JPS59109813A - Measuring device of optical displacement - Google Patents

Abstract

PURPOSE:To improve the precision of displacement measurement independently of the precision of the grating pitch of an optical scale, by focusing plural pulse images due to the movable optical scale onto two image sensors and integrating the average value of the extent of movement. CONSTITUTION:An optical displacement scale 20 which is moved together with a moving object as one body is irradiated with light through optical guide paths corresponding to two image sensors 44a and 44b by an optical projector 42 of a fixed optical system 40, and plural image pulses due to the reflected light are focused onto sensors 44a and 44b. This focusing is scanned periodically, and positions of image pulses are detected as addresses of sensors 44a and 44b by pulse image position detectors 60a and 60b; and every time when scale-over occurs, image pulses are specified successively by a pulse image specifying part 90 of a data processor 80. The extent of movement of specified image pulses in every scanning cycle and an average extent of movement are calculated, and the displacement quantity of the moving object is measured easily with a high precision in an average movement extent integrating part 94 independently of the precision of the grating pitch of the optical scale.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an optical change measuring device that uses a light beam as a probe to measure the amount of displacement of a moving object. More specifically, by observing the movement of the optical lens (turn) in proportion to the relative movement between the moving part and the fixed device, and integrating the amount of change in the optical lens, - a specific pattern. When the scale overflow occurs, the following optical pattern CT is focused, the integral value of the amount of change is calculated, and these values are continuously added to the Iln order. Let's measure the relative momentum between the optical system and the internal displacement measuring device.

Conventionally, an Ic optical displacement measuring device using light has been constructed using, for example, a linear displacement measuring device, an optical displacement scale fixed to a movable base, and an optical measuring device. There is also a system consisting of a system. The optical displacement scale represents a grating pattern in which the optical coefficients, such as reflection coefficients or transmission coefficients, vary in the direction of relative movement in a rectangular periodic manner at close intervals. On the other hand, the optical system includes a light source, a lens that focuses the light emitted from the light source and projects it onto a periodic optical pattern with an optical displacement scale, a lens that adjusts the light, and a lens that adjusts the light that is reflected or transmitted from the optical pattern. It consists of a lens that condenses light and a photoelectric conversion element that receives the reflected light. The photoelectric conversion element is composed of a single pixel, and captures a bright and dark image based on transmitted light or reflected light from the optical displacement scale passing over the photoelectric conversion element. The method for determining the amount of displacement was to convert the image into electrical pulses and count them to detect the amount of displacement of the object per unit time.

Since conventional devices adopt such a method, they have the disadvantage that errors in the pattern period of the optical displacement scale are reflected in the measurement accuracy. Therefore, the pitch interval of the optical displacement scale must be manufactured with extremely high precision.However, it is extremely difficult to form a periodic lattice-like reflecting plate or transmitting plate over a long range with an interval width of 0°1mm. If the movement M of the object ranges from 1 m to several meters, it is necessary to create an optical displacement scale of 1 m to several meters.

However, there are drawbacks such as changes in the pattern pitch due to culm lengthening the optical displacement scale, distortion of the scale, thermal expansion or contraction, etc., and it is difficult to improve accurate measurement accuracy. was there.

Therefore, the present invention has been made to improve such conventional drawbacks, and is an optical displacement scale that can improve displacement measurement accuracy regardless of the pitch accuracy of the grating pattern of the optical displacement scale. The purpose of the present invention is to provide a displacement measuring device that can be used to measure displacement.

That is, the present invention provides an optical displacement scale in which the optical characteristic shifts to the right in a pulse-like pattern that periodically fluctuates with respect to a relative displacement direction, and an optical pulse-like pattern that the optical displacement scale has is received. at least two image sensors, each of which images an optical pulse-like pattern of the optical displacement scale on each image sensor at any time by relative movement with the optical displacement scale; an optical system having an optical projector for obtaining two moving pulse images; and a pulse signal corresponding to the pulse image is inputted from each of the image sensors by scanning it at regular intervals; At least two pulse image position detecting devices for detecting positions on the image sensor; and a data processor 1!1' for measuring the displacement of a movable object by processing output signals from the respective pulse position detecting devices every scanning period. !
The data processing device specifies, for each scanning period, a pulse image from which the amount of movement of the pulse image during at least one past scanning period can be determined based on the positional relationship of the pulse image on each image sensor. a pulse image specifying section; a movement amount calculation section that focuses on the pulse image specified by the pulse image identification section and calculates a pulse image movement amount for each scanning period; an average value calculation unit that calculates the average value of each corresponding pulse image movement amount, and sequentially controls the pulse image identification unit and movement amount calculation unit for each scanning period, and calculates the average value of the pulse image movement amount to the treetop frame height. 111 of the pulse image as the scanning period progresses.
This optical displacement measuring device is characterized in that it sequentially pays attention to subsequent pulses for one unsuccessful movement and cumulatively adds the average value of each pulse image movement amount on each image sensor.

