JP2008009899A - Automatic teaching system and method for assembly work robot - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an automatic teaching system for automatically teaching an assembly operation to an assembly work robot. <P>SOLUTION: The teaching system for teaching an assembly procedure of a plurality of works to the assembly work robot comprises a storage means 1, a determination means 2 and a conversion means 3. The storage means 3 stores work information as unit information for each work (including object information within the work) D1, mutual arrangement information D2 for assembly work sets each of which is a combination of a plurality of works and assembly supplementary information D3. The determination means 2 determines an assembly work content for an assembly work set which is a group of a plurality of works composed of a predetermined kind of combination, among the assembly work sets, in which the respective works are arranged in a predetermined state in a work area, based on the work information, the mutual arrangement information and the assembly supplementary information. The conversion means 3 converts the determined assembly work content D4 to the assembly work procedure D5 for the assembly work robot. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a system and method for teaching a robot an assembly procedure for a plurality of workpieces.

  Conventionally, an assembly robot has been used as an apparatus for automatically assembling a plurality of workpieces (parts) into one. Such assembly robots include, for example, multi-joint robots having a plurality of joints that enable various movements depending on the application, and such assembly robots teach the assembly operation in advance. It is used by.

  When assembling multiple workpieces with this assembly robot, the operator picks up and assembles workpieces from the workpiece area for each workpiece placed in the work place (work area) used for assembly in advance. The procedure is performed by teaching the robot the assembly flow at the place (working area) where the work is performed. To give specific examples, work acquisition operation to acquire the workpieces in the assembly order, placement operation according to the mutual arrangement relationship with other workpieces when placing the acquired workpiece in the work area, The robot is instructed to specify the tool to be used and the end effector when connecting the workpieces and the operation when using the tool.

  This teaching work is generally time-consuming, and it is necessary to temporarily stop the production line during the teaching work when the teaching work is performed by direct teaching that performs the teaching work on site. For this reason, off-line teaching has emerged in which the environment of an actual production line is created on a computer and operations similar to actual teaching work can be simulated on the computer. The appearance of this offline teaching has made it possible to perform teaching work without stopping the production line.

Among the techniques using this offline teaching technique, there is a technique that further simplifies the actual work. An off-line teaching device that completes teaching work by tracing the contour of a work on a monitor screen. This offline teaching device is a device that performs a predetermined operation by tracing the outline of a workpiece to be processed displayed on a monitor screen with a mouse and inputting the traced position and order through the mouse. The machining operation according to the input position and order is enabled (Patent Document 1).
Japanese Patent Laid-Open No. 10-58363

As described above, offline teaching has appeared as one of techniques for performing teaching work.
However, normal off-line teaching is simply a real environment built on a computer, so the operating speed increases when the robot is moved, etc., but the teaching work procedure is performed in the same way as direct teaching. Therefore, a lot of human hands are involved in the teaching work.

  In addition, the apparatus disclosed in the above-mentioned patent document specifies the position of the workpiece to be chamfered and the order of the chamfering with a mouse operation on the monitor screen, and performs a predetermined simple operation in accordance with the designation. It is an automated device that teaches a dedicated processing device. For this reason, an apparatus teaching operation that is used more generally and requires complex operations, for example, an assembly operation in which a plurality of operations are arbitrarily combined in accordance with the type of the workpiece and the workpiece must be assembled. It is unsuitable for teaching work such as robots.

  Accordingly, the present invention provides an automatic teaching system that automatically teaches assembly work corresponding to a combination of a plurality of workpieces for assembly to a robot for assembly work that operates by arbitrarily combining a plurality of operations, and a teaching method therefor. With the goal.

In order to solve the above problems, the present invention is configured as follows.
One aspect of the automatic teaching system of the present invention is that workpiece information (including object information within a workpiece), a plurality of pieces of workpiece information, on the premise that an assembly operation procedure for a plurality of workpieces is taught to an assembly robot. A storage means for storing mutual arrangement information of assembly work sets, which is a combination of workpieces, and assembly supplementary information; and a set of a plurality of works consisting of combinations of predetermined types of the assembly work sets. Determination means for determining the assembly work content of the assembly work set arranged in a predetermined state in the work area based on the work information, the mutual arrangement information, and the assembly supplement information, and the assembly work content as the assembly work Converting means for converting into a robot assembly work procedure.

  In addition, on the premise of the above configuration, there is further provided a first recognition means for recognizing the type of each work placed in the work area, and the determination means covers all assembly work sets having the mutual arrangement information. Assembling work contents of an assembly work set composed of workpieces of a type recognized by the first recognition means are determined based on the workpiece information, the mutual arrangement information, and the assembly supplement information. Is preferred.

