JP3694272B2 - Absolute position detection method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method for detecting a current absolute position without moving a detector by using a liquid crystal or other electro-optic effect. The present invention includes, for example, a numerical control machine, a machine, a semiconductor substrate transfer device, a three-dimensional drawing device, an endoscopic surgical device, a μ machine manufacturing device, an articulated robot, a gene therapy drug manufacturing device, a position control non-contact shape measurement It can be used for position control of drive system actuators such as a device, an optical disc inspection device, a DNA inspection / analysis device.
[0002]
[Prior art]
Conventionally, as an apparatus for detecting an absolute position, for example, a scale on which information on a position is recorded at a fixed period, and a sensor element that moves relative to the scale and reads position information recorded on the scale Are known. Current sensor elements are mostly optical and magnetic methods. The former is a method that detects changes in light transmission and reflection characteristics using LEDs and LDs, and the latter is a method that detects magnetism from a magnetic material with a coil or the like. This is a detection method. In both cases, in order to achieve high resolution, a conventional method that causes a phase shift in the sensor signal due to the arrangement of multi-element detectors, which is multiplied by a logic element, or that the detected analog signal is quantized. Is used. In the position detection in the nano-order, this multiplication method is important, and it can be said that the cause of the error due to the processing on the software depends on the quality of the output. At present, the method that is said to have the highest detection resolution is an optical multiplication method, which increases the resolution by quantizing an analog signal from the pickup.
[0003]
[Problems to be solved by the invention]
However, optical position detection generally has the effect of disturbance of diffracted light, and the finer the patterning of scale position information, the more difficult it is to avoid adverse effects on accuracy due to diffracted light interference caused by transmission or reflection. I can't. In addition, in order to increase the resolution, the accuracy of multi-element arrangement is required, and the position-to-signal output characteristics from the pickup are not linear, and the reproducibility is poor. The point that the above burden is very large remains as a problem.
[0004]
In terms of materials, consideration is required for the quality of the scale, which is an element of the position detector, in particular, heat and environmental temperature. For example, in the nanoscale manufacturing method, if micromachining technology or vapor deposition technology is used, it is possible to produce a small precision scale, but it is difficult to give sufficient thickness, so it can be easily deformed by subtle residual stress or thermal stress. Therefore, sufficient reliability cannot be obtained due to, for example, dynamic deformation when driving at high speed. Therefore, the conventional optical detection has a limit in high resolution and high reliability, and a new scale material has been demanded. In the case of the magnetic system, there is a change in magnetic flux density with time, and the detection sensitivity does not reach that of the optical system.
[0005]
In addition, when using a scale in which position information is recorded at a fixed period, it is possible to measure the position within one period, but the position detection (movement amount detection) over a plurality of periods is detected. It is necessary to accumulate the number of signal repetitions. This position detection over a plurality of cycles is likely to cause an integration error of the number of signal repetitions, and is not very reliable. Moreover, when creating a two-dimensional (XY axis) scale, if two absolute one axes are combined, two dimensions are possible, but cumulative errors are generated, so alignment and other adjustments are very time consuming and yield Was bad.
[0006]
Furthermore, conventionally, the current position cannot be known when the sensor element is stopped. By moving the sensor element, the encoded coordinates created on the scale can be read, thereby detecting the current position. However, since moving the mechanical unit may cause a runaway, it is necessary to provide a plurality of sensors and limiting circuits installed near the origin, CW and CWW limits, near the origin, etc. in order to ensure safety. there were. In addition, since a movement for knowing the present location is not necessary originally, the work tact time becomes longer than necessary. Therefore, a more complicated process requires a method for knowing the present location in a non-movable manner.
[0007]
Therefore, an object of the present invention is to provide a new method of detecting the current absolute position without movement by eliminating the method of counting the sensor element from the end by mechanically returning the sensor element to the origin. It is another object of the present invention to provide a method that enables absolute detection on a two-dimensional plane, which cannot be achieved by a conventional method.
[0008]
[Means for Solving the Problems]
The present invention has been made to solve the above problems, and according to the present invention, the present position can be easily known by using an electro-optic effect such as liquid crystal. That is, for example, when an electro-optical shutter effect of a liquid crystal matrix disposed on a surface light source is used, an arbitrary position can be turned on by opening / closing a liquid crystal shutter pixel. Scanning is performed by sequentially opening and closing the liquid crystal pixels without mechanical driving such as return to the origin of the optical sensor installed above the position where the liquid crystal pixels are located. When the pixel at the position where the photosensor is installed opens, the output of the photosensor increases at that time. Thereby, the position of the optical sensor can be known. The same applies to the XY direction, that is, two-dimensional scanning. This makes it possible to know the absolute position without mechanically returning the optical sensor or liquid crystal matrix connected to the drive system to the origin.
