JP2011029490A - Device for mounting component, and method of mounting component - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a device for mounting a component which accurately mounts a component A on a component B, and method of mounting component; and a method of mounting a component. <P>SOLUTION: A control unit 5 of this device for mounting a component comprises: an image recognition part 101; an image recognition part 102; a reference position coordinate conversion part 103; a fixation parameter storage part 104; a calibration part 105; a frame parameter storage part 106; a first correction amount calculation part 107; a second correction amount calculation part 108; a third correction amount calculation part 109; and a mounting coordinate calculation part 110. Cameras 2-3 and an arm drive part 6 are connected to the control unit 5. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a component mounting apparatus and a component mounting method.

  In the component mounting apparatus, as shown in FIG. 2 of Patent Document 1, when the semiconductor element 2 (component A) is mounted on the lead frame 9 (component B), the component A is on the rail, and the component A is mounted on the component. It is adsorbed by the collet 8 and mounted on the component B. In this case, the mounting accuracy depends on the positional relationship between the position of the component B placed on the rail and the collet 8 for mounting the component, and the position of the component B. For this reason, the position of the component A and the component mounting collet 8 before being mounted by the component mounting collet 8 are detected by image recognition. Thereby, the component mounting collet 8 can adsorb the component A with high accuracy. Next, the center position of the component B is detected by image recognition, and the detected value is compared with the image center position of the camera that captures the component B to detect the amount of deviation of the center position of the component B. Based on the detected deviation amount of the center position of the component B, a method of mounting the component A attracted by the collet 8 on the component B with high accuracy has been proposed.

JP 2004-311569 A

  However, in the conventional component mounting apparatus, the difference between the position of the component A and the predetermined start point of the collet is detected using the captured image before being picked up by the collet, and the detected displacement amount is obtained. Based on this, the collet was moved and adsorbed. In other words, since the amount of deviation in the state where the part A is actually attracted to the collet is not detected, there is a problem that the part A may not be accurately mounted on the part B if there is a deviation after adsorption. It was. Further, the amount of deviation of the center position of the component B is detected by comparing the center position of the component B on which the component A is mounted with the image center of the camera that captures the component B. If the relationship between the collet and the starting point of the collet does not match, the collet may not be moved to the target coordinates to be mounted, and there is a problem that the part A may not be accurately mounted on the part B. .

  The present invention has been made in view of the above-described problems, and an object of the present invention is to provide a component mounting apparatus and a component mounting method for accurately mounting the component A on the component B.

The component mounting apparatus of the present invention is an electronic component mounting apparatus in which a second component is mounted on a first component using a movable arm.
A first deviation amount calculation unit that calculates a first deviation amount indicating how much the first component is deviated from a predetermined reference state based on an image obtained by imaging the first component;
A second shift for calculating a second shift amount indicating how much the second component is shifted from a predetermined reference state based on an image obtained by imaging the movable arm in a state where the second component is held. A quantity calculator;
A third deviation amount calculation unit that calculates a third deviation amount indicating how much the length and the inclination angle of the movable arm are deviated from a predetermined reference value based on an image obtained by imaging the movable arm;
The movable arm corrects the position where the second component is mounted on the first component based on the first deviation amount, the second deviation amount, and the third deviation amount, and the movable arm An arm drive unit for driving
It is characterized by providing.

  The component mounting apparatus according to the present invention is characterized in that the first deviation amount calculation unit calculates the first deviation amount in a coordinate system of the movable arm.

  In the component mounting apparatus according to the present invention, the third deviation amount calculation unit is based on information obtained by extracting the position and angle of the movable arm from images obtained by imaging the movable arm at at least three different angles. A third deviation amount representing how much the length and the inclination angle of the movable arm deviate from a predetermined reference value is calculated.

Further, the present invention provides a component mounting method in which the second component is mounted on the first component using the movable arm.
A first deviation amount calculating step of calculating a first deviation amount indicating how much the first component is deviated from a predetermined reference state based on an image obtained by imaging the first component;
A second shift for calculating a second shift amount indicating how much the second component is shifted from a predetermined reference state based on an image obtained by imaging the movable arm in a state where the second component is held. A quantity calculation step;
A third deviation amount calculating step for calculating a third deviation amount indicating how much the length and the inclination angle of the movable arm are deviated from a predetermined reference value based on an image obtained by imaging the movable arm;
The movable arm corrects the position where the second component is mounted on the first component based on the first deviation amount, the second deviation amount, and the third deviation amount, and the movable arm An arm driving process for driving
It is characterized by having.

  According to the present invention, the first deviation amount indicating how much the first component is deviated from the predetermined reference state is obtained, and the second component is in a state where the movable arm holds the second component. A second deviation amount indicating how much the deviation is from the predetermined reference state is obtained, and further, a third deviation amount from the reference value of the length and inclination angle of the movable arm is obtained. By using the obtained three displacement amounts, the displacement amount of the mounting position of the second component is corrected and the movable arm is driven, so that the mounted component is accurately mounted on the component to be mounted. It becomes possible.

It is the schematic which shows the structural example of the whole system of the component mounting apparatus which concerns on embodiment of this invention. It is a block diagram which shows an example of a structure of the component mounting apparatus which concerns on the same embodiment. It is the schematic of the arm of the component mounting apparatus which concerns on the embodiment, the attaching part with which the front-end | tip is mounted | worn, and a calibration jig | tool. It is the schematic of the mounting robot of the component mounting apparatus which concerns on the same embodiment. It is a figure which shows an example of the data hold | maintained at the fixed parameter memory | storage part which concerns on the embodiment. It is a figure showing an example of data held in a schematic diagram of a mounting frame concerning the embodiment, and a frame parameter storage part. It is an example of the image of the state which the arm 8 by the camera 4 which concerns on the embodiment adsorb | sucked the connector 11. FIG. It is an example of the flowchart of the calibration of the component mounting apparatus which concerns on the same embodiment. It is an example of the flowchart of the calibration of the component mounting apparatus which concerns on the same embodiment. It is the schematic of the captured image of the camera 4 which concerns on the same embodiment, and the schematic of the state of a mounting robot. It is the schematic of the picked-up image of the cameras 2 and 3 which concern on the embodiment, and the schematic of the state of a mounting robot. It is the schematic explaining each coordinate system of the cameras 2, 3, and 4 which concern on the embodiment. It is the schematic explaining how to obtain | require the deviation | shift amount of the inclination of the arm 8 at the time of the calibration which concerns on the same embodiment. It is the schematic explaining how to obtain | require the deviation | shift amount of the length of the arm 8 which concerns on the embodiment. It is the schematic explaining how to obtain | require the deviation | shift amount of the length of the arm 8 which concerns on the embodiment. It is the schematic explaining how to obtain | require the deviation | shift amount of the length of the arm 8 which concerns on the embodiment. It is the schematic explaining the conversion of the reference point which the cameras 2 and 3 which concern on the embodiment imaged. It is a figure which shows an example of the corrected coordinate in each row | line | column of the mounting frame 12 which concerns on the embodiment. It is the schematic explaining conversion of the reference point of the connector which the camera 4 concerning the embodiment imaged. It is a figure which shows an example of the coordinate of the mounting position to each row | line | column of the mounting frame which instruct | indicates the mounting robot which concerns on the embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS. In addition, this invention is not limited to the embodiment which concerns, A various change is possible within the range of the technical thought.