FIG. 1 is a block diagram conceptually showing the configuration of the present invention. The optical displacement scale 20 is an optical displacement scale having a pattern in which the transmission coefficient or the reflection coefficient changes periodically and in the form of a square wave pulse in the direction of relative movement.
- It is le. The pattern of the optical displacement scale can be formed, for example, in the form of a grid, dots, or the like. The period of this pulse pattern is generally 0.1 to 0.2 mm apart, but is not limited to these values. This optical displacement scale is manufactured, for example, by vacuum-depositing chromium on the surface of a flat glass plate, and etching away the vacuum-deposited chromium film to the above-mentioned intervals using a process. The optical system/IO (well, it consists of an optical projector/12 and a plurality of image sensors 4J-4, 4a, 44b, etc. that receive pattern images of optical displacement scales. What is the optical projector 42? , a light source, a lens, a half mirror, a prism, etc. It uses a light guide path such as an optical fiber, and the light from the light source is projected onto an optical displacement scale, and the reflected light or transmitted light is imaged on an image sensor. The above-mentioned light guide path has at least two systems, and each system's light beam illuminates each part of the optical displacement scale. Image sensor 44 that receives light
a, 44b, etc.) are a large number of A to D arranged in a one-dimensional array, or kit/rear storage type COD, B.
BD etc. can be used.

Here, the image pattern that can be imaged on each image sensor must be configured so that at least two cycles of the pulse pattern on the optical displacement scale are always projected.

The plurality of pulse image position detection devices 60a, 160b, etc.
A clock scanning pulse is applied to each image sensor to sequentially extract the video signal in full and send out the scanning signal.

Then, the video electrical signal corresponding to the pulse periodic pattern on each image sensor extracted by the scanning signal is
A plurality of pulse image position detection devices 60a and 60, respectively.
Enter it in b etc.

The pulse image position detecting devices 60a and 60b detect pulses from the P1-pulse train output for each pixel of each image sensor.
Obtain each pulse-like image pattern using the optical displacement system. and - detecting the positions of said at least two pulsed image patterns on an image sensor. Here, the position is a pixel address expressed as an ordinal number from one end of pixels arranged one-dimensionally on the Y image sensor.

The position data of at least two pulse images on the image sensor detected by each pulse image position detection device 60 a z , 6011 etc. is processed by a data processing device 80 .
Enter. The data processing device 80 includes a pulse image specifying unit 90 for specifying a pulse of interest in order to measure the moving color of the pulse image that moves in a short scanning period among the at least two pulse images, and a pulse image specifying unit 90 for specifying a pulse of interest, Specific part 9
0, the movement extraction unit 92, which calculates the amount of movement of the specified pulse in one scanning cycle interval, calculates the average value of each movement (A) determined for each image. It consists of a movement amount averaging section 93 to find and an average movement amount accumulating section 94 that cumulatively increases this average movement amount for each scanning period.

Then, the final displacement amount is given as a numerical value indicated by the average movement amount integrating section 94.

Next, the principle of measuring the movement of an object using the apparatus of the present invention will be explained.

FIG. 2 is a conceptual diagram explaining this principle. In Figure 2, the pulse video transmission is l'c, which is explained in principle.
In order to show the pulse position, the pulse is described as a narrow pulse, and the amount of movement is determined using an image sensor. In addition, the numerical value of the scale So''S4 on the horizontal axis indicates the number indicating the pixel position on the image sensor, and at the same time corresponds to the scanning time axis of the sirino full data extracted from the image sensor by the scanning signal. First, at time ]-〇, the pulse video signal obtained by scanning the image sensor 1 is as shown in Fig. 2 (a), which is the younger one in terms of one scanning time (in terms of pixel position number). The first pulse image No. 1g (hereinafter simply referred to as [pulse-1)] is indicated by the symbol P1, and the second pulse that appears at the later position on the time axis and the larger pixel position number is indicated by P2. The pulse P1 and the pulse P2 can be considered as the center position of the pulse image obtained by the pulse image position detection device 60a.Then, the second scanning time 1
”+L is the detected pulse pattern diagram as the second
Accepted in figure (b).

Here, the movement of Pl is measured together with Dl.

It should be noted that the movement distance during one scanning period in the drawing is drawn to be extremely large compared to the actual movement. Here, if we assume that the pulse for measuring the amount of movement is the first pulse P1, then the pulse shown in Fig. 2 (
As shown in C), the pulse P1 moves from right to left on the image sensor in the drawing.

And each scan lap! Move to y3 phrase m D + , l) 2
, D 3 and add this i- to I! The amount of movement of the pulse image is measured from time i to time T3. While the pulse P1 appears on the image sensor, the amount of movement can of course be measured. However, when moving from scanning time T3 to the next scanning time T4, the second
The pulse P1 written on the leftmost side of the diagram (C) is overscaled from the left end on the image sensor. However, on the image sensor, at any time,
It can be seen that a pulse appears at the right end because the two pulse images are designed to be projected. If this is reversed, the leftmost pulse in the diagram is guaranteed to be the first pulse, so the second pulse P2 in Figure 2 (C) is the second pulse.
In the diagram (d), it is regarded as the first pulse P1. Movement of the pulse image m I,L from time T3 to time T4.

