JP2000146494A - Impact position standardizing and unexploded bomb identifying apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately estimate without influence to weather by receiving a flying sound or an explosion sound of an artillery shell by at least four acoustic sensors separately installed in advance and estimating an impact point or the like of the shell by utilizing a change in a Doppler frequency or time difference of the like for an explosion sound to arrive at a plurality of sensors. SOLUTION: The impact position standardizing and unexploded bomb identifying apparatus comprises at least four acoustic sensors 7a to 7d separately installed in advance. Their detection signals are input to signal processors 10a to 10d, passed through a band-pass filter, then A-D converted into digital signals, the digital signals are further converted into an RF band, which are in turn transmitted to a signal processor through transmitters 11a to 11d. The data received by the processor are transferred to signal processors 14a to 14d, and rising times of explosion sounds are measured. Then, a temporary impact position Pa is obtained by an arithmetic circuit 15 by utilizing an explosion sound arriving time difference, a change in the Doppler frequencies or the like between the systems, and corrected based on influences of an air temperature and a wind to estimate an impact position Px.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an impact position locating method and an unexploded object discriminating method for estimating a drop position and an explosion position of a high-speed flying object such as a shell in a training ground and determining an explosion of the shell. It concerns the device.

[0002]

2. Description of the Related Art In a conventional impact position estimating device using sound, a method of using a hyperbolic characteristic based on a rising time difference between explosive sounds, receiving the sound of a shell shot or an explosion sound by a plurality of acoustic sensors. Thus, two-dimensional sound source position estimation could be performed. There has also been an impact position estimation device using infrared rays.

[0003]

In the conventional position estimation device utilizing sound, the altitude estimation when a shell explodes in the air and the use of a bombshell that does not generate an explosion sound when struck by an unexploded ordnance do not occur. It is difficult to estimate the impact position.
Further, it is difficult to estimate a position in the case of poor visibility such as fog by using a landing position estimating device using infrared rays.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problem, and has as its object to estimate a three-dimensional position of a bullet by using sound waves generated by a bullet regardless of explosion, misfire, or weather conditions. I have.

[0005]

In order to solve the above-mentioned problems, at least four or more acoustic sensors which are separately arranged for detecting a time difference between sound waves generated from a shell and a signal received by each sensor. The obtained signal is wirelessly transferred to a signal processing unit described later, and an appropriate processing method is selected for each type of sound wave, and the impact position of the shell is three-dimensionally (latitude, longitude, elevation) based on each processing result. An impact position locating and unexploded ordnance discriminating device, comprising: means for estimating and displaying the information on an operation unit to be described later; and means for determining the explosion of a shell.

[0006] Also, a bullet-jumping sound generated from a shell in flight or a sound wave of a single-frequency sound source whose frequency changes continuously with time is received by an acoustic sensor, and the obtained signal is subjected to fast Fourier transform. (Hereinafter referred to as FFT), a band-pass filter for removing out-of-band frequency components, a means for storing the filtered signal in a memory circuit via an A / D converter, and an FFT by a method described later. Doing, means for detecting the characteristic portion of the signal that changes in time and frequency of the sound wave as a line,
Means for performing a wavelet transform on a signal near the line disappearance time with reference to the detected line to remove noise components, and a function corresponding to a change in frequency versus time of the line.
(X) is created, and the cross-correlation between the noise-removed signal fα (X) and the function ψξ (X) is obtained at each sampling interval.
A circuit having means for accurately measuring the line disappearance time is provided in four systems of A, B, C, and D, and a circuit is also provided for calculating a disappearance position of a sound source from a line disappearance time difference between the respective systems. 2. A method for estimating a sound source position, comprising:

Further, a characteristic feature of a signal which changes in time and frequency which indicates a change in Doppler frequency present in the shell shot sound by means of the same means as in claim 2 can be used for the shell shot sound generated from the shell in flight. A means for detecting a line and a means for performing a least-squares approximation process on the detected line are respectively performed for four systems A, B, C, and D, and a line (Doppler shift curve) and a trajectory calculation in each system are performed. 2. The apparatus according to claim 1, further comprising means for comparing the terminal trajectory obtained by the above-mentioned method and estimating the trajectory and impact position of the shell.

[0008] Also, a sound wave and the like at the time of the explosion of the shell are received by an acoustic sensor, transferred to a signal processing unit described later via an A / D converter, and a circuit for measuring the rising time of the sound wave is provided. 2. A sound source position estimating method according to claim 1, further comprising means for estimating a sound source position in three dimensions (latitude, longitude, altitude) by utilizing a difference in arrival time of sound waves between the systems. Landing location and unexploded ordnance identification device.

In addition, the above-described impact position locating and unexploded ordnance discriminating apparatus, which corrects the temperature difference of the air in which the sound wave propagates in the region where the position is estimated and the refraction of the sound wave due to the wind to reduce the position estimation error. Sound wave correction means.

[0010] A wind direction and wind speed measuring means relating to the above-mentioned impact position locating and unexploded ordnance discriminating device for measuring the wind direction and wind speed of a sound wave propagation path using sound waves.

