JP2812716B2 - Sensor device for automatic alignment measurement - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial application field) The present invention mainly relates to automatic alignment for detecting a rotational direction position and an XY axis position of a workpiece such as a semiconductor wafer in an industrial production line. BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sensor device for measurement, and more particularly to a sensor device for automatic alignment measurement capable of high-speed and high-accuracy detection.

2. Description of the Related Art Conventionally, for example, when performing alignment of a workpiece (for example, a semiconductor wafer) in an industrial production line on a table plane in a rotational direction and an XY-axis direction, light from the workpiece is used as a television camera. And send the image data from this TV camera to the XY tracker, and move the table on which the work piece is placed according to the signal from this XY tracker to position the work piece in the rotational direction and the XY axis. I do it.

(Problems to be Solved by the Invention) Since the conventional sensor device for positioning using a television camera needs to scan on a two-dimensional screen, the capturing speed is 1/30 second or 1/60 second. Need. This is very slow while the servomotor used to move the table on which the workpiece is placed operates at a high speed of 1000 PPS or more (speed of 1000 pulses per second).

In order to increase the processing speed by the combination of the sensor device and the servomotor, a high-speed processing device is required to cover the low speed of taking in the sensor device, which increases the cost. In addition, the conventional sensor device requires a lot of processing time to increase the detection accuracy of the position in the rotation direction, and a difficult problem of processing time versus accuracy is always caused. In addition, the resolution of a TV camera is about 50
Assuming that one side of the field of view is 1 cm, for example, the resolution of one line is 20 microns by simply calculating, and the substantial accuracy is considered to be 40 to 50 microns.
This greatly reduces accuracy, for example, in detecting a separation line (dicing pattern) of a semiconductor wafer.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has a high data acquisition speed and a high processing speed, can keep costs low, and particularly has an automatic alignment measurement with high rotational direction position detection accuracy and high resolution. It is an object to provide a sensor device for use.

[Configuration of the invention]

(Means for Solving the Problems) In order to solve the above-mentioned problems, a sensor device for automatic alignment measurement according to the present invention includes a pair of one-dimensional optical sensors orthogonal to an optical path from an object and a positive side. While providing in parallel on the minus side, another pair of one-dimensional optical sensors orthogonal to the optical path and the pair of one-dimensional optical sensors are provided in parallel on the plus side and the minus side to constitute a sensor unit, and A signal from the paired one-dimensional optical sensor is input, and the averaged data of the pixel count at the change point and the averaged data on the plus side and the negative side of the paired one-dimensional optical sensor are averaged. And a processing unit that detects the rotational position and the XY-axis position of the object on a plane orthogonal to the optical path by calculating difference data for the displacement.

(Operation) When detecting the position of an object on a plane orthogonal to the optical path, a pair of one-dimensional optical sensors provided in parallel to the optical path and parallel to the one-dimensional optical sensor and the optical path are used. It is sufficient to scan only a pair of one-dimensional optical sensors provided, and it is not necessary to scan the entire two-dimensional screen unlike a television camera. Therefore, the data acquisition speed and the processing speed are increased, and it is possible to cope with a high-speed servo motor or the like. Further, since there is no need to provide a high-speed processing device, the cost can be reduced. Further, the rotational direction position of the object on a plane orthogonal to the optical path can be obtained with high accuracy and high speed from the pixel count of the change point of the signal input from each one-dimensional optical sensor. Since the sensor unit is configured by the one-dimensional optical sensor, high resolution can be obtained by increasing the number of pixels.

(Embodiment) An embodiment of a sensor device for automatic alignment measurement according to the present invention will be described with reference to the accompanying drawings.

1 and 2 show an example in which the present invention is used as a sensor device for automatic alignment measurement of a semiconductor wafer (silicon substrate). In FIG. 2, a semiconductor wafer (hereinafter, referred to as a wafer) 1 as a workpiece is placed on a table (not shown), and can be moved in the XY-axis direction and rotated around a central axis.

Light emitted from a light source (not shown) and reflected by the wafer 1 is split in two directions by a half mirror 2 provided in the sensor unit A. One of the split lights is incident on a sensor plate 3a for position detection in the X-axis direction. The sensor plate 3a is provided with the optical path 4 of the reflected light incident from the wafer 1 orthogonal thereto, and the sensor plate 3a is provided with a pair of one-dimensional optical sensors 5a and 5b orthogonal to the optical path 4 and parallel to each other. The detection axis directions of these one-dimensional optical sensors 5a and 5b optically coincide with the X-axis direction of the table on which the wafer 1 is mounted.