  Further, in this configuration, the apparatus further includes second recognition means for recognizing the state (position and posture) of each work placed in the work area, and the conversion means is the assembly robot from the work area. The operation procedure of the workpiece acquisition by the second recognition means is obtained on the basis of the state of each workpiece recognized by the second recognition means, and the assembly work content is converted into the assembly work procedure of the assembly robot. More preferably.

  In the above configuration, the hand unit of the assembly work robot further includes an imaging unit, and the first recognition unit is placed in the work area based on the image information obtained by the imaging unit. Recognizing a plurality of types of workpieces, the second recognition means determines the status of each workpiece on the workpiece area based on the image information obtained by the imaging means and the status information of the hand portion of the assembly robot. It is desirable to be configured to recognize (position and orientation).

  Further, in each of the above-described configurations, the work information and the mutual arrangement information of the assembly work set are extracted from CAD data and given to the storage means, and the assembly supplement information is input to the storage means. And means.

  Further, in each of the above-described configurations, the assembly supplementary information includes coupling means selection supplemental information in which information indicating a combination of works and information on means to be used for coupling between works by the combination of the works are associated with each other. Handling means selection supplementary information that associates information with information on the means for handling the workpiece, and the determination means includes at least the combination means selection supplementary information based on the combination of the recognized types of workpieces. It is preferable that the inter-work connecting means is selected from among them, and the tool to be used and the end effector are selected from the handling means selection supplementary information based on the selected inter-work connecting means.

  In each of the above configurations, the assembly work content includes the assembly order of the workpieces included in the assembly workpiece set, the assembly direction of the workpieces, the coupling means between the workpieces, and the assembly workpiece set. The tool used when assembling and the end effector are included, and the assembly work procedure includes the replacement of the tool used and the end effector, and the work area of each work including all the works of the assembly work set. Preferably, a series of assembly operation procedures by the assembly robot between the work areas is included.

  One aspect of the teaching method of the present invention is based on the premise that an assembly operation procedure for a plurality of workpieces is taught to an assembly robot, workpiece information that is single piece information of a workpiece, and an assembly workpiece set that is a combination of a plurality of workpieces. The mutual arrangement information and assembly supplementary information are stored in a memory, and a set of a plurality of workpieces of a combination of a predetermined type among the above-mentioned assembly work sets, and each workpiece is arranged in a predetermined state in the work area. The assembly work content of the assembly work set is determined based on the work information, the mutual arrangement information, and the assembly supplement information, and the assembly work content is converted into an assembly work procedure of the assembly robot. .

  In the present invention, assembly supplementary information is configured and a series of assembly work contents required for assembly can be automatically determined. The determined assembly work content is converted into an assembly work procedure of the robot for assembly work by arbitrarily combining the partial operations that the robot can perform. Therefore, it is possible to automatically teach the assembly operation according to the assembly work set to the robot.

  According to the present invention, it is possible to automatically teach a robot an assembly operation according to a combination of workpieces for assembly. For this reason, the teaching work by the operator is greatly shortened.

  Furthermore, when the configuration is such that the type and state of the work placed in the work area can be recognized, even if the work is randomly placed in the work area, the work is automatically identified and The assembly operation can be automatically taught to the robot. For this reason, the teaching work of assembly work sets consisting of arbitrary combinations of workpieces can be fully automated, and when products of various types and small lots are frequently replaced and assembled, or products that have entered in the course of production are replaced during production. Thus, when assembling, the assembly operation can be changed smoothly, and the production efficiency can be increased.

Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings.
The automatic teaching system according to the present invention targets an assembly work robot. An example of a typical assembly robot is an articulated robot having a plurality of joints. In a series of operations by the articulated robot, the workpieces are picked up from a workpiece region where a plurality of workpieces (parts) are placed, and they are assembled in a working region different from the workpiece region.

  In the following, an automatic teaching system for automatically teaching an assembly work procedure of an assembly work set in which workpieces such as main parts and connection parts (for example, screws) constituting a product are arbitrarily gathered together will be described. .

FIG. 1 shows a main part of an automatic teaching system according to the present invention.
The storage means 1 shown in the figure includes work information D1 including general information such as shape information and material information of each work piece, configuration information (object information) unique to the work piece such as a screw hole, and the like. For example, mutual arrangement information D2 indicating the order of the workpieces to be assembled and the mutual assembly positions of the workpieces, and the means for connecting the workpieces (for example, screwing or adhesive fixing) Etc.) and assembly supplementary information D3 including information necessary for determining the tool to be used and the like are stored.

