JPH06126671A - Robot positioning method - Google Patents

Abstract

PURPOSE:To secure a robot positioning method that is able to extremely reduce any mounting dislocation to be produced by the expansion and head inclination of a robot in an electronic parts mounted machine. CONSTITUTION:This method consists of a teaching process preteaching a reference mark 4 serving as a basis for discriminating any dislocation of a robot position, a recognizing process recognizing the reference lark 4 with a base plate recognizing camera 3 mounted on a head 2 each time a continuous position operation is carried out for a specified period, a calculating process calculating a compensation value of the robot position from a difference between the data recognized by the camera 3 and the teaching data, a renewal process renewing the compensation value to the latest value and a positioning process setting the specified position on the basis of compensation.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a robot positioning method based on mark recognition by an electronic component mounting machine for mounting electronic components on a printed circuit board.

[0002]

2. Description of the Related Art In recent years, in electronic component mounters in the form of orthogonal robots, there are many robot positioning methods in which a linear scale or the like is installed to reduce the amount of expansion and contraction of the robot in order to reduce the mounting position displacement due to expansion and contraction of the robot. It is used.

An example of expansion / contraction correction of a robot in a conventional electronic component mounting machine will be described with reference to FIGS. 5 and 6. In general, an electronic component mounter in the form of an orthogonal robot includes a robot arm 7 that moves in the front-back direction, a robot arm 8 that supports the robot arm 7, and a robot arm 7 that is attached to the robot arm 7 and moves in the left-right direction to print electronic components. It has a suction head 2 to be mounted on a substrate 6, and the head 2 is equipped with a substrate recognition camera 3 for recognizing a substrate mark 9 on the printed substrate 6 in order to correct the positional deviation of the printed substrate 6. ing. As a means for correcting the expansion and contraction of the robot, a linear scale 10 is provided parallel to the robot arm 7, and the head 2 is a linear scale.
It will work along 10.

The head 2 reads the position on the linear scale 10 when operating in the left-right direction, and sends the movement amount data to the control unit 5 provided in the mounting machine. The control unit 5 obtains the amount of expansion and contraction of the robot arm 7 by referring to the sent data of the movement amount. As a result of the robot arm 7 expanding and contracting due to temperature changes over time, the head 2 at the mounting position
When the deviation is 0.1 mm, there occurs a phenomenon that an IC with a lead pitch of 0.2 mm or less cannot be mounted as shown in FIG. Lead pitches for electronic components such as TAB (tablation) mounting are becoming narrower and lead pitches are being reduced to zero.
The chances of using ICs of 2 mm or less are increasing. Therefore, a means for compensating the mounting position displacement due to the expansion and contraction of the robot will be required more and more in the future.

[0005]

However, in the robot positioning method for correcting the amount of expansion and contraction of the robot as described above, when the linear scale itself expands and contracts, an accurate amount of expansion and contraction of the robot arm cannot be obtained. There was a problem that the accuracy of the mounting position deteriorates. In addition, since the positional deviation with respect to the case where the head is tilted in the left-right direction due to the bending of the robot arm is not taken into consideration, there is a problem that a deviation remains between the corrected mounting position and the actual mounting position of the head. is there.

Further, there is a problem in that the accuracy of mounting varies due to the change in expansion and contraction of the robot with time even if the same device performs the mounting operation. In view of the above problems, it is an object of the present invention to provide a robot positioning method capable of reducing a mounting position shift caused by expansion / contraction of a robot of an electronic component mounting machine and inclination of a head as much as possible.

[0007]

In order to solve the above problems, a robot positioning method according to the present invention is a method for picking up and holding a component by a head mounted on an orthogonal robot and positioning it at a predetermined position. A teaching step of teaching a reference mark for identifying a displacement of the robot position in advance, a recognition step of recognizing the reference mark by a camera mounted on the head every time a continuous positioning operation is performed for a certain period, and a camera A calculation process for calculating the correction amount of the robot position from the difference between the recognized data and the teaching data, an updating process for updating the correction amount to the latest value, and a positioning process for determining a predetermined position based on the correction amount are provided. It is a thing.

[0008]

According to the present invention, with the above-described steps, highly accurate positioning can be performed by using the recognition camera mounted on the head portion which actually performs the mounting operation in the orthogonal robot positioning. Further, by performing mark recognition every time a continuous positioning operation is performed for a certain period, it is possible to grasp the constantly changing expansion / contraction state of the robot and perform highly accurate positioning correction from the latest data. Furthermore, if a board recognition camera, which is standardly equipped with an electronic component mounter in the form of an orthogonal robot, is used as the recognition means, special equipment such as a linear scale can be used as standard equipment.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will be described with reference to FIGS. In FIG. 1, reference numeral 1 denotes an orthogonal robot, which recognizes a plurality of reference marks 4 installed below the head 2 by using a board recognition camera 3 mounted on a suction head 2 on which electronic components are mounted. The recognized data is sent to the control unit 5, the difference between the reference mark 4 and the teaching data is calculated, and the correction amount is obtained. The mounting position on the printed circuit board 6 is determined based on the obtained correction amount.

