JP2009279719A - Teaching device and teaching method for robot - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique capable of suppressing an error caused by a camera arrangement position and teaching a robot with an operation posture with high accuracy.
A master part in which a simulated part 8 simulating an assembly part of a second part is provided on a pallet 4 having a camera 10 provided at a position corresponding to the assembly part of a first part is fixed to 6 parts. In this state, the target position of the second part is calculated from the first image taken by the camera 10, and the pallet 4 in a state where the master part 6 is not fixed is arranged at the arrangement position of the first part and the second part. The actual position of the second part is calculated from the second image captured by the camera 10 with the robot 16 held by the robot 16, and the deviation between the target position of the second part and the actual position is smaller than a predetermined value. Is taught to the robot 16 as an assembly posture in which the second component is assembled to the first component.
[Selection] Figure 1

Description

  The present invention relates to a teaching technique for teaching an operation posture to a robot. In particular, the present invention relates to a teaching technique in which a first part is arranged with respect to a robot, a second part is held by the robot, and the robot is operated to teach the robot an operation posture when the second part is assembled to the first part.

  Patent Document 1 describes a teaching technique for teaching a robot an operation posture using a master part and a camera. In the technique of Patent Document 1, a master part is placed at a working position of a robot, the master part is photographed by a camera, and a robot operation program is created using coordinates extracted from the photographed master part image. .

Japanese Patent Laid-Open No. 5-6214

For example, when the robot is caused to insert a second component having a diameter substantially the same as the diameter of the recess into the recess formed in the first component, a strict accuracy requirement is imposed on the posture taught to the robot. It is done. When teaching an assembly operation that requires precision as described above using a teaching technique using a master part and a camera, the accuracy of the posture of the taught robot is affected by the position where the camera is installed. For example, when the camera is installed at a position where the concave part of the first part and the second part are bird's-eye views, the tip of the second part inserted into the concave part and the concave part of the first part due to the influence of distortion of the camera lens It becomes impossible to accurately grasp the positional relationship with the part. In this case, there is a possibility that a correct operation posture cannot be taught to the robot.
In view of the above, the present invention provides a technique capable of suppressing an error caused by a camera arrangement position and teaching a robot with an operation posture with high accuracy.

The present invention teaches a teaching operation in which a first part is arranged with respect to a robot, a second part is held by the robot, and the robot is operated to teach the robot an operation posture when the second part is assembled to the first part. It is embodied in a teaching device for. This teaching device can be arranged at the arrangement position of the first component, and can be fixed to the pallet having a camera provided at a position corresponding to the assembly part of the first component to which the second component is assembled. The simulated part imitating the assembly part of the second part assembled to the assembly part of the first part is provided at a position corresponding to the target position of the assembly part of the second part with respect to the assembly part of the first part. And a processing device for processing an image taken by the camera.
The processing device calculates the position of the simulated part from an image of the simulated part taken by the camera with the master part fixed to the pallet, and the pallet in a state where the master part is not fixed is arranged at the arrangement position of the first part. The position of the second part is calculated from the image of the second part taken by the camera while the second part is held by the robot, and the deviation between the calculated position of the simulated part and the position of the second part is calculated. Calculate

  The shapes of the first part, the assembly part of the first part, the second part, and the assembly part of the second part are not particularly limited. For example, the first component may be a plate-shaped component, an insertion hole may be formed in the plate-shaped component, and the second component may be a rod-shaped component that is inserted into the insertion hole. In this case, the pallet is provided with a camera at a position corresponding to the position where the insertion hole is formed in the plate-shaped part, and the master part imitates at least a tip portion inserted into the insertion hole of the bar-shaped part. Simulated parts are formed. Further, the pallet may be formed in the same shape as the plate-like component, may have a shape including a jig for mounting the plate-like component, or may have a shape greatly different from them.