It is sufficient to pay attention to the P2 pulse in FIG. 2(C') and the P1 pulse in FIG. 2(d) and measure their movement amount D4. In this way, by accumulating the movement m, the total movement amount from time 10 to time ]-4 can be obtained. That is, L=DI+D2+D'3+D4. By repeating this operation, even if the determined pulse exceeds the scale from the left end of the image sensor, the amount of movement of the subsequent pulse can be calculated by accumulating the movement m.
The total amount of movement can be calculated.

The pulse image specifying section 90 is a section that specifies which pulse to focus on among at least two pulses appearing on the image sensor. This is mainly the rough 1 of the calculator.
~It is easiest to realize this through clothing. However, by assembling a logic circuit or the like, it can also be realized in the Birdow J near field. To explain the identification method, in principle, attention will be paid to the first pulse P1.

The left end scale over of the first pulse can be determined as J, which is a sudden increase m in the number of times of the first pulse.

That is, the position of the first pulse increases rapidly in Figure (d) compared to Figure (C).

Similarly, when the pulse image moves to the right, overscaling of the second pulse from the right edge can be detected by a sudden decrease in the position of the first pulse. Therefore, when overscaling is detected, the initial movement amount is the position difference between the new 1st pulse and the old 2nd pulse when moving to the left, and the position difference between the new 2nd pulse and the old 1st pulse when moving to the right. It is sufficient to take the position difference as the amount of movement.

However, since an actual pulse image has a large width, it is not possible to accurately determine the center position of the pulse at the edge on the image sensor. To determine the exact pulse center position and use this value, determine the virtual school end and M@setting node. In short, in the relationship between the target position on the image sensor and the pulse position, when a specific pulse satisfies a predetermined positional relationship, successive pulses are noted as specific pulses. described a method using one light guide path and one image sensor. Measuring dish above]! In P, it has been assumed that the image pulse is ideal and only the position can be specified, without changing the width. However, in reality, as shown in FIG. 7(a), the image pulse on the image sensor has a maximum width of 1/4 of the length of the image sensor. To find the center position from this wide pulse video, generally, the video signal is cut at a certain threshold value Vo, shaped into a rectangular waveform as shown in Figure 7 (1)), and then the shaped rectangular wave is The center position of is being calculated. In addition, a light beam (FIG. 3(a)) 25 is applied to the optical displacement scale 20 by the optical projector 42.
(7) Bright [C] As shown in Figure 3 (b17), it has a Gaussian distribution. Furthermore, since the light beam is oriented at a constant angle with respect to the optical scale, the Gaussian distribution of its intensity is also asymmetric. Therefore, the pulse image on the image display also does not have a uniform waveform as shown in FIG. 3(C). Furthermore, the image sensor that receives the pulsed pattern also has non-uniform sensitivity among each pixel. Furthermore, the optical characteristics of the one-pulse pattern of the optical displacement school can also be made to have an ideal pulse waveform.

Because of these non-uniformities, the same pulsed image reflected from the same pulsed pattern on the optical displacement scale and projected onto the image sensor, and the relative movement of the pulsed pattern, i.e., the pulsed image on the image sensor. As the position of the pulse image changes, the shape of the pulse image may change. for example,
The pulsed image 26 shown in FIG. 3(d) is distorted like the pulsed image 27 as it moves on the image sensor. Therefore, when we cut the video pulse at a certain level and shape it into a rectangular wave, we find the center position of the pulse, as shown in Figure 3(e).
The center position shown in the figure becomes 26a127a, giving an error in the measurement of the pulse position.

In order to reduce this error, the present invention employs the following means.

At least two systems of projection light are provided on the optical displacement scale 20, and the reflected light is received by each image sensor, and the movement of the pulse image is determined independently for each scan, and the six movements i! The configuration is such that the actual amount of movement is determined by averaging llff1. This averages out the errors mentioned above. As a first example, Figure 4 (a
), two light sources Ll and 12 are installed, and the light sources illuminate two different locations on the optical displacement scale 20,
The configuration can be such that these reflected lights are received by the image sensor 81 and $2.

In the second example, as shown in FIG. 4(b), the luminous flux emitted from a single light source is
It is possible to separate the beam into two systems using, for example, a half mirror M as the beam separating means, and as shown in the figure, the beam 4q having a symmetrical light distribution can be received by the image sensors S1 and 82 and averaged as described above.

In this case, two systems of luminous flux are obtained from a single light source, and since the distribution of the illuminance is symmetrical to each other, the movement of the pulse image determined by J to the image sensors S1 and 82 It is possible to mutually compensate for the errors in (a).