The other light split by the half mirror 2 is incident on a sensor plate 3b for position detection in the Y-axis direction. The sensor plate 3b is provided orthogonal to the optical path 4 of the reflected light incident from the wafer 1. The sensor plate 3b is provided with a pair of one-dimensional optical sensors 5c and 5d orthogonal to the optical path 4 and the one-dimensional optical sensors 5a and 5b and parallel to each other. The detection axis directions of these one-dimensional optical sensors 5a and 5b optically coincide with the Y-axis direction of the table on which the wafer 1 is placed.

One-dimensional charge-coupled device (CCD) sensors are used as the one-dimensional optical sensors 5a to 5d. A one-dimensional CCD sensor has a large number of pixels (pixels) (not shown) (not shown) (500 to 10,000, usually preferably about 1000) which are linearly arranged as a light detection unit. The strength is read and stored electrically. As the one-dimensional optical sensors 5a to 5d, a moss-type solid-state image sensor (MO
S) or CPD may be used.

FIG. 3 shows the sensor plate 3b cut along a cutting plane orthogonal to the detection axes of the one-dimensional optical sensors 5c and 5d, and a condenser lens 6 is arranged on the light incident side of each of the primary optical sensors 5c and 5d. Is established. As the condenser lens 6, for example, a cylinder lens (kamaboko type) is used. Further, a rod lens (round bar) may be used. In FIG. 3, the left side shows the case where parallel light is incident, and the right side shows the case where emitted light is incident.

The use of the condenser lens 6 is for averaging a slight curve or the like of the cut line when detecting the cut line or the like of the wafer 1. That is, the condenser lens 6 is used to positively collect light beams perpendicular to the detection axes of the respective one-dimensional optical sensors 5c and 5d. Further, the position detection in the Y-axis direction is performed by the pair of parallel one-dimensional optical sensors 5c and 5d in order to perform the alignment in the rotational direction with high accuracy or the alignment in the Y-axis direction with high accuracy. This is because the detection error is rounded with high precision. Note that the sensor plate 3a for detecting the position in the X-axis direction has the same configuration, and the description is omitted.

FIG. 3 shows a two-line one-dimensional optical sensor including a plus-side one-dimensional optical sensor 5c and a minus-side one-dimensional optical sensor 5d.
Although the case where 5c and 5d are arranged in parallel is shown, it is not necessarily limited to one line on the plus side and one line on the minus side. For example, when pursuing more accuracy in the sense of reading error and manufacturing process difference, A one-dimensional optical sensor may be provided with two or more lines on the plus side and two or more lines on the minus side.

4 (A) and 4 (B) use a slit 7 instead of the condenser lens 6. FIG. In the case of the slit 7, averaging by light collection is not performed, so that the narrower the gap, the higher the positional accuracy.

Although the sensor plates 3a and 3b are actually arranged so as to be orthogonal to each other, when each of the one-dimensional optical sensors 5a to 5d is optically projected onto the wafer 1, as shown in FIG. The one-dimensional optical sensors 5a to 5d are arranged on each side of the square. In FIG. 5, there is a gap between the ends of each of the one-dimensional optical sensors 5a to 5d, but the ends of each of the one-dimensional optical sensors 5a to 5d are extended and intersected, so that they are arranged in a so-called grid shape. Is also good. However, the intersection is an optical projection, and in practice, the one-dimensional optical sensors 5a to 5d are provided as shown in FIG. 1 and do not intersect.

This is where the half mirror 2 is provided. Even if it is impossible to physically intersect the one-dimensional optical sensors 5a to 5d with each other, the optical projections are arranged so as to intersect in a grid pattern. It becomes possible.

When the one-dimensional optical sensors 5a to 5d are arranged so as not to intersect with each other as shown in FIG. 5, it is not always necessary to provide the half mirror 2, and as shown in FIG. Alternatively, the one-dimensional optical sensors 5a to 5d may be provided, and the reflected light from the wafer 1 may directly enter the one-dimensional optical sensors 5a to 5d without passing through the half mirror 2.