  First, the structure of the component mounting apparatus in this embodiment is demonstrated using FIGS. FIG. 1 is a schematic diagram showing a configuration example of the entire system of the component mounting apparatus in the present embodiment. FIG. 2 is a block diagram showing an example of the configuration of the component mounting apparatus according to this embodiment. FIG. 3 is a schematic view of a calibration jig and an attachment portion attached to the arm tip of the component mounting apparatus in the present embodiment. FIG. 4 is a schematic view of a mounting robot of the component mounting apparatus in the present embodiment. FIG. 5 is a diagram illustrating an example of data held in the parameter storage unit in the present embodiment. FIG. 6 is a diagram illustrating a schematic diagram of the mounting frame and an example of data on the center of gravity of each column of the mounting frame in the present embodiment. FIG. 7 is an example of an image in a state where the arm by the camera in this embodiment sucks the connector.

  1, 2, and 5, the component mounting apparatus includes a mounting robot 1, a camera 2, a camera 3, a camera 4, a control unit 5, an arm driving unit 6, and a stage 7. The connector 11 sucked from the connector forming apparatus 9 by the mounting robot 1 is mounted on a mounting frame 12 on the stage.

  The mounted robot 1 includes an arm 8. As shown in FIG. 3, the mounting robot 1 moves the calibration jig 13 attached to the tip of the arm 8 to a predetermined position by the arm driving unit 6. Further, the mounting robot 1 moves the connector 11 attracted by the mounting portion 211 attached to the tip of the arm 8 to the predetermined mounting position calculated by the control unit 5 by the arm driving unit 6, and moves the connector 11. Mounted on the mounting frame 12.

  The camera 2 and the camera 3 capture an image of the calibration jig 13 attached to the tip of the arm 8 of the mounted robot 1. In addition, the camera 2 and the camera 3 capture an image of the mounting frame 12 after the mounting frame 12 placed on the stage 7 stops at a predetermined position. Further, the camera 2 and the camera 3 send the captured image to the control unit 5. The camera 2 and the camera 3 are installed on the upper side with respect to the mounting frame 12 as shown in FIG.

  The camera 4 captures an image of the calibration jig 13 when the calibration jig 13 attached to the tip of the arm 8 of the mounting robot 1 comes to a predetermined position. Further, the camera 4 is attached to the attachment portion 211 attached to the distal end of the arm 8 after the attachment portion 211 attached to the distal end of the arm 8 of the mounted robot 1 attracts the connector 11 and stops at a predetermined position. An image of the attracted connector 11 is taken. Further, the camera 4 sends the captured image to the control unit 5. Moreover, the camera 4 is arrange | positioned below with respect to the connector 11 like FIG. The camera 4 is attached by rotating 90 ° clockwise with respect to robot coordinates described later.

  The control unit 5 stores in advance in the control unit 5 a setting value (hereinafter referred to as a design value of the mounting frame 12) related to position information of a plurality of components in the mounting frame 12. The control unit 5 receives images taken from the camera 2, the camera 3, and the camera 4. Moreover, the control unit 5 detects the information of the imaged part position from the imaged image. Further, the control unit 5 sends a signal for controlling the arm 8 to a predetermined position to the arm drive unit 6 during the calibration process. Further, when the connector 11 is mounted, the control unit 5 instructs the arm driving unit 6 to move the connector molding device 9 in order to attract the connector 11. Further, when the connector 11 is mounted, the control unit 5 is mounted on the tip of the arm 8 using the image received from the cameras 2 to 4 and the stored design value of the mounting frame 12 according to the procedure described later. The correction value of the movement destination of the connector 11 attracted by the attachment unit 211 is calculated. The control unit 5 calculates the movement coordinates obtained by calculating the movement destination coordinates on the mounting frame 12 of the connector 11 attracted by the mounting portion 211 attached to the tip of the arm 8 based on the calculated correction value. The previous coordinates are sent to the arm drive unit 6.

  The arm driving unit 6 moves the calibration jig 13 attached to the tip of the arm 8 to a predetermined position in the screens of the camera 2, the camera 3, and the camera 4 by the control from the control unit during the calibration process. Moreover, the arm drive part 6 moves the attachment part 211 with which the arm 8 was mounted | worn to the predetermined position of the connector shaping | molding apparatus 9 which shape | molds the connector 11 by control from a control unit, and adsorb | sucks the connector 11 . Further, the arm drive unit 6 moves to a predetermined position in the screen of the camera 4 under the control of the control unit after attracting the connector 11. The arm driving unit 6 receives a control signal for the mounting position calculated by the control unit 5, and drives the arm 8 of the mounting robot 1 and the mounting unit 211 attached to the mounting position based on the received control signal.

  The stage 7 stops the mounting frame 12 placed on the stage 7 at a predetermined position.

  The arm 8 is a component that can be removed during maintenance work of the mounted robot 1, and is mounted on the mounted robot 1. Further, as shown in FIG. 3, the arm 8 has a calibration jig 13 attached to the tip at the time of calibration processing, and an attachment portion 211 attached to the connector 8 when the connector 11 is mounted on the mounting frame 12.

  The connector molding device 9 is a device in which the connector 11 is molded by cutting or the like using a mold.

  Next, the configuration of the control unit 5 will be described using the blocks in FIG. The control unit 5 includes an image recognition unit 101, an image recognition unit 102, a reference position coordinate conversion unit 103, a fixed parameter storage unit 104, a calibration unit 105, a frame parameter storage unit 106, and a first correction amount. The calculation unit 107, the second correction amount calculation unit 108, the third correction amount calculation unit 109, and the mounted coordinate calculation unit 110 are configured.

  The image recognition unit 101 receives images captured by the camera 2 and the camera 3. In the calibration process, the image recognition unit 101 detects the reference mark 213 of the calibration jig 13 attached to the tip of the arm 8 from the received image and the angle with respect to the position and the reference direction according to the camera coordinates. Further, the image recognition unit 101 stores the detected position of the reference mark 213 in the camera coordinates and the angle with respect to the reference direction in the fixed parameter storage unit 104. When the connector 11 is mounted, the image recognition unit 101 detects the reference point of the mounting frame 12 from the received image and the angle with respect to the position in the camera coordinates and the reference direction. As the reference direction, a unique direction such as a horizontal direction or a vertical direction is used in each camera image.

FIG. 3 is a schematic view of the arm 8, the attachment portion 211 attached to the tip thereof, and the calibration jig 13. In FIG. 3A, when the connector 11 is mounted on the mounting frame 12, the mounting portion 211 is mounted on the tip of the arm 8. Further, a suction part 212 that sucks the connector 11 is provided below the attachment part 211. In addition, in FIG. 3A, the reference points of the mounting frame 12 are the center of gravity G _D1 (x _D1 , y _D1 , θ _D1 ) of the first column and the center of gravity G _D30 (x _D30 , y _D30 , θ _D30 ).
In FIG. 3B, the calibration jig 13 is attached to the tip of the arm 8 during calibration. Further, the calibration jig 13 is provided with a cross hole (reference mark) 213 whose position is detected during calibration.

The image recognition unit 102 receives an image captured by the camera 4. Further, during the calibration process, the image recognition unit 102 detects the reference mark 213 of the calibration jig 13 attached to the tip of the arm 8 from the received image and the angle with respect to the position and the reference direction according to the camera coordinates. Further, the image recognition unit 102 stores the detected position of the reference mark 213 in the camera coordinates and the angle with respect to the reference direction in the fixed parameter storage unit 104.
When the connector 11 is mounted on the mounting frame 12, the image recognition unit 101 detects the reference point of the connector 11 from the received image and the angle with respect to the position and the reference direction based on the camera coordinates.

