JP2940234B2 - Velocity distance measuring device using spatial filter - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for measuring a velocity and / or a distance using a spatial filter, which has been improved to improve measurement accuracy.

[0002]

2. Description of the Related Art Conventionally, in the automatic traveling of an automatic guided vehicle or the like, a traveling guide is formed by laying an electromagnetic induction wire or an optical reflecting tape on a traveling path, or an encoder or a tachogenerator is attached to an axle or a measuring wheel. There is a method of measuring the speed and moving distance of an unmanned vehicle from a pulse or an analog voltage according to the rotation of a wheel.

However, these methods require construction work for laying a reflective tape or the like on the road surface, so that the work must be performed every time the traveling path is changed, which is troublesome. Accurate measurement was not possible due to wheel wear.

For this reason, as a method of measuring the moving distance in a non-contact manner without the need for external guidance, a velocity distance measuring method using a spatial filter has been developed.

[0005] This distance measuring method using a spatial filter is performed by the apparatus configuration shown in FIG. That is, as shown in the figure, the reflected light from the road surface 1 is detected by a one-dimensional line sensor (CCD) 3 via an optical system 2, sampled at a predetermined cycle, and converted into an electric signal. The optical system 2 and the line sensor 3 are mounted on the bottom surface of a vehicle body (not shown). The line sensor 3 is one in which optical conversion elements are arranged one-dimensionally along the traveling direction, and signals according to light and dark are sequentially output.

An output signal P from the line sensor 3 passes through a read circuit 4 and is converted by an A / D converter 6 into a digital signal CCD.
After being converted into data, it is input to the spatial filter operation unit 5. The spatial filter operation unit 5 integrates this output signal within a predetermined range, thereby extracting an arbitrary predetermined frequency component from the reflected light composed of optically random frequencies. This is for obtaining a vector corresponding to the moving distance in the phase space shown in FIG.
That is, the CCD data D _i (i = 1, 2,... N) from the line sensor 3 composed of N pixels is represented by a weight function S _i (i = 1, 2).
2,... N), and the products of the same suffix are sequentially added N times to perform a product-sum operation to obtain a product-sum value b _n . Here, the load function _Si is shown in FIG.
As shown in FIG. 7, the product is the product of the sine wave function and the Hanning function. This sine wave function has n in the length L of the line sensor.
And the wavelength p (= L / n)
Is called a filter pitch.

Next, when the load function C _i (i = 1, 2,... N) is read, the CCD data D _i (i = 1, 2,.
Between N), the product of equal subscript product-sum operation is performed by the added successively N times, Sekiwachi a _n is obtained. Here, the weight function C _i is a product of the cosine wave function and the Hanning function, as shown in FIG.

Further, since the sum-of-product values a _n , b _n are obtained for each sampling, the sum-of-product values of the previous sampling are calculated as a
_n-1 and b _n-1, and the product-sum value of the current sampling is a _n and b _n . In this manner, the vectors A ₁ , at each sampling time in the phase space in which the product sum values a _n−1 , b _n−1 , a _n , b _n are two-dimensional coordinates.
A ₂ is shown as FIG. Accordingly, the phase difference [Delta] [phi _n between previous and current sampling phase rotation calculating unit 7
Is obtained as in the following equation (1).

[0009]

[Number 1] _{Δφ n = arctan {(b n} -1 · a n -a n-1 · b n) / (a n-1 · a n -b n-1 · b n)} ... formula (1)

Here, the moving distance x ₀ and the phase difference Δφ _n
Is proportional to the filter pitch p
(= L / n) and 2π, the moving distance x ₀ and the velocity V are obtained as in the following equations (2) and (3). However, the phase difference between one sampling is within one rotation.

[0011]

X ₀ = (p / 2π) · ΣΔφ _n Equation (2)

[0012]

V = (p / 2π) · Δφ _n Equation (3)

Therefore, in the speed-distance calculating section 11, the phase difference Δφ _n is summed up by the accumulator 8 and (p / 2π) is integrated by the multipliers 9 and 10, and the moving distance x ₀ and the speed are respectively calculated. V is required.

[0014]

As described above, the velocity-distance measuring device using the spatial filter has a feature that non-contact and high-precision measurement is possible.

However, in a place where the reflected light from the road surface changes suddenly or becomes irregular, there is a problem that the accuracy of detecting the moving distance and the speed is reduced.