In the apparatus according to the present invention, a camera is provided at a position corresponding to the assembly part of the first part to which the second part is assembled. Therefore, when the master part is fixed to the pallet and the simulated part provided on the master part is photographed by the camera, the simulated part can be captured at the center of the photographing range. In addition, when the pallet in which the master part is not fixed is arranged at the position where the first part is arranged and the second part held by the robot is photographed by the camera, the second part is captured at the center of the photographing range. Can be taken. Therefore, the influence of distortion aberration hardly occurs on the image taken by the camera. Accordingly, it is possible to accurately grasp the position of the simulated part and the position of the second part from the photographed image by the camera, and it is possible to accurately teach the robot the operation posture in which the second part is held at the target position.
According to this apparatus, it is possible to suppress an error caused by the arrangement position of the camera, and to teach the robot the operation posture with high accuracy.

It is preferable that the above-described processing apparatus gives an operation command to the robot by using a deviation between the calculated position of the simulated part and the position of the second part.
According to this configuration, the teaching work can be automatically performed between the teaching device and the robot.

The present invention is also embodied in a teaching method for a robot using the teaching device described above. In this teaching method, a master part is fixed to a pallet, a first photographing process for photographing a simulated part with a camera, and a target for calculating the position of the simulated part from an image of the simulated part photographed in the first photographing process by a processing device. Position calculation process and second imaging process in which the pallet in a state where the master part is not fixed is replaced with the first part and arranged with respect to the robot, the second part is held by the robot, and the second part is imaged by the camera And an actual position calculation step of calculating the position of the second component from the image of the second part imaged in the second imaging step by the processing device, and the actual position and actual position calculated by the processing device in the target position calculation step. A deviation calculating step for calculating a deviation from the position of the second part calculated in the position calculating step is provided.
According to this method, it is possible to suppress an error caused by the arrangement position of the camera and to teach the robot the motion posture with high accuracy.

  According to the present invention, it is possible to accurately teach the robot the operation posture when the second part is assembled to the first part. Even when strict accuracy requirements are imposed on the operation posture taught to the robot, the assembly posture when the second component supported by the robot is assembled to the first component can be accurately taught to the robot.

The main features of the embodiments described below are listed.
(Feature 1) The camera is fixed to the pallet, and the simulated part and the second part are photographed from the same position on the pallet under the same photographing condition.
(Feature 2) The processing device extracts the position and contour shape of the assembly part of the simulated part or the second part from the image photographed by the camera, and the position of the assembly part of the second part becomes the position of the simulated part. While matching, the robot's posture when the contour shape of the assembly part of the second part is superimposed on the contour shape of the simulated part is stored in the robot.
(Characteristic 3) When assembling a plurality of second parts at the same time, the pallet is provided with a camera for each second part.
(Characteristic 4) The processing device instructs the robot to operate based on the deviation between the position of the second part calculated from the image of the second part and the position of the simulated part calculated from the image of the simulated part. Calculate The processing device gives an operation command to the robot based on the calculated operation amount and operation direction, and operates the robot so that the second part held by the robot moves to the target position calculated from the image of the simulated part.

  Embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows a schematic configuration of the teaching device 2 and the articulated robot 16. The teaching device 2 arranges the first part (not shown) with respect to the articulated robot 16, holds the second part (not shown) by the articulated robot 16, and operates the articulated robot 16. This is used for teaching work that teaches the multi-joint robot 16 the operation posture when the second part is assembled to the first part. In this embodiment, the first component is a component that is mounted on the pallet and has a set of insertion holes, the second component is in the shape of a rod, and the tip portion is the insertion hole of the first component. Is to be inserted.

  The teaching device 2 includes a pallet 4, a master part 6, and an image processing device 28. The pallet 4 has the shape of a pallet on which the first component is mounted. A set of CCD cameras 10 is provided on the pallet 4. The cameras 10a and 10b are connected to the image processing device 28, and images taken by the cameras 10a and 10b are processed by the image processing device 28. As shown in FIG. 1, during the teaching work of this embodiment, the pallet 4 is fixed at a position where the pallet 4 is arranged when the first part of the fixed plate 12 is arranged in the work area of the articulated robot 16. Is done. The master part 6 can be detachably fixed to the pallet 4. The master part 6 is provided with a set of simulated parts 8. Each simulated part 8a, 8b imitates the tip of a second part assembled to the first part. The configuration of the pallet 4 and the master part 6 will be described in detail later.

