A passive sonar buoy underwater maneuvering weak target adaptive detection method and system

Abstract

This invention belongs to the field of passive underwater target detection technology, and discloses an adaptive detection method and system for underwater maneuvering weak targets using passive sonar buoys. First, a cooperative detection buoy group is automatically constructed using a dual-maximum method. This group is used to locate underwater acoustic signals and eliminate underwater noise and interference outside the cooperative detection area. Second, the segmented sampling and processing of the positioning information is projected in both the horizontal and vertical directions. Pre-detection tracking theory and the Hough detection algorithm set are used to adaptively detect the position information on both planes, extracting points that satisfy the dual accumulation thresholds from noise and interference. If detection fails, the segmented sampling and processing are repeated, and the projection and detection process is repeated to obtain initial detection points. Finally, multi-level motion constraints are used to fuse the points on both planes to obtain a high-precision true target trajectory, thus achieving passive detection of underwater weak targets.

Description

Technical Field

[0001] This invention relates to the field of signal processing in underwater acoustic detection, and in particular to an adaptive detection method and system for underwater maneuvering weak targets based on a passive sonar buoy array. Background Technology

[0002] In recent years, countries have been comprehensively utilizing various means to continuously advance the development of underwater target acoustic stealth capabilities. Airborne sonar buoys are an important symbol of the modernization of airborne early warning and detection, and a crucial means of countering the threat of underwater vehicles. Airborne sonar buoys can passively receive the radiated noise of underwater vehicles for covert detection; they can also actively emit sound waves to detect underwater vehicles and receive target echoes for precise detection. Therefore, compared to other detection equipment, airborne sonar buoys are the most important instruments used in airborne early warning and detection. However, facing the complex marine environment, and with the increasing low detectability of underwater targets, passive detection by airborne sonar buoys is becoming increasingly difficult.

[0003] In recent years, underwater target detection technology and applications have developed rapidly. Domestic and international research indicates that Track Before Detect (TBD) technology is an effective method for detecting underwater targets with low observability. Compared to classic sonar target identification techniques, its advantage lies in the fact that TBD does not directly detect a single received signal, but rather accumulates multiple consecutive received signals without correlation, increasing the amount of target information in the detection data and thus improving the detection probability. Among various TBD algorithms, the Hough Transform TBD (HT-TBD) algorithm exhibits excellent detection performance in underwater target detection. At present, research on HT-TBD algorithm mainly focuses on the detection of linear moving targets ([1] Wang, Xuemin, Lu Renwei, Li Wenhai. Underwater target passive detection method based on Hough transform track-before-detect[J]. Journal of Physics Conference Series JPhCS, 2022, 2258(1): 1-6. [2] Wang Xuemin, Zhang Xiangyu, Wu Minghui, et al. Underwater target detection method based on cross-location and Hough transform before detection tracking[J / OL]. Systems Engineering and Electronics Technology: 1-10[2022-05-07].), or the detection of individual typical curved moving targets ([3] Wang Xuemin, Tan Shuncheng, Yu Hongbo. Passive Detection Method of Underwater Maneuvering Target Based on Random Parabolic Hough Transform[C] / / 2022 5th International Conference on Information Communication and Signal Processing, Shenzhen, China, 2022: 534-538. [4] Wang Xuemin, Wu Fang, Zhang Xiangyu, et al. A passive detection method for underwater targets with small rudder angles [J / OL]. Systems Engineering and Electronics: 1-10 [2023-05-31].), while the adaptability is poor under unknown motion conditions. When performing tasks such as cruising, entering ambush areas and escaping danger zones, underwater targets ensure their underwater navigation safety by changing speed, direction and depth.However, existing linear HT-TBD algorithms and typical curve-based HT-TBD algorithms mainly focus on the horizontal plane, while some passive directional sonar buoys lack or fail to fully utilize pitch information, making passive detection of underwater maneuvering weak targets in three-dimensional space more complex. Summary of the Invention

[0004] This invention proposes an adaptive detection method and system for underwater maneuvering weak targets using a passive sonar buoy array. This method aims to overcome the limitations of underwater weak target determination and maneuvering detection applications, address the impact of underwater weak target depth variations on the detection of passive directional sonar buoy arrays, and thus expand the adaptability of underwater weak target adaptive passive detection.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] An adaptive detection method for underwater maneuvering weak targets using a passive sonar buoy array includes:

[0007] Step 1: The collaborative detection buoy group is automatically constructed using the maximum energy method and the minimum spacing method. The collaborative detection buoy group is used to locate the underwater acoustic signal and eliminate underwater noise and interference outside the collaborative detection area to obtain the information of the segment to be detected. The buoys in the passive directional sonar buoy array are passive directional sonar buoys with the function of measuring azimuth and pitch angles.