On the other hand, as shown in FIG. 6, the wafer 1 has an orientation flat 10 formed at one end as a reference line, and separation lines 11 are drawn in a grid pattern on the substrate plane. The square portion surrounded by the separation line 11 becomes a chip after separation and has a side of several millimeters to 1 cm. Also, the width of the separation line 11 is about 20-500 microns. With respect to such processing of the wafer 1, an automatic alignment measuring sensor device is used for alignment when cutting along the separation line 11, alignment of beam exposure, alignment of bonding pads, alignment of a prober, and the like. Can be

Optically projected one-dimensional optical sensors 5a to 5a shown in FIG.
When the set 5d (hereinafter referred to as an optical sensor) 5 is superimposed on the wafer 1 shown in FIG. 6, the result is as shown in FIG. The optical sensor 5 is fixed in position, and the wafer 1 can move in the rotation direction and the XY-axis direction with respect to the optical sensor 5.

FIG. 8 is an enlarged view showing a cross section of the separation line 11 of the wafer 1, and the separation line 11 is formed by an inverted trapezoidal recess. When a parallel ray is applied to the separation line 11, the light 14 hitting the inclined surface 11a does not return.
In the signal level of 15, the portion corresponding to the inclined surface 11a decreases. Further, the signal level of the portion corresponding to the bottom surface 11b also slightly decreases. Thereby, the separation line 11 can be detected.

If the cross section of the separation line 11 is not trapezoidal and the inclined surface 11a is vertical or overhanging, such as a square, rectangle or trapezoid, the separation line 11 is clearly formed by irradiating not a parallel ray but a radiation ray. Can be detected.

For example, when the wafer 1 and the optical sensor 5 are arranged as shown in FIG. 7, the signal levels of the one-dimensional optical sensors 5c and 5d for detecting the position in the Y-axis direction are as shown in FIG. Become. That is, the signal level of the portion corresponding to the separation line 11 is clearly reduced (arrow A), and the signal level is significantly reduced at the border of the orientation flat 10 (arrow B). Since the color of the wafer 1 is usually silver, if the color of the table is black, the position of the orientation flat 10 can be detected more clearly.

The position of the wafer 1 is shifted in the rotation direction,
If the lines of 11 and the orientation flat 10 are not orthogonal to the one-dimensional optical sensors 5c and 5d, as shown in the bad example, the positions of the parts where the signal level decreases (arrows A and B) are on the plus side and the minus side. At the same time, the width of the portion (arrow A) where the signal level decreases is increased. Arrow C is Y of the separation line 11
An arrow D indicates a difference in spatial frequency, and an arrow E indicates a difference in detection of the orientation flat 10 in the Y-axis direction.

Each one-dimensional optical sensor detects a grid-like separation line 11
The detection signals 15 of 5a to 5d are as shown in FIG. In this figure, since the waveforms of the signal level 15 on the plus side and the minus side match, there is no displacement in the rotational direction.

FIG. 11 shows the configuration of the processing unit B shown in FIG.
An X-axis arithmetic processing unit 17 for inputting detection signals from the one-dimensional optical sensors 5a and 5b for position detection in the X-axis direction and performing arithmetic processing;
A Y-axis arithmetic processing unit 18 is provided for inputting detection signals from the one-dimensional optical sensors 5c and 5d for axial position detection and performing arithmetic processing.

Since the XY-axis arithmetic processing units 17 and 18 have the same configuration, the Y-axis arithmetic processing unit 18 will be described as an example. As shown in FIG. 12, a clock generator 20 is provided in the Y-axis arithmetic processing unit 18, and the one-dimensional optical sensors 5c and 5d are operated by a clock one pixel at a time by a signal (not shown) from the clock generator 20. . The detection signals from the one-dimensional optical sensors 5c and 5d are input to an analog unit 21 as a signal level change point detecting unit.

As shown in FIG. 13, detection signals from the one-dimensional optical sensors 5c and 5d are input to the analog unit 21 and the external data processor 22
Is compared with the threshold value input through the D / A converter 23 from
A comparator 24 is provided to remove small signal level changes.

That is, the detection signal 15 input to the analog unit 21 is
FIG. 14 (A) shows the detected waveform of the separation line 11
It is assumed that a detection waveform 15b of a circuit pattern on the wafer 1 appears in addition to the detection waveform 15a. A small waveform of the detection signal 15 which does not reach the threshold level as a threshold value input from the external data processor 22 is removed (FIG. 3B), and the detection signal passed through the A / D converter 25 is removed. 15 has the waveform shown in FIG.