  The image processing device 28 includes a position coordinate calculation unit 30, a deviation calculation unit 32, a storage unit 36, an operation unit 38, and a display unit 40. The image processing device 28 is connected to the robot controller 42 so as to be communicable. The position coordinate calculation unit 30 calculates the positions of the captured parts based on the images captured by the cameras 10a and 10b. For example, if the simulated parts 8a and 8b are captured images, the position coordinate calculation unit 30 calculates the position coordinates of the simulated parts 8a and 8b in the image, respectively. Further, the position coordinate calculation unit 30 calculates the position of the second component in the image from the image in which the second component is captured. The deviation calculation unit 32 calculates a deviation between the position of the simulated component 8 and the position of the second component calculated by the position coordinate calculation unit 30. The calculated deviation can be transmitted to the robot controller 42. The storage unit 36 stores image data captured by the cameras 10a and 10b, data representing numerical values calculated by the calculation units 30 and 32, and various programs necessary for operating the image information processing apparatus 28. Yes. The image processing device 28 is realized by an information processing device such as a personal computer executing the above-described program. That is, the calculation units 30 and 32 described above are realized by reading and executing a program stored in the storage unit 36 by a CPU of a computer or the like. The operation unit 38 includes operation means such as a keyboard, and the operator can input various information to the image processing device 28 by operating the operation unit 38. The display unit 40 can display various information processed by the image processing device 28.

  The articulated robot 16 is an arm type robot fixed to the fixed plate 12 by a fixing jig (not shown). The multi-joint robot 16 includes arms 17 and 19, arm joints 18 and 20, a substrate support joint 22, a chuck substrate portion 24, and a set of chuck portions 26. The operation of the articulated robot 16 is controlled by a robot controller 42 connected to the articulated robot 16. The robot controller 42 includes an operation unit 44, an operation amount calculation unit 45, a joint control unit 46, and a storage unit 48. The storage unit 48 stores the rotation angles of the joints 18, 20, and 22 of the articulated robot 16. In addition, the storage unit 48 stores information regarding the position where the pallet 4 is disposed with respect to the articulated robot 16. Further, the storage unit 48 stores various programs necessary for operating the articulated robot 16. The motion amount calculation unit 45 calculates a motion amount such as a movement amount and a movement direction of the articulated robot 16 from the deviation transmitted from the image information processing device 28. The calculated motion amount is input to the joint control unit 46. The joint control unit 46 controls the operations of the joints 18, 20, and 22. In addition, the joint control unit 46 executes a process of calculating the movement amount of the multi-joint robot 16 (the rotation amounts of the joints 18, 20, and 22). The operation unit 44 allows the operator to operate the articulated robot 16. For example, when the operator operates the operation unit 44, a driving instruction for the joints 18, 20, and 22 can be input. The input instruction is output to the joint control unit 46, and the joint control unit 46 drives the joints 18, 20, and 22 according to the instruction. In addition, the joint control unit 46 can drive the joints 18, 20, and 22 based on the data stored in the storage unit 48. Further, the joint control unit 46 can drive the joints 18, 20, and 22 according to the motion amount calculated by the motion amount calculation unit 45. In addition to the operation of the articulated robot 16, the operation unit 44 can input an instruction to store the posture of the articulated robot 16 in the storage unit 48, an instruction to start an actual assembly operation, and the like.

