A method and device for passive tracking of underwater targets

Abstract

This invention relates to the field of passive underwater target tracking technology, specifically to a passive underwater target tracking method and apparatus. It addresses the problem of high system complexity and cost in achieving desired directional accuracy. This invention employs a special device to filter and amplify the received underwater acoustic signal, converting it into a digital underwater acoustic signal. Then, it performs refined spectrum estimation to obtain the frequency corresponding to the maximum signal amplitude within a specific frequency band and calculates the signal-to-noise ratio at that frequency. Target detection is then performed; Doppler frequency shift is calculated and the target azimuth angle is estimated; the relative motion trend to the target is estimated; the rudder angle is calculated and sent to the main control computer of the AUV platform via a communication interface circuit. The main control computer then transmits the rudder angle command to the servo motors in the AUV's power and energy unit to control the AUV's navigation and tracking.

Description

Technical Field

[0001] This invention relates to the field of passive underwater target tracking technology, and specifically to a passive underwater target tracking method and apparatus. Background Technology

[0002] Aircraft or important underwater equipment are usually equipped with acoustic beacons that emit signals at specific frequencies as a target location indicator, commonly known as "black boxes". In special circumstances, searchers can use detection equipment to search for and track targets by utilizing the underwater acoustic signals emitted by these beacons.

[0003] AUV is short for Autonomous Underwater Vehicle. Passive target tracking methods based on AUV platforms have long been an important research direction in signal processing engineering and sonar detection engineering. Typically, in tracking missions, AUVs first need to estimate the target's location, and then track it based on the positioning results.

[0004] Underwater target localization methods based on AUV platforms are mainly divided into two categories: active localization methods and passive localization methods. Active localization methods primarily utilize the active sonar array on the AUV's nose for detection, locating the target based on the reflected signal information. Passive localization methods primarily utilize the sonar array for direction finding, locating the target based on multiple direction-finding angles. When using a passive sonar array for target detection and localization, it is usually necessary to mount a specially structured array of multiple hydrophones on the AUV, combining beamforming and other algorithms for target direction finding. To achieve the desired orientation accuracy, there are certain requirements for the channel consistency and aperture size of the hydrophone array, resulting in a larger AUV size, higher system complexity, and higher system cost. Summary of the Invention

[0005] In view of this, in order to solve the problem of high system complexity and cost in order to achieve the desired orientation accuracy, the present invention provides an underwater target passive tracking method and device.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A passive underwater target tracking method, characterized by the following steps:

[0008] 1) After filtering and amplifying the received underwater acoustic signal, it is converted into an analog underwater acoustic digital signal. Then, a refined spectrum estimation is performed to obtain the frequency corresponding to the maximum signal amplitude in a specific frequency band, and the signal-to-noise ratio of the signal at that frequency is calculated.

[0009] 2) Perform target detection;

[0010] 3) Calculate the Doppler frequency shift and estimate the target azimuth angle;

[0011] 4) Estimate the motion trend relative to the target;

[0012] 5) Calculate the rudder angle and send it to the main control computer of the AUV platform through the communication interface circuit. The main control computer then transmits the rudder angle command to the servo motor in the AUV's energy and power unit to control the AUV's navigation and tracking.

[0013] The specific processing method is as follows:

[0014] S1. Calculate the frequency f corresponding to the maximum amplitude of the underwater acoustic digital signal within a specific frequency band. m (Main frequency) and the signal-to-noise ratio (SNR) at the main frequency m ;

[0015] S2. Set the execution cycle of step S1 to T0, and continuously accumulate the f calculated in step S1. m and SNR m The data constitutes the main frequency and signal-to-noise ratio data sequence:

[0016]

[0017]

[0018] When a target signal is received, set the threshold f required for target detection. min f max SNR min Given μ (0 < μ < 1), calculate China satisfies The number of elements n1 and China satisfies The number of elements n2, when n1≥μn and n2≥μn, is considered to have detected a target and proceeds to step S3; otherwise, no target is detected and returns to step S1.

[0019] S3, the main frequency data sequence calculated based on S2 Calculate the Doppler frequency shift Δf of the received signal relative to the center frequency f0 of the target transmitted signal, and take... China satisfies The elements constitute a new main frequency data sequence The Doppler frequency shift calculation method is as follows:

[0020]

[0021] In the formula, f0 is the center frequency of the signal transmitted by the underwater target;

[0022] The target azimuth angle θ is estimated based on the Doppler frequency shift Δf, and the calculation method is as follows:

[0023]

[0024] In the above formula, v is the speed of the AUV and c is the speed of sound underwater.

[0025] If the target azimuth angle θ is within the range of [0°, 90°], and the calculated value of θ exceeds the normal range due to interference from harmonic signals, the maximum value of θ should be limited to 90°.

[0026] S4. An algorithm for estimating the relative motion trend of the target based on the Doppler frequency shift principle. The estimation results are: (1) The AUV has passed the target; (2) The AUV has not passed the target and is sailing towards the target; (3) The AUV has not passed the target and is sailing away from the target; (4) The AUV has not passed the target and is sailing close to the target.

[0027] If the estimation result is the first type, the program stops running, the mission ends, the vehicle rises and transmits GPS signals; otherwise, proceed to step S5.

[0028] S5, Calculate rudder angle

[0029] Set the K value coefficient, and calculate K based on the estimation result of step S4. k The calculation method is as follows:

[0030]

[0031] In the above formula, K k-1 If the K value coefficient of S5 in the previous execution cycle is used, then the rudder angle is calculated as follows:

[0032]

[0033] In the above formula, k m k r and For the pre-calibrated weight coefficients and maximum rudder angle threshold, Roll is the roll angle of the AUV measured by the navigation sensor unit of the AUV platform at this time.