The change point of the detection signal 15 is detected by the comparator 26. That is, the data from the A / D converter 25 is compared with the data of the immediately preceding pixel for each pixel. If the signal level is higher than the immediately preceding signal level, an up signal is output. The data order determination circuit 27 determines the data order of these signals, and outputs an up signal and a down signal to the counter unit. In the case of equal, no signal is output.

The signal from the analog section 21 is input to latches 29a and 29b provided in the counter section 28, as shown in FIGS. On the other hand, the clock of the pixel rate is input from the clock generator 20 to the counter 30 as a signal, and the latches 29a and 29b use the value input from the counter 30 as the pixel count when the signal is input from the analog unit 21 as an absolute value. Hold in the form of The latches 29a and 29b include a latch 29a that holds a pixel count when an up signal is input and a latch 29b that holds a pixel count when a down signal is input.

The pixel counts from the latches 29a and 29b are subjected to a subtraction process by a subtractor 32 provided in a width calculation unit 31, and the data is output as an external data processor as width data of a waveform (FIG. 14C).
The data is transferred to 22 or the like, and a process of creating basic data for making a GO / NOGO determination or the like is performed.

On the other hand, after the pixel counts from the latches 29a and 29b are added by the adder 34 provided in the center point calculation unit 33, the average data divided by the divider 35 is calculated, and from each of the average data, The center point of the waveform (FIG. 14 (C)) is obtained, and the center point data and the remainder are input to the processing operation unit 36 shown in FIG. 12 and FIG.

The signals from the plus and minus dividers 35 are input to a subtractor 37 provided in a processing operation unit 36, subjected to a subtraction process, and then added to the other detection points by an adder 38,
By dividing by the divider 39 according to the number of additions,
The difference between the plus side and the minus side is obtained as difference data.

That is, in order to cope with an application in which the number of detection points is increased in order to increase the accuracy of detection, the difference between the positive side and the negative side of the pixel count of each detection point is added to the adder 38.
, And the result is divided by the number of detection points by a divider to obtain an average value of the difference data.

The signal from the divider 35 is compared by the comparator 40, and the smaller one is the direction signal. In order to use these difference data and direction signals as basic data for controlling the position in the rotational direction, the two-axis correlation operation processing unit shown in FIGS. 11 and 17 is used.
Transferred to 41 for processing.

On the other hand, the signal input from the divider 35 to the subtractor 42 and the comparator 43 is compared with a reference point from the external data processing unit 22 to obtain a deviation (difference data) and a direction signal.

These difference data and direction signal are input to the power section C shown in FIG. The buffer 44 receives the ENB from the comparator 46 provided in the two-axis correlation operation processing unit 41.
When a signal is input, difference data and a direction signal are output. That is, first, the position in the rotation direction is controlled, and when the displacement in the rotation direction disappears, the difference data and the direction signal are output to control the position in the XY axis direction.

As shown in FIGS. 11 and 17, difference data and direction signals from the divider 39 and the comparator 40 provided in the X-axis arithmetic processing unit 17, and the Y-axis correlation arithmetic processing unit 41, as shown in FIGS. The difference data and the direction signal from the divider 39 and the comparator 40 provided in the axis operation processing unit 18 are input. XY axis divider
The difference data from 39 is input to the adder 47 and added,
The average value is obtained by the divider 48 and becomes the internal data of the total rotation axis deviation of the XY axes.

Normally, only the operation of the comparator 46 to which the direction signal from the comparator 40 of the XY axis is input is compared with the scanning data of the next optical sensor 5, and if there is a difference, the power unit C is used until the direction and the deviation value disappear.
The rotation axis driver 50 drives the motor 51 for the rotation axis to rotate the table on which the wafer 1 is mounted. When the data becomes the same as the previous scan data in the comparator 46, an ENB signal is output to the buffer 44 to perform position control in the XY axis direction.

If a deceleration command is required or a direct function with the mechanical unit D is required, the deviation value data from the rotary axis encoder of the mechanical unit D and the external data processor 22 are subtracted.
52, the direction is obtained by the comparator 46, the data is output to the power unit C until the difference is 0 and the direction signal becomes 0, and the motor for the rotating shaft of the mechanical unit D is output by the rotating shaft driver 50.
Drive 51 to rotate the table on which the wafer 1 is placed,
Align in the rotation direction. When the difference becomes 0, the comparator 46 outputs an ENB signal to the buffer 44 to control the XY axes.