[0011]

This device receives at least four or more acoustic sensors installed in isolation beforehand, which receives the sound of a flying or explosive sound from a shell, corrects the refraction of sound waves due to the temperature of the sound wave propagation path and the wind, and adjusts the Doppler frequency. By using the time difference when the change or explosion sound reaches multiple sensors or the time difference when the flying sound disappears with multiple sensors,
It is characterized by estimating the trajectory of a shell, the point of impact, and the altitude at the time of an explosion, regardless of whether or not there is a misfire.

According to the second aspect of the present invention, a sound from a projectile or a single-frequency sound source generated from a projectile during flight is received by an acoustic sensor so that the frequency of the sound changes continuously with time, and an FFT is performed from the acquired signal. Is performed, a band-pass filter is passed to remove out-of-band frequency components, the filtered signal is stored in a memory circuit via an A / D converter, and FFT is performed in the following manner. The signal f to be processed when performing FFT
When the frequency of (X) is f ₀ ± 1000 (Hz),
Sampling is performed at about 2 × (f ₀ +1000) (Hz), an appropriate window function is applied for each data number Ap point, and the frequency resolution f is such that f ≦ (f ₀ +1000) / (Ap). The FFT is performed by using an FFT size and inserting zeros other than data. However, the second and subsequent FFT
Represents the number of data points (Ap-Bp) overlapped with the previous data, and combined with the newly acquired Bp point data, Ap
FFT is performed as a point (however, Ap> Bp and Ap and Bp use the optimum value for the time-frequency change of the signal to be processed). Thus, the signal f
The FFT is performed for the entire time of (X). This method aims to improve the frequency resolution when analyzing time-frequency changes. Normally, if the sampling interval is constant, the frequency resolution doubles if the number of samples doubles, but if there is a time-to-frequency change due to the increase in the time width of the window function, examine the frequency change every minute. Can not be done. However, in the above method, data for the Ap point is used for the frequency resolution and data for the Bp point is used for the time resolution, so that time-frequency analysis with high resolution is possible. However, in this method, a single-frequency sound source is time-dependent due to the Doppler effect or the like.
This is effective for analysis in a case where the frequency change continuously changes below the FFT resolution. When the frequency level changes with time, correction is required for the following reasons. For example, when the frequency of the signal g (X) changes with time and the level of the signal is constant, the frequency near the center of the window function on the time axis of the signal g (X) to which the window function has been applied is changed after the FFT is performed. The level in the frequency domain becomes maximum,
If the center of the window is shifted in the time direction, the frequency of the signal g (X) existing near the center of the window becomes the maximum level, so that the frequency for each time shift can be analyzed. However, the signal f whose frequency level changes at each time as shown in FIG.
In the case of (X), the signal at the center of the window does not reach the maximum level. Therefore when carrying out the FFT is subjected to window function signal f (X), the frequency having the maximum level for the frequency of the window center is a frequency f _M or more shown in FIG. 7 to be lower, and increased in the following f _M Become. The shift between the frequency at the center of the window and the maximum level frequency after FFT increases as the time-level change becomes steeper. Therefore, the window function w (X) is added to the signal f (X).
If the applied, is multiplied as signal present at the center of the window function f (X) is the maximum level function w _{F (X)} is a signal f (X), always on the time axis of the window function w (X) 3 near the center of
One for the level of function of the product _{f (X) w (X)} w F (X) is the maximum, when the FFT the signal f (X), f near the center of the time axis of the window function w (X) The level of the frequency according to (X) becomes maximum. In order to check the frequency level, the function w _F (X) is removed, and the FF is again obtained in the same time domain.
By performing T, the level of the frequency with respect to the original signal f (X) can be obtained. Therefore, it is possible to perform a one-hour frequency analysis every (Bp / 2) / (f ₀ +1000) seconds without lowering the frequency resolution. In this manner, a means for detecting a characteristic portion of a temporally and frequency-changed signal indicating a Doppler frequency change present in a shell shot sound as a line, a means for performing least-squares approximation on the detected line, Time change of frequency from line and signal f
Means for measuring the extinction time of (X). However, the time information is lost by the FFT within the time width of the window function.
It is impossible to measure whether (X) has disappeared by the above method. Therefore, with reference to the line detected by the above means, an appropriate wavelet is selected from a library of several types of orthonormal bases for a signal near the line disappearance time, and a wavelet transform is performed on the signal f (X).

[0013]

(Equation 1) And means for removing noise components of signals near the line disappearance time. Here, ψ φ _(X) is the wavelet function, performed efficiently removing noise from a signal f (X) by performing the selected wavelet transform optimal wavelet from a library of wavelets. Also,
A function ψξ (X) corresponding to the frequency-time change of the line is created, and the noise-removed signal fα (X) and the function ψξ
A cross-correlation of (X) is obtained at each sampling interval, and four systems of A, B, C, and D are provided for accurately measuring the line disappearance time. It can be accurately estimated.

