JPH02249914A - Position detecting device - Google Patents

Abstract

PURPOSE:To detect the present position of a moving object with high accuracy by correcting the deflection length calculated by a deflection arithmetic circuit from the reference length stored in a storage circuit by a correcting circuit. CONSTITUTION:On a metallic scale plate 10 placed along a running path of a moving object 1, a slit 11 is pierced at a prescribed pitch. In this state, when the object 1 moves, a CCD sensor 14 photodetects parallel rays from a light source 13 through the slit 11, and by a binarizing circuit 15, four kinds of limited patterns C are outputted. In this state, the present position of the object 1 is detected by these four kinds of patterns. In this case, the present position of the object 1 can be detected as a position shifted in the advance direction from a '0' point of the scale plate 10 by a value obtained by adding or subtracting the deflection length calculated by a deflection arithmetic circuit 16 from the reference length stored in a storage circuit 17 by a correcting circuit 19.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a position detection device that non-contactly measures the amount of movement of an object moving on a travel path and moves the object a desired distance.

[Conventional technology and its issues]

In an unmanned multi-storey warehouse or the like, a moving object (1) such as a stanker crane as shown in FIG. 5 is automatically moved along a travel path. There is a rail (2) below the moving object (1).
is arranged, and a power supply line (3) is installed on the upper side in parallel with the rail (2). Furthermore, a position detection device for measuring the travel distance of the moving object (1) is installed on the outside of the moving object (1).

Position detection devices include those that use a rotary encoder, those that use a Magnescale, and those that optically measure travel distance to detect a position.

Position detection devices using rotary encoders include those in which the moving object (1) is equipped with a rotary encoder, and a roller that moves on the floor is fixed to the rotary shaft of this rotary encoder, and the moving object (1) is mounted on the floor. There is one in which a rank is arranged along the running path, and a pinion that meshes with the rack is fixed to the rotating shaft of a rotary encoder.

The rotational speed of the roller and pinion is converted into a linear distance by a rotary encoder, the travel distance of the moving object (1) is measured, and the position of the moving object (1) is detected.

However, if the roller slips on the floor, errors will occur and
Moreover, if the slip continues, errors will be accumulated, resulting in a problem that the distance determined by the rotary encoder and the actual distance traveled will be significantly different. Furthermore, when a rank and pinion are used, there is a problem in that the vehicle jumps when the vehicle suddenly stops, making it impossible to measure the actual distance traveled.

A position detection device using Magnescale has a magnetic recording surface formed on a hard, long scale, and a magnetic sensor that runs while maintaining a small distance from the scale when moving an object (1).
It is fixed to.

In order to read magnetic records accurately, the scale is made hard, but a hard scale has the disadvantage of being easily broken. Furthermore, the magnetic sensor must be moved while maintaining a small distance from the scale, but since the moving object (1) oscillates laterally, it is difficult to use it in practice.

Therefore, in order to overcome the drawbacks of the above two types of contact-type position detection devices, a non-contact optical position detection device has been developed.

An optical position detection device, for example, consists of depositing a large number of slit-like thin metal wires in the length direction on a long glass plate, and counting the number of the thin metal wires with an optical sensor attached to a moving object. The current position of a moving object is detected based on the count value.

In order to increase the accuracy of this optical position detection device, the width of the thin metal wires can be made thinner and the number of thin metal wires can be increased.
Since thin metal wires are formed by vapor deposition, there is a limit to how thin the width can be, and even if the width of thin metal wires can be made thin, there may be problems in use such as measurement errors when dust adheres to them.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a position detection device that is easy to manufacture, has good dust resistance, and can detect the current position of a moving object with high precision.