Similarly, in the control of the XY axes, the scan data before the optical sensor 5 is normally output from the comparator 43 to the X-axis driver 53 and the Y-axis driver 54 of the power unit C, and the X-axis motors 55 and Y of the mechanical unit D are output. The axis motor 56 is driven to perform positioning in the XY axis direction.

Further, when performing a comparison operation with the data of the encoder of the mechanical unit D and the external data processing unit 22, etc., the difference value data is obtained by the subtracter 42, the direction is obtained by the comparator 43, and the difference is 0 when the difference is 0. Output to XY axis doiraba 53, 54 until the signal becomes 0,
When the value becomes 0, a stop signal (INT) is generated and the entire process is completed. However, even if the operation of the XY axis is completed, if a deviation on the rotation axis occurs due to the influence, the ENB signal from the comparator 46 becomes invalid, and the operation starts again.

As described above, according to the above embodiment, the orientation flat of the wafer 1 can be obtained only by scanning the four-line one-dimensional optical sensors 5a to 5d.
10. The position in the rotation direction and the position in the XY axis direction such as the separation line 11 can be detected. Therefore, it is not necessary to scan hundreds of scanning lines arranged on a two-dimensional plane as in the related art, and it is possible to dramatically improve the capture speed (possibly 1/15000 second). Motor 5
It is possible to cope with the case where a high-speed servomotor or the like is used as 1,55,56, and it is not necessary to perform time-consuming arithmetic processing, so that the processing speed can be greatly improved. Since there is no need to provide a high-speed processing device, the cost can be reduced.

Further, the position of the wafer 1 in the rotation direction can be obtained with high accuracy and high speed in the same manner as the position in the XY axis direction, and by increasing the number of pixels of the one-dimensional optical sensors 5a to 5d, a high resolution (500 ~ 10000 decomposition). Therefore, the field of view versus resolution,
The conflicting factors of resolution versus speed and speed versus field of view can be solved.

FIG. 19 shows another embodiment of the present invention, and is a projection arrangement diagram of the one-dimensional optical sensors 5a to 5d corresponding to FIG. In this embodiment, a one-dimensional optical sensor for detecting a position in the X-axis direction is used.
5a and 5b are arranged in parallel on the plus side and two lines in parallel on the minus side. Similarly, the one-dimensional optical sensors 5c and 5d for detecting the position in the Y-axis direction are similarly arranged in parallel on the plus side and the minus side by two lines. It should be noted that one-dimensional optical sensors 5a to 5d of three or more lines may be provided side by side on the plus side and the minus side, respectively. According to this embodiment, higher detection accuracy can be obtained than in the above embodiment.

FIG. 20 shows still another embodiment of the present invention, and is a projection arrangement diagram of the one-dimensional optical sensors 5a to 5d corresponding to FIG. In this embodiment, the one-dimensional optical sensors 5a and 5b for detecting the position in the X-axis direction have four lines on the plus side and four lines on the minus side.
It is arranged on the line straight line. Similarly, the one-dimensional optical sensors 5c and 5d for detecting the position in the Y-axis direction are similarly arranged on four lines on the plus side and the minus side, respectively. Note that two lines, three lines, or five or more lines may be arranged on the plus side and the minus side, respectively.

According to this embodiment, simultaneous scanning is performed by each of the one-dimensional optical sensors 5a to 5d, so that the scanning time for one line is 4 times.
The scanning for the lines can be performed, and the data capturing speed can be greatly improved. In addition, since the number of pixels can be greatly increased, the resolution and the detection accuracy can be significantly improved.

In the above embodiment, a sensor device for automatic alignment measurement of a semiconductor wafer has been described as an example. However, the present invention is not limited to this, and an alignment device for a voidless chip mount IC (wiring without using a wire bonder). A wide range of applications, such as a position detection device that performs die bonding and die bonding at once, a roundness inspection device for circular objects such as cans, a marking position detection device, an automatic tracking device, and a printing plate superposition position detection device can do.