According to a third aspect of the present invention, the sound of a shell shot from a flying shell is received by four acoustic sensors, and a temporal processing indicating a Doppler frequency change present in the shell shot sound is performed by the same processing method as described above. And a process of detecting a characteristic portion of a signal that changes in frequency as a line, and performing a least-squares approximation process on the detected line for each of the four systems A, B, C, and D. Shift curve) and the terminal trajectory value obtained by the trajectory calculation,
The trajectory and the impact position can be estimated from the sound of the shell shot.

According to a fourth aspect of the present invention, there is provided a circuit for receiving a sound wave or the like at the time of explosion of a shell by an acoustic sensor, transferring the sound wave to an after-mentioned signal processing unit via an A / D converter, and measuring a rising time of the sound wave. Each of the four systems is provided, and the sound source position can be estimated in three dimensions (latitude, longitude, altitude) by deriving and solving six equations using the sound wave arrival time difference between the systems.

According to the fifth aspect of the present invention, a sound source position estimation error can be reduced by correcting a temperature difference of air in which sound waves to be observed propagate and a refraction of sound waves due to wind.

In the sixth invention, the wind direction and the wind speed of the region to be observed are calculated from the sound wave propagation time and are adopted in the above-mentioned correction method, whereby the error in estimating the sound source position can be reduced.

[0018]

FIG. 3 relates to an embodiment of an impact position locating and unexploded ordnance discriminating apparatus. The principle thereof will be described below with reference to FIG. 1a to 1d in the figure are acoustic sensors, 2a
2d denotes a temperature sensor, 3a and 3b denote calibration sound sources. The area surrounded by A, B, C and D is the landing, P1-P
Numeral 2 indicates the trajectory of the shell, and P2 is the point of impact. Also, K
1, K2 and K3 are arbitrary points on the trajectory of the shell. If the portion surrounded by A, B, C, and D is a rectangle,
Acoustic sensors [1a] at arbitrary positions near the point,
The acoustic sensor [1b], the temperature sensor [2b], and the calibration sound source [3b] are installed at arbitrary positions near the point B with the temperature sensor [2a] and the calibration sound source [3a]. An acoustic sensor [1c] and a temperature sensor [2] near the midpoint
c] is installed. A method of estimating a landing point will be described by taking, as an example, a case in which a shell that flies from P1 and passes through K1, K2, and K3 and hits P2 using a straight line extending in the vertical direction as the Z axis. When a shell flies below the speed of sound in the air, the air at the head of the shell vibrates and generates a sound wave.When this sound wave is observed on any ground along the path of the shell, the frequency of the sound wave depends on the observation point and the shell It changes depending on the angle with the traveling direction and the speed of the shell. Now
The shell flies from P1, passes through K1, K2, and K3, and
When hitting at two points, the acoustic sensor [1a] shown in FIG.
The frequency of the sound wave observed by the acoustic sensor [1d] is observed as K1>K2> K3 due to the Doppler effect, but the observed frequency and the frequency change rate per unit time are determined by the acoustic sensor [1d]. 1a] -Acoustic sensor [1
d] and the angle of the shell in the advancing direction. The time required for the sound wave generated at the point K1 to reach the acoustic sensors [1a] to [1d] varies depending on the distance to each acoustic sensor and the temperature, wind direction and wind speed. FIG. 4 shows the angle of the shell [5] and the distance from the acoustic sensor [4] when the shell [5] passes through the point K1, and the acoustic sensor [4] with reference to FIG.
A description will be given of the frequency of the flying sound and the arrival time of the sound wave observed in the above. Now, on the X, Y, and Z coordinates, the trajectory of the shell [5] when the shell [5] passes through the point K1 from the point P1 and hits the point P2 is represented by a broken line connecting the points P1 and P2. This dashed line is assumed to be on the X, Z plane, one point on the trajectory of the shell [5] is point K1, and its coordinates are (Xp, Y
p, Zp), the angle between the tangent of the trajectory of the shell at point K1 and the X axis is γ (°). The installation point of the acoustic sensor [4] is point S1, the coordinates are (Xs, Ys, Zs), the frequency of the sound wave generated at point K1 is F ₀ (Hz), and the advance of the shell [5] at point K1. Direction K Assuming that the degree is β (°), the velocity of the shell [5] is Vs (m / S), and the temperature in FIG. 4 is T (° C.), the sound wave F generated when the shell [5] passes through the point K1. The frequency f (Hz) and the sound wave arrival time ST (Sec) when ₀ (Hz) is observed by the acoustic sensor [4] are as follows.

[0019]

(Equation 2) (L2, M2, N2), the angle β between the two straight lines
(°) is β = Cos ⁻¹ (L1 · L2 + M1 · M2 + N1 · N
2) A = Xp-Xs, B = Yp-Ys, C = Zp-Z
s, D = -Cosγ, E = 0, F = Sinγ Next, in FIG. 3, sound waves generated when the shell flies from the point P1 to the point P2 are transmitted to the acoustic sensor [1b] and the acoustic sensor [1].
5A and 5B show the change of the frequency when the observation is performed in [c]. However, the frequency characteristic of the sound wave observed by the acoustic sensor [1b] is A, and the frequency characteristic of the sound wave observed by the acoustic sensor [1c] is B.