[Means to solve the problem]

In order to achieve the above object, the present invention is arranged along the traveling path of a moving object, and by drilling slits at a constant pitch, the center of the slit part and the center of the non-slit part are alternately set as a reference point at a constant pitch. a scale plate that appears at , and has one end set as the 0 point; a CCD sensor that is fixed to a moving object, has one end as the origin, and has a length capable of recognizing at least two or more reference points from the origin; A light source that emits parallel light to the CCD sensor through a slit, and a binarization circuit that binarizes the output of the CCD sensor and distinguishes a slit part and a non-slit part from an "H" signal as an "L" signal. The deviation is determined by identifying the intermediate position from the origin to the inverted part of the first "H" signal and "L" signal and the next inverted part as the detection line, and calculating the distance from this detection line to the origin as the deviation length. A storage circuit that stores the distance from the 0 point of the scale to the reference point currently detected as the detection line as a reference length, and a storage circuit that stores whether the origin of the output signal of the CCD sensor is an "H" signal or an "L" signal. Recognizes the current detection pattern based on the number of signals and inverted parts, and compares it with the previously recognized pattern to determine whether the currently recognized reference point has shifted in the forward or backward direction, and determines that it has shifted. In this case, add or subtract the length of 1 pitch of the reference point from the previous reference length to create a new reference length.
The apparatus includes a pattern recognition circuit that is stored in the storage circuit, and a correction circuit that corrects the reference length based on the deviation length calculated by the deviation calculation circuit and outputs the result as the current position of the moving object. be.

[Effect]

The current position of the moving object is determined from the 0 point on the scale plate by the CCD.
Calculated as the distance to the origin of the sensor. This current position can be detected by the correction circuit subtracting or adding the deviation length calculated by the deviation calculation circuit from the reference length stored in the storage circuit.

〔Example〕

An embodiment of the present invention will be described with reference to FIGS. 1 to 4.

In the figure, (10) is a metal scale plate arranged along the traveling path of the moving object (1), for example, 1.
5 (tm) wide slit (11) with a width of 3.0 (n
) at a constant pitch, and as shown in Figures 2 and 3, a reference point (
Assume A). In addition, as shown in Fig. 3, the left end of the scale plate (1o) is set as 0 point (0), and the first reference point (A) and the second reference point (A, The reference point (A) is 1.5, such as point (A2).
(w) so that it appears at a constant pitch.

If the reference points (A) appear at a constant pitch, you are pickpocketed.

The widths of the gates (11) do not have to match. (12) is a moving object (
1), with a slit (
11), the light source (13) and the CCD sensor (14)
Bury it facing each other.

From the light source (13), towards the CCD sensor (14),
Emits parallel light (L). As shown in FIG. 3, the CCD sensor (14) has, for example, the left end as the origin (14a), and can recognize at least two reference points (A) from this origin (14a). The length of [crane]
For example, there are 128 cells (not shown) inside and 32 cells (not shown).
They are encapsulated at a pitch of [μm]. (15) is the above C
The output from the CD sensor (14) is binarized, and as shown in Fig. 3, for example, a binarization circuit outputs a slit part as an "H" signal and a non-slit part as an "L" signal. be.

(16) is a deviation calculation circuit which inputs the signal binarized by the binarization circuit (15), and as shown in FIG. ”The detection line (
B) and from this detection line (B) to the origin (1
4a) is calculated as the deviation length (x). Since light is diffracted, the inverted part of the "H" signal and "L" signal may not coincide with the edge of the slit (11), but since the light is similarly diffracted at the edge part, the detection line (
B) can be accurately specified and the deflection length (X) can be accurately calculated.

(17) is a storage circuit that stores the distance from the zero point (0) of the scale plate (10) to the currently detected detection line (B) as a reference length (y). (18) inputs the output signal of the binarization circuit (15) and receives the moving object (
1), the pattern recognition circuit recognizes the pattern (C) shown in FIG. 3 <a> to (d) and outputs the reference length (y) to the memory circuit (17). Recognized pattern (C) and currently recognized pattern (C)
It is determined whether the currently recognized reference point (A) has shifted in the forward or backward direction, and if it is determined that it has shifted, the reference point (A) is set to the previous reference length (y). Add or subtract the length (a) of 1 pitch to create a new standard length (
y), stored in the storage circuit (17) as (
19) is a correction circuit that calculates the current position (z) of the moving object (1), and this correction circuit (19) is connected to the memory circuit (
The current position (z) is calculated by subtracting or adding the deviation length (X) calculated by the deviation calculation circuit (16) from the reference length (y) stored in 17).