  The reference position coordinate conversion unit 103 converts the position of the reference mark 213 of the calibration jig 13 attached to the tip of the arm 8 detected by the image recognition unit 101 and the image recognition unit 102 and the angle with respect to the reference direction of each camera coordinate. Receive by value. Further, the reference position coordinate conversion unit 103 uses the coordinates of the reference mark 213 received from the image recognition units 101 and 102 and the angles with respect to the reference direction with reference to the position cr (x, y) determined by the mounted robot 1. Convert to the robot coordinate value. The reference position coordinate conversion unit 103 stores the coordinates of the reference mark 213 converted into robot coordinates and the angles with respect to the reference direction in the fixed parameter storage unit 104.

  FIG. 4 is a schematic diagram of the mounting robot 1 in the present embodiment. In FIG. 4, the mounting robot 1 includes a mounting robot main body 201, a movable unit 202, and an arm 8. The arm 8 is driven to a predetermined position by the movable part 202 being moved by a drive signal from the arm driving part 6. In FIG. 4, cr (x, y) is set at a unique position between the linear movable unit 202 and the mounting robot body 201 with the arm 8 positioned perpendicular to the mounting robot body 201. Yes.

  The fixed parameter storage unit 104 has an angle with respect to the position and reference direction of the reference mark 213 detected by the image recognition unit 101 and the position and reference direction of the reference mark 213 detected by the image recognition unit 102 with respect to the position and reference direction. The angle, each angle of the reference mark 213 converted to the robot coordinate by the reference position coordinate conversion unit 103 with respect to each position and reference direction according to the camera coordinates, the length of the arm 8 to be described later, the tilt deviation amount of the arm 8, and the calibration The coordinate difference and inclination angle difference between the motion jig 13 and the attachment portion 211 are held.

  FIG. 5 is a configuration example of data held in the fixed parameter storage unit 104 in the present embodiment. The fixed parameter storage unit 104 stores coordinate values of the center position of the reference mark 213 captured by the camera 2 on the camera 2 and coordinate values obtained by converting the center position of the reference mark 213 captured by the camera 2 into robot coordinates. And the resolution of the camera 2, the coordinate value of the center position of the reference mark imaged by the camera 3 on the camera 3, and the coordinate value obtained by converting the center position of the reference mark 213 imaged by the camera 3 into robot coordinates. The resolution of the camera 3, the coordinate value on the camera 4 of the center position of the reference mark 213 captured by the camera 4, and the coordinate value obtained by converting the center position of the reference mark 213 captured by the camera 4 into robot coordinates The mounting angle of the camera 4, the resolution of the camera 4, the length of the arm 8 actually measured in step S <b> 2, and the tilt deviation amount of the arm 8 to be described later are associated and held. To have.

  The calibration unit 105 transmits a control signal for moving the arm 8 to a predetermined position to the arm driving unit 6 at the time of calibration in the procedure described later. Further, the calibration unit 105 transmits the coordinates of the movement destination of the arm 8 to the third correction amount calculation unit 109 based on the control signal transmitted to the arm driving unit 6.

As shown in FIG. 6A, the frame parameter storage unit 106 uses the center of gravity G (x, y, θ) of the mounting frame as the origin and the center of gravity position of each column in a plurality of child parts constituting the mounting frame 12. Each coordinate and each tilt angle are held. Further, the data held in the frame parameter storage unit 106 holds values obtained from the design values for the mounting frame 12. FIG. 6B is an example of a data structure held in the frame parameter storage unit 106. The coordinates G _Di (x _Di , y _Di ) of the center of gravity of each column of the mounting frame 12 and the inclination θ _Di are held. For the x axis, the upward direction from the bottom is defined as the positive direction, for the y axis, the right direction from the right is defined as the positive direction, and for the angle, clockwise is defined as positive.

When the connector 11 is mounted on the mounting frame 12, the first correction amount calculation unit 107 uses the reference points G _D1 (x _D1 , y _D1 , θ _D1 ) and G _{D30 of} the mounting frame 12 detected by the image recognition unit 101. (X _D30 , y _D30 , θ _D30 ) is received as a position in camera coordinates and an angle with respect to a reference direction. Further, after being received as the positions of the reference points G _D1 and G _D30 and the angle with respect to the reference direction, using the data stored in the fixed parameter storage unit 104 and the data stored in the frame parameter storage unit 106, The first correction amount is calculated according to the procedure described later. In addition, the first correction amount calculation unit 107 transmits the calculated first correction amount to the mounting coordinate calculation unit 110.

That is, the first correction amount calculation unit 107 determines the left and right sides of the mounting frame 12 detected from the design values of the mounting frame 12 held in the fixed parameter storage unit 104 and the images actually captured by the cameras 2 and 3. by using the reference point _{G D1} and _{G D30,} and calculates the first correction amount which is the amount of deviation of mounting frame 12.

The second correction amount calculation unit 108 uses the reference points G _C (x) of the plurality of connectors 11 detected by the image recognition unit 102 after the connector 11 is adsorbed by the attachment unit 211 attached to the tip of the arm 8. _{C 1} , y _C , θ _C ) is received as a position in camera coordinates and an angle with respect to a reference direction. Further, after receiving the angle with respect to the position and reference direction of the reference point G _C, using the data stored in the fixed parameter storing unit 104, calculates a second correction amount according to the procedure described below. Further, the second correction amount calculation unit 108 transmits the calculated second correction amount to the mounting coordinate calculation unit 110.

  FIG. 7 is an example of an image in a state where the arm 8 by the camera 4 in this embodiment attracts the connector 11. As shown in FIG. 7, the suction part 212 on the lower side of the attachment part 211 attached to the arm 8 attracts ten connectors 11 at a time. Under the control of the control unit, the arm 8 simultaneously mounts the ten connectors 11 on each column of the mounting arm 12 described with reference to FIG.

That is, the second correction amount calculating unit 108 uses the reference point G _C of connectors 11 in a state of being adsorbed to the mounting portion 211 mounted on the arm 8, which is detected from the image captured by the camera 4, connected A second correction amount that is a shift amount of the child 11 is calculated.

  The third correction amount calculation unit 109 detects each position and each position according to the camera coordinates of the reference mark 213 of the calibration jig 13 attached to the arm 8 detected by the image recognition unit 101 and the image recognition unit 102 during calibration. The tilt angle and the coordinates of the movement destination of the arm 8 are received from the calibration unit 105. In addition, the third correction amount calculation unit 109 uses the position and angle of inclination of the received reference mark 213 in the camera coordinates and the coordinates of the movement destination of the arm 8 to describe the amount of inclination deviation of the arm 8 by a method described below. And the length of the arm 8 is calculated as a third correction amount. In addition, the third correction amount calculation unit 109 transmits the calculated third correction amount to the mounting coordinate calculation unit 110. The third correction amount calculation unit 109 stores the calculated third correction amount in the fixed parameter storage unit 104.

That is, the third correction amount calculating unit 109 uses the reference point G _C of the reference marks 213 of the calibration jig 13 mounted on the arm 8, which is detected from the image captured by the camera 4, displacement of the arm 8 A third correction amount that is an amount is calculated.

The mounted coordinate calculation unit 110 includes a first correction amount calculated by the first correction amount calculation unit 107, a second correction amount calculated by the second correction amount calculation unit 108, and a third correction amount calculation unit 108. The third correction amount calculated by is received. Also, the mounting coordinate calculation unit 110 calculates the mounting position of the arm 8 that has attracted the connector 11 on the mounting frame 12 using the received correction amount. In addition, the mounting coordinate calculation unit 110 transmits the calculated mounting position to the arm driving unit 6.
Above, description of the control unit 5 is complete | finished.