  The arm joint 18 is connected to the upstream end of the arm 19 and supports the arm 19 in a swingable manner. The downstream end of the arm 19 is connected to the arm joint 20. The arm joint 20 is connected to the upstream end of the arm 17 and supports the arm 17 so as to be swingable. The downstream end of the arm 17 is connected to the substrate support joint 22. The substrate support joint 22 is connected to the chuck substrate portion 24 and supports the chuck substrate portion 24 so that the chuck substrate portion 24 can rotate and move in the vertical direction. The chuck substrate portion 24 is provided with a set of chuck portions 26. The multi-joint robot 16 grips the second component using the chuck portion 26.

FIG. 2 shows an enlarged side view of the pallet 4 and the master part 6 shown in FIG. FIG. 3 is a plan view of the pallet 4. The cameras 10a and 10b of the pallet 4 are fixed to the camera support 56 by a fixture (not shown). The cameras 10a and 10b are positioned at positions corresponding to the respective insertion holes of the first part where the second part is assembled. In FIG. 3, the positions of the insertion holes of the first component are indicated by broken lines 58a and 58b. In this embodiment, the cameras 10a and 10b are arranged such that the centers 60a and 60b of the lenses 54a and 54b substantially coincide with the centers of the insertion holes 58a and 58b of the first component, respectively. That is, the optical axes of the cameras 10a and 10b substantially coincide with the axis of the insertion hole of the first component. For this reason, in this embodiment, the cameras 10a and 10b are provided with second parts (simulated parts 8a and 8b when the master part 6 is fixed to the pallet 4) assembled in the insertion holes of the first parts. Capture from the bottom right in the center of the shooting range. That is, the images photographed by the cameras 10a and 10b include an assembly surface of the second component or the simulated components 8a and 8b (the end surface 51 of the second component 50 in FIG. 4 and the simulated component in FIG. 2) in the central region of the image. 8 end face 52). For this reason, the image taken by the camera 10 is hardly affected by distortion. The position of the simulated part 8 and the position of the second part can be accurately grasped from the photographed image by the camera 10, and the articulated robot 16 is accurately instructed of the operation posture in which the second part is held at the correct assembly position. Can do.
The positional relationship between the simulated component 8 and the camera 10 realized when the master component 6 is fixed to the pallet 4 is a surface direction perpendicular to the imaging axis of the camera 10, that is, the simulated component 8 captured by the camera 10. It is only necessary to accurately reproduce the positional relationship when the tip of the second component is at the target position with respect to the insertion hole of the first component on a plane parallel to the end surface 52 of the first component. On the other hand, the positional relationship in the direction parallel to the imaging axis of the camera 10, that is, the distance d1 between the end face 52 and the lens 54 of the camera 10 is the tip of the second part and the first part when placed at the target position. It does not have to match the distance of the insertion hole. The distance d1 shown in FIG. 3 can be appropriately set according to the performance of the camera 10 such as the focal length.

  As described above, the simulated parts 8a and 8b formed on the master part 6 are imitations of the assembly part of the second part assembled in the first part (in this embodiment, the tip part of the bar-shaped part). is there. When the master part 6 is fixed to the pallet 4, the simulated parts 8 a and 8 b are fixed at a target position where the tip of the second part should be positioned with respect to the insertion hole of the first part. From this, by teaching the robot 16 the operating posture so that the position of the second part held by the robot 16 matches the position of the simulated part 8 when the master part 6 is fixed to the pallet 4, The articulated robot 16 can be instructed of a posture for accurately realizing the target position of the second part when the second part is assembled to the first part.

  FIG. 5 is a flowchart showing the work process of teaching work for teaching the joint robot 16 using the teaching device 2. In step S2, the pallet 4 is positioned and fixed at the arrangement position of the fixing plate 12 (see FIG. 1). The pallet 4 is fixed to the fixing plate 12 by a plurality of pallet fixing tools 14. As described above, in this embodiment, the pallet 4 has the shape of a pallet on which the first component is mounted. Therefore, the pallet 4 can be easily positioned on the fixed plate 12 by fixing the pallet 4 with the pallet fixing tool 14 attached to the fixed plate 12 in advance. When the pallet 4 is fixed to the fixed plate 12, the positions where the cameras 10a and 10b are arranged are the positions where the insertion holes for the first component are arranged when the pallet for mounting the first component is fixed to the fixed plate 12. They are almost identical (see FIG. 3).

