阅读说明：本技术 一种晶圆自动纠偏方法 (Automatic deviation rectifying method for wafer ) 是由 徐方 凌霄 杨奇峰 谭学科 朱洪彪 姚东东 于 2019-11-08 设计创作，主要内容包括：本发明属于真空机器手运输晶圆过程中自动校正手指上晶圆位置的控制技术,具体涉及一种晶圆自动纠偏(AWC)方法,主要包含AWC方法的实现方式以及AWC算法。本发明包括以下步骤：AWC数据采集：采集晶圆通过激光传感器时,激光传感器所记录的坐标；AWC标定：通过机器手手指中心坐标以及晶圆半径,计算得到激光传感器在机器手基坐标下的坐标,用于以此坐标为基准进行自动纠偏功能；AWC偏差计算：通过激光传感器在机器手基坐标下的坐标、手指中心在机器手基坐标下的坐标、晶圆中心的坐标以及触发点相对于机器手基坐标原点的夹角,计算得到偏差值。本发明实现了机器手在传输晶圆过程中的自纠偏功能,保证了机器手传输晶圆的精度。(The invention belongs to a control technology for automatically correcting the position of a wafer on a finger in the process of transporting the wafer by a vacuum robot, and particularly relates to an automatic wafer deviation correcting (AWC) method which mainly comprises an implementation mode of the AWC method and an AWC algorithm. The invention comprises the following steps: AWC data acquisition: collecting coordinates recorded by a laser sensor when a wafer passes through the laser sensor; AWC calibration: calculating to obtain the coordinate of the laser sensor under the robot hand base coordinate through the robot hand finger center coordinate and the wafer radius, and performing an automatic deviation rectifying function by taking the coordinate as a reference; AWC deviation calculation: and calculating to obtain a deviation value through the coordinate of the laser sensor under the robot base coordinate, the coordinate of the finger center under the robot base coordinate, the coordinate of the wafer center and an included angle of the trigger point relative to the original point of the robot base coordinate. The wafer self-correcting device realizes the self-correcting function of the robot in the wafer transmission process, and ensures the precision of the robot in wafer transmission.)

1. An automatic deviation rectifying method for a wafer is characterized by comprising the following steps:

1) AWC data acquisition: collecting the center coordinates of the fingers of the robot hand recorded by the laser sensor when the wafer passes through the laser sensor;

2) AWC calibration: calculating to obtain the coordinate of the laser sensor under the robot hand base coordinate through the robot hand finger center coordinate and the wafer radius, and performing an automatic deviation rectifying function by taking the coordinate as a reference;

3) the AWC data was collected again: collecting the center coordinates of the fingers of the robot hand recorded by the laser sensor when the wafer passes through the laser sensor;

4) AWC deviation calculation: and obtaining a deviation value through the coordinate of the laser sensor under the robot base coordinate, the coordinate of the finger center under the robot base coordinate, the coordinate of the wafer center and the included angle of the trigger point relative to the original point of the robot base coordinate, so as to realize the automatic deviation rectifying function of the robot.

2. The automatic deviation rectifying method for the wafer according to claim 1, wherein the step 1) is as follows: the wafer blocks the laser sensor A, B and 4 sets of coordinates of the finger center of the robot hand from t1 to t4 at 4 triggering times generated when the robot hand leaves the laser sensor A, B during the wafer transferring process.

3. The method as claimed in claim 2, wherein the positions of the laser sensors A and B are not symmetrical with respect to the movement track of the robot.