  The determination means 2 in FIG. 1 is a set of a plurality of workpieces composed of a combination of workpieces of a predetermined type or a combination of workpieces specified by external input in the assembly workpiece set, and each workpiece is The assembly work contents of the assembly work sets arranged in a predetermined state in the work area (for example, in a state where they are respectively placed upright and placed in a predetermined position) are determined. The workpiece information D1, the mutual arrangement information D2, and the assembly supplementary information D3 include a series of information necessary for sequentially assembling each workpiece included in the assembly workpiece set (for example, each workpiece). Information for determining logically, in what order and in what position, and in what type of connecting means and using which tool. For this reason, in the determining means 2, for example, based on the workpiece information D1, the mutual arrangement information D2, and the assembly supplementary information D3, the assembly order of the workpieces included in the assembly workpiece set, the assembly of the workpieces The assembly work content determination information D4 can be created by logically constructing a series of assembly work contents such as direction, connecting means between the workpieces, and end effector (and tool used) when assembling the assembly work set. can do.

  The converting means 3 in FIG. 1 converts the determined assembly work content information D4 into assembly work procedure information D5 that allows the assembly work robot to operate smoothly. Here, the partial operation commands that are possible in the assembly robot are arranged in an appropriate order, and the assembly work content is converted into series information of the commands. If the end effector needs to be replaced or a tool is required, the command is added to an appropriate position in the series information.

  Each of the means 1, 2, 3 includes a memory such as a ROM (Read Only Memory) or a RAM (Random Access Memory) in which a storage area for each of the information D1, D2, D3 is allocated to the storage means 1, a determination means 2 and The conversion means 3 can be configured by connecting an ASIC incorporating an algorithm for realizing the above-described function or a general-purpose computer configured to execute a program corresponding to the ASIC to the memory. Then, for example, the teaching process is started by inputting a start signal of the process from an input means (not shown), and the assembly work procedure information (command program) D5 executed by the assembly work robot is obtained from the conversion means 3. Output. If this command program is registered in the control unit of the assembly work robot via communication or a portable recording medium, teaching of the assembly operation of the assembly work set to the assembly work robot is automatically completed. be able to.

  As described above, the automatic teaching system can automatically determine all assembly work contents necessary for assembly and can automatically convert the assembly work contents into the assembly work procedure of the assembly robot. Thereby, it is possible to automatically teach the assembly operation according to the assembly work set to the robot.

As described above, the automatic teaching system can automatically teach the robot the assembling operation corresponding to the type of the workpiece for assembling, so that the teaching work by the operator is greatly shortened.
In the above description, the type of work included in the assembly work set is determined in advance (or externally input from a keyboard or the like, for example, to specify the type of work). However, it may be configured to automatically recognize the type of each workpiece of the assembly work set placed in the work area by configuring a recognition means. In this case, a plurality of works are placed in the work area. Even when placed randomly in any combination, these workpieces can be automatically identified and their assembly operations can be taught to the robot fully automatically. With this configuration, it is possible to change the assembly operation smoothly, such as when frequently changing and assembling products of a large variety of small lots, or when replacing and assembling products that have entered during production. Thus, the production efficiency can be increased.
(Example 1)
In the present embodiment, a fully automatic teaching system for teaching an assembly operation procedure to an articulated robot having multiple joints in a fully automatic manner is shown.

FIG. 2 is an overall configuration diagram of the fully automatic teaching system in the present embodiment.
The fully automatic teaching system 5 teaches an assembly operation to a 3D-CAD device 50 storing workpiece design information, an assembly robot 51 composed of a multi-joint joint robot, and an assembly robot. A teaching device 52 is configured.

  The 3D-CAD device 50 includes work information including shape information and material information of each work piece, object information such as screw holes formed on the work, and assembly of a work set such as an assembly. Stores general work design information (CAD data) D6 including mutual arrangement information including information such as order and mutual assembly position between works.

  The assembly robot 51 includes an arm 510 having multiple joints (three joints in this example), a mounting portion 512 for mounting an end effector at the tip (hand portion), and a robot control device 514. It is a robot. This robot incorporates a plurality of motors (servo motors, stepping motors, etc.) at each joint point, and is configured to be able to rotate 360 degrees around the arm axis from the joint point to the tip and to bend at each joint point. ing. Each of the motors is connected to the robot control device 514 of the assembly work robot 51 through a signal line 516-1. The motor is rotated by transmitting a control signal from the robot control device 514, and the arm Operations such as the position, orientation, and traveling direction of the tip are appropriately changed. This control signal is a signal indicating forward rotation, reverse rotation, and rotation angle for controlling the hand unit to a target state based on the position or posture of the hand unit obtained from the feedback signal of each motor. Since the assembly work robot 51 has three joints, various movements are possible in the work space.

  In this example, a conventional camera (imaging camera) 518 provided with an imaging element such as a CCD (Charge Coupled Device) is further attached to the tip of the arm (hand portion). The imaging camera 518 is fixed to the arm tip so that the direction of the arm tip (that is, the direction of the rotation axis of the arm tip) matches the direction of the optical axis of the imaging camera, and the robot control device is configured by the signal line 516-2. 514.