Next, in step S4, the master part 6 is fixed to the upper part of the pallet 4 fixed to the fixing plate 12 (see FIGS. 1 and 2). When the master part 6 is fixed to the pallet 4, the simulation parts 8 a and 8 b of the master part 6 are arranged at positions where the end faces 52 a and 52 b face the cameras 10 a and 10 b of the pallet 4, respectively. In step S <b> 6, the simulated part 8 is photographed by the camera 10 while the master part 6 is fixed to the pallet 4. In the first image taken by the camera 10a in this step, the simulated part 8a is shown in the center of the image, and in the first image taken by the camera 10b, the simulated part 8b is shown in the center of the image. .
In step S8, the position coordinate calculation unit 30 calculates the positions of the simulated components 8a and 8b from the first images captured by the cameras 10a and 10b, respectively. In this step, the position coordinates of the simulated part 8 may be calculated by converting the position on the image into space coordinates. For example, the contour of the simulated component 8 shown in the first image may be extracted, the coordinate value of the extracted contour may be calculated, or the coordinates of the center point of the extracted contour may be calculated. The position coordinates calculated by the position coordinate calculation unit 30 are stored in the storage unit 36 as target position coordinates of the tip part of the second component. Here, when the calculated positions of the simulated parts 8a and 8b are deviated from the center of each first image, the positions of the cameras 10a and 10b are adjusted, and the process may be executed again from the photographing process in step S6.

  When the master part 6 is removed from the pallet 4 (step S10), in step S12, the second part 50 used for actual assembly is gripped by the chuck portion 26 of the articulated robot 16 (see FIG. 4). For example, the operator can operate the operation unit 44 of the robot controller 42 to cause the articulated robot 16 to grip the second component.

In step S <b> 13, the joint controller 46 calculates the motion amounts of the joints 18, 20, and 22 from the position of the pallet 4 and the current posture of the articulated robot 16. In this step, the second part is grasped from the information regarding the fixed position of the pallet 4 stored in the storage unit 48 of the robot controller 42 and the information regarding the position of the chuck unit 26 calculated from the current posture of the articulated robot 16. The amount of movement of the articulated robot 16 necessary for the chuck portion 26 positioned above the pallet 4 to be calculated is calculated. In step S14, the articulated robot 16 is operated based on the calculated operation amount. Specifically, the joint control unit 46 drives the joints 18, 20, and 22 according to the movement amount calculated in step S13, and moves the chuck unit 26 (see FIG. 4).
If the robot controller 42 does not hold information regarding the fixed position of the pallet 4, the operator operates the operation unit 44 instead of the processing in step S 13 described above, and the chuck unit 26 is moved to the pallet by eye. 4 may be moved upward.

When the second parts 50a and 50b are moved above the pallet 4 by the operation of the articulated robot 16 in step S14, the tip parts of the second parts 50a and 50b are arranged above the cameras 10a and 10b of the pallet 4. Is done. In step S <b> 16, the second component 50 is photographed by the camera 10. In this step, photographing is performed under the same conditions (positions of the cameras 10a and 10b, focal lengths, etc.) as when the simulated parts 8a and 8b were photographed. The end face 51a of the second component 50a is shown in the second image taken by the camera 10a, and the end face 51b of the second part 50b is shown in the second image taken by the camera 10b.
In step S18, the position coordinate calculation unit 30 calculates the positions of the second components 50a and 50b from the second images captured by the cameras 10a and 10b, respectively. In this step, similarly to the calculation of the target position in step S8, the position coordinates of the second component 50 may be calculated by converting the position on the image into space coordinates. Further, the contour of the second component 50 shown in the second image may be extracted, and the coordinate value of the extracted contour may be calculated, or the coordinates of the center point of the extracted contour may be calculated. The position coordinates calculated by the position coordinate calculation unit 30 are stored in the storage unit 36 as actual position coordinates of the tip of the second component 50.