[0034] Calculate the rudder angle Then, the rudder angle is sent to the main control computer of the AUV platform via the communication interface circuit. The main control computer transmits the rudder angle command to the servo motor and controls the AUV to track the navigation, and finally returns to step S1.

[0035] Step S1 above analyzes and processes the underwater acoustic digital signal using a refined spectrum estimation method, as detailed below:

[0036] S1-1, Signal Modulation Frequency Shift: Assume the lowest frequency in a specific frequency band is F. min The highest frequency is F max Then the center frequency is f k =(Fmax -F min ) / 2, by multiplying the analog signal by a factor Frequency shifting of a signal, for a discrete-time series, is represented as:

[0037]

[0038] In the above formula, x0(n) is the original underwater acoustic digital signal, f s This refers to the sampling frequency of the digital-to-analog converter circuit in the signal acquisition and processing unit.

[0039] S1-2, Low-pass filtering: The frequency-shifted signal is first low-pass filtered, and then resampled. The cutoff frequency of the low-pass filter is:

[0040] f c =F max -f k

[0041] S1-3, Resampling

[0042] After modulation and filtering, the number of sampling points is reduced, and a lower sampling frequency f′ is used. s Resampling x′(n) yields signal x″(n), with zero-padding to ensure the same number of sampling points. The frequency resolution is then:

[0043]

[0044] Where N is the number of FFT points;

[0045] S1-4, FFT Transform

[0046] After the steps, the signal x″(n) is a complex signal. Performing an N-point FFT on the signal will yield the refined frequency domain signal x″(w).

[0047] S1-5, Calculate the main frequency f m

[0048] Find the refined signal amplitude spectrum |x″(w)| in a specific frequency band [F min F max The frequency with the largest amplitude within the range is taken as the main frequency f. m The calculation method is as follows:

[0049] w m =argmax|x″(w)|(w∈[2πF) min / f s ,2πF max / f s ])

[0050]

[0051] S1-6. Calculate the signal-to-noise ratio (SNR) of the signal at the dominant frequency. m

[0052] Determine f m After that, it is necessary to determine based on f m Calculate the signal-to-noise ratio (SNR) m The calculation method is as follows:

[0053]

[0054]

[0055] The condition for continuous accumulation in step S2 is that the execution period of steps S2-S5 is T (T=nT0,n>1 and n∈Z), and the accumulation continues within the execution period T seconds of steps S2-S5.

[0056] The specific method for step S4 above is as follows:

[0057] S4-1. According to the Doppler frequency shift formula, when the AUV passes the target, the received frequency is less than the frequency f0 of the signal emitted by the sound source target. Set a threshold ε (0 < ε < 1) to determine whether the AUV has passed the target.

[0058] The main frequency data sequence was calculated in step S3. calculate In which f′ is satisfied m (i) The number of elements ≤ f0 is n3. When n3 ≥ εn′, it is determined that the AUV has crossed the target, the algorithm ends and proceeds to step S5; otherwise, the AUV has not crossed the target and proceeds to step S4-2.

[0059] S4-2, Set the minimum error angle θ min When the included angle θ calculated in step S3 is less than or equal to θ min If the AUV is determined to have not crossed the target and is sailing directly toward the target, the algorithm ends and proceeds to step S5; otherwise, it proceeds to step S4-3.

[0060] S4-3. When the AUV is not sailing directly towards the target, it is necessary to determine whether the AUV is approaching or deviating from the target, based on the calculation in step S3. Calculate the rate of change of the current period frequency The calculation formula is as follows:

[0061]

[0062] In the formula, T0 is the execution cycle of step S1, and a frequency change rate threshold ω (ω < 0) is set. When the condition is met... If the AUV does not cross the target and deviates from it, it is determined that the AUV does not cross the target and is approaching it.

[0063] In step S1-2 above, the filtering is performed directly in the frequency domain. The specific filtering method is as follows:

[0064] S1-2-1. Perform a standard FFT transform on the signal x(n) obtained in step S1-1 to obtain the frequency domain signal x(w);

[0065] S1-2-2. Frequency domain filtering is performed using a rectangular window, i.e., x′(w)=x(w)*h(w), where h(w) is the frequency domain filter function:

[0066]

[0067] S1-2-3. Perform IFFT transform on the filtered frequency domain signal x′(w) to obtain the filtered time domain signal x′(n).

[0068] An apparatus used in a passive underwater target tracking method comprises an AUV platform and an underwater target detection device.

[0069] The AUV platform consists of a main control computer and related data interface units, navigation sensor units, and energy and power units. The energy and power units consist of batteries, motors, and servo motors. The main control computer is used to control the entire AUV operating system.

[0070] The underwater target detection device consists of a single-channel hydrophone unit and a signal acquisition and processing unit. The signal acquisition and processing unit comprises a signal preprocessing circuit, a digital-to-analog converter circuit, a signal processing circuit, and a communication interface circuit. The signal preprocessing circuit is connected to the single-channel hydrophone unit, and the data communication interface circuit is connected to the main control computer data interface unit in the AUV platform.

[0071] The single-channel hydrophone receives underwater acoustic signals and filters and amplifies the analog signals through a signal preprocessing circuit. After the signal is acquired, the signal acquisition and processing unit converts the analog signals into digital signals through a digital-to-analog converter circuit. The signal processing circuit performs a series of processing on the digital signals and transmits the processing results to the AUV's main control computer through a communication interface circuit.

