature amount of the target work), the program generation time can be further reduced. Further, even when the shape of the target work becomes new, it is only necessary to input a teaching point data calculation formula for only the additional model (additional part) using the feature amount of the target work. In other words, the rules of the target work can be easily added without changing the system.

In the robot program generating apparatus according to the fourth aspect of the present invention, in addition to the above-described effects, the main body model and the part model are classified and registered. Also, since only the changed portion needs to be registered, the registration of the target work having the new shape can be performed in a shorter time.

Further, in the robot program generating apparatus according to the fifth aspect of the present invention, in addition to the above-described effects, since it is easy to register and change the teaching rule, it is possible to give more detailed point designations, so that the actual The teaching rule can be appropriately changed (adjusted) according to the coating.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram showing an embodiment of a robot program generation device for executing a robot program generation method of the present invention.

FIG. 2 is a diagram illustrating an example of a basic target work (wire model) constituting a target work (wire model) for generating a robot program.

FIG. 3 is a diagram illustrating an example of a mounting part (wire model) constituting a target work (wire model) for generating a robot program.

FIG. 4 is a diagram showing an example of a target work (wire model) for generating a robot program.

FIG. 5 is a view showing an example of a paint spraying pattern from a spray nozzle of a painting robot, wherein (a) is a front view,
(B) is a side view.

6A and 6B are diagrams for explaining a procedure for creating a teaching point, wherein FIG. 6A is a diagram for explaining calculation of a feature point, and FIG. 6B is a diagram for explaining calculation of a teaching point. .

FIG. 7 is a diagram illustrating an example of a robot program.

FIG. 8 is a table showing an example of attitude master data.

FIG. 9 is a chart showing an example of layout data.

FIG. 10 is a diagram for specifically explaining point names.

FIG. 11 is a view showing a list of values of a point 7 shown in FIG. 2;

FIG. 12 is a diagram for explaining a procedure for creating teaching data by combining teaching point data and teaching posture data;

FIG. 13 is a diagram for explaining a procedure for rearranging teaching data according to an input movement order.

FIG. 14 is an input screen of a formula for calculating a corner point of a target work.

FIG. 15 is an input screen for a formula for calculating a corner point of an attachment part.

FIG. 16 is an overall configuration diagram showing another embodiment of the robot program generation device of the present invention.

FIG. 17 is a diagram illustrating an example of the shape of a target work (wire model) used in the description of the other embodiment illustrated in FIG. 16;

FIG. 18 is a diagram showing a state where a main body registration file is called on a screen.

FIG. 19 is a diagram showing data examples of four types of body models having a common basic shape and registered using the body registration file shown in FIG. 18;

FIG. 20 is a diagram showing a state where a component registration file is called on the screen.

21 is a diagram showing an example of data of three types of component models having a common basic shape registered using the component registration file shown in FIG. 20;

FIG. 22 is a diagram showing two component models.

FIG. 23 is a diagram illustrating one movement pattern of a spray nozzle.

FIG. 24 is a diagram showing a state in which a table for registering teaching rules is called on the screen.

FIG. 25 is a diagram illustrating another movement pattern of the spray nozzle.

FIG. 26 is a diagram showing a state where a table for registering a teaching rule is called on the screen.

FIG. 27 is a diagram showing a path through which a spray nozzle passes through a corner of a point 8;

[Explanation of symbols]

 Reference Signs List 1 master data 1a model data 1b teaching rule 1c teaching rule 1d attitude master 1e tool data 1f layout data 1g robot geometric data 1h robot language data 2 input means 3 origin / inclination / characteristic point calculating means 4 teaching point calculating means 5 attitude data calculation Means 6 Data combining means 7 Rearranging means 8 First converting means 9 Second converting means 11 File storage unit 11a Main body registration file 11b Parts registration file 11c Teaching rule registration file 12 Registration means

Claims (5)

[Claims] 1. A plurality of models for classifying a target work on the basis of its characteristics and teaching point data based on each model, and calculating teaching point data based on another new additional model. A teaching rule, a posture master for determining the posture of the machining tool, and master data in which conversion data for coordinate conversion are registered, and one of the plurality of models registered in the master data is included. A step of selecting a model, calculating a teaching point based on the feature amount input to the selected model and the teaching rule of the master data, and corresponding to the feature amount and the additional model registered in the master data. A procedure for calculating a teaching point based on a teaching rule; A procedure for calculating the attitude data at each teaching point based on the attitude master of the master data, a procedure for combining the calculated teaching point data and the attitude data, and arranging the data obtained by the integration according to the input instruction. A method for generating a robot program, comprising: a step of replacing the data; and a step of converting the rearranged data into data based on the robot origin. 2. The method according to claim 1, further comprising a step of converting the converted data based on the robot origin into a program language defined for the robot. 3. Teaching for calculating teaching point data based on a plurality of models obtained by classifying the target work based on its characteristics and each model, and calculating teaching point data based on another new additional model. A rule, an attitude master for determining the attitude of the processing tool, and master data in which conversion data for coordinate conversion is registered, and one model is selected from the plurality of models registered in the master data. Input means for inputting a feature amount to the selected model, inputting a teaching rule based on another new additional model, and inputting an instruction of a rearrangement order; and a feature input by the input means. The teaching point is calculated based on the amount and the teaching rule of the master data. Teaching point calculation means for calculating a teaching point based on the feature amount and a teaching rule corresponding to the additional model registered in the master data; and posture data at each teaching point calculated by the teaching point calculation means. Attitude data calculation means for calculating based on the attitude master of the master data, Data combining means for combining the teaching point data calculated by the teaching point calculation means and the attitude data calculated by the attitude data calculation means, Rearranging means for rearranging the data combined by the data combining means in accordance with an instruction input by the input means; and converting the data rearranged by the rearranging means into data based on the robot origin. Robot program generating apparatus characterized by comprising a switch means. 4. A plurality of body models and component models obtained by classifying a target work based on its characteristics, teaching rules for calculating teaching point data based on each body model and each part model, and a posture of a machining tool. A master data in which a posture master for determination and conversion data for coordinate conversion are registered, and a target work are classified into a plurality of body models and component models based on their characteristics, and the classified body models are classified. Using a main body registration file storing a table for registering the characteristic amounts of the parts and a part registration file storing a table for registering the characteristic amounts for the classified part models; To register newly registered body model and part model in the master data And selecting a main body model and a part model from the plurality of main body models and part models registered in the master data, inputting a feature amount to the selected main body model and part model, and sorting Input means for inputting the instruction of the above, teaching point calculating means for calculating a teaching point based on the feature amount input by the input means and the teaching rule of the master data, and each of the teaching points calculated by the teaching point calculating means. Attitude data calculating means for calculating attitude data at the teaching point based on the attitude master of the master data, and combining the teaching point data calculated by the teaching point calculating means with the attitude data calculated by the attitude data calculating means. Do Data combining means; sorting means for sorting the data combined by the data combining means in accordance with the instruction input by the input means; and data sorted by the sorting means based on the robot origin. A robot program generation device, comprising: a conversion unit for converting data into data. 5. A teaching rule registration file storing a table for registering teaching rules, and registration means for registering a newly registered teaching rule in the master data using the teaching rule registration file. Item 5. The robot program generation device according to item 3 or 4.