  In step S20, the deviation calculator 32 calculates a deviation between the target position calculated from the first image in step S8 and the actual position calculated from the second image in step S18. As described above, the second component when the second components 50a and 50b are assembled at the assembly position of the first component according to the positions of the simulated components 8a and 8b calculated from the first images taken by the cameras 10a and 10b. The correct positions of 50a and 50b are shown. For this reason, the deviation calculated by the deviation calculating unit 32 becomes smaller as the second parts 50a and 50b are arranged closer to the correct position.

In step S <b> 21, the deviation calculated by the deviation calculation unit 32 is transmitted from the image information processing device 28 to the robot controller 42. The transmitted deviation is input to the operation amount calculation unit 45.
In step S22, it is determined whether or not the deviation input in step S21 is equal to or less than a threshold value. The threshold value can be arbitrarily set according to the accuracy required for teaching the assembly position. For example, if there is a very strict accuracy requirement regarding assembly position teaching, does the actual position calculated for the second part match the target position calculated for the simulated part (ie, the deviation is zero)? It may be determined whether or not.
In step S20 described above, the deviation between the target position and the actual position is calculated from both the image captured by the camera 10a and the image captured by the camera 10b. Therefore, in step S22, determination processing may be executed for each deviation, or determination processing may be executed for a value obtained by integrating them. For example, the determination process may be executed for the average deviation.

  If the deviation is equal to or smaller than the threshold value (YES in step S22), the articulated robot 16 realizes a correct assembly posture. Therefore, the posture is stored in the storage unit 48 of the robot controller 42 as an assembly posture when the second component is assembled to the first component (step S26), and the teaching work is completed. In the process of step S26, for example, information regarding the deviation and the comparison comparison process of the deviation is displayed on the display unit 40 of the image information processing apparatus 28. It may be executed by inputting an instruction to store the posture.

If the deviation is larger than the threshold value (NO in step S22), the articulated robot 16 does not yet satisfy the required accuracy condition, and requires further posture correction. For this reason, in step S24, the movement amount calculation unit 45 calculates the movement amount and movement direction of the chuck unit 26 based on the deviation. The calculated motion amount is input to the joint control unit 46.
In step S <b> 25, the joint controller 46 calculates the motion amount of each indirect 18, 20, 22 of the articulated robot 16 from the input movement amount and direction of the chuck unit 26 and the current posture of the articulated robot 16. Is done. In step S14, the joint control unit 46 operates the articulated robot 16 according to the calculated operation amount. For example, the center point of the end face 52 of the simulated part 8 calculated from the first image and the center point of the end face 51 of the second part 50 calculated from the second image are coincident, but the contour does not overlap. It is understood that the distance d1 (see FIG. 2) from the camera 10 to the simulated part 8 and the distance d2 (see FIG. 4) from the camera 10 to the second part are not the same distance. In such a case, NO is determined in step S22, and the distance d2 from the camera 10 to the second part is corrected by the operation of the articulated robot 16 in step S14. Then, the deviation between the actual position of the second part after the operation and the target position is calculated again and evaluated. The contours of the end surface 52 of the simulated component 8 and the end surface 51 of the second component 50 extracted from the first image and the second image overlap so that the actual position where the second component is located satisfies the desired accuracy at the target position. A series of processing is repeated until approaching (that is, YES is determined in step S22).

  According to the teaching device 2 described above, errors due to the position of the camera 10 can be suppressed, and the articulated robot 16 can be taught the operation posture with high accuracy by automatic teaching. The assembling posture when assembling the second part to the first part can be accurately taught to the articulated robot 16.

  In the above embodiment, an example of teaching that teaches the assembly posture to the articulated robot 16 has been shown, but the embodiment of the present invention is not limited to this. For example, the motion amount may be calculated by the image information processing apparatus 28 and an instruction to the joint control unit 46 may be transmitted to the robot controller 42. Alternatively, the joint controller 46 may execute the process executed by the motion amount calculator 45.

