JPH03179285A - Target tracking apparatus - Google Patents

Abstract

PURPOSE:To accurately display a real target azimuth by calculating the declination rate of a target from the difference between the target azimuth up to the previous time and target azimuth of this time calculated by a target tracking processing part and automatically selecting the optimum time constant corresponding to the declination rate. CONSTITUTION:When the target azimuth calculated by a target tracking processing part 9 is stored in the first and second regions of a target azimuth memory means 13, a declination rate calculation means 14 reads the newest target azimuth from the second region and also reads the target azimuth up to the previous time from the first region and calculates the declination rate of the target azimuth from the difference between the target azimuth up to the previous time and the target azimuth of this time. Whereupon, a time constant selection means 15 selects the time constant corresponding to the declination rate from a memory to output the same to an averaging processing part 6.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention relates to a target tracking device, and particularly to a time constant for signal strength averaging thereof.

[Prior Art] Regarding a conventional target tracking device, a passive target tracking device used as a sonar, for example, will be described.

This target tracking device receives sound waves from a sound source (hereinafter referred to as a target) with a receiver equipped with an acoustic wave receiving element, and includes an amplification section, an A/D conversion section, a phasing processing section, and a signal strength calculation section. , an averaging processing section, a data thinning means, and a peak azimuth calculation means to track the sound source and display it on a display, etc. In particular, the averaging processing section calculates the coefficients of the calculation formula described below by the operator. was determined empirically and obtained the target signal buried in noise.

FIG. 5(a) is a diagram illustrating the signal strength in the θ direction calculated by the signal strength calculation unit, and the signal strength 5(t) calculated at a predetermined time (1) is expressed as the time n θ
It is shown every 0 intervals.

This figure shows the signal strength calculated at a predetermined time (1) for each time, and for example, the signal strength 5(t) is set at a level of about 30 (θ n ).

This figure shows that, for example, no sound source signal was obtained between t and t5, and a sound source signal was obtained at time t""t12.

FIG. 6 shows the positional relationship between the receiver of the present device and the target object corresponding to the case where the signal strength shown in FIG. 5(a) is obtained. T(
t - t ) is the 15th point at time t - t5
1 The position of the target object, T (t6 to t12), and T (t13 to) are respectively e 12 after time 1 −1 1
'13 is the position of the target object, indicating that the target object exists in the θ direction only at time t-t12.

FIG. 5(b) is a diagram illustrating that the average signal strength was obtained by exponential integration.

This figure shows the signal strength 5(t)θ n in Figure 5(a) averaged by exponential integration, and the averaging coefficient (α) of exponential integration is set to 0.3. It shows.

Also, this calculation formula is S (t)-(1-α)・5o(tn)θ
It is expressed as n + α·100 (t) (α≦1) θ n-1 , and if there is a signal of 18 levels or higher (hereinafter referred to as effective level), it is determined that there is a signal.

In addition, the peak calculation means detects the peak data from the average signal strength for each direction calculated based on the above formula through the data thinning means, and interpolates it using the left and right data. I had obtained the true peak direction.

Next, a case where noise is averaged based on the above calculation formula will be described below.

FIG. 5(c) is a diagram illustrating a case where the calculated signal strength includes noise that randomly fluctuates over time.

In the figure, (40) and (41) are noises of about 27 levels calculated at times t 1 and t4, for example.

FIG. 5(d) is a diagram obtained by averaging the signal intensities in FIG. 5(c) by exponential integration. In the figure, (42) to (46) are the results of averaging noises (40) and (41) by exponential integration.

This figure shows that when the averaging coefficient (α) of exponential integration is set to about 0.3, when noise (40) and (41) are around 27 levels, exponential integration is performed with fast response, and the averaged result is This shows that (42) and (45) also end up at the valid level.

Figure 5(e) is a diagram calculated by increasing the averaging coefficient (α) of the exponential integral compared to Figure 5(d), for example, when the averaging coefficient (α) is 0.7. be.

In the figure, (47) to (50) are noise (40) and (
41) is averaged by exponential integration. This indicates that the noise is below the effective level by reducing the averaging coefficient.

However, this figure shows that if the averaging coefficient (α) is made too large to slow down the response speed, as shown in figure (C), there is no signal in the θ direction at time t13; In this case, the signal will be about 18 levels, or 1. 1
This also shows that even though there is a signal, the averaged result does not reach the effective level.