  The robot control device 514 controls the operation of the assembly robot 51 by, for example, sequentially executing a pre-registered command program or a command program fetched from outside with an internal computer. Further, the robot control device 514 of this example has a built-in function for photographing a work on a work area from a plurality of directions with a predetermined procedure when the work recognition command is input. Upon receiving the workpiece recognition command, the robot control device 514 operates the arm 510 of the assembly work robot 51 to set the work area as an imaging range, for example, at an angle of 45 degrees from above with respect to a predetermined position in the work area. The imaging camera 518 is directed from multiple directions in a predetermined order and arrangement, such as from the side, from the side, from the front. The robot controller 514 operates the imaging camera 518 at each timing when the imaging camera 518 is arranged in each direction to receive an image signal from the imaging camera 518, and the received image signal is transmitted through a signal line (not shown). To the teaching device 52.

  The teaching device 52 includes an extraction unit 520, an input unit 521, a storage unit 522, a determination unit 523, a conversion unit 524, an image processing unit 525, and a multi-directional imaging instruction unit 526, and teaches the robot controller 514 about an assembling operation. To do.

  The configuration of the figure is a diagram showing a system board mounted on the teaching device 52 in a divided manner. This system board includes a memory such as a ROM (Read Only Memory) and a RAM (Random Access Memory), An ASIC formed by incorporating an algorithm (assembly procedure calculation sequence, etc.) described later, or a general-purpose computer configured to execute a program that implements a function corresponding to the ASIC is connected to the memory by a bus, etc. It is configured.

  The storage means 522 is a memory such as ROM or RAM to which storage areas M1, M2, and M3 for work information (including object information in the work) D1, mutual arrangement information D2, and assembly supplementary information D3 are allocated. Various types of information are stored. Among the information stored here, the workpiece information D1 and the mutual arrangement information D2 are extracted from the CAD data (design information) D6 stored in the 3D-CAD device 50 through the extraction means 520, This is given (stored) to the storage means 522. For example, the CAD data D6 is acquired via a signal cable (not shown), only necessary information is extracted from the CAD data D6, and the predetermined storage area M1 is converted into a format that can be read by the determination means 523 described later. , Remember to M2.

  Further, the assembly supplement information D3 may be stored in the predetermined storage area M3 in the above-described format in advance, or may be read via the input means 521 connected with, for example, a keyboard (or read from a storage medium). Or the like) may be stored in the above-described format in the predetermined area M3 of the memory.

  In the case of this example, the assembly supplement information D3 is a combination in which information indicating a combination of workpieces is associated with information on means (screwing, adhesive fixing, etc.) used for coupling between workpieces when the workpieces are combined. Means selection supplement information, and handling means selection supplement information in which workpiece information and information on means (such as an end effector and a tool used) for handling the workpiece are associated with each other. The coupling means selection supplementary information is used to determine the coupling means between the workpieces, and the handling means selection supplemental information is used to determine the end effector and tool used by the robot.

  The image processing means 525 is a first recognizing means 525-1 for recognizing the type of each work placed in the work area by image analysis, and the state (position and orientation) of each work placed in the work area as an image. It is comprised by the 2nd recognition means 525-2 recognized by analysis. A signal of an image photographed by the photographing camera 518 is transferred from the robot control device 514 to each recognition means via a signal line (not shown), and the first recognition means 525-1 selects the type of work based on the image signal. The information indicating the type of the workpiece is recognized and transferred to the determining means 523 via the data bus. The second recognizing means 525-2 recognizes the state of the work based on the image signal and the position signal of the photographing camera (position signal of the robot hand) transmitted from the robot control device 514 together with the image signal. The information indicating the state of is transferred to the conversion means 524 via the data bus.

  The determination means 523 uses the work information (including the in-work object information) D1, the mutual arrangement information D2, and the assembly supplement information D3 stored in the storage means 522, thereby the first recognition means 525-1. The assembly work contents of the assembly work set including the recognized types of work are determined. In this case, the contents of the assembly work are determined within the range of all assembly work sets having the mutual arrangement information D2.

  The converting means 524 converts the assembly work content information (assembly work content determination information) D4 determined by the determination means 523 into assembly work procedure information D5 that allows the assembly work robot 51 to operate smoothly, and provides a robot controller. Output to 514 via a signal line (not shown). In the conversion processing of this example, the operation procedure for acquiring the workpiece by the assembly robot 51 from the workpiece area is obtained based on the state of each workpiece recognized by the second recognition means 525-2, and the assembly robot 51 While assembling the instructions of possible partial operations in an appropriate order, the contents of the assembly work are converted into sequence information including those instructions. If the end effector needs to be replaced or a tool is required, the command is added to an appropriate position in the series information.