4. The automatic deviation rectifying method for the wafer according to claim 1, wherein the step 2) is as follows: calculating the coordinate of the laser sensor A or B under the machine hand base coordinate through the coordinate obtained in the step 1) and the wafer radius R, wherein the calculation formula is as follows:

(x2-a)^2+(y2-b)^2＝R^2 (1)

(x3-a)^2+(y3-b)^2＝R^2 (2)

finishing to obtain:

(C2^2+1)*b^2+(2*x2*C2-2*C1*C2-2*y2)*b+x2^2-2*x2*C1+C1^2+y2^2-R^2＝0

C1＝(x3^2-x2^2+y3^2-y2^2)/(2*(x3-x2))

C2＝(y3-y2)/(x3-x2)

solving the formulas (1) and (2) can obtain the value of b, and substituting the formula can obtain the value of a;

wherein, (x2, y2), (x3, y3) are the center O of the robot finger respectively₂、O₃Coordinates under the robot hand base coordinate, (a, B) are coordinates of the laser sensor a or B under the robot hand base coordinate, and C1, C2 are constants converted from the coordinates acquired in step 1).

5. The automatic deviation rectifying method for the wafer according to claim 1, wherein the step 4) is as follows: calculating to obtain an offset value according to the coordinates of the laser sensor A, B under the robot base coordinate obtained in the step 2), the coordinates of the centers of the 4 groups of robot fingers obtained in the step 3), the coordinates of the wafer center at the time of t 1-t 4 and an included angle of the trigger point relative to the robot base coordinate origin at the time of t 1-t 4, wherein the calculation formula is as follows:

X1＝x1+d*cos(θ1+ψ-π/2)

Y1＝y1+d*sin(θ1+ψ-π/2)；

X2＝x2+d*cos(θ2+ψ-π/2)

Y2＝y2+d*sin(θ2+ψ-π/2)

X3＝x3+d*cos(θ3+ψ-π/2)

Y3＝y3+d*sin(θ3+ψ-π/2)

X4＝x4+d*cos(θ4+ψ-π/2)

Y4＝y4+d*sin(θ4+ψ-π/2)

(X1-Xa)^2+(Y1-Ya)^2＝R^2 ①

(X2-Xb)^2+(Y2-Yb)^2＝R^2 ②

(X3-Xb)^2+(Y3-Yb)^2＝R^2 ③

(X4-Xa)^2+(Y4-Ya)^2＝R^2 ④

the coordinates (a, b) of the laser sensor A, B under the robot base coordinate system are respectively (Xa, Ya), (Xb, Yb), R is the wafer radius, (X1, Y1), (X2, Y2), (X3, Y3), (X4, Y4) are the coordinates of the robot finger center at times t1 to t4, (X1, Y1), (X2, Y2), (X3, Y3), (X4, Y4) are the coordinates of the wafer center at times t1 to t4, θ 1, θ 2, θ 3, θ 4 are the included angles of the trigger points at times t1 to t4 relative to the robot base coordinate system origin, d is the deviation distance between the wafer center and the finger center, and ψ is the deviation angle between the wafer center and the finger center relative to the finger coordinate system;

and solving a group of deviation values d and psi according to any three groups of equations from the first to the fourth to obtain 4 groups, respectively calculating the distance between the wafer center (X1, Y1) and the sensor A (Xa, Ya) at the triggering moment by using the 4 groups of deviation values and the corresponding trigger points to be equivalent to the calculated value of the radius of the wafer, selecting a group of deviation values with the calculated radius of the wafer being closest to the actual radius of the wafer as deviation-correcting deviations, and converting the deviations into joint deviation values to correct the deviation.

Technical Field

The invention belongs to a control technology for automatically correcting the position of a wafer on a finger in the process of transporting the wafer by a vacuum robot, and mainly comprises an AWC method implementation mode and an AWC algorithm.

Background

The domestic robot hand does not have the function of automatically adjusting the position of the wafer on the finger of the robot hand in the process of transmitting the wafer, but the wafer deviates from the center position of the finger of the robot hand to cause the phenomenon of wafer falling or wafer collision in practical application after multiple transmissions, so that the robot hand needs to develop a function of automatically correcting the position of the wafer on the finger in order to ensure that the wafer is always kept within the allowable error range of the center of the finger in the process of transmitting the wafer.

Disclosure of Invention

The invention realizes the method for automatically correcting the position of the wafer on the finger in the process of transferring the wafer by the robot hand.