In other words, in Figure (e), there is a time lag between the actual position of the target object and the target direction displayed by the device. This deviation becomes a practical problem as the target moves faster.

In addition, as another method, signal strength averaging processing is performed using the moving average calculation formula shown below. For example, if the number of signal strength samples is k, the sample time is t, and the signal strength of the target object at each time in the θ direction is 5(t), θ n , then the moving average of the signal strength S ( t,) is θ
n t′″tn (k≧1). However, if k is made too small, noise etc. will also be calculated based on the above formula.

Furthermore, if the value is too large, the true signal strength may no longer be included.

Note that α and k function similarly to the time constants in a CR open circuit, and are collectively referred to as time constants in this specification.

Therefore, when randomly fluctuating underwater noise that is always present in the ocean is received by a receiver, the underwater noise overlaps with the true signal, making it difficult to see the signal from the true target being buried in the noise. Ta.

For this reason, the time constant manually input to the average value calculation means must be manually determined by the operator while constantly judging the situation, requiring experience and skill.

[Problems to be Solved by the Invention] The conventional target tracking device as described above calculates and averages signal strength, but requires an operator to determine and input an averaging coefficient (time constant). If the averaging coefficient (time constant) is set too large, noise, etc. will be averaged based on the averaging coefficient (time constant), so the direction of the detected peak will be different from the actual position of the target object. There have been problems in that the peak direction may be displayed behind the peak, or a clear peak direction may not be detected.

In addition, if the averaging coefficient (time constant) is made too small, the response to the movement of the peak direction will improve, but weak signals will be buried in background noise and will be lost, or a clear peak direction will not be detected. There was a problem.

Furthermore, the averaging coefficient (time constant) must be entered by the operator while constantly checking the situation, which requires experience and skill, and there is a problem in that it is not easy to input.

The present invention was made in order to solve such problems, and provides a target tracking device that can automatically calculate the optimal averaging coefficient (time constant) from the target deflection rate and detect a clear peak direction. The purpose is to obtain.

[Means for Solving the Problems] A target tracking device according to the present invention collects acoustic signals received via a receiver having a plurality of acoustic wave receiving elements in each direction, and After calculating the signal strength of the signal,
The averaging processing means uses a predetermined time constant to average the signal, calculates the maximum azimuth from the averaged signal, and the target tracking processing means tracks how the azimuth moves along with the 111f period. In a target tracking device that outputs the target direction of a sound source to a display unit every time, a first area and a second area for storing the target direction calculated by the target tracking processing means are set to -H.
and a drifting direction storage means in which the newest target direction is stored in a second area and the previous target direction is stored in the first area, and a first area stored in the target direction storage means. A deflection rate calculation means for calculating the deflection rate of the sound source from the difference between the previous target direction stored in the second area and the direction, and a deflection rate of the deflection rate calculation means. and time constant selection means for selecting a specific number according to the time constant and outputting the selected time constant to the averaging processing means.

[Operation] In the present invention, when the target azimuth calculated by the target tracking processing means is stored in the first area and the second area of the target azimuth storage means, the variation rate calculation means stores the newest target azimuth in the second area.
, the previous target azimuth is read from the first area, and the displacement rate of the target azimuth is calculated from the difference between the previous target azimuth and the current target azimuth.

Then, the time constant selection means selects a time constant corresponding to the change angle rate from the memory and outputs it to the averaging processing section.

[Embodiment] FIG. 1 is a conceptual diagram of a target tracking device showing an embodiment of the present invention.

In the figure, (1) is a receiver, for example, a plurality of N acoustic receiving elements are arranged in an array, and the elements convert the received acoustic signals into electrical signals, each of which has N channels of independent channels. It is output as an analog electrical signal.

(2> is the amplification section, and the N
It amplifies the analog electrical signal of the channel.

(3) is an A/D converter that digitally converts the N-channel analog electrical signals output from the amplifier (2) based on a predetermined sampling period.

(4) is a phasing processing unit, which inputs the N-channel digital signals output from the A/D converter (3), and
It performs phasing for each direction and outputs the absolute value of the amplitude of the arriving signal as a signal (hereinafter referred to as a phasing signal) for each M direction.

Here, phasing means, for example, that each input signal of N channels is interposed with a variable delay element, and the signal arrival time from the sound source in the target direction through each acoustic receiving element and variable delay element is all equal. (In other words, all the input signals of N channels are in the same phase), and the outputs of the variable delay elements are added together.