  The multidirectional imaging instruction means 526 transmits a workpiece recognition command to the robot controller 514 via a signal line (not shown). Here, for example, when a predetermined input is received from a user interface such as a keyboard, the work recognition command is transmitted to the robot controller 514. In addition, when a sensor is installed in the work area and an object such as an assembly work set is placed on the work area, the multi-directional shooting instruction means 526 sends the work to the robot controller 514. A recognition command may be transmitted.

The operation of the fully automatic teaching system will be described below with an example of an assembly work set.
FIG. 3 is an example of an assembly work set.
In the following description, the teaching method for the assembly robot will be described in detail by taking as an example an assembly work set composed of the workpiece A, workpiece B, and workpiece C shown in FIG.

  Work A is a stop (Tomedai) with two screw holes. This screw hole is formed at a position H1 on the AX surface of the workpiece A. The workpiece B is a fastener (Tomechanag) having two through holes for screws. This through hole penetrates from the upper surface (BX surface) to the lower surface (BY surface) of the workpiece B. Work C is two screws (bis C1 and bis C2). The tightening portion of the workpiece C protrudes from the lower surface (CY surface) of the head. In this assembly work set, work A and work B are main parts, and work C is equivalent to a connection part.

4 to 6 show examples of data formats of various information D1, D2, and D3 stored in the storage means 522. FIG.
FIG. 4 is a diagram showing the contents of the work information D1 in the form of a table. FIG. 4A shows the general information and FIG. 4B shows the object information.

In each figure, in order to make the data configuration easy to understand, data items constituting the data are arranged on the left, and the contents of data included in the data items are shown on the right.
In the data example shown in FIG. 5A, a work corresponding to the data is represented by a work symbol immediately adjacent to the right.

Further, the object information in FIG. 5B is information attached to the work A.
FIG. 5 is a diagram showing the contents of the mutual arrangement information (assembly information) D2 in a table format.
As in FIG. 4, the data items constituting the data are arranged on the left and the contents of the data included in the data items are shown on the right, as in FIG. 4.

  This mutual arrangement information D2 is established between work A, work B, and screw C, which is work C, and the data items in the figure show the "assembly order", "assembly position", and "tolerance". As can be seen from the contents, specific restriction information when each is arranged is described.

FIG. 6 is a diagram showing the contents of the assembly supplement information D3 in a table format.
FIG. 4A shows the combination means selection supplement information. As handling means selection supplement information, end effector selection supplement information is shown in FIG. 5B, and tool selection supplement information is shown in FIG.

  In the combination means selection supplementary information in FIG. 5A, the type of the combination means to be selected is shown in the left column, and the detailed combination method of the combination means and the conditions for selecting the combination means are shown in the right column. Was listed.

  In the end effector selection supplement information of FIG. 5B, the left effector name to be selected is shown in the left column, and the specifications that are the conditions for selecting the end effector are listed in the right column.

  In the used tool selection supplemental information in FIG. 5C, the left column shows the name of the used tool to be selected, and the right column shows the conditions for selecting the used tool (in this example, the coupling means and the used parts). ) Was listed.

FIG. 7 is a diagram showing an example of the teaching method of the automatic teaching system performed under the above configuration.
Assembly supplementary information D3 is stored in a predetermined storage area M3 of a memory as storage means 522 by input from an input device (input means) 521 such as a keyboard, and is arranged mutually with design information D6 of CAD device 50 and work information D1. With the information D2 extracted and stored in the predetermined areas M1 and M2 of the memory, processing in the automatic teaching system is started.

  First, a set of works necessary for assembly (assembly work set) is randomly arranged on the work area, and a work recognition command signal is sent to the robot controller 514 via the multi-directional imaging instruction means 526 by sensor detection or an external input instruction. By transmitting the image signal, the multi-directional photographing by the imaging camera 518 is executed, the image signal is taken into the image processing means 525, and the image is analyzed (S1). In this image analysis, the first recognizing means 525-1 extracts an edge indicating the outer shape of each work from the image, and each work is stored in the work information D1 stored in the memory. A suitable work type is selected by shape matching, statistics are obtained for each selected type, and the type and number of various types of work included in the image are obtained. Further, the second recognizing means 525-2 determines in what state each kind of work obtained by the first recognizing means 525-1 is (in what position and posture in the work area). Ask. The position of the work in the work area is determined by, for example, attaching a predetermined mark to a predetermined coordinate in the work area, and capturing each work in the work area together with the mark. The work position may be specified by image analysis based on the above. In addition, the posture may be obtained by analyzing, for example, the arrangement (tilt or the like) of the edge of the type of workpiece recognized by the first recognition unit 525-1 in the image. In the above image analysis, the information on the work in the work area can be obtained individually from each of the captured images, and if the information on the work cannot be obtained from one image, Information on the workpiece can be obtained by using another image taken from the direction. Further, by using these plural images, it becomes possible to accurately obtain information on the workpiece. In this case, since the way each workpiece is reflected differs depending on the shooting direction, image analysis is performed by acquiring the hand position at the time of shooting of the assembly robot 51 from the robot controller 514 and specifying the shooting direction of each image. I do.