The position detection device according to the present invention is configured as described above, and next, the current position (2), that is, the scale plate (10), (7)
From 0 point (0) to CCD-t':/su (14) origin (14
The means for measuring the distance to a) will be explained.

When the moving object (1) moves, the COD sensor (14) receives parallel light (L) from the light source (13) through the slit (11), and the binarization circuit (15) converts it into a third parallel light.
Only four kinds of limited patterns (C) as shown in FIGS. (a) to (d) are output. The first one shown in Figure 3(a)
The pattern (C1) has an "L" signal at the origin (14a) and has two inverted parts. The second one shown in Figure 2(b)
In the pattern (C2), the origin (14a) is the "L" signal,
It has three inverted parts. In the third pattern (C8) shown in FIG. 2(c), the origin (14a) is an "H" signal,
It has two inverted parts. In the fourth pattern (C4) shown in FIG. 2(d), the origin (14B) is the "H" signal, 3
It has an inverted part. The pattern (C) shown in FIG. 2(e) is the same as the first pattern (C) shown in FIG. 2(a).

The current position (Z) of the moving object (1) is detected as follows using these four types of limited patterns (C). The moving object (1) is traveling in the direction [rightward in the drawing]
From the state where the origin (14a) of the COD sensor (14) coincides with the zero point (0) of the scale plate (10),
As shown in FIG. 3(a), it is assumed that the detection line (B1) of the first pattern (C1) moves until it detects the first reference point (A1). The current position (zl) of the moving object (1) at this time is determined by the deviation length ( The scale board (10
) can be recognized as a position shifted from the 0 point (0) in the direction of travel.

When the moving object (1) moves further in the advancing direction, a second pattern (C2) with a shorter deflection length (x2) is output, as shown in FIG. 2(b). The second pattern (
The detection line (B2) of the first pattern (C2)
), the first reference point (8) is detected. Therefore, the current position (z2) is the same as the deviation length (yl) from the reference length (yl). When the moving object (1) moves further in the direction of movement, a third output pattern (C8) as shown in FIG. 2(c) is output. The detection line (B3) of the third output pattern (Cs) detects the second reference point (A2), and the pattern recognition circuit (18) detects the first reference point detected by the second output pattern (C2). A storage circuit (19) determines that the deviation has occurred from the point (A1) in the traveling direction and stores a new reference length (ya) obtained by adding the length (a) of one pitch to the reference length (yl). Therefore, the current position (z3) is stored at the new reference length (ya
) by subtracting the new deviation length (xl). In this way, the pattern recognition circuit (18) recognizes the pattern (C) and determines the reference length (y).
The current position (Z) can be detected by adding or subtracting the length (a) of one pitch to or from the current position (Z).

Whether to add or subtract the length (a) by one pitch to the reference length (y) will be explained with reference to FIG. 4. When changing from the first pattern (C) to the second pattern (C2),
And the third pattern (Ca) to the fourth pattern (
C4) and vice versa, the reference length (y) stored in the storage circuit (17) is changed from the second pattern (C2) to the third pattern (
C3) and the fourth pattern (C4)
When changing from to the first pattern (C1), the reference length after the change (yn) is the value obtained by adding the length of one pitch (a) to the reference length before the change (ynt). .

Conversely, from the first pattern (CI) to the fourth pattern (
C4) and the third pattern (Ca
) to the second pattern (C2), the moving object (1) has retreated, and the reference point (A) detected by the detection pattern (B) is close to the 00 point on the scale plate (10). The reference length (y n) after the change is the length of one pitch from the reference length (ynt) before the change.
Take the value obtained by subtracting .

By the way, the CCD sensor (14) is, for example, 4 [Tsuru]
Since 128 cells (not shown) are enclosed at a pitch of 32 μm within the width of
) is 1.93 (Tsuru) and the standard length (yl) is 2.25
(n), and in this case,
The current position (zl) is (zt) = ()'1) -
(xt), 0.3 from the 0 point of the scale plate (10)
2 (Tsuru) It can be seen that the position is shifted in the traveling direction [to the right in the drawing]. Therefore, even if the width of (11) and the distance between the reference points (A) are 1.5 (m) , the current position (2) can be detected on the order of 0.01 (m).