Therefore, by setting the delay time of the variable delay element in advance for each specific direction, an output signal phased in a specific direction can be obtained.

Also, the plurality of directions M refers to, for example, 360 degrees in all directions with 10
It is divided into sectors for each degree, and 36 degrees of divided directions are provided.

The divided orientations in this case may or may not overlap with each other.

In other words, even if it is 0 degrees to 10 degrees or 10 degrees to 20 degrees, 3
It may also be 59 degrees to 11 degrees or 9 degrees to 21 degrees.

(5) is a signal strength calculation unit, which calculates the phasing signal output from the phasing processing unit (4) for a short time (for N samples) for each direction.
The signal strength of each direction is obtained by adding the signal strength, and the signal strength of 1#I is output every time the phased signal for N samples is input.

This data is all the signal strengths calculated for each direction.

(6) is an averaging processing section, which inputs the signal strength for each direction output from the signal strength calculation section (5) for a predetermined period of time, and calculates the average as a time constant or , is calculated by moving average or exponential integration using a time constant input by the operator, extracts the target signal buried in noise, etc., and outputs the data (hereinafter referred to as averaged data) for each direction. be. In this case, the explanation will be given assuming that the calculation was performed by exponential integration.

(7) is a data thinning means, and each time the averaged data of the averaged signal strength outputted from the averaging processing unit (6) for each direction is manually inputted to group A, one set of the latest averaged data is generated. This outputs the following.

This latest averaged data means that if the target is a ship whose moving speed is not very fast, the repetition period of the tracking process described later may be slow, so the inputted averaged data is thinned out at a predetermined timing. That's true. Therefore, the load on the CPU is reduced.

(8) is a peak direction calculation means, and a data thinning means (
7) A true peak azimuth (hereinafter referred to as true azimuth data) is calculated by a method described later from the latest averaged data outputted from 7), and is used as the target azimuth.

(9> is the target tracking processing section, and the peak direction calculation means (
Compare the current true azimuth data from 8) with the azimuth data of multiple past targets stored in the target azimuth storage means described later, and determine which of the current peak azimuth data is the same as the past target. In other words, it determines whether a past target can be considered to have moved there, and outputs current direction data (tracking direction) for each target.

Furthermore, if there is an initial direction instruction for target tracking, the peak closest to the specified direction is set as the current target direction, and the same target tracking process as described above is performed in subsequent processes.

(10) is an image processing unit that processes the target video signal output from the target tracking processing unit (9) and the averaging processing unit (6).
Predetermined image processing is executed in order to display the video signal of the averaged data output from the video signal.

(11) is an image memory, which stores video data output from the image processing section (10) at a predetermined location.

(12) is a CRT, which displays video data output from the image memory (11) as a video.

(13) is a target direction storage means, which is a target tracking processing unit (
The target tracking azimuth data calculated in step 9) is stored, and the first area stores the past (until the previous) target tracking azimuth data and the second area stores the current target tracking azimuth data. have.

(14) is a variation rate calculation means, and a target direction storage means (
13) Read the previous target tracking azimuth data stored in step 13) and the current target tracking azimuth data stored in step 13), and calculate the azimuth in which the target has moved (referred to as the deflection rate) based on the calculation formula described later. do.

In this case, the current target tracking direction data is θt, and the past target tracking direction data in seconds is θ(t −Δ
t), the angle variation rate is calculated by Δ t .

(15) is a time constant selection means, which stores a time constant corresponding to a variation rate in a memory (not shown), and a variation rate calculation means (
The stored time constant is output to the averaging processing section (6) according to the angle variation rate calculated in step 14). As for the time constant corresponding to this variation rate, a small variation rate is stored with a large time constant, and a large variation rate is stored with a small time constant.

In addition, when the data is initialized manually (for example, the second set of data), the stored initial time constant is sent to the averaging processing unit (6).
Output to.

FIG. 2 is a diagram illustrating the processing of the peak azimuth calculation means. This figure shows the values of the latest averaged data in each direction thinned out by the data thinning means (7) on the vertical axis and the direction on the horizontal axis, detecting the highest peak data (8a), and detecting multiple data or This shows that the true peak azimuth (8b) is calculated by performing interpolation from the left and right data.