The characteristics of the frequency characteristic due to the difference in the installation points of the acoustic sensors and the method of estimating the landing point using the difference in the arrival time of the flying sound from the shell will be described with reference to the block diagram shown in FIG. . Before estimating the impact position, data such as ammunition type, launch angle, initial velocity, and temperature are input to the calculator [23] from the input device [19] of the operation unit, and signal processing is performed by the transmitter [11f]. To the department. The data received by the receiver [12k] of the signal processing unit is sent to the arithmetic circuit [15] via the control circuit [18], and performs a trajectory calculation to predict the terminal trajectory of the shell. The arithmetic circuit [15] calculates the frequency characteristics of the sound waves that will be observed by the sensors [7a] to [7d] in consideration of the influence of the wind and the temperature, and stores them in the storage circuit [16].
The flying sound data actually observed by the acoustic sensors [7a] to [7d] are respectively subjected to A / D after passing through the band-pass filters in the signal processing circuits [10a] to [10d] of the sensor unit. After the conversion and the conversion of the digital signal to the RF band, the transmitter [11a] to the transmitter [11a]
d] to the signal processing unit. The data received by the receivers [12e] to [12h] of the signal processing unit are converted to the signal processing circuits 2 [14a] to 2 [14].
d), a filter for performing a FFT by the method according to claim 2 and extracting a curve indicating a Doppler shift of the flying sound from the converted frequency domain data, and a minimum for performing smoothing of the extracted data. The square approximation is performed, and the result is stored in the storage circuit [16]. The calculated value data thus obtained is compared with actual data to calculate a trajectory error. If the error from the calculated value is equal to or greater than X, the direction and distance of the deviation between the actual trajectory and the calculated trajectory and the impact position are predicted from the error information, and the acoustic sensor [1a] is again used from the predicted impact position. Calculate the frequency characteristics observed by the acoustic sensor [1d]. This process is repeated by gradually reducing the calculation range until the error becomes equal to or less than X, and finally the impact point is estimated. The above has described the method of estimating the landing point using the frequency change of the shell flying sound generated while the shell is flying.

Next, a method of estimating a landing point using a time difference when the sound of the shells disappears will be described with reference to the block diagram shown in FIG. The flying sound data observed by the acoustic sensor [7a] to the acoustic sensor [7d] is converted into a signal processing circuit [10a] to a signal processing circuit [10] of the sensor unit.
d], the digital signal is A / D converted after passing through the band-pass filter and converted into the RF band, and then transmitted to the transmitter [11].
a] through the transmitter [11d] to the signal processing unit.
The data received by the receivers [12e] to [12h] of the signal processing unit are transferred to the signal processing circuits 2 [14a] to 2 [14d], and the FFT is performed by the method according to claim 2. Then, a filter for extracting a curve indicating the Doppler shift of the flying sound from the converted frequency domain data is applied and stored in the storage circuit [16]. A wavelet transform is performed by the method according to claim 2 with reference to a Doppler shift curve of the stored data to remove noise components from sound waves observed by the acoustic sensors [7a] to [7d]. The disappearance time of the flying sound caused by the shell is measured by the method described in Item 2, and the flying sound disappearing time difference between the acoustic sensors is calculated by the arithmetic circuit [15]. FIG. 6 shows an installation example of an acoustic sensor according to the present invention. Hereinafter, a method of estimating a landing position using a flight sound extinction time difference between acoustic sensors will be described with reference to FIG. S1~S4 represents an acoustic sensor, those _{coordinates, S1: (X 0, 0} , Z 1), S2: (0, -
_{Y 0, Z 2), S3} : (0, Y 0, Z 3), S4: (-
X ₀ , 0, Z ₄ ), P is the impact position, and its coordinates are P:
(X, Y, Z), and the distance from each of the acoustic sensors S1 to S4 to the impact position P is PS1 = r1, PS2 = r2.
If PS3 = r3, PS4 = r4, and the sound speed is C, the following four equations are derived.

[0022]