[0009]
That is, according to the present invention, optical means comprising a projector and a light receiver are arranged on a scale having m (m ≧ 2) transmission part shutter pixels in n columns (n ≧ 2 ), and any number of columns n the opening and closing sequentially the shutter pixels on the m 1 row, estimates the position of the optical means is arranged in among the m shutter pixels as addresses, the next row corresponding to the estimated address by opening and closing the adjacent shutter pixels, to compare the magnitude of the output value of the light receiver which is located adjacent the shutter pixels, thus the absolute position detecting method characterized by detecting the position of said optical means Provided.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
The scale used in the present invention has m (m ≧ 2) pixels arranged in at least one dimension, the width of the pixels is 10 to 100 μm, and the pitch is 10 to 100 μm. The dimensions (columns) in which the pixels are arranged are not particularly limited. However, when the pixels are arranged in multiple dimensions (n columns), the pattern is preferably staggered, but is not limited thereto. In the case of two dimensions (n = 2), it is preferable to shift the half wavelength from the one dimension pattern.
[0011]
The width of the pixel is 5 to 500 μm, preferably 20 to 100 μm. This is because when the pixel width is 5 μm or less, the influence of the interference of the diffracted light cannot be ignored by the Fraunhofer rule as in the diffraction grating, and this becomes disturbance light and the precise position cannot be specified. On the other hand, if the width of the pixel is larger than 500 μm, the linearity characteristic of the position versus output is distorted. Furthermore, the pitch is 5 to 500 μm, preferably 20 to 100 μm. This is due to the same reason as the pixel width described above.
[0012]
Further, specifically, the scale of the present invention is configured by facing a pair of transparent substrates and enclosing a liquid crystal therebetween. A transparent electrode film is deposited on one side of the transparent substrate with a width of 5 to 500 μm and at a pitch of 5 to 500 μm, and a similar transparent electrode film is deposited on the opposite transparent substrate in a pair with one side, Further, the light shielding film is formed by short-circuiting around the transparent electrode film. The width and pitch are limited to the above values as described above.
[0013]
Here, examples of the transparent substrate include, but are not limited to, a glass substrate and a polyacrylic substrate. The transparent electrode film is preferably an ITO (InSnO ₃ ) film, for example, but is not limited thereto. The transparent electrode film is deposited by a CVD method or the like.
[0014]
Furthermore, the liquid crystal may be, for example, a birefringent liquid crystal element, a transmission / scattering liquid crystal element, a TN (twisted nematic) liquid crystal, an STN (super TN) liquid crystal, a ferroelectric liquid crystal element, an antiferroelectric liquid crystal, or a polymer dispersed liquid crystal. However, it is not limited to these. The distance (cell gap) between the opposing substrates is 1 to 100 μm, preferably 5 to 20 μm.
[0015]
Further, as the light shielding film, for example, an aluminum vapor deposition film can be used, but the material is not limited as long as light wraparound can be prevented. For example, as a method for producing a light shielding film, it is possible to form a film by dry or wet etching after chromium deposition or screen printing of gold paste and baking to form a film in addition to the aluminum deposition method. Vapor deposition, screen printing, etc. are also possible for the ITO film corresponding to the window. The transparent substrate on the side where the light shielding film is formed is preferably on the light receiver side of the optical means described later.
[0016]
When the cell gap of the scale is 5 to 20 μm, the applied voltage to the transparent electrode is a frequency of DC to 1 kHz, preferably DC to 300 Hz, and a rectangular wave of 3 to 50 V, preferably 3 to 15 V is applied. Is preferred. By changing the driving voltage within such a range, the height and / or width of the signal waveform can be adjusted. This is because the liquid crystal refractive index distribution, viewing angle, tilt angle, and transmittance can be adjusted by adjusting the driving voltage of the liquid crystal. The driving voltage of the liquid crystal may be changed independently for each pixel, or may be changed collectively for a plurality of pixels as a set.