  Next, using various pieces of information (work information D1, mutual arrangement information D2, and assembly supplement information D3) stored in the memory 522 for the combination of works specified by the first recognition means 525-1. The contents of assembly work are determined by the determining means (S2).

FIG. 8 is a processing flow when determining the contents of the assembly work.
Here, the assembly work set in FIG. 3 will be described as an example.
First, it is checked whether or not a connection part is included in the assembly work set specified by the first recognition means 525-1 (S2-1). Since the first recognition means 525-1 recognizes the work A, the work B, and the work C, the determination means 523 retrieves the information of each work in the work information D1, and connects to the work (for example, a screw). Or a screw, a seal, or a part designated for bonding belongs to the group of connected parts). Since the assembly work set of this example includes screws, it is determined that there is a connecting part.

  If there is a connecting part, the condition of the combination means selection supplementary information is referred to and a combination means that matches the condition is selected (S2-2). In this example, since the connecting part is a screw, the condition “used part has a screw” described in the first column of the coupling means selection capture information in FIG. In order for the coupling means “screw tightening” corresponding to this to apply, the condition “there is a screw hole in the part to be used” must be satisfied, so the assembly work set specified by the first recognition means 525-1 It is further checked whether there is a part with “screw hole” in the workpiece information. In the assembled work set of this example, since the object information of work A has “screw hole” information, all the conditions of joining means “screw tightening” are satisfied, and means for joining work C to work A “Screw tightening” is selected.

  Next, information indicating the order of assembling the workpieces is extracted from the “assembly order” of the mutual arrangement information D2 (S2-3). When there are a plurality of types of mutual arrangement information D2 for each assembly work set, information indicating the assembly order is extracted from the mutual arrangement information D2 belonging to the assembly work set recognized by the first recognition means 525-1. The corresponding mutual arrangement information D2 is extracted by, for example, associating each piece of mutual arrangement information with information on a combination of works corresponding to each piece of mutual arrangement information, and the mutual arrangement information D2 belonging to all the works obtained by the first recognition means 525-1. The information indicating the assembly order is extracted therefrom.

  Next, the assembly direction of each workpiece when assembling the workpiece in the assembly order is determined (S2-4). Here, the direction of the combined surface when the workpieces are combined with each other at the assembly position of the workpiece set in the “assembly position” of the mutual arrangement information D2 is determined. In this example, the assembly direction is determined so that the number of combined surfaces increases in the horizontal direction.

  First, for work A and work B to be assembled first and second, the “Assembly position” item of the mutual arrangement information D2 is “the AX side of work A and BY side of work B match”, “ There are two conditions that "the center point of A matches the center point of work B". Therefore, the direction in which the AX plane and the BY plane are horizontal is determined as the direction of the combined plane.

  Next, for the work B and screw C1 that are assembled second and third, the "Assembly position" item of the mutual arrangement information D2 "the BX surface of work B and the CY surface of the head of screw C match" There is a condition. Therefore, the direction in which the BX surface of the workpiece B and the CY surface of the head of the screw C1 are horizontal is determined as the direction of the combined surface.

  Next, the screw C2 to be assembled fourth has a condition that “the BX surface of the work B and the CY surface of the head of the screw C match” in the “assembly position” item of the mutual arrangement information D2. Therefore, the direction in which the BX surface of the workpiece B and the CY surface of the head of the screw C2 are horizontal is determined as the direction of the combined surface.

  Next, a tool to be used for connecting the connecting parts with the determined coupling means is determined (S2-5). Here, the corresponding tool to be used is extracted from the tool selection supplementary information using the determined combination means and the connection component (used component) used for the combination as keys. In this example, “screw tightening” is selected as the coupling means and the part used is “screw”, so “bit C driver” is selected in the tool selection capture information.

  Next, an end effector is selected (S2-6). Predetermined information such as “shape”, “material”, “mass”, “surface roughness”, “surface treatment” and the like from the workpiece information D1 for each workpiece of the assembly workpiece set identified by the first recognition means 525-1. Is extracted, and an end effector that matches the “specification” of the end effector selection supplementary information is selected. For the workpiece from which the tool used is extracted in step S2-5, predetermined information such as “shape”, “material”, “mass”, “surface roughness”, “surface treatment”, etc., corresponding to this tool is used. An end effector that is extracted from a tool information table (not shown) and conforms to the “specification” of the end effector selection supplementary information is selected.