[0072] The aforementioned single-channel hydrophone is a miniature hydrophone based on a semiconductor MEMS manufacturing platform.

[0073] Compared with the prior art, the present invention has the following advantages and effects:

[0074] 1) This invention enables AUVs to perform passive underwater target tracking by installing only a single single-channel hydrophone, without relying on traditional multi-hydrophone arrays. This effectively reduces the size of the AUV, lowers system complexity, and thus reduces system development costs.

[0075] 2) The method of this invention does not require the use of the phase information of the signal. Therefore, instead of using a common FIR filter for time-domain filtering, filtering is performed directly in the frequency domain to minimize spectral aliasing and reduce computation. Attached Figure Description

[0076] Figure 1 This is a flowchart of the underwater target passive tracking method of the present invention.

[0077] Figure 2 This is a flowchart of the algorithm for estimating relative running trends in this invention.

[0078] Figure 3 This is a structural diagram of the underwater target passive tracking device of the present invention.

[0079] Figure 4 This is a flowchart of the refined spectrum estimation algorithm of the present invention.

[0080] Figure 5 This is a schematic diagram of a scenario for an example.

[0081] Labeling: 1. AUV platform; 2. Underwater target detection equipment;

[0082] 1.1 Related data interface unit, 1.2 Navigation sensor unit, 1.3 Energy and power unit,

[0083] 1.3.1 Battery; 1.3.2 Motor; 1.3.3 Servo; 1.3.4 Rudder; 1.3.5 Propeller;

[0084] 2.1 Single-channel hydrophone unit; 2.2 Signal acquisition and processing unit;

[0085] 2.2.1 Signal preprocessing circuit, 2.2.2 Digital-to-analog conversion circuit, 2.2.3 Signal processing circuit, 2.2.4 Communication interface circuit. Detailed Implementation

[0086] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0087] An underwater target passive tracking device comprises an AUV platform 1 and an underwater target detection device 2.

[0088] The aforementioned AUV platform 1 consists of a main control computer and related data interface units 1.1, navigation sensor units 1.2, and energy and power units 1.3. The energy and power units 1.3 consist of a battery 1.3.1, a motor 1.3.2, and a servo motor 1.3.3. The main control computer 1.1 is used to control the entire AUV working system, and the servo motor is used to control the rudder.

[0089] The aforementioned underwater target detection equipment includes one single-channel hydrophone unit 2.1 and one signal acquisition and processing unit 2.2. The single-channel hydrophone unit 2.1 is used to receive underwater acoustic signals. The signal acquisition and processing unit 2.2 mainly consists of a signal preprocessing circuit 2.2.1, a digital-to-analog converter circuit 2.2.2, a signal processing circuit 2.2.3, and a communication interface circuit 2.2.4. The signal preprocessing circuit 2.2.1 is connected to the single-channel hydrophone unit 2.1, and the data communication interface circuit 2.2.4 is connected to the main control computer data interface unit 1.1 in the AUV platform 1.

[0090] The single-channel hydrophone 2.1 of the aforementioned underwater target detection device receives underwater acoustic signals, and then the analog signals are filtered and amplified by the aforementioned signal preprocessing circuit 2.2.1;

[0091] The signal acquisition and processing unit 2.2 acquires signals and then converts the analog signals into digital signals via the digital-to-analog converter circuit 2.2.2. The signal processing circuit 2.2.3 performs a series of processing steps on the digital signals and sends the processing results to the communication interface circuit 2.2.4.

[0092] The signal processing circuit of the above-mentioned signal acquisition and processing unit 2.2 performs a series of processing on the digital signal according to the method of the present invention, and transmits the final calculated rudder angle result to the main control computer of the above-mentioned AUV via the communication interface circuit;

[0093] The main control computer transmits rudder angle commands to the servo motors in the AUV's energy and power unit to control the AUV's navigation and tracking.

[0094] The aforementioned single-channel hydrophone unit 2.1 is a miniature hydrophone based on a semiconductor MEMS manufacturing platform.

[0095] A passive underwater target tracking method, comprising the following steps:

[0096] 1) After filtering and amplifying the received underwater acoustic signal, it is converted into an analog underwater acoustic digital signal. Then, a refined spectrum estimation is performed to obtain the frequency corresponding to the maximum signal amplitude in a specific frequency band (hereinafter referred to as "master frequency"), and the signal-to-noise ratio of the signal at that frequency is calculated.

[0097] 2) Perform target detection;

[0098] 3) Calculate the Doppler frequency shift and estimate the target azimuth angle;

[0099] 4) Estimate the motion trend relative to the target;

[0100] 5) Calculate the rudder angle and send it to the main control computer of the AUV platform via the communication interface circuit. The main control computer then transmits the rudder angle command to the servo motor in the AUV's energy and power unit to control the AUV's navigation and tracking.