  Next, the mounting procedure of the component mounting apparatus in the present embodiment will be described with reference to FIGS. First, a calibration procedure performed before mounting the connector 11 on the mounting frame 12 will be described with reference to FIGS. 8 and 9 are examples of a flowchart of calibration of the component mounting apparatus in the present embodiment. FIG. 10 is a schematic diagram of a captured image of the camera 4 and a schematic diagram of the state of the mounted robot in the present embodiment. FIG. 11 is a schematic diagram of captured images of the cameras 2 and 3 and a schematic diagram of the state of the mounted robot in the present embodiment. FIG. 12 is a schematic diagram illustrating coordinates in the cameras 2 to 4 in the present embodiment. First, the calibration jig 13 is attached to the tip of the arm 8 of the mounting robot 1 (S1).

  Next, the user of the component mounting apparatus actually measures the length L [mm] of the arm 8 with a measure or the like, and stores the actually measured value in the fixed parameter storage unit 104 of the control unit 5 (S2).

  Next, according to an instruction from the calibration unit 105 of the control unit 5, the arm driving unit 6 moves the arm 8 on which the calibration jig 13 is mounted to a predetermined position at a predetermined angle (S3). Hereinafter, the arm 8 moves as the arm driving unit 6 is controlled by a control signal from the calibration unit 105.

First, the tilt angle θ of the arm 8 is set to the center coordinate of the camera 4 instructed from the calibration unit 105, and the arm driving unit 6 moves the movable unit 202 and the arm 8 variably. After the arm 8 has moved to the instructed position, the camera 4 acquires imaging data (S4). As shown in FIG. 10, the inclination angle of the arm is θ = 0 ° in the initial state in FIG. 4 and a minus angle in the clockwise direction. FIG. 10A is a schematic diagram of the state of the movable unit 202 and the arm 8 when the tilt angle θ of the arm 8 is −135 °, and a schematic diagram of an image displayed on the camera 4. Further, as shown in FIG. 10A, the image on the camera 4 installed on the lower side of the arm 8 is obtained by the calibration jig 13 attached to the tip of the arm 8 when θ = −90 °. It is installed at an angle that is projected horizontally.
Next, the tilt angle θ of the arm 8 is set to the center coordinate of the camera 4 instructed from the calibration unit 105, and the arm driving unit 6 moves the movable unit 202 and the arm 8 while changing them. After the arm 8 has moved to the instructed position, the camera 4 acquires imaging data (S4). FIG. 10B is a schematic diagram of the state of the movable unit 202 and the arm 8 when the inclination angle θ of the arm 8 is −135 °, and a schematic diagram of an image displayed on the camera 4. The reason why the tilt angle of the arm 8 is reversed left and right in the image of the camera 4 is that the camera 4 is installed on the lower side of the arm 8 and images from the lower side.
Next, the tilt angle θ of the arm 8 is set to the central coordinate of the camera 4 instructed from the calibration unit 105, and the arm driving unit 6 moves the movable unit 202 and the arm 8 in a variable manner. After the arm 8 has moved to the instructed position, the camera 4 acquires imaging data (S4). FIG. 10C is a schematic diagram of the state of the movable portion 202 and the arm 8 when the tilt angle θ of the arm 8 is −45 °, and a schematic diagram of an image displayed on the camera 4.

Next, the arm drive unit 6 moves the movable unit 202 and the arm 8 in a variable manner to the center coordinates of the camera 2 instructed from the calibration unit 105. After the arm 8 moves to the instructed position, the camera 2 acquires imaging data (S4). FIG. 11A is a schematic diagram of the state of the movable unit 202 and the arm 8 at the left reference position _GD1 of the mounting frame 12, and a schematic diagram of an image displayed on the camera 2. FIG. Note that since the shooting magnification of the camera 2 is higher than that of the camera 5, the cross hole of the reference mark of the calibration jig 13 is copied to the camera 2.
Next, the arm driving unit 6 moves the movable unit 202 and the arm 8 in a variable manner to the center coordinates of the camera 3 instructed from the calibration unit 105. After the arm 8 moves to the instructed position, the camera 3 acquires imaging data (S4). FIG. 11B is a schematic diagram of the state of the movable unit 202 and the arm 8 at the right reference position _GD30 of the mounting frame 12, and a schematic diagram of an image displayed on the camera 3. FIG.

Next, the image recognition unit 101 receives images captured by the cameras 2 and 3, and the center coordinates c ₁ (x ₁ , y ₁ ) of the reference mark 213 of the calibration jig 13 and the received image (FIG. 11) and c ₃₀ (x ₃₀ , y ₃₀ ) is detected (S5). Further, the image recognition unit 102 receives an image captured by the camera 4, and the center coordinates p <b> 1 of the reference mark 213 of the calibration jig 13 from the image of θ = −90 ° (FIG. 10A) among the received images. ( _XP1 , _yP1 ) is detected (S5).
Each camera coordinate is demonstrated using FIG. FIG. 12 is a schematic diagram for explaining the coordinate systems of the cameras 2, 3, and 4 in the present embodiment. The center coordinates c ₁ (x ₁ , y ₁ ) and c ₃₀ (x ₃₀ , y ₃₀ ) and p 1 (x _P1 , y _P1 ) of the reference mark 213 detected in FIGS. 11 and 12 are as shown in FIG. The coordinates in the camera, with the lower right as the origin, cameras 2 and 3 have a maximum value of 479 pixels on the vertical axis and a maximum value of 639 pixels on the horizontal axis, and camera 4 has a maximum value of 1023 pixels on the vertical axis and a maximum on the horizontal axis The value is 767 pixels. In this embodiment, the vertical axis is defined as the x axis, the upward direction from the origin is defined as the positive direction, the horizontal axis is defined as the y axis, and the left direction from the origin is defined as the plus direction.
In the present embodiment, the center coordinates c ₁ (x ₁ , y ₁ ) = (314.082, 249.794) of the reference mark 213 are detected from the image captured by the camera 2, and the reference mark is detected from the image captured by the camera 3. The central coordinates c ₃₀ (x ₃₀ , y ₃₀ ) = (297.818, 198.282) of the 213 are detected, and the central coordinates p 1 (x of the reference mark 213 are detected from the θ = −90 ° image captured by the camera 4. _{P1, y P1)} (508.181,412.719) is detected.
However, since the camera 4 is attached by rotating 90 ° clockwise relative to the robot coordinates, the image acquired from the camera 4 has a coordinate relationship as shown in FIG.
Further, the image recognition unit 101 and the image recognition unit 102 store the center coordinates of each reference mark 213 detected from each image in the fixed parameter storage unit 104.

  Next, the image recognition unit 101 and the image recognition unit 102 transmit the detected center coordinates of the reference mark 213 to the reference position coordinate conversion unit 103. In FIG. 10, the calibration unit 105 calculates the third correction amount of the coordinates b1 (x, y), b2 (x, y), and b3 (x, y) of each arm 8 in each calibration state. To the unit 109. The third correction amount calculation unit 109 receives the base coordinates of each arm 8 in each calibration state from the calibration unit 105 (S6).

Next, the reference position coordinate conversion unit 103 converts the received center coordinates of the reference mark 213 into robot coordinates (S7). In the present embodiment, as shown in FIG. 4, the robot coordinates are expressed in units within the movable range of the arm 8 with a unique coordinate cr (x, y) predetermined for the mounted robot 1 as an origin (reference point). It is defined as expressed in mm. Thereby, in the present embodiment, the center coordinates (314.082, 249.794) [pixel] of the reference mark are converted to (170.970, −59.424) [mm] from the image captured by the camera 2, The center coordinate c (297.818, 198.282) [pixel] of the reference mark is converted into (170.980, −249.755) [mm] from the image captured by the camera 3, and θ = From the −90 ° image, the center coordinates (508.181, 412.719) [pixel] of the reference mark are converted into (241.490, −107.815) [mm].
Further, since the reference position coordinate conversion unit 103 is attached by rotating the attachment angle of the camera 4 by 90 ° clockwise with respect to the robot coordinates, the inclination is −90 °, that is, −π / 2 = −1.571. To do. The reference position coordinate conversion unit 103 stores the converted coordinates and the attachment angle of the camera 4 in the fixed parameter storage unit 104.