(Equation 3) Also, R ₁₂ = r1−r2, R ₁₃ = r1−r3, R
_{14 = r1-r4, R 23} = r2-r3, R 24 = r2
_-R4, When R 34 = r3-r4, it can lead to six equations below. ^{_{4Z 2 (Z 1 2 + Z}} 2 2 -R 12 2 -2Z 1 Z 2) + 4Z (-X 0 2 Z 1 -Y 0 2 Z 2 -Z 1 3 -Z 2 3 + 2XX 0 Z 1 -2YY 0 Z 2 ^{_{+ Z 1 R 12 2 + Z}} 2 R 12 2 + X 0 2 Z 2 + Z 1 2 Z 2 -2XX 0 Z 2 + Y 0 2 Z 1 + Z 1 Z 2 2 +2 YYoZ 1) + X 0 4 + Y 0 4 + Z 1 4 + Z 2 ^{_{4 + 4X 2 X 0 2 +}} 4Y 2 Y 0 2 + 2X 0 2 Z 1 2 + 2Y 0 2 Z 2 2 -4XX 0 3 + 4YY 0 3 -4XX 0 Z 1 2 + 4YY 0 Z 2 2 -4X 2 R 12 2 -4Y 2 ^{_{R 12 2 -2X 0 2 R 12}} 2 -2Y 0 2 R 12 2 -2Z 1 2 R 12 2 -2Z 2 2 R 12 2 + 4XX 0 R 12 2 -4Y Y 0 R 12 2 + R 12 4 -2X 0 2 Y ₀ ² -2 ^{_{0 2 Z 2 2 -4X 0 2}} YY 0 - 2Y 0 2 Z 1 2 -2Z 1 2 Z 2 -4YY 0 Z 1 2 + 4XX 0 Y 0 2 + 4XX 0 Z 2 2 + 8XX 0 YY 0 = 0 4Z 2 (Z ^{_{1 2 + Z 3 2 -R 13}} 2 -2Z 1 Z 3) + 4Z (-X 0 2 Z 1 -Y 0 2 Z 3 -Z 1 3 -Z 3 3 + 2XX 0 Z 1 + 2YY 0 Z 3 + Z 1 R 13 2 ^{_{+ Z 3 R 13 2 + X}} 0 2 Z 3 + Z 1 2 Z 3 -2XX 0 Z 3 + Y 0 2 Z 1 + Z 1 Z 3 2 -2 YY 0 Z 1) + X 0 4 + Y 0 4 + Z 1 4 + Z 3 4 + 4X ^{_{2 X 0 2 + 4Y 2 Y}} 0 2 + 2X 0 2 Z 1 2 + 2Y 0 2 Z 3 2 -4XX 0 3 -4YY 0 3 -4XX 0 Z 1 2 -4YY 0 Z 3 2 -4X 2 R 13 2 -4Y 2 R ₁₃ ² -2X ₀ ^{2 _{13 2 -2Y 0 2 R 13 2}} -2Z 1 2 R 13 2 -2Z 3 2 R 13 2 + 4XX 0 R 13 2 + 4Y Y 0 R 13 2 + R 13 4 -2X 0 2 Y 0 2 -2X 0 2 Z 3 ^{_{2 + 4X 0 2 YY 0 -}} 2Y 0 2 Z 1 2 -2Z 1 2 Z 3 2 + 4YY 0 Z 1 2 + 4XX 0 Y 0 2 + 4XX 0 Z 3 2 -8XX 0 YY 0 = 0 4Z 2 (Z 1 2 + Z 4 ^{_{2 -R 14 2 -2Z 1 Z 4}} ) + 4Z (4XX 0 Z 1 -4X X 0 Z 4 -Z 1 3 -Z 4 3 + Z 1 R 14 2 + Z 4 R 14 2 + Z 1 2 Z 4 + Z 1 Z 4 ^{_{2) + 16X 2 X 0 2}} + Z 1 4 + Z 4 4 -8XX 0 Z 1 2 + 8XX 0 Z 4 2 - 4X 2 R 14 2 -4X 0 2 R 14 2 -4Y 2 R 14 2 -2Z 1 2 R 14 2 -2Z _{4 ^{R 14 2 + R 14 4 -2Z}} 1 2 Z 4 2 = 0 4Z 2 (Z 2 2 + Z 3 2 -R 23 2 -2Z 2 Z 3) + 4Z (-4YY 0 Z 2 +4 YY 0 Z 3 -Z 2 3 ^{_{-Z 3 3 + Z 2 R 23}} 2 + Z 3 R 23 2 + Z 2 2 Z 3 + Z 2 Z 3 2) + 16Y 2 Y 0 2 + Z 2 4 + Z 3 4 + 8YY 0 Z 2 2 -8YY 0 Z 3 2 -4X 2 ^{_{R 23 2 -4Y 0 2 R 23}} 2 -4Y 2 R 23 2 -2Z 2 2 R 23 2 -2 Z 3 2 R 23 2 + R 23 4 -2Z 2 2 Z 3 2 = 0 4Z 2 (Z 4 2 + Z ^{_{2 2 -R 24 2 -2Z 2 Z}} 4) + 4Z (-X 0 2 Z 4 -Y 0 2 Z 2 -Z 4 3 -Z 2 3 -2XX 0 Z 4 -2YY 0 Z 2 + Z 4 R 24 2 + Z ^{_{2 R 24 2 + X 0 2}} Z 2 + Z 2 Z 4 2 ^{_{+ 2XX 0 Z 2 + Y 0}} 2 Z 4 + Z 2 2 Z 4 +2 YY 0 Z 4) + X 0 4 + Y 0 4 + Z 4 4 + Z 2 4 + 4X 2 X 0 2 + 4Y 2 Y 0 2 + 2X 0 2 Z 4 2 + 2Y 0 ^{_{2 Z 2 2 + 4XX 0 3}} + 4YY 0 3 + 4XX 0 Z 4 2 + 4YY 0 Z 2 2 -4X 2 R 24 2 -4Y 2 R 24 2 -2X 0 2 R 24 2 -2Y 0 2 R 24 2 -2Z 4 2 ^{_{R 24 2 -2Z 2 2 R 24}} 2 -4XX 0 R 24 2 -4Y Y 0 R 24 2 + R 24 4 -2X 0 2 Y 0 2 -2X 0 2 Z 2 2 -4X 0 2 YY 0 - 2Y 0 2 ^{_{Z 4 2 -2Z 2 2 Z 4}} 2 -4YY 0 Z 4 2 -4XX 0 Y 0 2 -4XX 0 Z 2 2 -8XX 0 YY 0 = 0 4Z 2 (Z 3 2 + Z 4 2 -R 34 2 -2Z ₃ Z _{4 ^{+ 4Z (-X 0 2 Z 4}} -Y 0 2 Z 3 -Z 3 3 -Z 4 3 -2XX 0 Z 4 + 2YY 0 Z 3 + Z 3 R 34 2 + Z 4 R 34 2 + X 0 2 Z 3 + Z 4 2 Z ^{_{3 + 2XX 0 Z 3 + Y}} 0 2 Z 4 + Z 3 2 Z 4 -2 YY 0 Z 4) + X 0 4 + Y 0 4 + Z 3 4 + Z 4 4 + 4X 2 X 0 2 + 4Y 2 Y 0 2 + 2X 0 2 Z 4 2 ^{_{+ 2Y 0 2 Z 3 2 +}} 4XX 0 3 -4YY 0 3 + 4XX 0 Z 4 2 -4YY 0 Z 3 2 -4X 2 R 34 2 -4Y 2 R 34 2 -2X 0 2 R 34 2 -2Y 0 2 R 34 2 ^{_{-2Z 3 2 R 34 2 -2Z 4}} 2 R 34 2 -4XX 0 R 34 2 + 4Y Y 0 R 34 2 + R 34 4 -2X 0 2 Y 0 2 -2X 0 2 Z 3 2 + 4X 0 2 YY 0 - 2Y ₀ ² Z _{4 ^{-2Z 4 2 Z 3 2 + 4YY}} 0 Z 4 2 -4XX 0 Y 0 2 -4XX 0 Z 3 2 + 8XX 0 YY 0 = 0 above six equations actually observed flying sound fading time difference S ₁₂ = (r1 −r2) / C, S ₁₃ = (r1−r3)
_{/ C S 14 = (r1-} r4) / C, S 23 = (r2-r3)
_{/ C S 24 = (r2-} r4) / C, S 34 = (r3-r4)
/ C, the distance of the difference between the acoustic sensors R ₁₂ = S ₁₂ · C, R ₁₃ = S ₁₃ · C, R ₁₄ = S
₁₄ · C R ₂₃ = S ₂₃ · C, R ₂₄ = S ₂₄ · C, R ₃₄ = S
Calculating the ₃₄ · C, and solving the simultaneous equations in the arithmetic circuit [15] shown in FIG. 1, the temporary bullet fixing position _{P a (X1, Y1, Z1} ) are determined, temperature and wind effect on the rear, wherein the correction to the bullet fixing position _{Px (X α, Y α,} Z α) is estimated.