[0101] The specific steps are as follows:

[0102] S1. This invention proposes a refined spectrum estimation algorithm. This algorithm can calculate the dominant frequency f of an underwater acoustic digital signal within a specific frequency band. m and the signal-to-noise ratio (SNR) at the main frequency. m ;

[0103] S2. This invention proposes a target detection method. The execution cycle of step S1 is set to T0 (executed once every T0), and the execution cycle of steps S2-S5 is T (T = nT0, n > 1 and n ∈ Z). Within the execution cycle T seconds of steps S2-S5, the f calculated in step S1 is continuously accumulated. m and SNR m The data constitutes the main frequency and signal-to-noise ratio data sequence:

[0104]

[0105]

[0106] When the target signal is received, the dominant frequency of the received signal and the signal-to-noise ratio (SNR) at that frequency should be within a reasonable range (the dominant frequency should not exceed the maximum Doppler frequency shift interval, and the SNR should not be lower than the limit value (this limit value is affected by environmental factors and needs to be measured in a specific underwater acoustic environment)). Set the threshold f required for target detection. min f max SNR min and μ (0 < μ < 1). At this point, it is necessary to calculate... China satisfies The number of elements n1 and China satisfies The number of elements is n2. When n1≥μn and n2≥μn, the target is considered to be detected, and the process proceeds to step S3; otherwise, the target is not detected, and the process returns to step S1.

[0107] S3. This invention proposes a method for calculating the Doppler frequency shift and estimating the target azimuth angle. Based on the dominant frequency data sequence calculated in S2... Calculate the Doppler frequency shift Δf of the received signal relative to the center frequency f0 of the target transmitted signal. Take... China satisfies The elements constitute a new main frequency data sequence The Doppler frequency shift calculation method is as follows:

[0108]

[0109] In the above formula, f0 is the center frequency of the signal transmitted by the underwater target.

[0110] The target azimuth angle θ is estimated based on the Doppler frequency shift Δf, and the calculation method is as follows:

[0111]

[0112] In the above formula, v is the speed of the AUV and c is the speed of sound underwater.

[0113] It is important to note that the target azimuth angle θ should be within the range of [0°, 90°]. If the calculated value of θ exceeds the normal range due to interference from harmonic signals, the maximum value of θ should be limited to 90°.

[0114] S4. This invention proposes an algorithm for estimating the relative motion trend of a target based on the Doppler frequency shift principle. There are four possible estimation results: (1) the AUV has passed the target; (2) the AUV has not passed the target and is heading directly towards it; (3) the AUV has not passed the target and is deviating from it; (4) the AUV has not passed the target and is approaching it. When the estimation result is the first one (the AUV has passed the target), the program stops running, the mission ends, the vehicle rises and transmits GPS signals; otherwise, proceed to step S5.

[0115] S5. This invention proposes a method for calculating rudder angle. The method involves setting the K-value coefficient, which requires calculating K based on the estimation results from step S4. k The calculation method is as follows:

[0116]

[0117] In the above formula, K k-1 This is the K value coefficient for S5 in the previous execution cycle. Therefore, the rudder angle calculation method is as follows:

[0118]

[0119] In the above formula, k m k r and For the pre-calibrated weight coefficients and maximum rudder angle threshold, Roll is the roll angle of the AUV measured by the navigation sensing unit of the AUV platform at this time.

[0120] Calculate the rudder angle Then, the rudder angle is sent to the main control computer of the AUV platform via the communication interface circuit. The main control computer transmits the rudder angle command to the servo motor and controls the AUV to track and navigate. Finally, the process returns to step S1.

[0121] Preferably, step S1 above proposes a refined spectrum estimation method for analyzing and processing underwater acoustic digital signals. This algorithm can effectively improve the frequency resolution of the signal within a specific frequency band, given a limited number of FFT points, thereby improving the accuracy of frequency calculation. The specific method is as follows:

[0122] S1-1, Signal Modulation Frequency Shift: Assume the lowest frequency in a specific frequency band is F. min The highest frequency is F max Then the center frequency is f k =(F max -F min ) / 2. Multiply the analog signal by a factor. This allows for frequency shifting of the signal, which is equivalent to shifting the signal at frequency f. k The spectral line corresponding to that point has moved to the origin. For a discrete-time series, it can be represented as:

[0123]

[0124] In the above formula, x0(n) is the original underwater acoustic digital signal, f s This refers to the sampling frequency of the digital-to-analog converter circuit in the aforementioned signal acquisition and processing unit;

[0125] S1-2, Low-pass filtering: Resampling must satisfy the time-domain sampling theorem to ensure that the signal does not aliasing in the frequency domain after FFT. Therefore, the frequency-shifted signal needs to be low-pass filtered before resampling. The cutoff frequency of the low-pass filter is:

[0126] f c =F max -f k

[0127] It's important to note that while ordinary FIR filters can guarantee linear phase, their filtering performance is affected by the filter order. Higher filter orders result in better filtering performance, but also require more computation.

[0128] Since the method of this invention does not require the use of signal phase information, it does not employ a conventional FIR filter for time-domain filtering. Instead, it performs filtering directly in the frequency domain to minimize spectral aliasing and reduce computational load. The specific filtering method is as follows:

[0129] (1) Perform a normal FFT transform on the signal x(n) obtained in step S1-1 above to obtain the frequency domain signal x(w);

[0130] (2) Frequency domain filtering is performed using a rectangular window, i.e., x′(w)=x(w)*h(w), where h(w) is the frequency domain filter function:

[0131]

[0132] (3) Perform IFFT transform on the filtered frequency domain signal x′(w) to obtain the filtered time domain signal x′(n);

[0133] S1-3, Resampling

[0134] After the signal is modulated and filtered, the number of sampling points will decrease. At this point, a lower sampling frequency f′ will be used. s Resampling x′(n) to obtain signal x″(n) will ensure the same number of sampling points by padding with zeros. Therefore, the frequency resolution is:

[0135]

[0136] Where N is the number of FFT points.

[0137] S1-4, FFT Transform

[0138] After the above steps, the signal x″(n) is a complex signal. Then, performing an N-point FFT on the signal will yield the refined frequency domain signal x″(w).