As shown in FIG. 13A, when θ = −135 °, the third correction amount calculation unit 109 stores the coordinates b2 (x, y) of the base of the arm 8 received from the calibration unit 105 and fixed parameters. Using the center coordinates p2 (x _p2 , y _p2 ) of the reference mark 213 stored in the unit 104, a straight line P2 is generated with the coordinates of the base of the arm 8 as the start point and the center of the reference mark 213 as the end point. In addition, when θ = −45 °, the third correction amount calculation unit 109 calculates the base coordinates b3 (x, y) of the arm 8 acquired from the calibration unit 105 and the center coordinates of the fixed parameter storage unit 105 reference mark. Using p3 (x _p3 , y _p3 ), a straight line P3 is generated with the coordinates of the base of the arm 8 as the start point and the center of the reference mark as the end point.
Next, as shown in FIG. 13B, the third correction amount calculation unit 109 performs an operation of overlapping the start point b2 (x, y) of the straight line P2 and the start point b3 (x, y) of the straight line P3.
Next, as illustrated in FIG. 13C, the third correction amount calculation unit 109 sets the end point p2 (x _p2 , y _p2 ) of the straight line P2 and the end point p3 (x _p3 , y _p3 ) of the straight line P3 in a straight line. Generate a connecting straight line.
Next, as shown in FIG. 13 (d), the third correction amount calculation unit 109 obtains an angle θ formed by the generated straight line connecting p2 and p3 and the x-axis from the equation (1).
θ = tan ⁻¹ ((y _p2 −y _p3 ) / | x _p2 −x _p3 |) Expression (1)
The angle θ obtained by the equation (1) corresponds to the amount of tilt deviation of the arm 8. In equation (1), if the solution is positive, it means a counterclockwise tilt shift, and if the solution is negative, it means a clockwise tilt shift.

  Next, the third correction amount calculation unit 109 determines whether or not the obtained inclination deviation amount θ of the arm 8 is within an allowable range of deviation amount = ± 0.5 ° (S9). As a result of the determination in step S9, if the deviation amount θ is larger than the allowable range (S9: No), the process returns to step S1 and the adjustment jig 13 is remounted on the tip of the arm 8, and steps S2 to S8 are repeated. As a result of the determination in step S9, when the deviation amount θ is within the allowable range (S9: Yes), the third correction amount calculation unit 109 stores the deviation amount θ in the fixed parameter storage unit 104, and determines the length of the arm 8 in step S10. Proceed to shift amount calculation processing. For example, the inclination of the arm 8 is calculated as θ = 0.271709 [deg], and the calculated inclination as shown in FIG. 5 is converted to a robot coordinate unit of angle, and 0.00473 [rad] is also fixed parameter. It is stored in the storage unit 103.

Next, how to determine the amount of deviation of the arm length will be described with reference to FIGS. 14 to 16 are schematic diagrams for explaining how to obtain the amount of deviation of the length of the arm 8 in the present embodiment. The coordinates are performed using robot coordinates. Further, the straight line P2 and the straight line P3 generated by the amount of tilt deviation of the arm 8 are also used. As shown in FIG. 14A, when θ = −90 °, the third correction amount calculation unit 109 stores the coordinates b1 (x, y) of the base of the arm 8 received from the calibration unit 105 and the fixed parameters. Using the center coordinates p1 (x _p1 , y _p1 ) of the part 105 reference mark, a straight line P1 is generated with the base coordinates of the arm 8 as the start point and the center of the reference mark as the end point.
Next, as shown in FIG. 14B, the third correction amount calculation unit 109 rotates the straight lines P1, P2, and P3 by 90 degrees counterclockwise about the end points p1, p2, and p3, respectively.
Next, as shown in FIG. 14C, the third correction amount calculation unit 109 superimposes the straight lines P1, P2, and P3 rotated by 90 °. When there is no deviation in the length of the arm 8, the coordinates of the end points p1, p2, and p3 of the straight lines P1, P2, and P3 rotated by 90 ° coincide with each other, so that the result is as shown in FIG. When there is a deviation amount, it is as shown in FIG. Hereinafter, the case where there is a deviation in the length of the arm 8 will be described with reference to FIGS.

In FIG. 15A, the midpoint (x _Q1 , y _p1 ) of p1 (x _p1 , y _p1 ) and p2 (x _p2 , y _p2 ) is expressed as in equation (2).
(X _Q1 , y _Q1 ) = ((x _p1 + x _p2 ) / 2, (y _p1 + y _p2 ) / 2) Equation (2)
In FIG. 15B, an angle θ _A formed by a perpendicular line passing through the midpoint (x _Q1 , y _Q1 ) and the y axis is expressed as shown in Expression (3).
θ _A = tan ⁻¹ ((y _p2 −y _p1 ) / (x _p2 −x _p1 )) (3)
Further, in FIG. _{15 (c), p2 (x} p2, y p2) and _p3 (x _{p3, y p3)} of the middle point _{(x Q2, y p2)} is expressed by formula (4).
(X _Q2 , y _Q2 ) = ((x _p2 + x _p3 ) / 2, (y _p2 + y _p3 ) / 2) Equation (4)
Further, in FIG. 15D, an angle θ _B formed by a perpendicular passing through the midpoint (x _Q2 , y _Q2 ) and the y axis is expressed as in Expression (5).
θ _B = tan ⁻¹ ((y _p3 −y _p2 ) / (x _p3 −x _p2 )) (5)
In addition, in FIG. 16A, a straight line having an inclination tan θ _A passing through the midpoint (x _Q1 , y _Q1 ) is expressed as in Expression (6).
y = x · tan θ _A + (y _Q1 −x _Q1 · tan θ _A ) (6)
Similarly, a straight line passing through the midpoint (x _Q2 , y _p2 ) and having an inclination tan θ _B is expressed as in Expression (7).
y = x · tan θ _B + (y _Q2 −x _Q2 · tan θ _B ) (7)
Further, in FIG. 16B, the intersection (x _c , y _c ) between the expressions (6) and (7) is expressed as the expressions (8) to (9).
_{x c = ((y Q2 -x} Q2 · tanθ B) - (y Q1 -x Q1 · tanθ A)) / (tanθ A -tanθ B)
_{= X Q1 + ((y Q2} -y Q1) - (x Q2 -x Q1) · tanθ B) / (tanθ A -tanθ B) ··· Equation (8)
y _c = x _c · tan θ _A + (y _Q1 −x _Q1 · tan θ _A ) (9)
_{Then, p1 (x p1, y p1} ) and the intersection point _(x c, _{y c)} from obtaining the length L of the arm.
L = y _p1 − (y _p1 −y _c )
= Y _p1 − (y _p1 −x _c · tan θ _A + (y _Q1 −x _Q1 · tan θ _A ))
= Y _p1 − (y _p1 −x _Q1 + ((y _Q2 −y _Q1 ) − (x _Q2 −x _Q1 ) · tan θ _B ) / (tan θ _A −tan θ _B ) · tan θ _A + (y _Q1 −x _Q1 · tan θ _A )) ... Formula (10)
The length L [mm] of the arm 8 obtained by the above calculation is stored in the fixed parameter storage unit 104 by the third correction amount calculation unit 109 replacing the measured value. For example, the length of the arm 8 is calculated as L = 112.125 [mm] by the above calculation, and is stored in the fixed parameter storage unit 103 as shown in FIG.
This completes the calibration process of the arm 8 of the mounted robot 1.