[0008] Step 2: Adaptively detect the position information of the horizontal and vertical planes using the pre-detection tracking theory and the Hough detection algorithm set, and obtain the points that satisfy the double accumulation threshold from noise and interference;

[0009] Step 3: By fusing dual-plane point tracks using multi-level motion constraints, a high-precision target trajectory is obtained, thus achieving passive detection of weak underwater targets.

[0010] Optionally, step one specifically includes:

[0011] Within the detection range of the coverage array, the coordinated detection buoy group is determined by detecting the position of each sonar buoy and the acquired signal strength, based on the maximum signal strength and minimum spacing method. Assume the sonar buoy coverage array consists of N rows and M columns of passive sonar buoys, where the spacing between adjacent rows and adjacent columns is D and d, respectively. In the Cartesian coordinate system Oxyz, the coordinates of the sonar buoy positions are (x...). si ,y si ,z si (i = 1, 2, ..., N × M), the set of measurement data of the i-th sonar buoy at time k is:

[0012] Z i ={z ki |k=1,2,…,K}

[0013] In the formula: zki ={(α) k(i)j ,β k(i)j ,e k(i)j |j=1,2,…,n j Let} be the dataset received by the i-th sonar buoy at time k. Where n j To measure the number, α k(i)j Let β be the azimuth angle of the j-th measurement. k(i)j Let e ​​be the pitch angle measured for the j-th time. k(i)j This represents the energy information for the j-th measurement.

[0014] Based on the azimuth and elevation angles measured by N×M passive sonar buoys at time k, their energy information e is characterized. k(i)j The set of energy values ​​received by N×M sonar buoys at time k is:

[0015] {e kij |i=1,2,…,N×M,j=1,2,…,n i}

[0016] The maximum energy value is searched using the maximum value method, and the corresponding sonar buoy is i. * .

[0017]

[0018] In the formula: the decision threshold is N s / 2,N s =N×M, where N is the number of buoys in the sonar coverage array and N is the buoy accumulation matrix. Based on the maximum energy value formula, further search is conducted for the two buoys i1 with the second-highest energy values ​​in the sonar coverage array. * and i2 * Calculate buoy i respectively * With buoy i1 * and buoy i2 * The spacing r(i) * i1 * ), r(i * i2 * Then, according to the minimum spacing formula min{r(i * i1 * ),r(i * i2 * Searching for the corresponding sonar buoy, i.e., the cooperative detection buoy. ** ;

[0019] In the rectangular coordinate system Oxyz, a collaborative detection buoy group (i * i ** Processing time k n j Individual measurement z ki ={(α) k(i)j ,βk(i)j ,e k(i)j |j=1,2,…,n j ,i=i * i ** The directional information α in} k(i)j Pitch information β k(i)j and energy information e k(i)j Cross-positioning technology was used to obtain the three-dimensional spatial location and energy information (x) of weak underwater targets. k(i*)j ,y k(i*)j ,z k(i*)j ,e k(i *)j), Based on the detection range of the coordinated detection buoy group and prior information on the movement of underwater weak targets, the time it takes for a maneuvering underwater weak target to pass through the sonar buoy array is estimated, thus obtaining the information of the segment to be detected. Optionally, step two specifically includes:

[0020] The information of the segment to be detected in three-dimensional space is projected onto horizontal and vertical information; the position and energy information of the horizontal and vertical layers are sampled and processed using the pre-detection tracking theory to obtain the set of points to be detected in the horizontal and vertical layers.

[0021] For the detection segment information sampling, assuming that the dimension of the set of corresponding position points at each time step in the same segment information is N, we arbitrarily select the set of position points Z within K consecutive time steps (K≥5). k For example, in the Cartesian coordinate system Oxyz, the set of points at time k is Z. k ={(x i*k ,y i*k ,z i*k ,e i*k (k = 1, ..., K). The set of location points Z... k Project onto the horizontal plane xy to construct the data space A xy ={Z1,…,Z K} and parameter space B xy and to B xy Initialization and discretization processing.

[0022] Construct the parameter cumulative matrix D xy and energy accumulation matrix E xy and D xy and E xy Initialization process.

[0023] A set of linear Hough detection algorithms was used to test A. xy and B xy Hough detection is performed based on the accumulation of both point count and energy thresholds. Initial target points that satisfy both point count and energy thresholds are calculated on the horizontal xy plane.

[0024] The location point set Zk Projecting onto the vertical plane yz, constructing the data space A yz and parameter space B yz Similarly, the initial target point is obtained on the horizontal plane yz.