In FIG. 1, data observed by the acoustic sensors [7a] to [7d] are converted into signal processing circuits [10a] to [10-] of the sensor unit.
d], the digital signal is A / D converted after passing through the band-pass filter and converted into the RF band, and then transmitted to the transmitter [11].
a] through the transmitter [11d] to the signal processing unit.
The data received by the receivers [12e] to [12h] of the signal processing unit are transferred to the signal processing circuits 2 [14a] to 2 [14d], and the rise time of the explosion sound is measured. utilizing the explosion sound arrival time differences between strains, and solving the simultaneous equations in the arithmetic circuit [15] in the same formula, provisional bullet fixing position _{P a (X1, Y1, Z1} )
Is obtained, by correcting the temperature and wind effect of post according bullet fixing position _{Px (X α, Y α,} Z α) is estimated.

When a shell is flying above the sound speed with respect to the atmosphere, a shock wave is generated by the shell. When the speed of the shell decreases and becomes lower than the sound speed, no shock wave is generated by the shell. Since the initial speed of the howitzer to be estimated by the device is much higher than the speed of sound, a shock wave is generated until the speed of the shell reaches subsonic speed. Therefore, when the airspeed of a shell passing near the acoustic sensor is higher than the sound speed,
The point of impact is estimated by the method described.

Next, a method of calculating the wind speed and direction by the sound wave will be described with reference to FIG. The calibration sound source [3 shown in FIG.
a), a sound wave of a certain frequency is generated by applying a trigger to the calibration sound source [3b], and the sound wave is received by the opposed acoustic sensor (the sound wave generated from the calibration sound source [3a] is an acoustic sensor [3]). 1c], the sound wave generated from the calibration sound source [3b] is received by the acoustic sensor [1d]). The time when the sound wave reaches the acoustic sensor [1c] and the acoustic sensor [1d] from the trigger that generated the sound wave is measured,
The time, the air temperature by the temperature sensor [2a] to the temperature sensor [2d], the distance between the calibration sound source [3a] and the acoustic sensor [1c], and the distance between the calibration sound source [3b] and the acoustic sensor [1d] are given to the computer. This makes it possible to calculate the average wind direction and wind speed on the plane of the impact area surrounded by the acoustic sensors [1a] to [1d]. In FIG. 3, a straight line connecting the calibration sound source [3a] and the acoustic sensor [1c] is X1 axis, a distance between the straight lines is X, and a straight line connecting the calibration sound source [3b] and the acoustic sensor [1d] is Y1 axis. If the distance between the straight lines is Y and the straight lines are orthogonal, the acoustic sensors [1a] to [1a] in FIG.
d], the wind speed of the landing area surrounded by F is F, the wind direction is θ, and the sound wave generated from the calibration sound source [3a] at a certain time reaches the acoustic sensor [1c] is St1, and the calibration sound source [ Assuming that the time when the sound wave generated from 3b] arrives at the acoustic sensor [1d] is St2 and the sound speed is C, the wind speed F and the wind direction θ in the impact area are as follows.