That is, in the calibration process, the calibration jig 13 mounted on the arm 8 is imaged by the cameras 2 to 4, and the reference provided in the calibration jig 13 from the images captured by the image recognition units 101 and 102. The position of the mark 213 and the inclination with respect to the reference direction are detected. In addition, the image recognition units 101 and 102 store the detected position of the reference mark 213 and the inclination with respect to the reference direction in the fixed parameter storage unit 104. In addition, the reference position coordinate conversion unit 103 converts the center coordinates of the reference mark 213 received from the image recognition units 101 and 102 into robot coordinates and stores them in the fixed parameter storage unit 104.
Further, the third correction amount calculation unit 109 uses the center coordinates of the reference mark 213 held in the fixed parameter storage unit 104 and the coordinates of the base of the arm 8 received from the calibration unit 105, so that the reference of the arm 8 is obtained. The deviation amount of the inclination with respect to the direction and the length of the arm 8 are calculated as the third correction amount and stored in the fixed parameter storage unit 104.

Next, registration of the design value of the mounting frame 12 will be described using the schematic diagram of the mounting frame of FIG. As shown in FIG. 6A, the mounting frame 12 includes 10 rows in the x direction on the vertical axis and 30 columns in the y direction on the horizontal axis, that is, 300 child components 301. As will be described later, the connector 11 attracted to the arm 8 simultaneously attracts the ten connectors 11 for one row and simultaneously mounts them on each row of the mounting frame 12.
The center of gravity G (x, y, θ) that is the center of the mounting frame 12 is used as the origin, and the coordinates G _Di (x _Di , y _Di ) (1 ≦ i ≦ 30) of the center of gravity of each column and each column with respect to the x axis Is stored in the frame parameter storage unit 106 in advance as a design value frame parameter (S11).

Next, the adjustment jig 13 attached to the tip of the arm 8 is removed, and the attachment portion 211 is attached as shown in FIG. Further, the coordinate difference between the position of the attachment portion 211 and the position of the reference mark 213 of the jig 13 is measured, and the user stores it in the fixed parameter storage unit 104 (S12). For example, the difference is x-axis component of the mounting portion 211 and the calibration jig 13 _{X OC = 0.270 [mm],} y -axis component _Y OC = 0.000 [mm], the slope component theta = 0. When 000 [rad] is measured, it is stored in the fixed parameter storage unit 104 as shown in FIG.

Next, returning to FIG. 6, the frame position correction process as the first correction process is performed (S13). Hereinafter, the frame position correction process will be described with reference to FIGS. FIG. 17 is a schematic diagram illustrating the conversion of the reference points captured by the cameras 2 and 3 in the present embodiment.
First, the mounting frame 12 on the stage 7 is moved to a predetermined position and stopped by a drive unit of the stage 7 (not shown). The camera 2 and the camera 3 use the reference points G _D1 (x _D1 , y _D1 ) and G _D30 (x _D30 , y _D30 ) that are the center of gravity of the first row and the 30th row set on the left and right of the mounting frame 12. The captured image data is sent to the image recognition unit 101.
The image recognition unit 101 detects the coordinates of the reference points G _D1 and G _D30 on the camera coordinates from the received image data, and sends them to the first correction amount calculation unit 107.
Next, the first correction amount calculation unit 107 compares the received coordinates of the reference points G _D1 and G _{D30 with} the coordinates of each camera position in the camera coordinate system stored in the fixed parameter storage unit 104 and shifts them. Find the amount.
For example, as illustrated in FIG. 17, the first correction amount calculation unit 107 determines that the reference point of the mounting frame 12 captured by the camera 2 is G _D1 (x _F1 , y _F1 , θ _F1 ) = (267.808, 245. 945,0.0000) and the center position G _D1 (x _F1 , y _F1 , θ _F1 ) = (314.082, 249.794) of the camera system reference mark 213 stored in the fixed parameter storage unit 104 .0000) and the difference (deviation amount) (x _MF1 , y _MF1 , θ _MF1 ) = (− 46.274, −3.849, 0.0000) is obtained. Similarly, the first correction amount calculation unit 107 determines that the reference point of the mounting frame 12 captured by the camera 3 is G _D30 (x _F30 , y _F30 , θ _F30 ) = (243.643, 193.202, 0.0000. ) And the center position G _D20 (x _F30 , y _F30 , θ _F30 ) = (297.818, 193.202, 0.0000) of the reference mark 213 of the camera system stored in the fixed parameter storage unit 104. Then, the difference (deviation amount) (x _MF2 , y _MF2 , θ _MF2 ) = (− 54.175, −5.080, 0.0000) is obtained.

Next, the first correction amount calculation unit 107 calculates the displacements (x _MF1 , y _MF1 , θ _MF1 ) of the resolutions M _F1 and M _{F2 of} each camera held in the fixed parameter 104 and the calculated mounting frame 12. Using the coordinates of (x _MF2 , y _MF2 , θ _MF2 ), the robot coordinates are converted using the following equations (11) to (16).
_{a 1 = (- y MF1 ·} M F1) cos (θ FF1) - (- x MF1 · M F1) sin (θ FF1) + x RF1 ··· formula (11)
_{b 1 = (- y MF1 ·} M F1) sin (θ FF1) - (- x MF1 · M F1) cos (θ FF1) + y RF1 ··· formula (12)
θ ₁ = Θ + θ _FF1 Formula (13)
_{a 30 = (- y MF2 ·} M F2) cos (θ FF2) - (- x MF2 · M F2) sin (θ FF2) + x RF2 ··· formula (14)
_{b 30 = (- y MF2 ·} M F2) sin (θ FF2) - (- x MF2 · M F2) cos (θ FF2) + y RF2 ··· (15)
θ ₃₀ = Θ + θ _FF2 (16)
For example, when the coordinates of the reference position imaged by the camera 2 are converted using the equations (14) to (16), a ₁ = 171.007 [mm], b ₁ = −58.974 [mm], θ ₁ = 0.0000 [rad]. Similarly, the coordinates of the reference position of the mounting frame 12 on the stage 7 captured by the camera 3 are also a ₃₀ = −171.0029 [mm], b ₃₀ = −249.237 [mm], and θ ₃₀ = 0, respectively. .0000 [rad].

Next, the first correction amount calculation unit 108 calculates the reference points (a ₁ , b ₁ ) and (a ₃₀ , b ₃₀ ) in the robot coordinates captured by the cameras 2 and 3 and the following formula ( The center of gravity of the mounting frame 12 in the robot coordinates is obtained using 17) to (19).
x _Frame = (a ₁ + a ₃₀ ) / 2 [mm] (17)
y _Frame = (b ₁ + b ₃₀ ) / 2 [mm] (18)
θ _Frame = −tan ⁻¹ ((a ₁ −a ₃₀ ) / (b ₁ −b ₃₀ )) [rad] Expression (19)
For example, _using Equation (13) ~ (15), x Flame = 171.018 [mm], y Flame = -153.105 [mm], θ Flame = 0.000116 [rad] is found. Centroid of the thus determined was mounting frame 12 specifically represent the _G between columns 15 and _{16 (x Flame, y Flame,} θ Flame).