[0139] S1-5, Calculate the main frequency f m

[0140] Find the refined signal amplitude spectrum |x″(w)| in a specific frequency band [F min F max The frequency with the largest amplitude within the range is taken as the main frequency f. m The calculation method is as follows:

[0141] w m =argmax|x″(w)|(w∈[2πF) min / f s ,2πF max / f s ])

[0142]

[0143] S1-6. Calculate the signal-to-noise ratio (SNR) of the signal at the dominant frequency. m

[0144] Determine f m After that, it is necessary to determine based on f m Calculate the signal-to-noise ratio (SNR) m The calculation method is as follows:

[0145]

[0146]

[0147] As a preferred approach, step S4 above proposes an algorithm for estimating relative motion trends based on the Doppler frequency shift principle. The specific method is as follows:

[0148] S4-1. According to the Doppler frequency shift formula, when the AUV passes the target, the received frequency will be less than the frequency f0 of the signal emitted by the sound source target. Therefore, a threshold ε (0 < ε < 1) can be set to determine whether the AUV has passed the target.

[0149] The main frequency data sequence was calculated in step S3 above. calculate In which f′ is satisfied m (i) The number of elements ≤ f0 is n3. When n3 ≥ εn′, it is determined that the AUV has crossed the target, the algorithm ends and proceeds to step S5; otherwise, the AUV has not crossed the target and proceeds to step S4-2.

[0150] S4-2, Set the minimum error angle θ min When the included angle θ calculated in step S3 above is less than or equal to θ min If the AUV is determined to have not crossed the target and is sailing directly toward the target, the algorithm ends and proceeds to step S5; otherwise, it proceeds to step S4-3.

[0151] S4-3. When the AUV is not sailing directly towards the target, it is necessary to determine whether the AUV is approaching or deviating from the target. This requires calculation based on step S3. Calculate the rate of change of the current period frequency The calculation formula is as follows:

[0152]

[0153] In the formula, T0 represents the execution cycle of step S1 above. A frequency change rate threshold ω (ω < 0) is set. According to the Doppler frequency shift principle, as the AUV moves away from the target, the received frequency gradually decreases; the faster the frequency decreases, the further the AUV deviates from the target. Therefore, when the following conditions are met... If the AUV does not cross the target and deviates from it, it is determined that the AUV does not cross the target and is approaching it.

[0154] Example:

[0155] The following example, using the tracking of an underwater black box beacon with a transmission signal frequency of 37.5 kHz, illustrates the detailed implementation steps and calculation methods of this invention:

[0156] like Figure 5The diagram illustrates the scenario of this example. A 37.5kHz black box beacon is stationary at a depth of 5 meters underwater. The AUV is set to a depth of 5 meters, a speed of v = 5 m / s, and a distance of 1500 meters from the starting point to the beacon. The speed of sound underwater is c = 1500 m / s. The AUV first follows a pre-defined route at a constant depth. Upon detecting the target, it locates and tracks the target using the method of this invention, autonomously guiding itself to the target location.

[0157] It should be noted that in this example, the microcontroller used in the signal acquisition and processing module is an STM32F429, the FFT number is N = 4096, and the sampling frequency of the digital-to-analog converter circuit is f. s =125kHz; the microcontroller used in the main control computer is an STM32F407; the underwater beacon's sound source level is 180dB, and the signal frequency is f0 = 37.5kHz; the AUV is powered by a single-push propeller, and four servos are installed at the tail to control the navigation attitude.

[0158] A passive underwater target tracking method, such as Figure 1 As shown, the steps include:

[0159] S1. This invention proposes a refined spectrum estimation algorithm, which can calculate the dominant frequency f of an underwater acoustic digital signal within a specific frequency band. m and the signal-to-noise ratio (SNR) at the main frequency. m ;

[0160] like Figure 3 As shown, the specific method is as follows:

[0161] S1-1, Signal Modulation Frequency Shift: Assume the lowest frequency of a specific frequency band is F. min The highest frequency is F max Then the center frequency is f k =(F max -F min ) / 2. Multiply the analog signal by a factor. This allows for frequency shifting of the signal, which is equivalent to shifting the signal at frequency f. k The spectral line corresponding to that point has moved to the origin. For a discrete-time series, it can be represented as:

[0162]

[0163] In the above formula, x0(n) is the original underwater acoustic digital signal, f s f is the sampling frequency of the digital-to-analog converter circuit in the aforementioned signal acquisition and processing unit. In this example, f s=125kHz. Since the underwater beacon can be considered stationary, the reasonable range of the receiving frequency can be calculated using the AUV speed v, the underwater speed of sound c, and the beacon's transmitted signal frequency f0, combined with the Doppler frequency shift formula.

[0164]

[0165]

[0166] f min with f max It should be included within the main frequency band, but in order to minimize the influence of other frequency components, the main frequency band cannot be set too wide. Taking all factors into consideration, F is set... min and F max If the frequencies are 37.3kHz and 37.7kHz respectively, then f k =(F max -F min ) / 2 = 37.5kHz.

[0167] S1-2, Low-pass filtering: Resampling must satisfy the time-domain sampling theorem to ensure that the signal does not aliasing in the frequency domain after FFT. Therefore, the frequency-shifted signal needs to be low-pass filtered before resampling. The cutoff frequency of the low-pass filter is:

[0168] f c =F max -f k

[0169] It's important to note that while ordinary FIR filters can guarantee linear phase, their filtering performance is affected by the filter order. Higher filter orders result in better filtering performance, but also require more computation.