Also, a third example is given in the preferred embodiment and shown in FIG. 4 (C
), as in the previous second example, two lights are emitted from a single light source.
Obtain the luminous flux of the system and move them upwards on the optical displacement scale.
The optical system is configured to irradiate the areas and the distribution of the irradiation intensity is mutually symmetrical, and the images are received by image sensors S1 and 82, and the amount of pulse image movement is averaged. It is structured like this. In this case, since the reflected images from the same part are captured and their illuminances are symmetrical, it is possible to compensate for errors with high precision.

In summary, the measurement principle of the present invention is to form at least two pulse images on each image sensor, focus on one pulse, and measure the amount of movement of that pulse within a measurable range. After that, the average value of the movement amount measured for each image sensor is calculated, and if it becomes impossible to measure at the next scanning time, the second
This method focuses on a pulse and accumulates the average value of the movement amount of the second pulse with the previous value to obtain the overall movement amount. Therefore, the accuracy of the total displacement does not depend on the accuracy of the pitch interval of the optical displacement scale that determines the period of the first and second pulses. Resolution is determined by the precision of the pixel width that forms the pixels on the image sensor and the magnification of the optical projector that projects the pattern on the optical displacement scale onto the image sensor. Measurement accuracy is determined by the resolution and the number of scans per unit movement.

Hereinafter, the present invention will be explained in more detail using Examples. FIG. 5 is a configuration diagram showing a specific embodiment of the optical displacement measuring device according to the present invention. The optical displacement scale 208 is made by vacuum-depositing chromium on a glass substrate to create a chromium thin film, etching parts 22 at 0.1 mm intervals by photo etching to remove the chromium, and removing the chromium from parts 21. This is an elongated grid-like pulse pattern constructed as a chromium thin film.

The optical displacement scale 20a is fixed to a movable base.

On the other hand, the optical system 40a is fixed on a fixed base 41, and includes two light sources 421a and 4'2111 that emit light onto the optical pattern of the optical displacement scale, and optical lenses 422a and 422b that emit these as parallel light beams. The image sensor 4.4a detects the light reflected from the optical pattern.
, 44b and magnifying lenses 423a, 42
It consists of 3b. The output of each image sensor is input to a pulse image position detecting device 60a, 60b, which will be described later, and the signal is input to a data processing device 80, which is comprised of a bullet meter system and processed according to software, which will be described later. Data office 3! l! The device 80 is connected to the medium heat processing device 82 via a generally known interface 88a, coupled with a memory 84 for storing predetermined blocks and data, and transmits the calculation results to an interface 88b. The data is output to the CRT display 86 via the J: 10.

Figure I6 is a block diagram showing a specific configuration of the pulse image position detection device 60a.

The output of the CCD 44a constituting the image sensor/14a passes through an amplifier 602 and enters the inverting input of a comparator 604. The non-inverting input of the comparator 604 is connected to the reference voltage V by resistance division.

is applied. The output of the comparator 604 is input to a sample hold circuit 606.

Here, the output voltage V1 of the amplifier 602 is shown in FIG.
．． It shows the light and dark pattern of 1mm imaged on COD. The vertical line conceptually represents a pulse waveform corresponding to one pixel signal of the CCD 44a. In reality, COD is composed of 2048 pixels, so it is composed of 512 pulses.The comparator 604 uses the noise margin level Vo as the reference level and In other words, the voltage V2 that has passed through the comparator 604 has a waveform as shown in FIG. 7(b). The square wave-shaped envelope formed by
The signal is sampled and held at the peak value of the pixel signal through the delay circuit 614 in synchronization with the clock pulse for driving the OD/I4a and in synchronization with the center of the pulse width of one pixel signal. As a result, the output voltage V3 is as shown in FIG. 7(C).
You will get two square waves like this.

Here, the horizontal axis represents the scanning time and also corresponds to the position in pixel units on the CCD.

The timing generation circuit 612 converts Star 1 to A1 to C.
Cf) It is input to the drive device 610 and provides the start of COD scanning. Then, following the star mill signal 4
A pulse train is scanned from the COD in synchronization with a clock signal of M, Hz.

On the other hand, the output voltage 3 of the sample and hold circuit 606 is applied to the periodic signal input terminal (Cp terminal) of the flip-flop circuit FF1 (hereinafter, the flip-flop circuit is referred to as rFFJ).
When input to , the polarity is reversed and input manually to the Cp terminal of FF2 (622).

In addition, V3 is input as is to the CI) terminal of FF3 (623), and is inverted and input to the cp terminal of FF4 (624). Each FF is a D-type circuit, and a positive potential is always applied to the D input of F[1.

Further, a start signal A1 is input to the Bricella l-temporary terminals of FF1 to FF4. And 1:F2
-FF4 is connected to J: where the Q terminal output of the previous stage is respectively input to the D input of the next stage. Further, each "[: circuit d output terminal is connected to the NAND circuit 631,
It is connected to one terminal of 632, 633, and 634.

Further, the 4MH2 clock pulse of the timing generation circuit 612 is input to the other terminal of each NAND circuit.