The positioning procedure of the first embodiment of the robot having the above construction will be described with reference to the flow chart shown in FIG. In step 1 (denoted as S1 in the drawing; the same applies hereinafter), teaching of a mark serving as a reference for the robot position is performed. In step 2, the camera mounted on the head recognizes the origin 0 of the robot and the fiducial mark A at the position moved by dx in the X direction from 0, and step 3
Then, as in step 2, the origin 0 and B at a distance dy moved from 0 in the Y direction are recognized. In step 4, the recognition data and the teaching data are compared, and the correction amounts tx, Y in the X direction are compared.
After calculating the direction correction amount ty, the correction amount obtained in step 5 is updated. In step 6, the positioning correction is performed based on the correction amount. X-axis positioning coordinate value Lx
Using the correction amount tx in the X direction obtained in step 4 and the movement amount dx at the time of mark recognition, correction is performed according to the formula (1).

[0011]

## EQU1 ## Lx = (1 + tx / dx) Lx The Y-axis positioning coordinate value Ly is similarly corrected. In step 7, the mounting operation is performed. In step 8, it is judged whether or not a condition for a certain period for performing mark recognition is reached (for example, whether a certain time has elapsed or whether a certain number of positioning operations have been completed). If YES, the process returns to step 2. Correction by mark recognition is performed, and if NO, the process returns to step 6 to perform positioning correction of the next component.

As described above, according to the first embodiment, high-accuracy positioning can be performed by using the recognition camera mounted on the head portion that actually performs the mounting operation in the orthogonal robot positioning. Further, by recognizing the mark and updating the correction amount at regular intervals, it is possible to perform correction using the latest correction amounts tx and ty with respect to a temporal change in the robot state. Further, by performing mark recognition by the board recognition camera mounted on the head, it is possible to perform highly accurate positioning without requiring special equipment.

Next, a second embodiment of the present invention will be described based on the flow chart shown in FIG. Step
At 11, teach the mark that serves as the reference for the robot position. At step 12, the camera mounted on the head recognizes the reference mark A at the position where the robot is moved from the origins 0 and 0 in the X direction by dx, and at step 13, as in step 12, the origins 0 and 0 are moved in the Y direction. Recognize B at the distance moved by dy. In step 14, the recognition data and the teaching data are compared with each other and the correction amount tx in the X direction and the correction amount ty in the Y direction are compared.
After calculating, the correction amount obtained in step 15 is updated. In step 16, the change tendency is learned from the change in the correction amount, and in step 17, the positioning correction is performed based on the correction amount. The X-axis positioning coordinate value Lx is corrected as shown in Formula 2 using the correction amount tx in the X direction obtained in step 14 and the movement amount dx at the time of mark recognition.

[0014]

## EQU2 ## Lx = (1 + tx / dx) Lx The Y-axis positioning coordinate value Ly is similarly corrected. Step
Mounting operation is performed at 18. In step 19, it is judged whether or not a condition for a fixed period for performing mark recognition has been reached (for example, whether a fixed time has elapsed, or whether positioning operation has been completed a fixed number of times), and if YES, step 12
Then, the correction based on the mark recognition is executed again. If NO, then in step 20, the correction amount is interpolated by judging the change tendency of the correction amount learned in step 16 and the elapsed time since the previous mark recognition, and the process returns to step 7 to position the next part. Make a correction.

As described above, according to the present embodiment, the mark recognition is performed at regular intervals to update the correction amount, and at the same time, the correction amounts tx and ty are interpolated from the tendency of change of the updated correction amount. It is possible to reduce the positional deviation caused by the change in the amount of expansion and contraction of the robot over time.

Next, a third embodiment of the present invention will be described based on the flowchart shown in FIG. Step
At 31, the mark that serves as the reference of the robot position is taught. In step 32, the camera mounted on the head recognizes the reference mark A at the position where the robot is moved from the origin 0 and 0 in the X direction by dx, and in step 33, as in step 32, the origin 0 and 0 is moved in the Y direction. Recognize B at the distance moved by dy. In step 34, the recognition data and the teaching data are compared with each other, and the correction amount tx in the X direction and the correction amount ty in the Y direction are compared.
After calculating, the correction amount obtained in step 35 is updated. In step 36, the change tendency is learned from the change of the correction amount, and in step 37, the change tendency of the correction amount is judged and the period until the next mark recognition is adjusted (for example, when the degree of change is large, the period is shortened. Yes). In step 38, positioning correction is performed based on the correction amount. The X-axis positioning coordinate value Lx is used as the correction amount t in the X direction obtained in step 34.
Using x and the amount of movement dx at the time of mark recognition, correction is performed as in Equation 3.