[0170] Since the method of this invention does not require the use of signal phase information, it does not employ a conventional FIR filter for time-domain filtering. Instead, it performs filtering directly in the frequency domain to minimize spectral aliasing and reduce computational load. The specific filtering method is as follows:

[0171] (1) Perform a normal FFT transform on the signal x(n) obtained in step S1-1 above to obtain the frequency domain signal x(w);

[0172] (2) Frequency domain filtering is performed using a rectangular window, i.e., x′(w)=x(w)*h(w), where h(w) is the frequency domain filter function:

[0173]

[0174] (3) Perform IFFT transform on the filtered frequency domain signal x′(w) to obtain the filtered time domain signal x′(n);

[0175] In this example, the cutoff frequency f of the low-pass filter c The result can be obtained from the calculation of S1-1: f c =F max -f k =200Hz.

[0176] S1-3, Resampling

[0177] After the signal is modulated and filtered, the number of sampling points will decrease. At this point, a lower sampling frequency f′ will be used. s Resampling x′(n) to obtain signal x″(n) will ensure the same number of sampling points by padding with zeros. Therefore, the frequency resolution is:

[0178]

[0179] Where N is the number of FFT points. In this example, f′ s =1kHz, N=4096, from which the frequency resolution Δf can be calculated. k =f s ′ / N=0.24Hz.

[0180] S1-4, FFT Transform

[0181] After the above steps, the signal x″(n) is a complex signal. Then, performing an N-point FFT on the signal will yield the refined frequency domain signal x″(w).

[0182] S1-5. Calculate the frequency f corresponding to the maximum amplitude of the signal within a specific frequency band. m

[0183] Find the refined signal amplitude spectrum |x″(w)| in a specific frequency band [F min F max The frequency with the largest amplitude within the range is taken as the main frequency f. m The calculation method is as follows:

[0184] w m =argmax|x″(w)|(w∈[2πF) min / f s ,2πF max / f s ])

[0185]

[0186] In this example, F min F max and f s The frequencies are 37.3kHz, 37.7kHz, and 125kHz, respectively.

[0187] S1-6. Calculate the signal-to-noise ratio (SNR) of the signal at the dominant frequency. m

[0188] Determine f m After that, it is necessary to determine based on f m Calculate the signal-to-noise ratio (SNR) m The calculation method is as follows:

[0189]

[0190]

[0191] It should be noted that in this embodiment, due to insufficient memory in the STM32F429 microcontroller used in the signal processing circuit, it is not possible to perform too many FFT calculations at a time. Therefore, in this embodiment, the average of the five results of executing S1-1 to S1-6 five times, which yield the main frequency and the signal-to-noise ratio at the main frequency, will be used as the final output f. m and SNR m This method can significantly increase the time gain of the signal, thereby improving the frequency estimation accuracy, without increasing the computational memory usage.

[0192] S2. Based on 5.2 above, this invention proposes a target detection method. The execution cycle of step S1 is set to T0 (executed once every T0 seconds), and the execution cycle of steps S2-S5 is T (T = nT0, n > 1 and n ∈ Z). Within the execution cycle T seconds of steps S2-S5, the f calculated in step S1 will be continuously accumulated. m and SNR m The data constitutes the main frequency and signal-to-noise ratio data sequence:

[0193]

[0194]

[0195] When a target signal is received, the main frequency of the received signal and the signal-to-noise ratio at that frequency should be within a reasonable range. Set the threshold f required for target detection. min f max SNR min and μ (0 < μ < 1). At this point, it is necessary to calculate... China satisfies The number of elements n1 and China satisfies The number of elements is n2. When n1≥μn and n2≥μn, the target is considered to be detected, and the process proceeds to step S3; otherwise, the target is not detected, and the process returns to step S1.

[0196] In this embodiment, the execution period T0 of step S1 is 0.1 seconds, and the calculation period T of steps S2-S5 is 1 second. The threshold f required for the target detection is... min f max The values ​​of 37.375kHz and 37.625kHz calculated in step S1-1 are taken respectively. The AUV is then driven at a distance of 1500 meters from the underwater beacon to measure the signal-to-noise ratio (SNR) of the received signal at the dominant frequency within the aforementioned specific frequency band. m =10dB, therefore the above threshold SNR is set. min =10dB. Setting the above threshold μ=0.8 means that a target is considered detected when 80% or more of the received signal is a valid signal.

[0197] S3. Based on 5.3 above, this invention proposes a method for calculating the Doppler frequency shift and estimating the target azimuth angle. The dominant frequency result calculated in S2... Calculate the Doppler frequency shift Δf of the received signal relative to the center frequency f0 of the target transmitted signal. Take... China satisfies The elements constitute a new main frequency data sequence The Doppler frequency shift calculation method is as follows:

[0198]

[0199] In the above formula, f0 is the center frequency of the underwater target's transmitted signal, f0 = 37.5 kHz.

[0200] The target azimuth angle θ is estimated based on the Doppler frequency shift Δf, and the calculation method is as follows:

[0201]

[0202] In the above formula, v is the speed of the AUV and c is the speed of sound underwater. v = 5 m / s, c = 1500 m / s.

[0203] It is important to note that the target azimuth angle θ should be within the range of [0°, 90°]. If the calculated value of θ exceeds the normal range due to interference from harmonic signals, the maximum value of θ should be limited to 90°.