The outputs of each NAND circuit are input to counters C1 (641) to C4 (644), respectively, and the number of pulse trains output from the NAND circuits is calculated. Each counter C1 to C4 transfers the binary parallel signal sent to Callan 1 in a transfer amount of 1! 1] are transferred to and stored in buffer registers R1 to R4, respectively, by A2. Each buffer register R1 to R4 is connected to an interface 8.
8a so that it can be read by a computer.

The count values of each counter C1 and C4 of this pulse position detection device 60 are L1 time and t2 in FIG.
2 time t33 time t41M time respectively 4M i-I
This is the value obtained by the Z clock. .

Next, regarding the operation until 11 o'clock at each counter,
This will be explained with reference to FIG. From the timing generation circuit, a signal to the 1 star 1 shown in FIG. 8(b) is applied to the terminal of each FF at time to, and the start signal causes
All FFs are reset to 1~.

The start signal is also input to the temporary terminals of each counter, and similarly, the signal pulse J2 is sent to the star 1, which is then reset 1~. Subsequently, when V3 is input to the Cp terminal of FF1 after reset, the voltage ■3 becomes a timing signal for the FF1, so the voltage changes to 8 at the rising time 11 of the first pulse P1 of ■3.
3 and changes to Sera 1~ state. Therefore, the Q terminal output of FFI becomes as shown in Fig. 8 (CG). That is, a waveform is obtained that is applied to Relen 1 at time to and applied to Sera I at time 11. The 01 terminal output of this FF1 is transmitted to the D terminal of the next stage FF2, and the inverted input of v3 is input to the Cpts child. , delayed at time t22, F
The 02 terminal output of F2 is in the 122 time set state as shown in FIG. 8(e). Hereinafter, since the Q terminal output of the previous stage becomes the D input of the next stage, the output to the Q output end 1 is delayed until it is synchronized with each timing signal. As shown in FIG. 8 (0), the output from the 03 terminal becomes a signal that is activated at the rising time t3 of the second pulse P2.The output from the Q terminal of the FF4 becomes a signal at t4
The output is set at J'3. Therefore, the d-terminal output of these FF circuits is inverted and becomes lζ
Figure 8 (d), (f), (h), respectively.
j). Each of these signals corresponds to each NA
Power is applied to one terminal of the ND circuit.

The NAND circuit 631 therefore has one terminal at a high level from time to to t1, thereby overriding the effect of passing the clock pulse of the other terminal input. and during that time = 1-
The clock of MHZ is counted by the counter C1 (641). Similarly, in the NANO circuit 632, the one terminal input of A is a high level signal until time to~(2) as shown in FIG. 8(f). ) The J counter C2 will be at 81:1 during that time. Similarly, NAND circuit 6
33 passes the clock pulses from [0 to [3], and the counter C3 passes the clock pulses between times 1 and 1.
The NAND circuit 634 avoids passing the clock pulse at time to-t44, and the counter C4 counts the clock pulses during that period. As a result, the buffer 7 registers R1 to R4 are stored at time t1, time t22, t33, and time t44 of the V3 voltage shown in FIG. q (a), respectively.
The time display changes to t°b. The pulse image position detection device 60b that processes the signal waveform obtained from the image sensor 441) of another system also has a similar configuration.

Next, the contents processed by the computer of the data processing device 80 will be explained with reference to FIGS. 9 and 10. 9th
The figure shows flowcharts 1 to 1 of the computer software. The calculation of the amount of movement of the pulse image detected by the first system of image sensors will be mainly described below. First, it is assumed that the pulse image is moving at a constant speed to the left in FIG. Data processing device 80
When the program is started, a predetermined program is executed, and in step 100, various parameters used in this program are initialized to O. Here 1
is a counter indicating the number of scans of the image Renly.

Pl and P2 are addresses for storing the center positions of the first and second video pulses, respectively. Xl and X2 are memories for storing and holding the positions of pulses obtained by scanning as values before one scanning. Q is a part that determines whether the position of the pulse from the previous scan is in the valid zone or the prohibited zone and stores that information, and R is a part that does not indicate the pulse position information from the current scan and stores the parameters. It is a memory that keeps remembering.

D is a displacement amount of 1J34 per scanning interval of the identified video pulse, and [p is a cumulative average displacement amount, that is, a quantity proportional to the amount of movement of the object. 2 is a memory that stores the moving speed per unit time, that is, the moving speed calculated at each scanning interval.