[0017]
Changes in the liquid crystal pixels are read by optical means. The optical means includes a projector that irradiates light to the scale and a light receiver that detects transmitted and scattered light from the scale. Examples of the projector include, but are not limited to, a light emitting diode, a laser, a lamp, and EL (electroluminescence). The wavelength range to be used may be in the ultraviolet to infrared range, preferably in the wavelength range of 300 to 1500 nm. The projector may be either a point light source or a backlight that irradiates the entire scale.
[0018]
As the light receiver, various infrared sensors such as an image pickup element (CCD, C-MOS, vision chip), photodiode, photomultiplier, solid-state light receiving sensor, pyroelectric sensor, and surf mobile can be used.
[0019]
Further, the projector and the light receiver may be integrated such as a photocoupler or a photointerrupter. Further, a plurality of light projectors and light receivers may be installed like a plurality of photo interrupters or PSD (posision sensing devices). In addition, when the projector is a single point or surface light source and only a plurality of light receivers are connected, the reverse case is also included.
[0020]
Further, a slit may be provided between the light projector and the liquid crystal display panel or between the liquid crystal display panel and the light receiver to prevent light from wrapping around. The slit width varies depending on the size of the liquid crystal pixel, but is, for example, 50 to 500 μm in the case of a scale. The scale of the present invention is not limited to liquid crystal, and a piezoelectric element such as PZT and a shutter array manufactured by micromachining can also be used.
[0021]
The present invention will be described more specifically below. First, a method of knowing a precise absolute position in one dimension will be described, and problems in that case and a solution to the problem will be described. In particular, if an electro-optical shutter such as a liquid crystal and a point light source are used, a precise position can be determined by using a halftone in a pixel. For example, if a pair of a light source having the same width as a pixel and a sensor that inspects the light is arranged in a sandwiched manner with an electro-optic shutter matrix, a halftone can be used to further refine the pixel. It is possible to detect a position by arbitrarily dividing a pixel. That is, the pixels in the matrix can be divided with higher accuracy. In that case, the pixels need to be alternately arranged with the same width where they only block the light. However, when the position of the sensor-light source pair is detected non-movably using electro-optic scanning, the output of the same level is binarized at two different positions in the pixel, so it is located in either position. I can not identify. Therefore, if another pixel column is provided on the staggered pattern, the binarization problem is solved.
[0022]
FIG. 1 shows the positional relationship between a scale pixel array and a light source-sensor pair. 1, (a) is a cross-sectional view of the first column of the scale pixel array 1, (b) is a diagram corresponding to a transmission portion (P1,... P4) of the pixel array 1, and (c) is a sensor point. Represents. In the figure, L is a light source, S is a sensor, and 2 is a signal amplifier. Although the detailed cross section is omitted, the scale pixel array 1 is configured by facing a pair of glass substrates, holding a space (for example, 5 μm) by a spacer, and enclosing a liquid crystal. Further, an ITO film is deposited on the glass substrate to form a liquid crystal drive electrode, and a voltage controlled by a control circuit is applied to the liquid crystal drive electrode. On one side of the glass substrate, a pattern for forming a 30 μm square light transmission window made of an ITO film is formed, for example, at a pitch of 30 μm (transmission portion P1,... P4). A 30 μm square pattern of the ITO film is manufactured in a pair with one side, and a chromium vapor deposition film for light shielding is formed by short-circuiting around the ITO film (light shielding portion).
[0023]
Let R be the sensor signal output when the center AA ′ of the light source L-sensor S pair comes to an arbitrary position X. If the signal output R receives the amount of light corresponding to the size Q of the light receiving portion of the sensor, AA ′ corresponds to the amount of signal accumulated and moved by Q, so the relationship between the sensor position and the signal output V FIG. 2 shows a trace of the output change when moving from the point X0 to the point Xe.
[0024]
Therefore, as a method of knowing the position X of AA ′ without mechanically operating the light source L-sensor S pair, the pixels are lit in order from the pixel P1 in order from the X0 point side. As a result, when P2 related to the place where the light source L-sensor S pair is stopped lights up, an output of R is obtained (FIG. 3). This means that if the address of the pixel is stored at the stage where the light is sequentially turned on, the stop position Pi can be determined conversely from the output of R. It is also possible to determine a precise position in the pixel from the output amount of R.
[0025]
However, although any pixel Pi can be determined, it is not possible to specify whether the sensor position is Xi or Xi + 1 from R (in the case of FIGS. 4A and 4B), and R is 0 level. In the case of (i.e., the state where the light source-sensor pair is hidden in the light-shielding portion), even the pixel cannot be specified.