[0204] S4. Based on 5.4 above, this invention proposes an algorithm for estimating the relative motion trend of a target based on the Doppler frequency shift principle. There are four possible estimation results: (1) the AUV has passed the target; (2) the AUV has not passed the target and is heading directly towards it; (3) the AUV has not passed the target and is deviating from it; (4) the AUV has not passed the target and is approaching it. When the estimation result is the first one (the AUV has passed the target), the program stops running, the mission ends, the vehicle rises and transmits GPS signals; otherwise, proceed to step S5.

[0205] like Figure 4 As shown, the specific method is as follows:

[0206] S4-1. After the AUV passes the target, according to the Doppler frequency shift formula, the received frequency will be less than the frequency f0 of the signal emitted by the sound source target. Set a threshold ε (0 < ε < 1) to determine whether the AUV has passed the target.

[0207] The main frequency data sequence was calculated in step S3 above. calculate China satisfies The number of elements is n3. When n3≥εn′, it is determined that the AUV has crossed the target, the algorithm ends and proceeds to step S5; otherwise, the AUV has not crossed the target and proceeds to step S4-2.

[0208] In this example, the threshold ε is set to 0.7. This represents the effective main frequency data sequence within the execution period T seconds of steps S2-S5. If 70% or more of the data are less than the center frequency f0 of the beacon transmission signal, it is considered that the target has been crossed.

[0209] S4-2, Set the minimum error angle θ min When the included angle θ calculated in step S3 above is less than or equal to θ min If the AUV is determined to have not crossed the target and is sailing directly towards it, the algorithm ends and proceeds to step S8; otherwise, it proceeds to step S5-3.

[0210] In this example, the threshold θ is set as described above. min =3 degrees. This means that when the target azimuth angle θ≤3°, the AUV is considered to be sailing directly towards the target, and in this case, straight-line navigation is sufficient.

[0211] S4-3. When the AUV is not sailing directly towards the target, it is necessary to determine whether the AUV is approaching or deviating from the target. This requires calculation based on step S3. Calculate the rate of change of the current period frequency The calculation formula is as follows:

[0212]

[0213] In the formula, T0 represents the calculation period of steps S1-S2 above. A frequency change rate threshold ω (ω < 0) is set. According to the Doppler frequency shift principle, as the AUV moves away from the target, the received frequency gradually decreases; the faster the decrease, the further the AUV deviates from the target. Therefore, when the following conditions are met... If the AUV does not cross the target and deviates from it, it is determined that the AUV does not cross the target and is approaching it.

[0214] In this embodiment, the frequency change rate threshold ω is set to -1.2.

[0215] S5. Based on 5.5 above, this invention proposes a method for calculating rudder angle. The method involves setting the K-value coefficient, which requires calculating K based on the estimation results from step S4. k The calculation method is as follows:

[0216]

[0217] In the above formula, K k-1 This is the K value coefficient for S5 in the previous execution cycle. Therefore, the rudder angle calculation method is as follows:

[0218]

[0219] In the above formula, k m k r and For the pre-calibrated weighting coefficients and maximum rudder angle threshold, Roll is the roll angle of the AUV measured by the navigation sensor unit of the AUV platform at this time; in this example, k is set to... m k r and The values ​​are 0.2, 0.3, and 20 degrees respectively.

[0220] Calculate the rudder angle Then, the rudder angle is sent to the main control computer of the AUV platform via the communication interface circuit. The main control computer transmits the rudder angle command to the servo motor and controls the AUV to track and navigate. Finally, the process returns to step S1.

[0221] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. A method of passive tracking of an underwater target, characterized in that: The steps are as follows: 1) After filtering and amplifying the received underwater acoustic signal, it is converted into an analog underwater acoustic digital signal. Then, a refined spectrum estimation is performed to obtain the frequency corresponding to the maximum signal amplitude in a specific frequency band, and the signal-to-noise ratio of the signal at that frequency is calculated. The specific filtering method is as follows: The obtained signal is subjected to a general FFT transform to obtain a frequency domain signal ; Frequency domain filtering is performed using a rectangular window, i.e. ,in For frequency domain filter functions: to the filtered frequency domain signal IFFT transformation to obtain the filtered time domain signal ; 2) Perform target detection; 3) Calculate the Doppler frequency shift and estimate the target azimuth angle; 4) Estimate the relative motion trend of the target; 5) Calculate the rudder angle and send it to the main control computer of the AUV platform through the communication interface circuit. The main control computer then transmits the rudder angle command to the servo motor in the AUV's energy and power unit to control the AUV's navigation and tracking. The underwater target passive tracking method uses an apparatus consisting of an AUV platform (1) and an underwater target detection device (2); The AUV platform (1) consists of a main control computer and a main control computer data interface unit (1.1), a navigation sensor unit (1.2), and an energy and power unit (1.3). The energy and power unit (1.3) consists of a battery (1.3.1), a motor (1.3.2), and a servo motor (1.3.3). The main control computer is used to control the entire AUV working system. The underwater target detection device consists of a single-channel hydrophone unit (2.1) and a signal acquisition and processing unit (2.2); the signal acquisition and processing unit (2.2) consists of a signal preprocessing circuit (2.2.1), a digital-to-analog conversion circuit (2.2.2), a signal processing circuit (2.2.3), and a data communication interface circuit (2.2.4); the signal preprocessing circuit (2.2.1) is connected to the single-channel hydrophone unit (2.1), and the data communication interface circuit (2.2.4) is connected to the main control computer data interface unit (1.1) in the AUV platform (1). The single-channel hydrophone unit (2.1) receives underwater acoustic signals and filters and amplifies the analog signals through the signal preprocessing circuit (2.2.1); the signal acquisition and processing unit (2.2) acquires the signals and converts the analog signals into digital signals through the digital-to-analog converter circuit (2.2.2); the signal processing circuit (2.2.3) performs a series of processing on the digital signals and transmits the processing results to the AUV's main control computer through the data communication interface circuit (2.2.4).