FIG. 9 is an example of a management table of work content information selected and extracted in the flow for determining the assembly work content.
In this example, work content information of an assembly work set composed of work A, work B, screw C1, and screw C2 is shown. Using these pieces of information, it is possible to logically determine on the computer an operation procedure for causing the assembly work robot 51 to assemble the assembly work set.

Returning to the flow of FIG. 7, the process of determining on the computer the operation procedure for assembling the assembly work set by the assembly robot 51 will be described.
In the processing following step S2, the assembly work content information is converted into assembly work procedure information of the assembly robot 51 (S3). In this example, using the workpiece state information obtained by the second recognition means 525-2 as a reference for moving the workpiece, the robot movement trajectory from the acquisition of the workpiece to assembly is determined in order, The parameter information is acquired and output to the robot controller 514 together with the main program for causing the robot to execute the operation.

  For example, according to the management table shown in FIG. 9, first, the assembly order parameter is set to “1”, and the coordinate value (coordinate value on the work area) and posture of the workpiece corresponding to the assembly order “1” of the assembly work content information Information (for example, information indicating the arrangement of the joint surface on the work area during assembly) is acquired.

  Furthermore, the shape information of the end effector corresponding to the assembly order “1” is acquired from an information table (not shown), and the movement path (the workpiece acquisition position from the fixed point and the fixed point from this position) Rotation information of each motor until it returns to) and the movement trajectory for assembling the workpiece (rotation information of each motor from the fixed point to the assembly position on the work area until returning to the fixed point) Add to information.

  Also, if there is a tool to be used, calculate the joining operation between the workpieces (rotation information of each motor from the fixed point to the assembly position, performing the predetermined coupling operation and returning to the fixed point) and adding it to the assembly work content information To do.

FIG. 10 shows processing steps of the main program, and FIG. 11 shows parameter information referred to by the main program.
These pieces of information are output to the robot controller 514 and executed during the assembly work of the present assembly work set.

  First, initialization processing is performed (MS1). In this initialization process, the used end effector and tool used attached to the hand portion of the assembly robot are returned to a predetermined storage location (changer), and the hand portion is moved to a fixed point.

  Next, “N (N = 1, 2, 3, 4,...)” Is set as the assembly order parameter (MS2). This N is initially set to 1 and is incremented by 1 each time step MS-7, which will be described later, ends.

Next, the end effector corresponding to the assembly order “N” is attached to the hand portion with a changer, and the fixed point is returned (MS3).
Next, the robot is driven in accordance with the information of “operation trajectory for acquiring the workpiece” (corresponding to “operation trajectory for workpiece acquisition”) in the assembly order “N”, so that the coordinate position is moved from the fixed point to the workpiece area. The workpiece with random orientation (posture) is acquired with the end effector, and then returns to the fixed point (MS4).

  Next, the robot is driven in accordance with the information of “operation trajectory for assembling the workpiece” (corresponding to “operation trajectory for workpiece assembly”) in the assembly order “N”, so that the robot moves from the fixed point to the work area at the coordinate position. Place the workpiece in the assembly orientation (posture) and return to the fixed point (MS5).

Next, if there is information on the tool used in the “tool used” in the assembly order “N”, the tool used corresponding to this is mounted with the changer, and the fixed point is returned (MS6).
Next, by driving the robot according to the information of "joining motion" (corresponding to "joining motion between workpieces"), the robot moves from the fixed point to the work area, joins the workpieces, and returns to the fixed point (MS7) .

Then, returning to the process of step MS-2, the process is repeated in the same manner as described above.
In this example, the processes up to the assembly order “4” are repeated. The processes in steps MS6 and MS7 are executed only in the assembly orders “3” and “4”, and are omitted in the other cases.

FIG. 12 sequentially illustrates how the assembly work set is assembled by the above assembly procedure.
FIG. 4A shows the state after the process of step MS-5 when the assembly order is “1”. As shown in the figure, the workpiece A is arranged at the coordinate position of the work area in the assembly direction (posture), with the AX surface as the upper surface.

FIG. 5B shows the state after the process of step MS-5 when the assembly order is “2”. As shown in the figure, work B is arranged on the AX surface of work A with the BX surface aligned.
FIG. 6C shows a state in the middle of the process of step MS-5 when the assembly order is “3”. As shown in the figure, the screw C1 is inserted into the screw hole of the workpiece A.

  FIG. 4D shows the state after the process of step MS-7 when the assembly order is “4”. As shown in the figure, the screws C1 and C2 are screwed into the two screw holes of the workpiece A through the through holes of the workpiece B, and the assembly of the assembly work set is completed.

  As described above, in this embodiment, it is possible to automatically teach an assembly operation according to a combination of workpieces for assembly to the robot. For this reason, the teaching work by the operator is greatly shortened.