Next, in step 102, the routine detects whether data is input to the buffer registers R1 to R4 of the pulse image position detecting device 60 and the registers are in a ready state. If no data is input, the ready bit is not set and the program waits until it becomes ready] - Overturn. When the machine 10 is ready, in step 104, the contents of buffer registers R1 to R4, in which the numerical values of each counter are stored, are transferred to the addresses of memories 01 to C4, respectively. Next, in step 106, in order to measure the middle part of the pulse, the center of the first and second pulses is calculated using the formulas Pl = (C1+02)/2, P2 = (C304)/2. Calculate the position and set it to P.
l, stored in P2. Then, in the next step 1108, it is determined whether Pl is larger or smaller than L. This is the lζ value set in advance, and is the length of the prohibition zone set at a certain distance from the left end of the image lens 1, that is, the younger pixel address, as shown in FIG. 10(a). C'
be. Here, the section at the top of the diagram is called a prohibited zone, and the remaining section from 81 to $4 is called a valid zone. Step 1
In step 08, it is determined whether the position of the first pulse is in the prohibited zone or in the valid zone. If you are in a prohibited zone,
Proceeding to step 110, the parameter R is set to 1. In other words, if the parameter R is 1, it means that the first
This means that a pulse is present. If it is not in the prohibited zone, a value of 2 is set for J:SR in step 112.

That is, when the value of 1 is 2, it means that the first pulse exists in the effective zone. Next, the process moves to step 114, and it is determined whether ■ is O or not. The process moves to step 116, and the positions of the first and second pulses just read are moved to memories x1 and x2 and stored.The position state of the first pulse is also stored in Q.Then, step 118
Increment the odor T-scanning number parameter (2) by 1.

Next, in step 119, system change processing is performed.

That is, data 1q received from the second image sensor 44b is processed in the same way. In the same figure, for easy understanding 1
Although the same character variables are used, the addresses to be stored are actually divided into two series: the first system and the second system. The following processing is also repeated alternately between the first system and the second system.

The signals read from the image sensor in the second scan are subjected to similar processing in steps 104 to 114, respectively. Therefore, in step 114, since this is the second reading, the value of ■ is not O, and the process moves to the next step 120. In step 120, the value of Q and the value of R are compared. That is, R is a parameter indicating the position of the first pulse measured this time, and Q is a parameter indicating the position state of the first pulse measured last time. Therefore, when Q is 2 and P is 2, it can be seen that the first pulse exists in the effective zone in both the previous scan and the current scan. Therefore, the first
If the pulse exists in the effective zone, the first pulse is used as a specific pulse, the deviation between the previous position and the current position is calculated, and in step 122, the value is stored as the value of D. If the first pulse obtained by scanning at time t 1 with respect to time t is on the left side of the drawing, the image moves toward the smaller address number on the frame, and the value of D shows a negative value. There is. Further, if P1 moves rightward in the drawing, the value of D shows a positive value.

Next, in step 1232, the movement ff1-
If D is based on the first system, in step 1234, Dl is set and the amount of C movement is stored. Next step 11
Return to step 102 via 6.118.119;
The movement by the second system is determined in the same way. Step 123
In step 1236, the amount of movement J to the second system is stored as D2. Next, the process moves to step 1238, and the average value of the movements ff1D1 and D2 obtained in the first system and the second system is stored in D as an average movement amount.

Next, in step 124, c15T, complete transfer 1rIIIff
l-Ll) Perform all calculations. That is, the average movement ff1D obtained for each scanning period is 8! Count i. Next, in step 126, the average movement fitD is divided by the scan interval time to determine the displacement speed. Then, in step 128, the amount of movement Lp up to the current time and the movement speed V at the current time are displayed, and the process returns to step 116. In step 116, the current pulse position is stored in Xl and X2, respectively, and the current position state information of the first pulse is stored in Q, and step 118
Then, I is incremented from 1 to 1 and the process returns to step 102.

In the third scan, calculations were performed in the same manner (), and this time the process moved to step 130. That is, at the Ti scan time, the first pulse existed in the effective zone because Q was 2. However, at 1-2 this time, R became 1, meaning that he had entered the prohibited zone. Figure 10 (
The positional relationship of the pulses is as shown in C). In this case, the process moves to step 132, where Pl of the first pulse at the current time is subtracted from the previous ×1, and when this value is smaller than the predetermined value MJ:, the shift is made to the left according to the normal scanning interval. It can be done without looking at things that are within the scope of what is possible. Therefore, step 132
The result is NO, and it can be determined that the first pulse is moving to the left in FIG. 10. In step 144, since the first pulse entered the forbidden zone, the second pulse
Switch the pulse to a specific pulse. That is, the moving color of the second pulse is determined here. That is, D is PI-X2. This relationship is shown in FIG. 10(C). After calculations are similarly made for the second system, the average movement amount is determined in step 1238, and then in step 124, the previous absolute movement amount L11 is cumulatively added. Hereafter, the same processing as the previous time is performed. Then, it returns to the beginning and performs the fourth scan.