The technical scheme adopted by the invention for realizing the purpose is as follows:

an automatic deviation rectifying method for a wafer comprises the following steps:

1) AWC data acquisition: collecting the center coordinates of the fingers of the robot hand recorded by the laser sensor when the wafer passes through the laser sensor;

2) AWC calibration: calculating to obtain the coordinate of the laser sensor under the robot hand base coordinate through the robot hand finger center coordinate and the wafer radius, and performing an automatic deviation rectifying function by taking the coordinate as a reference;

3) the AWC data was collected again: collecting the center coordinates of the fingers of the robot hand recorded by the laser sensor when the wafer passes through the laser sensor;

4) AWC deviation calculation: and obtaining a deviation value through the coordinate of the laser sensor under the robot base coordinate, the coordinate of the finger center under the robot base coordinate, the coordinate of the wafer center and the included angle of the trigger point relative to the original point of the robot base coordinate, so as to realize the automatic deviation rectifying function of the robot.

The step 1) is as follows: the wafer blocks the laser sensor A, B and 4 sets of coordinates of the finger center of the robot hand from t1 to t4 at 4 triggering times generated when the robot hand leaves the laser sensor A, B during the wafer transferring process.

The positions of the laser sensors A and B cannot be symmetrical relative to the movement track of the robot hand.

The step 2) is as follows: calculating the coordinate of the laser sensor A or B under the machine hand base coordinate through the coordinate obtained in the step 1) and the wafer radius R, wherein the calculation formula is as follows:

(x2-a)^2+(y2-b)^2＝R^2 (1)

(x3-a)^2+(y3-b)^2＝R^2 (2)

finishing to obtain:

(C2^2+1)*b^2+(2*x2*C2-2*C1*C2-2*y2)*b+x2^2-2*x2*C1+C1^2+y2^2-R^2＝0

C1＝(x3^2-x2^2+y3^2-y2^2)/(2*(x3-x2))

C2＝(y3-y2)/(x3-x2)

solving the formulas (1) and (2) can obtain the value of b, and substituting the formula can obtain the value of a;

wherein, (x2, y2), (x3, y3) are the center O of the robot finger respectively₂、O₃Coordinates under the robot hand base coordinate, (a, B) are coordinates of the laser sensor a or B under the robot hand base coordinate, and C1, C2 are constants converted from the coordinates acquired in step 1).

The step 4) is as follows: calculating to obtain an offset value according to the coordinates of the laser sensor A, B under the robot base coordinate obtained in the step 2), the coordinates of the centers of the 4 groups of robot fingers obtained in the step 3), the coordinates of the wafer center at the time of t 1-t 4 and an included angle of the trigger point relative to the robot base coordinate origin at the time of t 1-t 4, wherein the calculation formula is as follows:

X1＝x1+d*cos(θ1+ψ-π/2)

Y1＝y1+d*sin(θ1+ψ-π/2)；

X2＝x2+d*cos(θ2+ψ-π/2)

Y2＝y2+d*sin(θ2+ψ-π/2)

X3＝x3+d*cos(θ3+ψ-π/2)

Y3＝y3+d*sin(θ3+ψ-π/2)

X4＝x4+d*cos(θ4+ψ-π/2)

Y4＝y4+d*sin(θ4+ψ-π/2)

(X1-Xa)^2+(Y1-Ya)^2＝R^2 ①

(X2-Xb)^2+(Y2-Yb)^2＝R^2 ②

(X3-Xb)^2+(Y3-Yb)^2＝R^2 ③

(X4-Xa)^2+(Y4-Ya)^2＝R^2 ④

the coordinates (a, b) of the laser sensor A, B under the robot base coordinate system are respectively (Xa, Ya), (Xb, Yb), R is the wafer radius, (X1, Y1), (X2, Y2), (X3, Y3), (X4, Y4) are the coordinates of the robot finger center at times t1 to t4, (X1, Y1), (X2, Y2), (X3, Y3), (X4, Y4) are the coordinates of the wafer center at times t1 to t4, θ 1, θ 2, θ 3, θ 4 are the included angles of the trigger points at times t1 to t4 relative to the robot base coordinate system origin, d is the deviation distance between the wafer center and the finger center, and ψ is the deviation angle between the wafer center and the finger center relative to the finger coordinate system;