[0017]

## EQU3 ## Lx = (1 + tx / dx) Lx The Y-axis positioning coordinate value Ly is similarly corrected. Step
The mounting operation is performed at 39. In step 40, whether the condition of a certain period for performing mark recognition in consideration of the period until mark recognition adjusted in step 37 is reached (for example, whether a certain time has elapsed, or a certain number of positioning operations are completed) It is determined whether or not), and if YES, the process returns to step 32 and correction by mark recognition is executed. If NO, the process returns to step 38 to correct the positioning of the next part.

As described above, according to the present embodiment, the mark recognition is performed at regular intervals to update the correction amount, and at the same time, the mark recognition period is adjusted based on the tendency of the change in the updated correction amount. It is possible to reduce the positional deviation caused by the change in the amount of expansion and contraction of the robot over time.

[0019]

As is apparent from the above-described embodiments, according to the robot positioning method of the present invention, correction is performed by using the recognition camera mounted on the head portion that actually performs the mounting operation in the orthogonal robot positioning. Thereby, highly accurate positioning can be performed, and electronic components with a narrow pitch can be mounted with high accuracy. Further, by recognizing the mark and updating the correction amount at regular intervals, it is possible to minimize the positional deviation caused by the temporal change of the robot expansion / contraction amount.

[Brief description of drawings]

FIG. 1 is a perspective view showing a state of being set in an electronic component mounter of a robot positioning method according to an embodiment of the present invention.

FIG. 2 is a flowchart of a positioning method according to the first embodiment of the present invention.

FIG. 3 is a flowchart of a positioning method according to a second embodiment of the present invention.

FIG. 4 is a flowchart of a positioning method according to a third embodiment of the present invention.

FIG. 5 is a perspective view showing an overall configuration of a robot positioning method in a conventional example.

FIG. 6 is a schematic diagram showing a mounting position shift of an electronic component due to expansion and contraction of a robot in a conventional example.

[Explanation of symbols]

1 ... Cartesian robot, 2 ... Head, 3 ... Board recognition camera, 4 ... Reference mark, 5 ... Control part, 6 ... Printed board, 7, 8 ... Robot arm, 9 ... Board mark, 10 ... Linear scale, 12 ... Land, 13 ... Lead.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Office reference number FI technical display location H05K 13/08 Q8315-4E

Claims (5)

[Claims] 1. A method for picking up and holding a component by a head and positioning it at a predetermined position, which comprises a teaching process of teaching a reference mark for identifying a displacement of a robot position in advance and a continuous positioning operation. The recognition step of recognizing the reference mark by the camera mounted on the head every time period is performed, the calculation step of calculating the correction amount of the robot position from the difference between the data recognized by the camera and the teaching data, and the correction amount A robot positioning method comprising an updating step of updating to a value and a positioning step of determining a predetermined position based on a correction amount. 2. The recognizing step is a recognizing step of recognizing a reference mark by a camera mounted on a head every time a continuous positioning operation has elapsed for a certain period of time.
The described robot positioning method. 3. The robot positioning method according to claim 1, wherein the recognizing step is a recognizing step of recognizing the reference mark by a camera mounted on the head every time a continuous positioning operation is performed a predetermined number of times. 4. A method for picking up and holding a component by a head and positioning it at a predetermined position, which comprises a teaching step of previously teaching a reference mark for identifying a displacement of a robot position and a continuous positioning operation. The recognition step of recognizing the reference mark by the camera mounted on the head every time period is performed, the calculation step of calculating the correction amount of the robot position from the difference between the data recognized by the camera and the teaching data, and the correction amount The updating process of updating to the value, the learning process of learning the change tendency by the transition of the correction amount obtained for each mark recognition, and the correction amount by judging the learned change tendency and the elapsed time from the previous mark recognition. A robot positioning method comprising an interpolation step of interpolating and a positioning step of determining a predetermined position based on a correction amount. 5. A method of picking up and holding a component by a head and positioning it at a predetermined position, which comprises a teaching step of teaching a reference mark for identifying a displacement of a robot in advance and a continuous positioning operation. The recognition step of recognizing the reference mark by the camera mounted on the head every time period is performed, the calculation step of calculating the correction amount of the robot position from the difference between the data recognized by the camera and the teaching data, and the correction amount An updating step of updating to a value, a learning step of learning a changing tendency by the transition of the correction amount obtained for each mark recognition, and an adjusting step of adjusting the fixed period for performing the mark recognition by judging the changing tendency of the correction amount A robot positioning method comprising a positioning step of determining a predetermined position based on a correction amount.