  In addition, since the type and state of the workpiece placed in the workpiece area can be recognized, even if the workpiece is randomly placed in the workpiece area, those workpieces are automatically identified and their assembly operations are performed. Can be automatically taught to the robot. For this reason, the teaching work of assembly work sets consisting of arbitrary combinations of workpieces can be fully automated, and when products of various types and small lots are frequently replaced and assembled, or products that have entered in the course of production are replaced during production. Thus, when assembling, the assembly operation can be changed smoothly, and the production efficiency can be increased.

1 shows a main part of an automatic teaching system according to the present invention. It is a whole block diagram of the fully automatic teaching system in a present Example. It is an example of an assembly work set. It is the figure which showed the contents of the work information D1 in the form of a table. It is the figure which showed the contents of the mutual arrangement information (assembly information) D2 in a table format. It is the figure which showed the contents of the assembly supplement information D3 in a table format. It is the figure which showed an example of the teaching method of this automatic teaching system. It is a processing flow when determining an assembly work content. It is an example of the management table of work content information. This is the main program processing step. This is parameter information referred to by the main program. The assembly order of the assembly work set.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Storage means 2 Determination means 3 Conversion means
D1 Work information
D2 mutual placement information
D3 Assembly supplement information
D4 Assembly work content information
D5 Assembly procedure information

Claims (8)

A system for teaching an assembly operation procedure of a plurality of workpieces to an assembly operation robot,
Storage means for storing workpiece information that is single piece information of a workpiece, mutual arrangement information of an assembly work set that is a combination of a plurality of workpieces, and assembly supplement information;
Among the assembly work sets, the assembly work contents of the assembly work set, which is a set of a plurality of works composed of combinations of a predetermined type and each work is arranged in a predetermined state in the work area, the work information, the mutual information Determining means for determining based on the arrangement information and the assembly supplement information;
Conversion means for converting the contents of the assembly work into an assembly work procedure of the assembly robot;
An automatic teaching system comprising: A first recognition means for recognizing the type of each work placed in the work area;
The determination means includes all assembly work sets having the mutual arrangement information as a range, the assembly work content of the assembly work set composed of the types of works recognized by the first recognition means as the work information, the mutual arrangement. To be determined based on the information and the above assembly supplement information,
The automatic teaching system according to claim 1. A second recognizing means for recognizing the state of each work placed in the work area;
The converting means obtains the assembling work contents from the work area by obtaining the work procedure of the work by the assembling work robot based on the state of each work recognized by the second recognizing means. Convert to assembly work procedure of assembly robot,
The automatic teaching system according to claim 2. Further comprising an imaging means in the hand portion of the assembly robot,
The first recognizing means recognizes a plurality of types of work placed in the work area based on the image information obtained by the imaging means,
The second recognizing means recognizes the state of each work on the work area based on the image information obtained by the imaging means and the state information of the hand portion of the assembly robot.
The automatic teaching system according to claim 3. Furthermore,
Extraction means for extracting the work information and the mutual arrangement information of the assembly work set from CAD data and giving the storage means;
Input means for inputting the assembly supplementary information into the storage means;
The automatic teaching system according to claim 1, wherein the automatic teaching system includes: The assembly supplement information includes coupling means selection supplemental information in which information indicating a combination of works and information on a means to be used for coupling between works in the combination of the works, and means for handling the work information and the work It is handling means selection supplementary information that is associated with the information of
The determining means selects an inter-work joining means from the joining means selection supplementary information based on at least a combination of the recognized types of workpieces, and performs the handling based on the selected inter-work joining means. Select the tool and end effector to use from the means selection supplemental information.
The automatic teaching system according to any one of claims 1 to 5, wherein The contents of the assembly work include the assembly order of the workpieces included in the assembly workpiece set, the assembly direction of the workpieces, the coupling means between the workpieces, and the tool and end used when assembling the assembly workpiece set. Including effectors,
The assembly work procedure includes the replacement of the tool used and the end effector, and a series of assembly of each work by the assembly work robot between the work area and the work area in the range of all works of the assembly work set. Including operating procedures,
The automatic teaching system according to any one of claims 1 to 6, wherein: A method of teaching an assembly work procedure of a plurality of workpieces to an assembly work robot,
Work information that is single piece information of a work, mutual arrangement information of an assembly work set that is a combination of a plurality of works, and assembly supplement information are stored in a memory,
Among the assembly work sets, the assembly work contents of the assembly work set, which is a set of a plurality of works composed of combinations of a predetermined type and each work is arranged in a predetermined state in the work area, the work information, the mutual information Determined based on the placement information and the above assembly supplement information,
Converting the assembly work content into the assembly work procedure of the assembly robot;
A teaching method characterized by the above.