and solving a group of deviation values d and psi according to any three groups of equations from the first to the fourth to obtain 4 groups, respectively calculating the distance between the wafer center (X1, Y1) and the sensor A (Xa, Ya) at the triggering moment by using the 4 groups of deviation values and the corresponding trigger points to be equivalent to the calculated value of the radius of the wafer, selecting a group of deviation values with the calculated radius of the wafer being closest to the actual radius of the wafer as deviation-correcting deviations, and converting the deviations into joint deviation values to correct the deviation.

The invention has the following beneficial effects and advantages:

the wafer self-correcting device realizes the self-correcting function of the robot in the wafer transmission process, and ensures the precision of the robot in wafer transmission.

Drawings

FIG. 1 is a scheme flow diagram of the present invention;

FIG. 2 is a schematic diagram of AWC data acquisition;

FIG. 3 is a schematic diagram of the AWC calibration principle;

FIG. 4 is a schematic diagram of wafer misalignment.

Detailed Description

The key points of the technology of the invention are as follows:

1. two sensors are mounted on the station.

2. The robot arm triggers the sensor during the telescoping process.

3. And calculating the calibration position through the coded disc value recorded at the triggering moment.

4. And calculating the deviation value between the wafer and the center of the finger according to the code disc value recorded at the triggering moment.

5. The robot automatically corrects the wafer position using the deviation.

The present invention will be described in further detail with reference to the accompanying drawings and examples.

As shown in fig. 1, the general scheme of AWC (active wafer centering) function includes: AWC data acquisition, AWC calibration and AWC deviation calculation.

As shown in fig. 2, AWC data is collected by two laser sensors, the robot arm will sequentially block the laser sensors A, B during wafer transportation, and the driver records the code wheel value of each axis of the robot arm at this moment while generating the trigger signal, and similarly, the trigger signal will also be generated and the code wheel value of each axis will be recorded when the wafer leaves the sensors. The robot hand records four code wheel values in total during the passage of the two laser sensors.

The robot hand needs to carry out AWC calibration before realizing the AWC deviation rectifying function, the AWC calibration aims at recording the standard position of the wafer on the finger of the robot hand, and the automatic deviation rectifying function is carried out by taking the position as the reference. In the AWC calibration process, the robot can record the code disc values of all axes at four triggering moments, and then transmits data to the robot controller through the bus. And the robot hand calculates the coordinate value of the sensor relative to the base coordinate of the robot hand at the moment through a calibration algorithm.

After calibration is completed, when the robot carries the wafer again and transmits the wafer to the designated station, the robot records the code disc values of all axes triggered four times again, and then transmits data to the robot controller through the bus. The robot hand calculates the deviation value between the wafer and the finger center of the robot hand by using the code disc values of all the axes at the 4-time triggering moment, and then converts the deviation into the joint deviation value of each axis, thereby realizing the automatic deviation rectifying function of the robot hand.

AWC calibration principle:

as shown in fig. 3, the solid black box is the position of the sensor A, B, which requires that the sensor position is not symmetrical with respect to the robot motion trajectory, ensuring that there is a time interval between the two triggers.

The calibration aims to calculate the coordinate value of the sensor under the robot base coordinate by recording the code disc value of the finger center twice and the wafer radius R. The principle is as follows: the center of the circle is found by two points and the radius on the known circle as the dotted line circle. The code wheel value of the center of the robot finger recorded when the wafer triggers the sensor B for the first time is O2, the distance from O2 to the sensor B is the wafer radius R at this time, the code wheel value of the center of the robot finger recorded when the wafer leaves the sensor B is O3, the distance from O3 to the sensor B is the wafer radius R at this time, and O2 and O3 can be regarded as two points on a circle (i.e., a dashed line circle) with the sensor B as the center and R as the radius. Knowing the coordinates and radius of the two points on the circle, the coordinates (a, B) of sensor B can be found. The coordinates of sensor a can be found in the same way.