That is, in a measuring device using a spatial filter, the line sensor 3
There is an area that deviates from the measurement range of, and a new area that becomes the measurement range, and the speed and the moving distance are detected on the assumption that the distributions of the spatial frequency components included in those areas are equal. Are different in their distribution. Therefore, the difference in the distribution appears as an error of calculation of the phase difference [Delta] [phi _n, it was tied to a decrease in measurement accuracy.

Therefore, the present inventors have already devised and applied for a moving distance measuring device with relatively high detection accuracy even when the reflected light from the road surface is not constant (see Japanese Utility Model Application No. 63-63). No. 145969, Japanese Patent Application No. 1-289980),
Nevertheless, there was a limit to improving accuracy.

The present invention has been made in view of the above prior art, and provides a measuring apparatus capable of accurately detecting a speed and a moving distance in a place where reflected light from a road surface changes suddenly or in an irregular place. The purpose is to do so.

[0019]

According to the present invention, a spatial distance measuring apparatus using a spatial filter comprises: a photoelectric conversion element for converting incident light into an electric signal; sampling from the photoelectric conversion element at a predetermined period; A space for extracting an arbitrary frequency component by separately multiplying a sine wave and a cosine wave as a weight function by an electric signal sequentially output from the read circuit for each sampling and integrating the electric signal within a predetermined range. A filter operation unit, and a phase difference Δφ _n between the previous and current samplings centered on the origin on the phase plane of a vector having two-dimensional coordinates of the two signals extracted by the spatial filter operation unit. And a phase difference Δφ calculated by the phase rotation calculating section.
_In a speed-distance measuring device with a spatial filter having a speed-distance calculating unit that calculates at least one of the speed and the moving distance proportional to _n ,

There are provided two spatial filter operation units using a load function whose cycle is shifted by a constant relative phase difference ψ as the spatial filter operation unit, and the two spatial filter operation units are extracted as the phase rotation operation unit. The phase difference Δφ _1n in the phase space based on the frequency signal,
And two phase rotation calculation units for calculating Δφ _2n , and the two signals extracted at the same sampling time by the two spatial filter calculation units are used as two-dimensional coordinates. A phase difference calculating unit for calculating the phase difference, and calculating the previous phase difference Δθ _n-1 and the current value Δθ _{n of} the phase difference obtained by the phase difference calculating unit, the relative phase difference ψ, and the two phase rotation calculating units. Using the obtained phase differences Δφ _1n and Δφ _2n , Δφ _n = (Δφ _1n + Δφ _2n )
A correction operation unit for obtaining a phase difference Δφ _n by a correction operation of ψ / (Δθ _n-1 + Δθ _n ), and at least one of the speed and the moving distance is obtained from the phase difference Δφ _n obtained by the correction operation unit; It is characterized in that the speed-distance calculation unit performs calculation.

[0021]

[Operation] Since there is a relative phase difference の 間 に between the load functions used by the two spatial filter calculation units, the device of the present invention
This is equivalent to two spatial filters being geometrically arranged in series at a distance D corresponding to ψ. Therefore, when the reflected light from the measurement object does not change suddenly, and is not irregular and does not change in the distribution of the spatial frequency component, the phase difference Δφ _1n and Δφ _2n obtained by the two phase rotation calculation units match. . In addition, both the previous value Δθ _n−1 and the current value Δθ _{n of} the phase difference obtained by the phase difference calculation unit match the relative phase difference ψ.

Conversely, when there is a change in the distribution of the spatial frequency component, generally, Δφ _1n ≠ Δφ _2n , Δθ _n-1 ≠ Δθ _n ≠
ψ. However, considering the average value, (Δθ _n-1 + Δ
θ _n ) / 2 and ψ: K = 2ψ / (Δθ _n−1 + Δθ _n )
(Δφ _1n + Δφ _2n ) / 2 should change in accordance with.

Therefore, Δφ _n = (Δφ _1n + Δφ _2n ) ψ
/ (Δθ _n-1 + Δθ _n ) and a correction operation is performed. As a result, even if the distribution of the spatial frequency component changes, it is possible to calculate the moving distance and speed with high accuracy.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to embodiments shown in the drawings. The same parts as those in the above-described embodiment are denoted by the same reference numerals and suffixes, and the description is omitted.