2. The method of passive tracking of underwater targets according to claim 1, characterized in that: The specific processing method is as follows: S1, the main frequency corresponding to the maximum value of the signal amplitude of the underwater acoustic digital signal in a specific frequency range and the signal-to-noise ratio at the main frequency ; S2, set the execution period of step S1 as , continuously accumulate the data calculated in step S1 and data, constitute the main frequency and signal-to-noise ratio data sequence: When a target signal is received, set the threshold required for target detection. , and ,calculate China satisfies number of elements as well as China satisfies number of elements ,when and If a target is detected, proceed to step S3; otherwise, if no target is detected, return to step S1. S3, the main frequency data sequence calculated based on S2 Calculate the center frequency of the received signal relative to the target's transmitted signal. Doppler frequency shift ,Pick China satisfies The elements constitute a new main frequency data sequence The Doppler frequency shift calculation method is as follows: In the formula is the center frequency of the signal transmitted to the underwater target; According to the Doppler shift Estimating target azimuth angle The calculation method is: In the above formulae is the speed of the AUV, is the underwater sound speed, Target azimuth exist Within the range, if interference from harmonic signals causes The calculated value is outside the normal range and should be... Maximize Limit; S4. An algorithm for estimating the relative motion trend of the target based on the Doppler frequency shift principle. The estimation results are: (1) The AUV has passed the target; (2) The AUV has not passed the target and is sailing towards the target; (3) The AUV has not passed the target and is sailing away from the target; (4) The AUV has not passed the target and is sailing close to the target. If the estimation result is the first type, the program stops running, the mission ends, the vehicle rises and transmits GPS signals; otherwise, proceed to step S5. S5, calculate rudder angle : The K value coefficient is set, and the estimated result of step S4 is calculated , and the calculation method is: In the above formula If the K value coefficient of S5 in the previous execution cycle is used, then the rudder angle is... The calculation method is as follows: In the above formula With is the weight coefficient of the early calibration and the maximum threshold of the rudder angle, is the roll angle of the AUV platform measured by the navigation sensor unit of the AUV at this time The rudder angle is calculated After that, the rudder angle is sent to the main computer of the AUV platform through the communication interface circuit, the main computer transmits the rudder angle command to the rudder and controls the AUV to track the voyage, and finally returns to step S1.

3. A method of passive tracking of an underwater target as claimed in claim 2, characterized in that: Step S1 involves analyzing and processing the underwater acoustic digital signal using a refined spectrum estimation method. The specific method is as follows: S1-1, Signal modulation frequency shift: Assuming the lowest frequency in a specific frequency band is... The highest frequency is The center frequency is By multiplying the analog signal by a factor Frequency shifting of a signal, for a discrete-time series, is represented as: In the above formula is the original underwater acoustic digital signal, is the sampling frequency of the digital-to-analog conversion circuit in the signal acquisition and processing unit; S1-2, Low-pass filtering: The frequency-shifted signal is first low-pass filtered, and then resampled. The cutoff frequency of the low-pass filter is: S1-3, Resampling After modulation and filtering, the number of points is reduced, with a lower sampling frequency To Resample the signal , with the same number of points, by zero-padding, the frequency resolution is Where N is the number of FFT points; S1-4, FFT Transform the signal after the step for a complex signal, performing N-point FFT on the signal again can obtain a refined frequency domain signal ; S1-5, calculating the main frequency Finding the refined signal amplitude spectrum In a specific frequency band The frequency with the largest internal amplitude is used as the main frequency. The calculation method is as follows: S1-6, compute signal to noise ratio at the primary frequency determining After that, it is necessary to calculate the signal-to-noise ratio of the signal The signal-to-noise ratio of the signal is calculated The calculation method is: 。 4. The method according to claim 2 or 3, characterized in that: The condition for continuously accumulating in the step S2 is that the execution period of the steps S2-S5 is is continuously accumulated in the execution period T seconds of the steps S2-S5.

5. The method of claim 4, wherein: The specific method of step S4 is as follows: S4-1. According to the Doppler shift formula, when the AUV passes the target, the received frequency is less than the frequency of the transmitted signal from the sound source target , a threshold is set to determine whether the AUV has passed the target The main frequency data sequence was calculated in step S3. ,calculate China satisfies number of elements ,when If the AUV has passed the target, the algorithm ends and proceeds to step S5; otherwise, if the AUV has not passed the target, proceed to step S4-2. S4-2, Set the minimum error angle When the included angle calculated in step S3 If the AUV is determined to have not crossed the target and is sailing directly toward the target, the algorithm ends and proceeds to step S5; otherwise, it proceeds to step S4-3. S4-3, when the AUV is not heading directly to the target, it is necessary to determine whether the AUV is approaching the target or deviating from the target, according to the step S3 Calculate the current period frequency change rate The calculation formula is as follows: in the formula Set a frequency change rate threshold for the execution cycle of step S1. When satisfied If the AUV does not cross the target and deviates from it, it is determined that the AUV does not cross the target and is approaching it.

6. The method of claim 1, wherein: The single-channel hydrophone unit (2.1) is selected from a micro hydrophone based on a semiconductor MEMS manufacturing platform.

Images