Let the coordinate values of O2 and O3 under the robot hand base coordinate be (x2, y2), (x3, y3), the wafer radius be R, and the center of the dotted line (i.e., the coordinate of sensor B) be (a, B).

The coordinate values of a and b can be obtained by the following formula:

(x2-a)^2+(y2-b)^2＝R^2

(x3-a)^2+(y3-b)^2＝R^2

finishing to obtain:

(C2^2+1)*b^2+(2*x2*C2-2*C1*C2-2*y2)*b+x2^2-2*x2*C1+C1^2+y2^2-R^2＝0

C1＝(x3^2-x2^2+y3^2-y2^2)/(2*(x3-x2))

C2＝(y3-y2)/(x3-x2)

solving a quadratic equation with one element can solve the value of b, substituting the quadratic equation into a formula can solve the value of a, and solving the coordinate value of the sensor A in the same way.

Principle of deviation calculation

The coordinate values of the two sensors under the base coordinate can be determined A, B through calibration, the sensors are triggered by the wafer 4 times during rectification, because the wafer has deviation on the finger, the coordinate value of the center of the finger recorded at the moment is not the coordinate value of the center of the finger at the calibration moment, and the deviation distance between the center of the wafer and the center of the finger is assumed to be d, and the angle is psi (the included angle between the line segment from the center of the wafer to the center of the finger and the positive direction of the x axis of the finger coordinate system). As shown in fig. 4. The coordinate value of the center of the wafer can be calculated through 4 coordinate values recorded during triggering, and the coordinates of the two sensors are set to be A (Xa, Ya) and B (Xb, Yb); the radius is R; the coordinate values of the finger centers of the 4 triggering sensors are a (X1, Y1), B (X2, Y2), C (X3, Y3) and D (X4, Y4), and the coordinate values of the corresponding wafer centers at the corresponding time are A (X1, Y1), B (X2, Y2), C (X3, Y3) and D (X4, Y4). Theta 1, theta 2, theta 3 and theta 4 are included angles between 4 trigger points a, B, c and d and a robot base coordinate (the included angles are included angles between the trigger points and the robot base coordinate origin and the positive direction of the x axis of the robot base coordinate), according to physical significance, the distance from the circle centers of A, D to the sensor A is the wafer radius R, and the distance from the circle centers of B, C to the sensor B is the wafer radius R. The standard equation for 4 circles can be listed. The formula is as follows:

X1＝x1+d*cos(θ1+ψ-π/2)

Y1＝y1+d*sin(θ1+ψ-π/2)；

X2＝x2+d*cos(θ2+ψ-π/2)

Y2＝y2+d*sin(θ2+ψ-π/2)

X3＝x3+d*cos(θ3+ψ-π/2)

Y3＝y3+d*sin(θ3+ψ-π/2)

X4＝x4+d*cos(θ4+ψ-π/2)

Y4＝y4+d*sin(θ4+ψ-π/2)

(X1-Xa)^2+(Y1-Ya)^2＝R^2 ①

(X2-Xb)^2+(Y2-Yb)^2＝R^2 ②

(X3-Xb)^2+(Y3-Yb)^2＝R^2 ③

(X4-Xa)^2+(Y4-Ya)^2＝R^2 ④

the method comprises the steps of solving a group of deviation values d and psi according to equations (i), (ii) and (iv), solving a group of deviation values d and psi according to equations (i), (iii) and (iv), calculating a calculated value of a radius at a trigger time by using 4 groups of deviation values and corresponding trigger points respectively, selecting a group of deviation values calculated to be the closest to the radius of a wafer as deviation-correcting deviations, and converting the deviations into joint deviation values to correct the deviation.