〔The invention's effect〕

The sensor device for automatic alignment measurement according to the present invention is provided with a pair of one-dimensional optical sensors orthogonal to an optical path from an object in parallel on the plus side and the minus side, while the optical path and the pair of one-dimensional optical sensors are provided. Another pair of one-dimensional optical sensors orthogonal to each other are provided in parallel on the plus side and the minus side to constitute a sensor unit, and signals from the paired one-dimensional optical sensors are input, and changes in the signals are performed. By adding and averaging the pixel count of the point and the averaged data on the plus side and the minus side of the paired one-dimensional optical sensor, or calculating the difference data of the deviation, on a plane orthogonal to the optical path. Processing unit that detects the position of the object in the rotation direction and the position in the XY-axis direction at, the pixel count of the change point of the signal from each one-dimensional optical sensor is simply averaged and averaged. Calculate the average data, compare the averaged data on the plus side and the minus side of the paired one-dimensional optical sensor, and perform averaging, or calculate the difference data for the deviation by a simple calculation that calculates the rotation direction of the object. Location and
XY axis position can be detected quickly and easily. In addition, angle measurement is not required to detect the rotational position and the XY axis position of the object, and only simple averaging and addition / subtraction are required, so that the data processing speed can be improved and the equipment can be simplified. In addition, the cost can be kept low, and the effect of particularly high detection accuracy and resolution of the rotational position can be obtained.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a sensor unit provided in an embodiment of a sensor device for automatic alignment measurement according to the present invention, and FIG. 2 is an overall view of the case where the above embodiment is applied to an automatic alignment device for semiconductor wafers. FIG. 3 is an enlarged cross-sectional view of the sensor plate in the above embodiment, and FIGS. 4A and 4B are a cross-sectional view and a perspective view showing an example in which a slit is used instead of a lens in the above embodiment. FIG. 5 is a projection arrangement diagram of the one-dimensional optical sensor in the above embodiment, FIG. 6 is a configuration diagram showing a semiconductor wafer, and FIG. 7 is a diagram in which the optical sensor shown in FIG. 5 is superimposed on the semiconductor wafer shown in FIG. FIG. 8 is a diagram showing a projection, FIG. 8 is a diagram showing a relationship between the shape of a separation line formed on the semiconductor wafer and a detection signal of the one-dimensional optical sensor, and FIG. 9 is a position detection in the Y-axis direction in FIG. Showing the detection signal of the one-dimensional optical sensor for the FIG. 11 is a diagram showing a waveform of a detection signal when a lattice-shaped separation line is detected by each one-dimensional optical sensor in the above embodiment, FIG. 11 is a block configuration diagram showing a processing unit in the above embodiment, and FIG. FIG. 13 is a block diagram showing a Y-axis arithmetic processing unit in FIG. 11, FIG. 13 is a block diagram showing an analog unit in FIG. 12, and FIG.
(B) and (C) are diagrams showing a circuit pattern removing process in the analog unit, FIG. 15 is a block diagram showing a counter unit, a width calculating unit and a center point calculating unit in FIG. 12, and FIG. 12 is a block diagram showing a processing operation unit in FIG. 12, FIG. 17 is a block diagram showing a two-axis correlation operation processing unit in FIG. 11, and FIG. 18 is a block diagram showing a power unit and a mechanical unit in the above embodiment. FIG. 19 is a projection arrangement view of a one-dimensional optical sensor showing another embodiment of the present invention, FIG.
The figure is a projection layout of a one-dimensional optical sensor showing still another embodiment of the present invention. 1. Semiconductor wafer, 2. Half mirror, 3a, 3b
Sensor plate, 4 ... optical path, 5 ... optical sensor, 5a, 5b,
5c, 5d: One-dimensional optical sensor, 6: Condensing lens, 7:
Slit, 10 ... Ori-Fla, 11 ... Separation line, 14 ...
Two-axis correlation operation processing unit, 15 detection signal, 17 X-axis operation processing unit, 18 Y-axis operation processing unit, A sensor unit, B
... Processing unit, C ... Power unit, D ... Mechanical unit.

Claims (1)

(57) [Claims] 1. A pair of one-dimensional optical sensors orthogonal to an optical path from an object are provided in parallel on the plus side and the minus side, and another one orthogonal to the optical path and a pair of the one-dimensional optical sensors. A pair of one-dimensional optical sensors are provided in parallel on the plus side and the minus side to form a sensor unit, and signals from the paired one-dimensional optical sensors are input, and the average data of the pixel count at the change point is input. And, by averaging the averaging data on the plus side and the minus side of the paired one-dimensional optical sensor, or by calculating the difference data of the deviation, the rotational direction position of the object on a plane orthogonal to the optical path and A sensor device for automatic alignment measurement, comprising a processing unit for detecting a position in the XY-axis direction.