The target tracking device configured as described above has a receiver <1>
When the acoustic signal converted into an electrical signal is output as an N-channel analog electrical signal by the N acoustic receiving elements of The converter (3) outputs each as a digital signal.

Then, the phasing processing section (4) manually processes the N-channel digital signals output from the A/D conversion section (3), and
The phasing is performed for each direction, and the incoming signal is outputted to the signal strength calculation section (5) as a regularization signal for each M direction.

Then, the signal strength calculation unit (5) adds the phasing signals output from the phasing processing unit (4) for each direction for a short time (for N samples) to obtain the signal strength for each direction, Each time a phased signal is input, a set of signal strength data is output.

Then, the averaging processing section (6) performs the signal strength calculation section (5).
The signal strength for each direction outputted from is manually calculated for a specified period of time,
The average value of the signal strength (averaged data) is determined by the following formula explained in the conventional example using a time constant input in advance (hereinafter referred to as the initial 81 constant) or a time constant from the time constant selection means (15). calculate.

Average value of signal strength (averaged data) in 100. (t), then S (t)-(α-1)・Sθ (tn)θ
Calculated based on the formula n + α・5(t) (α≦1) θ n−1, extracts the signal of the target object buried in noise, etc., and calculates the averaged data for each direction by means of peak direction calculation means. Output to (8).

In this case, for example, it is assumed that the signals at times t 1 and t2 are calculated using an initial time constant, and the initial time constant is statistically calculated and is ideal.

Then, the data thinning means (7) converts the averaged data into A
Every time a set is input, predetermined data is thinned out and one set of the latest data is output to the peak azimuth calculation means (8) for each azimuth.

This peak direction calculation means (8) is a data thinning means (7).
), the peak data (8a) is detected from the averaged data for each direction, and the true peak direction (8b) is calculated by interpolation using multiple data or left and right data, and target tracking is performed. Output to the processing section (9).

Then, the target azimuth storage means (13) compares the current true azimuth data from the peak azimuth calculation means (8) from the azimuth data of a plurality of past targets stored in the first area. It determines which of the current peak azimuths is the same as a past target, that is, whether it can be considered that the previous target has moved there, and calculates and outputs current tracking azimuth data for each target.

The image processing unit (10) also performs predetermined image processing using the azimuth data output from the target tracking processing unit (9) and the averaging processing unit (6), and transmits the data to the CRT via the image memory (11). (12) is displayed in association with the intensity and direction of the sound source.

Therefore, each time the target tracking processing section (9) outputs the direction data, the angle change rate is calculated from the previous tracking direction data and the current tracking direction data stored in the target storage means (14). By choosing the corresponding time constant,
It becomes possible to always automatically output an optimal time constant to the averaging processing means (6).

FIG. 3 is a block diagram illustrating the present invention in detail. In the figure, the phasing processing section (4), the image memory (II), and the CRT (12) are the same as those in FIG.

(20a) to (2On) are acoustic receiving elements, which output the sound from the target as analog electrical signals (hereinafter referred to as acoustic signals), and are equipped with a plurality (N channels), and the receiver (1) This corresponds to

In this case, the Nth acoustic receiving element is (20n).

(21a) to (21n) are low-pass filters (hereinafter L
(referred to as PF), and acoustic wave receiving elements (20a) to (2O
N channels of audio signals outputted from N channels are respectively inputted, and outputted as an audio signal with high frequencies cut. In this case, the Nth LPF is (2In).

(22a) to (22n) are amplifiers (hereinafter referred to as AMP), which amplify the N-channel acoustic signals output from LPF (21a) to (21n), respectively, and correspond to the amplification section (2). It is something to do. In this case, the Nth AMP is set to (22n).

Also, the LPF is the receiver (1) in Figure 1 or the amplifier section (
2) is supported.

(23a) to (23n) are sample and hold circuits, which simultaneously sample and hold the acoustic signals output from AMP (22a) to (22n) on all channels using the hold signal output from the timing controller (28). It is something. In this case, the Nth sample and hold circuit is designated as (23n).

(24a) to (24n) are A/D conversion circuits, which convert the held acoustic signals from the sample and hold circuits (23a) to (23n) into digital signals according to a conversion instruction signal and output the digital signals. . In this case, the Nth A/D conversion circuit is assumed to be (24r+).