In this way, the current position (z) of the moving object (1) is set to 0.
01 (tm), and when it is determined that the moving object (1) has traveled a desired distance, it outputs a signal to stop the drive source (not shown) of the moving object (1).

Note that the 0 point (0) on the scale plate (10) is, for example, 2.
5 (tm) width slits (hereinafter referred to as reference slits) can be bored every 1.8 (m) and set at the left edge of this reference slit. Moving object (1
) travels 10 (m), 0 point (0) becomes 1
．． Since it is updated every 8 (m), the moving object (1)
Even if an error occurs such as failing to detect the slit (11) due to sudden movement of the reference slit, this error will not be accumulated. The current position (Z) can be detected by memorizing this number and adding the reference length (y) to the value multiplied by the distance between the reference slits, and subtracting the deviation length (x) from this. .

The above describes one embodiment according to the present invention, and the present invention is not limited to this embodiment, and the design can be changed within the gist of the present invention. For example, the origin of the CCD sensor can be set opposite to the CCD sensor. When setting at the side end, the correction circuit adds rather than subtracts.

〔Effect of the invention〕

According to the present invention, since the width of the slit is relatively large, the scale plate can be manufactured easily and the manufacturing cost can be reduced. Moreover, if the width of the slit is large, the slit will not be clogged with dust, and erroneous measurements can be eliminated. Furthermore, since the measurement is performed without contact, the position can be accurately detected without being affected by vibration.

Furthermore, since it is measured with a CCD sensor, 0.01 (m)
It becomes possible to detect the highly accurate position of the order.

[Brief Description of the Drawings] Fig. 1 is a circuit configuration diagram including a partial schematic perspective view of the position detection device according to the present invention, Fig. 2 is a partial sectional view of the detection section, and Fig. 3 is a scale plate and CCD sensor. FIG. 4 is a diagram showing the output pattern of the CCD sensor that changes depending on the positional relationship between the pattern and the reference length. FIG. 5 is a schematic perspective view of a moving object as an example to which the position detection device is applied. (1”)--Moving object, (10)--Scale plate, (11)--Slit, (13)
-・Light source, (14) ・-=CCD sensor, (14a
) - Origin, (15) - Binarization circuit, (16
)--Deflection calculation circuit, (17)--Storage circuit, (B) (L) (X) ---Pattern recognition circuit, ---Correction circuit, (A) ---Detection line, ( C) --- Parallel light, (0) --- Deflection length, (y) ---- Current position. -Reference point 1, -pattern, -0 point, -reference length, patent applicant Hokuyo Electric Co., Ltd.

Claims (1)

[Claims] (1) By placing slits along the travel path of a moving object and drilling them at a constant pitch, the center of the slit part and the center of the non-slit part appear alternately at a constant pitch as a reference point, and one end A scale plate set as the 0 point, a CCD sensor fixed to a moving object, with one end as the origin, and long enough to recognize at least two reference points from the origin; A light source that emits parallel light, a binarization circuit that binarizes the output of the CCD sensor and distinguishes the slit part and non-slit part as an "H" signal or an "L" signal, and the first "H" signal from the origin. A deviation calculation circuit that identifies the intermediate position between the inverted part of the "L" signal and the next inverted part as a detection line, and calculates the distance from this detection line to the origin as the deviation length; A memory circuit that stores the distance from the 0 point to the reference point currently detected as the detection line as the reference length, and whether the origin of the CCD sensor output signal is an "H" signal or "L" signal and the number of inverted parts. Recognizes the current pattern and compares it with the previously recognized pattern to determine whether the currently recognized reference point has shifted in the forward or backward direction.If it is determined that the reference point has shifted, the previous reference length is A pattern recognition circuit which adds or subtracts the length of one pitch of the reference point from the reference point and stores it in the storage circuit as a new reference length, and the deviation length calculated by the deviation calculation circuit, A position detection device comprising: a correction circuit that corrects a reference length and outputs the corrected reference length as the current position of a moving object.