[0026]

(Equation 4) However, θ assumes that the X1 axis is 0 and the counterclockwise direction with respect to the X1 axis is the positive direction.

Next, a method of correcting refraction of a sound wave due to wind will be described with reference to FIG. The wind speed and wind direction are measured by the wind speed and wind direction measurement method or the wind direction / anemometer, and then the wind direction and wind speed around the landing area are checked. Then, the acoustic sensors [1a] to [1d]
Is calculated based on the time difference or frequency of the sound waves generated from the shell obtained in step (1). This calculated value includes an error due to the effect of wind. For this reason, the deviation of the coordinates due to the wind is predicted from the coordinates of the sound source calculated, and a temporary sound source is determined and ± SPN (m) in the X-axis direction and ± SPN in the Y-axis direction.
(M), the area of ± SPN (m) is determined in the Z-axis direction and divided into n, and the refraction of the sound wave due to the influence of wind and temperature is assumed on the assumption that a sound wave is generated from the center point of each of the n divided areas. Calculate and calculate the sound wave arrival time from each area to each sensor. Next, the difference in sound wave arrival time between the sensors based on the calculated value is calculated and compared with the actually measured value data to determine the error time between the sensors. This calculation is repeated to determine the coordinates (Xmin, Ymin, Zmin) of the area when the error between the actually measured value data and the calculated value is minimum in the area. Next, the SPN value is made smaller than the previous time, and ± SPN (m) in the X-axis direction and ± SPN
(M), an area of ± SPN (m) is determined in the Z-axis direction and divided into n, and calculation is continued assuming that a sound wave is generated from the center of each of the divided n areas. Determine the coordinates of the area where the error is minimized. The above calculation is repeated until the error between the actually measured value data and the calculated value becomes Min or less, and the coordinates when the error becomes Min or less are corrected coordinates of the sound source refracted by wind and temperature.

Next, the operation of the system will be described with reference to a block diagram (FIGS. 1 and 2) and a flow chart (FIG. 8) according to an embodiment of a device for locating and identifying an unexploded ordnance. The sensor unit shown in FIG. 1 is installed around the landing, and the signal processing unit is installed in an exercise hall several km away from the landing. The operation unit is installed near a shell shooting point in the training ground, and the analysis unit shown in FIG. 2 is installed at an arbitrary location. When the power of the system is turned on, the control circuit [18] of the signal processing unit and the control circuit [25] of the operation unit in FIG. 1 are initialized and system diagnosis is executed (step 32). ] (Step 52). Temperature sensor [8a] at regular intervals if system operation is normal
-Data acquisition from the temperature sensor [8d] and wind direction / wind speed calculation performed by the wind direction / wind speed calculation circuit [17] of the signal processing unit using sound waves from the calibration sound source [9a] and the calibration sound source [9b] (step) 33) The status of the observation area is updated in real time with these data, and used for correcting the landing data. When starting the landing estimation, necessary data is input in advance by the input device [19] of the operation unit (step 35), and the trajectory calculation of the shell is performed by the arithmetic circuit [15] of the signal processing unit based on the data. Perform (step 36)
Estimate the flight time until the shell hits (step 37)
A measurement start command from the input device [19] of the operation unit or a signal from the emission sound acquisition sensor [26] is used as a trigger to prepare for observation for performing landing estimation, and the calibration sound source [9] is prepared.
a] and the calibration sound wave from the calibration sound source [9b] are stopped, and the wind direction and speed are calculated using the latest data (step 3).
8, 39, 40). At the estimated time of arrival of the shell, each data from the acoustic sensors [7a] to [7d] is subjected to signal processing by the signal processing circuits 2 [14a] to 2 [14d] of the signal processing unit, and the An explosion or misfire is determined, and a signal processing method is selected (steps 41, 42, 43, 44). 5. When a shell has exploded, the rise time of the explosion sound is detected by the acoustic sensors [7a] to [7d], and the time difference between the acoustic sensors is measured to estimate the impact position by the method according to claim 4. Then, after storing the result in the storage circuit [16], the result is output to the operation unit (steps 45, 46, 47, 48, 49, 50). If it is determined that a shell has not been misfired, it is predicted from the calculated trajectory whether or not the speed of the shell in the observation section exceeds the speed of sound. , To the operation unit (steps 55, 61, 62,
63, 64, 65). When the speed of the shell in the observation section is lower than the speed of sound, the position of the shell is estimated by the method of claim 3 and the method of claim 2, and the result is stored in the storage circuit [1].
6], and then output to the operation unit (steps 56 and 5).
7, 58, 59, 60). This concludes the description of the estimated position of the shell.