[0026]
As shown in FIG. 5, in order to identify Xi or Xi + 1 according to the first column, or to identify which pixel the light-shielding part on the left side of the light source-sensor pair is hidden, a pixel array in the second column is provided. Solved by. FIG. 5A shows an arrangement example of two columns of pixels, and the width of the W pixel is the same as the sensitive width Q of the light source sensor pair in the same manner as P. That is, for example, in order to specify X2 or X3 when there is an output R when P2 is turned on as shown in FIG. 5B, P2 is turned off and W2 and W3 are It is possible to light up alternately and compare the magnitudes of the outputs T2 and T3 corresponding to each (FIG. 5 (c)).
[0027]
This is because when R ≠ 0, the relationship of T (2)> T (3) always holds as shown in FIG. 5C at an arbitrary position where the center of the light source-sensor pair is missing the right side of P2. The missing right means that the light source-sensor center AA ′ is on the left side of the triangle. Incidentally, since the vertex of the triangle is a singular point that is not binarized from the R by the pixel P, it can be uniquely obtained from the R output. In the case of R = 0, if the light is turned on in order from W1, it is possible to know which light shielding portion is located. That is, at the singular point that overlaps the light shielding portion between P1 and P2, T2 is maximum and all other T values are at the zero level. If Wi can be specified, the position of AA ′ is inevitably located at the center of the light-shielding portion of the pixel P. Therefore, the distance is calculated when a quarter wavelength is added.
[0028]
Further, the simplest configuration of the light source-sensor pair may be provided with only a single element pair so that both the first row and the second row can be detected as shown in FIG. If a twin pair is used, the algorithm of turning off the first row and turning on the second row can be omitted.
[0029]
As described above, since the binarization problem is solved, it is possible to determine a minute current position in the pixel from the sensor analog output intensity corresponding to the halftone. The desirable arrangement of the second pixel rows forming the staggered pattern must be arranged with a phase shift of ¼ wavelength.
[0030]
A flow sheet of the scanning method for opening and closing the pixels is shown in FIG. Scan the opening and closing of a row of pixels. The pixel position i-th where the output of the photo sensor exceeds a certain threshold is detected, a binarized value corresponding to the output is calculated, and the distance between the two candidates is selected. This threshold level has the same meaning as the aforementioned zero level. For example, this is the output level when the SN ratio = 1, and the minimum detection resolution. Next, in the second column, both the i-th and i + 1-th are alternately opened and closed, and it is possible to determine which side of the apex of the triangular wave is large or small. That is, if W _i > W _{i + 1} , the position is the i-th position, and if W _i <W _{i + 1} , the position is the i + 1-th position, and the distance can be obtained as shown in FIG.
[0031]
【The invention's effect】
According to the present invention, the current absolute position can be detected without moving without mechanically returning the sensor element to the origin and counting from the end.
[Brief description of the drawings]
FIG. 1 is a diagram showing a positional relationship between a scale pixel array and a light source-sensor pair.
FIG. 2 is a diagram illustrating a change in output of a sensor.
FIG. 3 is an output diagram when P2 related to a place where the light source-sensor pair is stopped is turned on.
FIG. 4 is a diagram showing that the position of a sensor cannot be specified.
FIG. 5 is a diagram in which a pixel array in the second column is provided.
FIG. 6 is a flowchart of a scanning method for opening and closing pixels.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Scale pixel array 2 ... Signal amplifier L ... Light source S ... Sensor

Claims (3)

An optical means including a projector and a light receiver is arranged on a scale having m (m ≧ 2) transmissive part shutter pixels in n columns (n ≧ 2 ), and m in any one column of n columns are sequentially arranged. by opening and closing the shutter pixels, to estimate the position of the optical means is arranged in among the m shutter pixels as addresses, by opening and closing the adjacent shutter pixels in the next row corresponding to the estimated address A method for detecting an absolute position, comprising: comparing the magnitudes of output values of light receivers located at adjacent shutter pixels , and detecting the position of the optical means. absolute position detection method according to claim 1, wherein are arranged so as not to overlap with the shutter pixels of columns each shutter pixel in n columns are adjacent. 3. The absolute position detection method according to claim 1, wherein the scale is made of liquid crystal, and the optical means includes a projector that irradiates the scale with light and a light receiver that detects transmitted or scattered light from the scale .

Images