FIG. 1 shows an embodiment of the present invention. As shown in the drawing, this embodiment has two types of spatial filter operation units 5a,
5b, phase rotation calculation units 7a and 7b, and a phase difference calculation unit 12, a correction calculation unit 13, and the like.

The spatial filter operation units 5a and 5b have the same basic structure as the spatial filter operation unit 5 in FIG. 3, but use a weight function having the same period but a relatively different phase difference ψ. Here, π / 4 was set as the relative phase difference ψ. That is, as compared with the weighting function of the spatial filter calculation unit 5a illustrated in FIG. 6A, the phase of the weighting function of the spatial filter calculation unit 5b illustrated in FIG. 6B is shifted by π / 4. Then, the computed sum of product value by the spatial filter operation section 5a and a _n, a b _n, respectively, and the computed sum of product value by the spatial filter operation portion 5b c _n, and d _n. Sekiwachi a _n, b _n is the preceding value _{a n-1, b n-} 1 with the phase rotation operation portion 7a, Sekiwachi c _n, d _n is the preceding value _{c n-1, d n-} 1
Along with the phase difference Δφ shown in FIG. 2 according to the following equations (4) and (5).
_1n and Δφ _2n are calculated.

[0027]

(Equation 4)

[0028]

(Equation 5)

The phase differences Δφ _1n and Δφ thus calculated
_2n should match if there is no change in the spatial frequency distribution, but in general, they often do not match because there is a change in the spatial frequency distribution.

On the other hand, the sum of products a _n ,
b _n, c _n, d _n are input to the phase difference calculator 12, the phase difference [Delta] [theta] _n by the following equation (6) is calculated.

[0031]

(Equation 6)

The current phase difference Δ thus calculated
theta _n and the phase difference [Delta] [theta] _n-1 of last time, if there is no change in the distribution of the spatial frequency, coincide with each other, also but should coincide in the relative phase difference ψ example [pi / 4 shown in FIG. 2, the actual Has a change in the spatial frequency distribution, and therefore does not often match the relative phase difference 上 記.

Then, the ratio of the average value of the phase differences Δφ _n-1 and Δθ _n to the relative phase difference ψ is expressed by the phase differences Δφ _1n ,
In order to correct Δφ _2n , the previous value Δθ _n-1 and the current value Δθ _n of the phase difference and the phase differences Δφ _1n and Δφ _2n are input to the correction calculation unit 13. The correction operation unit 13 corrects and obtains the phase difference Δφ _n according to the following equation (7).

[0034]

Δφ _n = (Δφ _1n + Δφ _2n ) · ψ / (Δθ _n-1 + Δθ _n ) Equation (7)

The phase difference Δφ _n thus corrected by the correction calculation unit 13 is then input to the speed-distance calculation unit 11 and moves through the accumulator 8 and the multipliers 9 and 10 shown in FIG. The distance x ₀ and the speed V are calculated.

Next, the validity of the above-described error correction will be described. Now, let the phase at the position x on the unmanned vehicle travel line be ψ (x), and the phase change f
Let (x) be the following equation (8).

[0037]

(Equation 8)

Where x _n is the sampling timing n
Where D is the distance between the spatial filters and d is the moving distance, the sampling timing n
The phase difference Δθn between the spatial filters in the above equation is _expressed by the following equation (9), and the phase difference Δφ _1n and Δφ _2n between one sampling of each spatial filter, that is, n−1 and n, is expressed by the following equation (10).

[0039]

(Equation 9)

[0040]

(Equation 10)

Then, f in the above equations (9) and (10)
When the term relating to the integral of (x) is approximated by a trapezoid, the following equation (1) is obtained.
1) and (12) are obtained.

[0042]

[Equation 11]

[0043]

(Equation 12)

From the above equations (11) and (12), D is the distance between the spatial filters and therefore corresponds to the geometric relative phase difference 間 の between the spatial filters, and d is the moving distance, so that Δφ _n
In consideration of the above, the following expression (13) is established. Therefore, the above-described error correction equation (7) is established.