(25) is a multiplexer, and A/D conversion circuit (2
The digital signals outputted from 4a) to (24n) are manually input, and one of the digital signals is output to the phasing processing section (4) according to a timing signal.
It is provided in the D conversion section (3>).

(26) is an integral processing unit, which calculates the signal strength of the phasing signal output from the phasing processing unit (4〉) for each direction, and generates one set of signals every time N samples of phasing signals are input. The data is output as intensity data, input for a predetermined period of time, and the average is calculated by moving average or exponential integration based on the time constant set in the register (29). It extracts signals and outputs the averaged data for each direction.

In this case, the explanation will be given assuming that the calculation was performed by exponential integration. This integral processing section (26) is the signal strength calculation section (26) described above.
5> and the averaging processing section (6>).

(27) is a CPU, which includes data thinning means (7), peak azimuth calculation means (8), target tracking processing section (9), target azimuth storage means (13), variation rate calculation means (14), and time constant selection. It performs an operation corresponding to means (15).

(28) is a timing controller, which uses a control signal output from the CPU (27) to control the Zumble hold circuit (2).
3a) - (23n), a conversion instruction signal to the A/D conversion circuits (24a) to (24n), a data selection signal to the multiplexer (25), and a phasing processing section. (4) A predetermined timing signal is output.

30 is a keyboard, which outputs a predetermined time constant etc. input by the operator to the CPU (27).

The operation of a concrete example of the target tracking device configured as described above will be explained using a flowchart.

FIGS. 4(a) to 4(e) are flowcharts specifically explaining the operation of a specific example.

As an initial setting, the CPU (27) sets an initial time constant in the register (29) of the integral processing section (26), and after initializing each section, sends a start signal to the timing controller (
28> (Sl).

By this initialization, each section enters a state for signal processing to be described later, and the receiver (1,) to phasing processing section (4) operate as shown in FIG. 4(a).

When a plurality of N acoustic wave receiving elements (20a) to (2On) receive an initialization signal, they input sound waves from a sound source, and
The acoustic signal received by the element is converted into an electrical signal, and each is output as an independent N-channel analog electrical signal as an acoustic signal (8101), each L P F
(22a) to (22n), the acoustic signal with the high frequency cut is A M P (21a) to A M P
(2fn).

And A M P (22a) ~A of N channel
M P (22n) amplifies these acoustic signals and sends them to sample and hold circuits (23a) to (23n), respectively.
) (S103).

Next, wait until the data sample time (S1
05), the timing controller (28) outputs a hold signal to the sample hold circuit (23), outputs a conversion instruction signal to the A/D conversion circuit (24), and further outputs a data select signal to the multiplexer (25). A timing signal is output to the phasing processing section (4).

Then, sample hold circuits (23a) to (23n)
sample and hold the amplified acoustic signals using the hold signal, and convert them into A/D conversion circuits (24a) and 24a.
(24n) (810B).

Then, the A/D conversion circuits (24a) to (24n) transfer the sampled and held acoustic signals to a timing controller (24n).
8>, the audio signals converted into digital signals are output to the multiplexer (25) (S107).

Then, the multiplexer (25) controls the timing: (
(28), the digitalized audio signal (24) from the A/D female converter circuit (24) is
(hereinafter referred to as data signals) are sequentially output to the phasing processing section (4) (S109).

Then, the phase processing unit (4) phase-phases the N-channel pig signals for each direction (5Lio) and outputs the amplitude value of the arriving signal as a phased signal (Sl12).

Next, if there is no end signal, the process continues from step 5101.

Next, the operation of the integral processing section (26) in FIG. 4(b) will be explained.

When the integral processing section (26) is initialized, it sets the input counter to O (S201).

Then, registers and the like are set to take in the phasing signal (S2 (13)).

Next, it is determined whether the phasing signal is manually inputted from the phasing processing unit (4>) (step 8205), and if not manually inputted, step 8203
transfer control to

When the phasing signal is input, the manual counter is updated 820
7) It is determined whether N samples of the phasing signal outputted from the phasing processing unit (4) have been input for each direction (S209).

If N samples are not input, control is executed at step 8203.
Move to.

If N samples are manually generated for each direction, the signal strength for each direction can be calculated by adding N samples of phased signals for each direction (
S211).

Then, for each direction, the time constant set in the register (29) is set to α, and the calculated signal strength is multiplied by (1-α) (S213).