Next, the first correction amount calculation unit 108 calculates the left position of the mounting frame 12 (position taken by the camera 2) (a ₁ , b ₁ , θ ₁ ) and the right position of the mounting frame 12 (camera 3 imaging _{position) (a 30, b 30,} θ 30) and the gravity center position _G (x _{Flame, y Flame,} using the value of θFlame), the inclination θ _i (1 ≦ _i ≦ in columns 1 to 30 of the mounting frame 12 30) is obtained using the following equation (20): θ _i = θ _Di + θ _Frame ... Equation (20)
For example, θ _i = −1.57068 is obtained.
Further, the coordinates (a _i , b _i ) of the columns 2 to 29 of the mounting frame 12 are calculated using the left and right coordinates (a ₁ , b ₁ ) and (a ₃₀ , b ₃₀ ) and the inclination θ _Flame .
As a result of these calculations, the corrected coordinates in each column of the mounting frame 12 are as shown in FIG.

FIG. 18 is a diagram illustrating an example of corrected coordinates in each column of the mounting frame 12 in the present embodiment. The center of gravity G _Di of each of the rows 1 to 30 of the mounting frame 12 is obtained. The data is corrected using the first correction amount.

Next, the first correction amount calculation unit 108 uses the value stored in the frame parameter storage unit 106 and the obtained deviation amount at the mounting position on the mounting frame 12 placed on the stage 7. A certain center of gravity is corrected.
First, the coordinates of the center of gravity of the mounting frame 12 that is the design value stored in the frame parameter storage unit 106 are (0, 0, −1.5708), and the position of the center of gravity obtained is (171.018). , −154.105, 0.000116). For this reason, the position of the center of gravity serving as a reference for the mounting position is obtained by adding both values (171.018, −154.105, −1.57068).
Next, based on this barycentric position, the barycentric value of each column of the mounting frame 12 stored in the frame / parameter storage unit 106 is corrected. As a result, the data (FIG. 6B) stored in the frame parameter storage unit 106 is corrected as shown in FIG.
Thus, the position correction process of the mounting frame 12 that is the first correction process is completed.

Next, returning to FIG. 6, the second correction process, which is the connector position correction, is performed (S14). Hereinafter, the connector position correction process will be described with reference to FIGS. FIG. 19 is a schematic diagram for explaining the conversion of the reference point of the connector 11 captured by the camera 4 in the present embodiment.
First, the arm drive unit 8 drives the mounting robot 1 so as to attract the connector 11 from the connector molding device 9 according to an instruction from the control unit 5.
Next, the suction portion 212 of the attachment portion 211 attached to the arm 8 sucks the connector 11 from the connector molding device 9. After attracting the connector 11, the arm driving unit 8 is driven by the instruction from the control unit 5 so that the centers of gravity of the plurality of connectors 11 attracted to the center position of the camera 4 come. For example, as shown in FIG. 7, when 10 connectors 11 are attracted to the attracting portion 212, G _C (x _C , y _C , θ _C ), which is the center of gravity of the 10 connectors 11, The arm driving unit 8 drives the arm 8 in accordance with an instruction from the control unit 5 at a predetermined position at the center.
Next, the camera 4 images the connector 11 sucked by the suction unit 212. Next, the image recognition unit 102 receives an image captured by the camera 4 and detects the barycentric coordinates of each connector 311. The second correction amount calculation unit 109 receives the detected center-of-gravity coordinates of each connector 311 and further obtains the center-of-gravity coordinates of the connector 11 including ten connectors 311. For the barycentric coordinates, for example, the least square method is performed on all barycentric coordinates of each connector 311 to calculate a linear equation y = a · x + b (a and b are constants). Next, the second correction amount calculation unit 109 calculates an average value of all x and y of each connector 311 and further substitutes it into the calculated linear equation to determine the barycentric coordinates of the connector 11. Further, the calculated inclination a of the linear equation y = a · x + b is the inclination θ of the center of gravity of the connector 11.
For example, as shown in FIG. 19, barycentric coordinates (x _FCT , y _FCT , θ _FCT ) = (506.790 [pixel], 382.060 [pixel], −0.180 [deg]) are obtained.
Next, the second correction amount calculation unit 109 reads the position of the reference mark 213 stored in the fixed parameter storage unit 104 and calculates the amount of deviation from the obtained barycentric coordinates.
For example, as shown in FIG. 19, the barycentric coordinates (x _MCT , y _MCT , θ _MCT ) = (− 1.391 [mm], −30.659 [mm], −0.0031 [rad]) are obtained.
Next, the second correction amount calculation unit 109 converts the deviation amount into the robot coordinate system using the following equations (21) to (18). Further, in the equations (16) to (18), the calculation is performed by replacing the tilt 90 ° of the camera 4 capturing the connector 12 with 0 °. In equations (21) to (23), M _CT is the resolution of the camera 4 and θ _RCT is the inclination of the center of gravity of the reference mark stored in the fixed parameter storage unit 104 (the inclination of the camera 4).
x _CT = ((− y _MCT ) cos (θ _RCT + π / 2) − (− x _MCT ) sin (θ _RCT + π / 2)) · M _CT (21)
y _CT = ((− y _MCT ) sin (θ _RCT + π / 2) − (− x _MCT ) cos (θ _RCT + π / 2)) · M _CT (22)
θ _CT = Θ + θ _RCT + π / 2 Formula (23)
For example, as a result of conversion using the equations (21) to (23), (x _CT , y _CT , θ _CT ) = (0.883, 0.040, −0.0031) is obtained as shown in FIG. It is done.
Next, the second correction amount calculation unit 109 uses the expressions (24) to (26) to calculate the deviation amount between the attachment part 211 attached to the arm 8 and the adjustment jig 13 used in the calibration process. to correct. As shown in FIG. 10, (x _OC , y _OC , θ _OC ) is the difference between the mounting part 211 and the adjustment jig stored in the fixed parameter storage part 104.
Δa = −x _CT + x _OC Expression (24)
Δb = −y _CT + y _OC Expression (25)
Δθ = −θ _CT + θ _OC Expression (26)
For example, as a result of correction using the equations (24) to (26), (Δa, Δb, Δθ) = (− 0.613, −0.040, 0.0031) is obtained as shown in FIG.
Thus, the position correction process of the connector 11 as the second correction process is completed.

Next, returning to FIG. 6, the mounting coordinate calculation unit 110 performs comprehensive correction processing that is position correction for mounting the connector 11 on the mounting frame (S <b> 15). The mounted coordinate calculation unit 110 uses the values obtained in the first correction process, the second correction process, and the third correction process so far, and uses the expressions (31) to (33) to (x, y, θ). Offset amounts (Δx _i , Δy _i , Δθ _i ) are obtained. As shown in FIG. 18, i (1 ≦ i ≦ 30) is a variable and represents each column on the mounting frame 12, and θi represents the inclination of the center of gravity (a _i , bi) in each column. Further, (ΔH _xi , ΔH _yi ) represents a value obtained by adding the inclination displacement amount θi of the mounting frame 12 actually mounted to the displacement amount of the attracted connector 11 in the attachment portion 211 attached to the arm 8, (ΔR _xi , ΔR _yi ) represents a correction value of the mounting position of the connector 11 in consideration of the arm length L.
ΔH _xi = Δa · cos (θ _i + Δθ) −Δb · sin (θ _i + Δθ) (27)
ΔH _yi = Δa · sin (θ _i + Δθ) + Δb · cos (θ _i + Δθ) (28)
ΔR _xi = −L · (cos (θ _i + Δθ) −cos (θ _i )) (29)
ΔR _yi = −L · (sin (θ _i + Δθ) −sin (θ _i )) (30)
For example, ΔH _xi = −0.042 [mm], ΔH _yi = 0.613 [mm], ΔR _xi = −0.3523 [mm], and ΔR _yi = −0.0006 [mm] are calculated.
Further, the mounting offset amount is calculated by equations (31) to (33) using the values calculated by equations (27) to (30).
Δx _i = ΔH _xi + ΔR _xi (31)
Δy _i = ΔH _yi + ΔR _yi (32)
Δθ _i = Δθ Formula (33)
For example, Δx _i = −0.394 [mm], Δy _i = 0.613 [mm], and Δθ _i = 0.0031 [rad] are calculated.