[0045]

Δφ _1n + Δφ _2n = (Δθ _n-1 + Δθ _n ) d / D = (Δθ _n-1 + Δθ _n ) Δφ _n / ψ Equation (13)

The speed / distance measuring device of the present embodiment having the configuration shown in FIG. 1 is implemented according to the flowchart of FIG. That is, the spatial filter operation unit 5a, a product sum value a _n by 5b, b _n, c _n, is d _n is calculated, then the above-mentioned phase rotation calculating unit 7a based on the expression, 7b, the phase difference computation unit 1
2, Δφ _1n , Δφ _2n , Δθ _n-1 , and Δθ _n are calculated. Then, the correction computing unit 13 Equation (7) is computed, corrected phase difference [Delta] [phi _n is determined. Then, from the phase difference [Delta] [phi _n, the distance calculating unit 11, the moving distance x ₀ by accumulation of the phase difference [Delta] [phi _n, The velocity V is calculated without accumulation.

[0047]

As described above, in the present invention, velocity and distance are measured using two spatial filters. In this case, by providing two spatial filter operation units using a load function whose cycle is shifted by the relative phase difference ψ, and obtaining the phase differences Δθ _n-1 and Δθ _n in the phase space at the same sampling time, Should be the relative phase difference ψ, assuming no error. However, where the reflected light changes abruptly or irregularly, errors tend to occur in the measurement accuracy, and the phase difference Δθ _n-1 and Δθ _n do not actually match the relative phase difference ψ. , The error is corrected and the phase difference Δ in the phase space between the previous and current samplings
The phase difference Δφ _n is obtained from the sum of θ _1n and Δφ _2n . Thereby, even if the distribution of the spatial frequency changes, for example, where the reflected light is not constant, the measurement error can be reduced.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a velocity distance measuring device using a spatial filter according to one embodiment of the present invention.

FIG. 2 is a diagram showing rotation of a vector in a phase space.

FIG. 3 is a diagram showing an example of a conventional apparatus.

FIG. 4 is a diagram showing rotation of a vector in a phase space.

FIG. 5 is a diagram showing an example of a load function.

FIG. 6 is a diagram showing an example of a load function used in one embodiment of the present invention.

FIG. 7 is a flowchart according to an embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Measurement object (road surface) 2 Optical system 3 Line sensor 4 Readout circuit 5a, 5b Spatial filter calculation part 6 A / D converter 7a, 7b Phase rotation calculation part 11 Speed distance calculation part 12 Phase difference calculation part 13 Correction calculation part

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-3-152468 (JP, A) JP-A-62-273064 (JP, A) JP-A-2-67214 (JP, U) JP-A 63-73068 142758 (JP, U) (58) Fields investigated (Int. Cl. ⁶ , DB name) G01C 22/00 G01B 11/00 G01P 3/36

Claims (1)

(57) [Claims] 1. A photoelectric conversion element for converting incident light into an electric signal, a readout circuit sampling from the photoelectric conversion element at a predetermined cycle and sequentially outputting the electric signal, and an electric signal sequentially output from the readout circuit for each sampling Sine wave and cosine wave are separately multiplied as a weighting function and integrated within a predetermined range, thereby obtaining a spatial filter operation unit for extracting an arbitrary frequency component, and two signals extracted by the spatial filter operation unit. a phase rotation calculator for calculating a phase difference [Delta] [phi _n between the previous and current sampling centered at the origin on the phase plane of the vector with two-dimensional coordinates, determined by the phase rotation calculating unit in the speed range finder according to a spatial filter and a distance calculating section for calculating at least one of the speed and the moving distance proportional to the phase difference [Delta] [phi _n, the spatial filter Starring The constant relative phase difference と し て
And two phase filter operation units each using a load function whose cycle is shifted only by a period, and a phase difference Δφ _1n in a phase space based on frequency signals extracted by the two spatial filter operation units as the phase rotation operation unit. , Δφ _2n , and a phase difference Δθ of a vector in a phase space in which two signals extracted by the two spatial filter calculators at the same sampling time are two-dimensional coordinates. And a phase difference calculation unit for calculating
Further, the previous value Δ of the phase difference obtained by the phase difference calculating section
Using θ _n−1 and the current value Δθ _n , the relative phase difference ψ, and the phase differences Δφ _1n and Δφ _2n obtained by the two phase rotation calculation units, Δφ _n = (Δφ _1n + Δφ _2n ) · ψ /
The phase difference Δφ _{n is} obtained by the correction operation (Δθ _n-1 + Δθ _n ).
The correction operation unit for obtaining the provided correction calculation unit speed distance measuring apparatus according to spatial filter the distance calculating unit at least one of the speed and the moving distance from the phase difference [Delta] [phi _n obtained is characterized by calculating the .