Next, the previous averaged data is multiplied by the time constant α (S21
5) Calculate the current averaged data from the above exponential integral averaging formula (S220) and store it in the memory for each direction. The averaged data here refers to the average signal strength calculated by an exponential integral averaging formula.

If this device has just been initialized, the previous averaged data is 100. Since (tn) is not stored, O is substituted.

Further, in this case, the time constant set in the register is the initial time constant.

Then, calculate the current average signal strength (current averaged data) for each direction based on the exponential integral averaging formula.
) and are stored for each direction (S222).

Then, the averaged data for each direction is calculated by CP for each direction.
Output to U (27) (S224).

Next, it is determined whether the end signal has been output (822B),
If it has not been output, control is transferred to step 5201 and the same process as above is executed for the next phasing signal.

Next, the operation of the CPU (27) (Fig. 4 (C)
) to (e) will be explained below.

As an initial setting, the output counter is set to 0 after executing the step St described in (a) (S2), and
Set the manual counter (S) to k (S3), integral processing section (2)
6), the averaged data is read for each direction (S5). Then, the input counter is updated (5ol) L (37), image data is created from the averaged data of each direction that has just been manually input, and is output to the image memory (11) (S9).

Next, it is determined whether the value of the human power counter (S) is greater than or equal to k (5ll), and if it is determined that it is greater than or equal to k, the human power counter is set to O (S13).

Then, it is determined whether there is an instruction to end tracking (515), and if there is an instruction to end tracking, the internal settings are changed to stop tracking the set target (817).

Also, if it is determined that there is no instruction to end tracking, step S
Transfer control to 19.

Next, calculate the averaged data that becomes the peak from the averaged data of each direction (5I9), and calculate and store the true peak direction by interpolation from the calculated peak data and the plurality of left and right direction data (S21). ).

Next, it is determined whether there is a target being tracked (523), and if it is determined that there is a target being tracked, the tracking direction data of the past target stored in the memory is compared with the current true direction. 525), the current direction (tracking direction) of each target is determined (S27). Then, it is determined whether there is an initial instruction for new target tracking (S29).

If it is determined that there is an initial instruction for new target tracking, the direction closest to the instructed direction is set as the current target tracking direction (
S31), and updates the output counter (S33). Further, if it is determined in step S29 that there is no new initial instruction for target tracking, control is shifted to step S33. in this case,
Suppose that it is determined that there is no such thing.

Next, if the output counter is updated, image data is created from the tracking direction data and output to the image memory (11>) (S35).

Then, it is determined whether the value of the output counter updated in step S33 has become 3 or more (S37).

If it is determined that the value of the output counter is not 3 or more, it is determined whether the orientation data is stored in the first area of the memory (S39). In this case, it is assumed that no orientation data is stored in the first area. Further, in this case, the azimuth data stored in the first area is the past azimuth up to the previous time.

Next, if it is determined that the data is not stored, the calculated orientation data is stored in the first area (S41).

Next, it is determined whether a time constant has been specified from the keyboard <30> (S43). In this case, it is assumed that it is determined that no instructions are given.

If it is determined that no instruction is given, an initial time constant is set in the register (29) of the integral processing section (2B>) (S45), and control is shifted to step S5.

Further, if it is determined in step S43 that a time constant has been instructed from the keyboard (30), the manually entered time constant is set in the register (29> of the integral processing section (26)) (S47
).

Then, the next data is read and the processes from step 85 to step S39 described above are executed, and since the orientation is stored in the first area in step S41, it is determined that the orientation is stored in the first area in step S39, The currently calculated orientation is stored in the second area (850).

Next, the direction stored in the first area (
(Past azimuth) and the azimuth (current azimuth) stored in the second area are substituted (552), and the deflection rate of the target is calculated (S54).

Then, the calculated angle change ratio is output to the image memory (1()) and displayed on the CRT (12) (35B).

Next, it is checked whether there is a time constant instruction from the keyboard (30) (1'l + 958), and if there is no time constant instruction, the time constant corresponding to the variation rate is read from the memory (sea).

Then, the specific number read is set in the register (29) of the integral processing section (26) (S62).

Next, by setting a time constant in the register (29) of the integral processing section (26), it is determined whether the process has ended or not (564). If the process has ended, the process is ended, and if not, the control moves to step S5.

Further, if a time constant is instructed in step 358, the manually inputted time constant is set in the register (29) of the integral processing section (26) (36B).