The image pattern obtained in this scan is similar to that shown in Figure 10 (C), and is obtained by moving the pulse position further to the left at time T22. Therefore, it is the same as the previous scan. Next, according to the fifth scan, the previous first pulse P1 has scaled over to the left. Therefore, what was the second pulse at time T3 is now the second pulse at time T4. will be detected as the first pulse. Therefore, in the scan at T4, the first pulse will be detected as the first pulse.
The pulse state memory 〇 is 1 and R is 2. Therefore, the process reaches step 140 through the same routine as described above. At step 140, J3 is determined as YES. In step 143, )) 1-X 1 is calculated and is greater than the predetermined value M. This indicates that the first pulse has scaled over from the left end. Therefore, step 14
1, and the value obtained by subtracting the position of the previous second pulse x 2 from the position P1 of the current first pulse is calculated as the amount of movement. At T5, which is the next sixth scanning time, in order to show a state diagram similar to that in FIG. 10(a),
) through the routine described in the figure. Therefore, the total moving wind when the image pulse continuously moves to the left in the image lens can be found by the above-mentioned repeated operation of calculating the average shift @ amount of the moving amount uniquely obtained by the first system and the second system. It turns out.

Next, the pulse image is displayed in the upper right direction of the image sensor, that is, the 11th
Let us consider the case of moving at a constant speed to the right in the figure. Assume that at time To, the positional relationship is as shown in FIG. 11(a). In this case as well, the pulse image to be noted is the first pulse existing in the effective zone. Therefore, in the scan at time To, the state memory (i) becomes R=2, and in the second scan J3, the pulse sequence becomes as shown in FIG. 11(b). Calculate the amount of displacement because the first pulse P1 will appear.The pulse that should be noted is the second pulse.
It must be pulse P2. Therefore, in this case, the status display memory shows Q=2 and R=1. Therefore, the determination in step 130 is YES, and the process moves to step 132. Here, in this case, xi-pi is g1
It can be seen that this value is larger than the predetermined maximum movement ff1M. That is, from Fig. 11<a) to (
b) shows that the first pulse P1 is actually the next counting pulse. Therefore, in such a case, it can be determined that the image is moving to the right. J, the determination in step 132 is YES, and the process moves to step 13/I. The pulse to be specified at time 1 is P2, and since the pulse to be specified at TO was Pl, the displacement MD must be calculated as P2-Xl.

At the next scanning time T3, the first pulse 1]1 is the first pulse
As shown in FIG. 1(C), the state has already entered the prohibited area, so there is no change in the state from FIG. 1(b), and Q-1 and R=1 are shown. Therefore, the determination in step 136 is YES, and the process moves to step 138, where the second pulse is calculated as the pulse to be specified. Therefore, the displacement I of the second pulse
Th I) 2-X2 is calculated as the amount of displacement.

At the next scanning time T3, a pulse positional relationship as shown in FIG. 11(d) is obtained. According to this, the first pulse moves from the prohibited zone to the valid zone, so the Q
-1, R=2. However, the determination in step 140 is YES, and the process moves to step 143. In step 143, it is determined whether the difference between the current first pulse position and the pulse position during the previous scan, that is, Pl-Xi, is greater than or smaller than a predetermined value M. If the pulse is large, it indicates that the pulse is moving to the left, and if it is small, it indicates that the pulse is moving to the right, so the process moves to step 138. That is, in step 138, the deviation is determined by focusing on the second pulse. At the next scanning time T4, a3 is shown in FIG. 11(e).
The pulse arrangement will be as shown in . In this case, the first pulse is in the effective zone during the previous scan and is also in the effective zone during the current scan. Therefore, Q=2, R
=2, the determination in step 120 is YES, and in step 122, the first pulse is focused on and its deviation is calculated. Then, the C sumitub 124 cumulatively adds them.

As described above, it is possible to measure leftward movement, rightward movement, and mixed left-right movement by combining these movements in the same manner.

Although the diagrams in FIGS. 10 and 11 schematically depict a large amount of movement, the amount of movement is actually very small. In the second embodiment below, one effective zone and one prohibited zone are defined for each image sensor υ, but the means for selecting a specific pulse for detecting displacement is in no way limited to this method. isn't it.

If the one-dimensional arrangement direction of the image sensor is shifted at an acute angle to the grid pattern of the optical scale, the distance is 0.1 m.
Of the optical patterns arranged at intervals of m, two patterns can also be arranged on the reimage pin υ'. It is also possible to arrange two images on the image sensor by enlarging them using a lens. In addition, in this example, displacement measurement in a straight line was used, but it is also possible to arrange a pulse pattern on the circumference and rotate it to detect rotational displacement and rotation speed using a similar method. It is possible. The point is that the present invention requires at least two pulse images to be projected at the same time, and the average value of each movement amount found in multiple systems is calculated and the average movement M is accumulated. Sometimes, attention is paid to the subsequent second pulse, and the average movement amount of that pulse is further cumulatively added.

According to the device according to the invention, the measured viscosity is therefore completely independent of the accuracy of the pulse period interval. Since the image detects the position of the pulse image on the laser using a clock of 4 Ml-lz, extremely accurate measurement is possible. Since one pixel of an image sensor is generally made up of 10-odd microns, the resolution is a magnification of 10-odd microns.