In addition, when a shooting target is set on the operation unit, the correction amount of the shooting angle for the next shot can be calculated and displayed from the acquired landing point data. The analysis unit shown in FIG. 2 analyzes shooting related data stored in the signal processing unit, performs statistical processing, and the like.

[0030]

As described above, since the position of the unexploded bomb due to the sound wave can be estimated in three dimensions, the effect according to the present invention can be obtained even in weather such as fog or rain. At this time, the search range can be reduced compared to the conventional case. Further, since the estimated time of the landing position is short, the correction amount of the gun can be calculated based on the deviation of the landing position with respect to the target, and the feedback to the shooting can be performed. From the data acquired by this system, it is possible to obtain information such as ballistic analysis, statistics, etc. based on the type of shell, temperature, weather, etc.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment relating to an impact landing and unexploded ordnance identification device according to the present invention.

FIG. 2 is a block diagram of an embodiment relating to an impact landing and unexploded ordnance discriminating apparatus according to the present invention.

FIG. 3 is a schematic diagram for explaining an example of installation of a sensor unit near a landing point and a method for estimating a landing position according to the present invention.

FIG. 4 is a schematic diagram illustrating a frequency when a sound wave generated by a gun head is observed at an arbitrary point.

FIG. 5 is a frequency characteristic comparison diagram in the case where sound waves generated at the time of shell flight are observed at arbitrary two points.

FIG. 6 is a schematic diagram illustrating a position estimating method for an impact position locating and unexploded bullet identification device according to the present invention.

FIG. 7 is a schematic diagram illustrating a time axis center and a maximum signal level position when a window function is applied to a signal whose level changes.

FIG. 8 is a flowchart of an embodiment of the apparatus for identifying a landed and unexploded ordnance according to the present invention.

[Explanation of symbols]

 1a to 1d, 7a to 7d Acoustic sensor 3a, 3b, 9a, 9b Sound source for calibration 10a to 10b Signal processing circuit 14a to 14d Signal processing circuit 2 15 Operation circuit 16 Storage circuit 17 Wind direction / wind speed calculation circuit

Claims (6)

[Claims] At least four or more acoustic sensors that are separately arranged to detect a time difference between sound waves generated from a shell and a signal received by each sensor is wirelessly transmitted to a signal processing unit described later. Then, select an appropriate processing method for each type of sound wave, and determine the impact position of the shell from the results of each processing.
An impact position locating and unexploded vehicle identification device, comprising: means for estimating in dimensions (latitude, longitude, altitude) and displaying it on an operation unit described later; and means for judging explosion of a shell. 2. An acoustic sensor receives a bullet-jumping sound or a single-frequency sound source whose sound frequency continuously changes with time by an acoustic sensor, and obtains a fast Fourier transform from the obtained signal. In performing FFT, a band-pass filter for removing out-of-band frequency components, means for storing the filtered signal in a memory circuit via an A / D converter, and FFT by a method described later Means for detecting a characteristic portion of a signal that changes over time and frequency of a sound wave as a line, and means for performing a wavelet transform on a signal near the line disappearance time with reference to the detected line to remove a noise component,
A function ψξ (X) corresponding to the line frequency versus time change is created, and the signal fα (X) and the function ψξ
A circuit having means for determining the cross-correlation of (X) at each sampling interval and accurately measuring the line disappearance time is represented by A,
2. A sound source position estimating method according to claim 1, further comprising: four circuits B, C, and D, and a circuit for calculating a sound source extinction position from a line extinction time difference between the respective systems. Location and unexploded ordnance identification device. 3. A characteristic portion of a signal which changes in time and frequency which indicates a change in Doppler frequency present in a shell shot sound by means of the same means as in claim 2, wherein the shell shot sound generated from the shell during flight is changed. A means for detecting as a line, and means for performing least square approximation processing on the detected line
Means for estimating the trajectory and impact position of the shell by comparing each line (Doppler shift curve) with the terminal trajectory obtained by trajectory calculation for each of the four systems B, C, and D 2. A device for locating and arriving at an impacted position according to claim 1, wherein: 4. A circuit for receiving a sound wave or the like at the time of explosion of a shell by an acoustic sensor, transferring the sound wave to an after-mentioned signal processing unit via an A / D converter, and measuring a rising time of the sound wave.
2. A method according to claim 1, further comprising means for estimating a sound source position in three dimensions (latitude, longitude, altitude) by utilizing a difference in arrival time of sound waves between the systems. Bullet identification device. 5. The impact position locating and unexploded ordnance discriminating apparatus according to claim 1, further comprising means for correcting a temperature difference of air in which the sound wave propagates and a refraction of the sound wave due to the wind. 6. The apparatus according to claim 1, further comprising means for measuring a wind direction and a wind speed of the sound wave propagation path using the sound wave.
The hit position locating and unexploded bullet identification device described in the above.