In this process, the peak direction is calculated for the third set of data and the output counter is updated to 3, so step S37
In this case, it is determined that the output counter has a value of 3 or more.

Then, the CPU (27) determines whether the first area of the memory is full, and if it is -full, clears the oldest orientation data (888).

Then, the orientation data stored in the second area is stored in the first area as the previous orientation data (S73). Further, if it is determined in step S8g that the first area is not full, control is transferred to step S73.

Next, the current direction data is stored in the second area (S7
5) The control is moved to step S52, where the past direction and the current direction are substituted into the variation rate calculation formula to calculate the variation rate, and the time constant corresponding to this variation rate is calculated by the integral processing unit (26). The time constant is set to , and the process described above is continued until the end signal is output.

Therefore, it is possible to automatically select a time constant according to the rate of change in the direction in which the target has moved.

In addition, since the angle change rate is displayed, when the operator inputs the time constant, it can be easily done manually.

In the above embodiment, the averaging formula is exemplified as using exponential integration, but the present invention is not limited thereto, and calculation may be performed using a moving average formula.

Furthermore, the phasing processing section (4) and the integration processing section (26) may be configured by hardware.

[Effect of four shots] As described above, according to the present invention, the target deflection rate is calculated from the difference between the previous (past) target orientation calculated by the target tracking processing unit and the current target orientation, The optimal time constant according to the rate of change is automatically selected and output to the averaging processing section, so the true target direction can be accurately displayed without the operator having to manually set the time constant. This effect has been obtained.

[Brief explanation of drawings]

FIG. 1 is a conceptual diagram of a target tracking device showing an embodiment of the present invention, FIG. 2 is a diagram explaining the processing of the peak azimuth calculation means, FIG. 3 is a block diagram explaining the present invention in detail, and FIG. Figure (a) ~
(e) is a flowchart specifically explaining the operation of the specific example, FIG. 5(a) is a diagram explaining the signal intensity for each direction, FIG. FIG. 5(C) is a diagram for explaining the signal strength including noise, FIG. 5(d) is a diagram for explaining that the signal strength in FIG. 5(c) is averaged by exponential integration, and FIG. ) is a diagram when the signal strength in FIG. 6(e) is averaged by increasing the averaging coefficient and using exponential integration, and FIG. 6 is a diagram showing the positional relationship between the receiver of the present device and the target object. In the figure, (1) is the receiver, (2> is the amplifier, (3)
is the A/D conversion section, (4> is the phasing processing section, (5) is the signal strength calculation section, (6) is the averaging processing section, (7) is the data thinning means, (8> is the peak azimuth calculation section) (9) is a target tracking processing unit, (10) is an image processing unit, (11) is an image memory, (12) is a CRT, (13) is a target direction storage unit, (
14) is a bending rate calculation means, (15) is a time constant selection means,
(20a) to (2On) are acoustic wave receiving elements, (21a>
~(2In) is a low-pass filter, (22a) ~(22
n) is the amplifier 1, (23a) to (23n) are sample and hold circuits, (24a) to (24n) are A/D conversion circuits, (25) is a multiplexer, (26) is an integration processing section, and (27) is CPU. (28) is a timing controller, (30> is a keyboard) 3 diagrams (3 diagrams) Concrete example L: Figure 3 (e) explaining v operation Qnuma 5 Figure Goko Calculus!? Figure 3 (b) to explain the average intensity based on Figure 5 Figure 3 (C) to explain the intensity of 'Buma' which includes noise Figure 5 (d) Figure 5

Claims (1)

[Claims] After phasing the acoustic signals received through a receiver having a plurality of acoustic wave receiving elements for each direction and calculating the signal strength of the plurality of signals, at a predetermined time. The averaging processing means uses a constant to average the signal, the maximum direction is calculated from the averaged signal, the target tracking processing means tracks how the direction moves over time, and calculates the sound source at each time. A target tracking device that outputs a target orientation to a display unit, comprising a first area and a second area for storing the target orientation calculated by the target tracking processing means, and the newest target orientation is stored in the second area. The target direction up to the previous time is stored in the first direction.
a target azimuth storage means stored in an area; and a difference between the previous target azimuth stored in the first area and the latest target azimuth stored in the second area stored in the target azimuth storage means. selecting a time constant according to the variation rate of the variation rate calculation means;
A target tracking device comprising: time constant selection means for outputting the time constant to the averaging processing means.