As a result, it can be considered that the resolution is on the order of several O9 microns. In addition, the pulse image measurement system has multiple stages,
In order to average the amount of movement, due to the non-uniformity of the illuminance of the light source, the non-uniformity of the pixel sensitivity of the image sensor, and the non-uniformity of the optical characteristics of one pulse pattern of the optical displacement scale,
As the pulse image moves, measurement errors caused by distortion of the pulse image are alleviated, allowing highly accurate measurement. In particular, the method described in the embodiment described above has a large error compensation (R effect).In this way, the device of the present invention can measure extremely precise displacement or velocity without any problem with the pitch interval of the optical scale. is possible.

Therefore, it has the characteristic that the optical scale can be manufactured easily.

[Brief explanation of drawings]

FIG. 1 is a block diagram illustrating the inventive concept of an optical displacement detection device according to the present invention. FIG. 2 is a principle explanatory diagram illustrating the displacement measurement principle of the device of the present invention. FIG. 3 is an explanatory diagram showing the non-uniformity of the pulse image on the image sensor I dog. FIG. 4 is an explanatory diagram showing the configuration of the optical system of the apparatus of the present invention. FIG. 5 shows the configuration of an optical displacement detection device according to a specific embodiment of the present invention. FIG. 6 is a block diagram 17 showing the detailed configuration of the internal pulse image position detection device of the same embodiment. FIG. 7 shows a signal waveform of the 1ζth signal, which explains the operation shown in FIG. 4. Figure 8 also shows the V of the device shown in Figure 6.
FIG. 9 is a flowchart showing the processing flow of the data processing device used in the same embodiment. FIG. 10 shows the concept of left movement processing. It is a conceptual explanatory diagram that covers the explanation.
The figure is also a conceptual explanatory diagram showing the rightward movement process. 20... Optical displacement scale 40... Optical system 42... Optical projector 44a, 4.4b... Image sensor (]-6
0a, 60b...Pulse video position detection device 8
0...Data processing device 90...Pulse video specifying unit 92...Movement focusing unit 93...Movement m averaging unit 94...・Average Moving Village Estimation Department Special R'F Applicant Toyota Machinery Co., Ltd. Agent Patent Attorney Hirodo Okawa Patent Attorney Shudo Fujitani Patent Attorney Akio Maruyama Figure 10 Figure 11

Claims (5)

[Claims] (1) An optical displacement scale whose optical characteristics have a pulse-like pattern that periodically varies with respect to the direction of relative displacement, and at least two images that receive the optical pulse-like pattern of the optical displacement scale. an optical pulse-like pattern of a sensor and said optical displacement scale is imaged onto each said image sensor such that at least two moving pulses are generated on each image sensor at any given time by relative movement with said optical displacement scale; an optical system that controls an optical projector for obtaining an image; and a pulse signal corresponding to the pulse image is input from each of the image sensors by scanning it at regular intervals, and the pulse signal corresponding to the pulse image is inputted to the image sensor V of the pulse image. at least two pulse image position detection devices that detect the position of the pulse position detection device; and a data processing device that processes the output signals of each of the pulse position detection devices every scanning period to measure the displacement of the movable object, and the data processing device The processing device includes a pulse image specifying unit that identifies a pulse image for which each movement amount of the pulse image during at least one past scanning period can be determined based on the positional relationship of the pulse image on each image sensor for each scanning circle 11; a movement calculating section which focuses on the pulse image specified by the pulse image specifying section and calculates the amount of movement of the pulse image for each scanning period; an average value extraction unit' that calculates the average value of the image movement amount; and a movement amount integration unit that sequentially controls the pulse image identification unit and the movement amount calculation unit for each scanning period and integrates the average value of the pulse image movement amount. 1il of the pulse image as the scanning period progresses.
The optical displacement side is characterized by cumulatively adding up the average value of each pulse image movement amount on each image sensor by paying attention to successive pulses in response to the scattered movement. Fixed device. (2) The optical displacement measuring device according to claim 1, wherein the optical displacement scale is an elongated optical displacement scale fixed to a moving stage that moves in a linear direction. . (3) The optical displacement scale is a love-imole circular scale disposed on the circumference of a disc, and the optical displacement scale is arranged such that the rotation of the optical displacement scale and the rotation of the rotating body are synchronously rotated. 2. The optical displacement measuring device according to claim 1, wherein the optical displacement measuring device is connected to a rotating body to measure rotational displacement of the rotating body. (4) The optical projector includes a single light source, a light beam separating means for separating a single light beam from the light source into two systems, and a light beam separated into the two systems by the optical scale. and the irradiation intensity distribution,
The optical system according to any one of claims 1 to 3, comprising an optical lens system that is symmetrical with respect to a direction of relative movement, and the irradiation pattern is received symmetrically by two image sensors. Displacement 11111 constant device. (5) With respect to the direction of relative movement, the light beams having two opposite intensity distributions are projected onto the optical scale at the same 81 ((Q) There is a patent claim characterized in that the illumination angles of the two systems of light beams are different (thereby, the two systems of light beams with different anti-sword angles are received by two image sensors). The optical 1 internal variation Bt measuring device according to item 4.