Next, the mounting coordinate calculation unit 110 uses the calculated mounting offset amount (Δx _i , Δy _i , Δθ _i ) and the coordinates (a _i , b _i , θ _i ) of each column of the mounting frame 12 to Equations (34) to (36) are used to calculate the coordinates (x _i , y _i , Θ _i ) of the mounting position of the connector 11 on each column of the mounting frame 12 instructed to the mounting robot.
x _i = a _i + Δx _i Expression (34)
y _i = b _i + Δy _i Expression (35)
Θ _i = θ _i + Δθ _i (36)
For example, when Expressions (34) to (36) are calculated for the first row of the mounting frame 12, x _i = 170.0613 [mm], y _i = −54.892 [mm], Θ _i = −1.56758 [Deg] (= −89.616 [deg]) is calculated.
FIG. 20 shows the coordinates of the mounting position (x _i , y _i , Θ _i ) on each column of the mounting frame 12 instructed to the mounting robot in the present embodiment calculated using the equations (34) to (36). It is a figure which shows an example.
The mounting position calculation process is thus completed.

  When the present invention is implemented, for example, in the conventional component mounting apparatus, the mounting accuracy for mounting the connector 11 on the mounting frame 12 is ± 0.06 [mm], but the component mounting apparatus of the present embodiment is used. As a result, the mounting accuracy was improved to ± 0.03 [mm].

As described above, first, calibration processing is performed, and the image recognition units 101 and 102 detect the coordinates of the reference points from the images captured by the cameras 2 to 4, and the detected coordinates and reference points of the respective reference points are detected. The inclination with respect to the direction is stored in the fixed parameter storage unit 104. Next, the reference position coordinate conversion unit 103 converts the coordinates detected by the image recognition units 101 and 102 into robot coordinates and stores them in the fixed parameter storage unit 104. In addition, the third correction amount calculation unit 109 uses the coordinates of the reference point held in the fixed parameter storage unit 104 during the calibration process, and the inclination and arm of the arm 8 as the third deviation amount with respect to the reference direction. 8 is obtained and stored in the fixed parameter storage unit 104.
Further, the first correction amount calculation unit 107 preliminarily stores the values stored in the frame parameter storage unit 106 in advance, the values stored in the fixed parameter storage unit 104, and the left and right in the mounting frame 12 on the rail 7. The coordinates detected from the captured image of the determined part position are compared with the robot coordinates, and the displacement amount of the mounting frame 12 as the first displacement amount is calculated.
Further, the second correction amount calculation unit 108 sucks the connector 11 and then the values held in the fixed parameter storage unit 104 and the centroids of the plurality of connectors 11 detected from the image captured by the camera 4. The position is compared with the robot coordinates, and a deviation amount with respect to the connector 11 as the second deviation amount is calculated.
Then, the mounting coordinate calculation unit 110 calculates the shift amount (first shift amount) with respect to the mounting frame 12 calculated by the first correction amount calculation unit 107 and the connector 11 calculated by the second correction amount calculation unit 108. The position at which the connector 11 is mounted on the mounting frame 12 using the shift amount (second shift amount) and the shift amount (third shift amount) with respect to the arm 8 calculated by the third correction amount calculation unit 109. And the arm 8 is driven to the calculated mounting position by the arm driving unit 6 so that the connector 11 can be mounted on the mounting frame with high accuracy.

  Further, in the present embodiment, the example in which the calibration jig 13 is mounted on the arm 8 and the calibration process is performed has been described. However, the reference mark 213 is provided in the mounting portion 211 to be mounted on the arm 8, and the reference mark is set to each The same effect can be obtained by detecting with the cameras 2 to 4.

  Furthermore, in the present embodiment, the example in which the mounting frame 12 is configured by a plurality of child components 301 has been described, but the mounting frame 12 may be configured by one or more components. In this case, the number of connectors 11 to be mounted may be the same as the number of components of the mounting frame.

  Furthermore, in the present embodiment, for the mounting frame 12 composed of a plurality of child parts 311, the leftmost column (first column) and the first column are used as reference point positions during calibration processing and mounting, respectively. Although an example using the right column (30th column) has been described, the position of the reference point is not limited to this, and may be another column that can be imaged by the cameras 2 and 3, and further, The same effect can be obtained even between.

  Furthermore, in the present embodiment, the method of imaging the mounting frame 12 using the cameras 2 and 3 has been described. However, depending on the size of the mounting frame 12 and the resolution of the camera, a single camera may be used.

  Note that all or some of the functions in FIG. 2 of the embodiment are implemented by a ROM (Read Only Memory), HDD (Hard Disk Drive), or HDD connected to a CPU (Central Processing Unit) (not shown) in the control unit 5. It can also be executed by a program stored in a USB memory or the like connected via a USB (Universal Serial Bus) I / F.

DESCRIPTION OF SYMBOLS 1 ... Mounting robot 2, 3, 4 ... Camera 5 ... Control unit 6 ... Arm drive part 7 ... Stage 8 ... Arm 9 ... Connector shaping | molding apparatus 11 ... Connector 12 ... Mounting frame 13 ... Calibration jigs 101, 102 ... Image recognition unit 103 ... Reference position coordinate conversion unit 104 ... Fixed parameter storage unit 105 ... Calibration unit 106 ... Frame parameter storage unit 107 ... First correction amount calculation unit 108 ... Second correction amount calculation unit 109 ... Third correction amount calculation unit 110 ... Mounted coordinate calculation unit

Claims (4)

In an electronic component mounting apparatus that mounts a second component on a first component using a movable arm,
A first deviation amount calculation unit that calculates a first deviation amount indicating how much the first component is deviated from a predetermined reference state based on an image obtained by imaging the first component;
A second shift for calculating a second shift amount indicating how much the second component is shifted from a predetermined reference state based on an image obtained by imaging the movable arm in a state where the second component is held. A quantity calculator;
A third deviation amount calculation unit that calculates a third deviation amount indicating how much the length and the inclination angle of the movable arm are deviated from a predetermined reference value based on an image obtained by imaging the movable arm;
The movable arm corrects the position where the second component is mounted on the first component based on the first deviation amount, the second deviation amount, and the third deviation amount, and the movable arm An arm drive unit for driving
A component mounting apparatus comprising: The component mounting apparatus according to claim 1, wherein the first deviation amount calculation unit calculates the first deviation amount in a coordinate system of the movable arm. The third deviation amount calculation unit has a predetermined length and inclination angle of the movable arm based on information obtained by extracting the position and angle of the movable arm from images obtained by imaging the movable arm at at least three different angles. The component mounting apparatus according to any one of claims 1 to 2, wherein a third shift amount representing how much the shift is from a reference value is calculated. In the component mounting method of mounting the second component on the first component using the movable arm,
A first deviation amount calculating step of calculating a first deviation amount indicating how much the first component is deviated from a predetermined reference state based on an image obtained by imaging the first component;
A second shift for calculating a second shift amount indicating how much the second component is shifted from a predetermined reference state based on an image obtained by imaging the movable arm in a state where the second component is held. A quantity calculation step;
A third deviation amount calculating step for calculating a third deviation amount indicating how much the length and the inclination angle of the movable arm are deviated from a predetermined reference value based on an image obtained by imaging the movable arm;
The movable arm corrects the position where the second component is mounted on the first component based on the first deviation amount, the second deviation amount, and the third deviation amount, and the movable arm An arm driving process for driving
A component mounting method characterized by comprising:

Images