Specific examples of the present invention have been described in detail above, but these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and changes of the specific examples illustrated above.
The technical elements described in this specification or the drawings exhibit technical usefulness alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. In addition, the technology illustrated in the present specification or the drawings achieves a plurality of objects at the same time, and has technical utility by achieving one of the objects.

It is a figure (1) showing schematic structure of the articulated robot and teaching apparatus in teaching work. It is a side view of a pallet and a master part. It is a top view of a pallet. It is a figure (2) showing schematic structure of the articulated robot and teaching apparatus during teaching work. It is a flowchart showing the work process of teaching.

Explanation of symbols

2: Teaching device 4: Pallet 6: Master parts 8, 8a, 8b: Simulated parts 10, 10a, 10b: Camera 12: Fixing plate 14: Pallet fixing tool 16: Articulated robot 17, 19: Arm 18, 20: Arm Joint 22: Substrate support joint 24: Chuck substrate unit 26, 26a, 26b: Chuck unit 28: Image processing device 30: Position coordinate calculation unit 32: Deviation calculation unit 36: Storage unit 38: Operation unit 40: Display unit 42: Robot Controller 44: Operation unit 45: Motion amount calculation unit 46: Joint control unit 48: Storage units 50, 50a, 50b: second parts 51, 51a, 51b: end faces 52, 52a, 52b of second parts: end faces of simulated parts 54, 54a, 54b: Lens 56: Camera support portion 58a, 58b: Position 60a, 60b of first part assembly portion (insertion hole): Center

Claims (3)

Teaching for teaching work in which the first part is arranged with respect to the robot, the second part is held by the robot, and the robot is operated to teach the robot the operation posture when the second part is assembled to the first part. Device,
A pallet having a camera that can be arranged at the arrangement position of the first part and is provided at a position corresponding to the assembly part of the first part to which the second part is assembled;
The simulated part that can be fixed to the pallet and imitates the assembly part of the second part that is assembled to the assembly part of the first part is at the target position of the assembly part of the second part with respect to the assembly part of the first part. A master part provided in the corresponding position;
A processing device for processing images taken by the camera;
The processing device calculates the position of the simulated part from an image of the simulated part taken by the camera with the master part fixed to the pallet, and the pallet in a state where the master part is not fixed is arranged at the arrangement position of the first part. The position of the second part is calculated from the image of the second part taken by the camera while the second part is held by the robot, and the deviation between the calculated position of the simulated part and the position of the second part is calculated. Teaching device characterized by calculating   The teaching device according to claim 1, wherein the processing device gives an operation command to the robot using a deviation between the calculated position of the simulated part and the position of the second part. A teaching method for teaching a robot an operation posture when assembling the first part and the second part by placing the first part with respect to the robot, holding the second part by the robot, and operating the robot.
It can be arranged with respect to the robot in place of the first part, and can be fixed to the pallet having a camera provided at a position corresponding to the assembly part of the first part to which the second part is assembled, A master part in which a simulated part simulating an assembly part of a second part assembled to an assembly part of one part is provided at a position corresponding to a target position of the second part when the second part is assembled to the first part. A preparation process for preparing parts and a processing device for processing an image captured by the camera;
A first photographing step of fixing a master part to a pallet and photographing a simulated part by a camera;
A target position calculating step of calculating the position of the simulated part from the image of the simulated part imaged in the first imaging step by the processing device;
A second imaging step in which the pallet in a state where the master part is not fixed is placed on the robot instead of the first part, the second part is held by the robot, and the second part is imaged by the camera;
An actual position calculating step of calculating the position of the second component from the image of the second component imaged in the second imaging step by the processing device;
A deviation calculating step of calculating a deviation between the position of the simulated part calculated in the target position calculating step and the position of the second part calculated in the actual position calculating step by the processing device;
Teaching method comprising.

Images