[0025] If the Hough detection point for the line class does not exist, construct the curve parameter space C on the horizontal plane xy. xy and C xy Initialization and discretization processing.

[0026] Construct the parameter cumulative matrix F xy and energy accumulation matrix G xy and for F xy and G xy Initialization process.

[0027] In the horizontal plane xy data space A xy Three data points are randomly selected from [Z]. i Z j Z k ](i,j,k=1,…,K), calculate the curve parameter set based on the set of equations for the random Hough transform of the curve.

[0028] A set of curve-based random Hough detection algorithms is used to analyze the data space A. xy and parameter space C xy Random Hough detection based on dual-threshold accumulation of points and energy is performed, and initial target points satisfying the dual accumulation thresholds are calculated on the horizontal plane xy.

[0029] The location point set Z k Projecting onto the vertical plane yz, constructing the data space A yz and parameter space B yz Similarly, underwater target points are acquired on the horizontal plane yz.

[0030] Optionally, step three specifically includes:

[0031] If the detected points exist in step two, point optimization processing is performed. Data points at any adjacent times k, k+1, and k+2 in the underwater target point points in the horizontal plane xy are selected, and the velocity, acceleration, heading, and rate of change of heading measured by the sonar buoy in the horizontal plane xy are calculated as constraints. Based on the prior information of underwater target motion, the initial points that meet the constraints are merged, and the initial points that do not meet the constraints are removed to improve the accuracy of the target points, that is, to obtain underwater target points with high fitting degree in the horizontal plane xy.

[0032] Similarly, underwater target points with high fitting degree are obtained on the vertical plane yz.

[0033] By merging sets of identical data points at the same time, spatial fusion of identical points in both the xy and yz planes is achieved, enabling passive detection of weak underwater targets. This invention also provides an adaptive detection system for underwater maneuvering weak targets using a passive sonar buoy array, the system comprising:

[0034] The passive buoy array preprocessing module automatically constructs a collaborative detection buoy group using the maximum energy method and the minimum spacing method. It uses the collaborative detection buoy group to locate underwater acoustic signals and eliminates underwater noise and interference outside the collaborative detection area to obtain information on the segment to be detected. The buoys in the passive directional sonar buoy array are passive directional sonar buoys with azimuth and pitch angle measurement functions.

[0035] The adaptive Hough detection processing module uses pre-detection tracking theory and Hough detection algorithm set to adaptively detect the position information of the horizontal and vertical planes, and obtains the traces that meet the double accumulation threshold from noise and interference.

[0036] The detection point optimization processing module uses multi-level motion constraints to fuse dual-plane point traces to obtain high-precision target real tracks, thus realizing passive detection of weak underwater targets.

[0037] Optionally, the passive buoy array preprocessing module specifically includes:

[0038] The collaborative detection buoy group acquisition unit automatically constructs the collaborative detection buoy group using the maximum energy method and the minimum spacing method; the buoys in the passive directional sonar buoy array are passive directional sonar buoys with the function of measuring azimuth and elevation angles.

[0039] Assume the sonar buoy coverage array consists of N rows and M columns of passive sonar buoys, where the spacing between adjacent rows and adjacent columns is D and d, respectively; in the rectangular coordinate system Oxyz, the coordinates of the sonar buoy positions are (x... si ,y si ,z si (i = 1, 2, ..., N × M), the set of measurement data of the i-th sonar buoy at time k is:

[0040] Z i ={z ki |k=1,2,…,K}

[0041] Optionally, the dataset received by the i-th sonar buoy at time k is...

[0042] z ki ={(α) k(i)j ,β k(i)j ,e k(i)j |j=1,2,…,n j}

[0043] Where, nj For the number of measurements, β k(i)j Let α be the azimuth angle of the j-th measurement. k(i)j Let e ​​be the pitch angle measured for the j-th time. k(i)j This represents the energy information for the j-th measurement.

[0044] Optionally, the set of energy values ​​received by N×M sonar buoys at time k is:

[0045] {e kij |i=1,2,…,N×M,j=1,2,…,n i}

[0046] The sound intensity acquisition unit is used to acquire the sonar buoy i corresponding to the maximum sound intensity at time k. * :

[0047]

[0048] Wherein, the decision threshold is N s / 2,N s =N×M is the number of buoys in the sonar coverage array, and N is the buoy accumulation matrix.

[0049] Based on the maximum energy value formula, further search for the two buoys i1 with the second-highest energy values ​​in the sonar coverage array. * and i2 * Calculate buoy i respectively * With buoy i1 * and buoy i2 * The spacing r(i) * i1 * ), r(i * i2 * Then, according to the minimum spacing formula min{r(i * i1 * ),r(i * i2 * Searching for the corresponding sonar buoy, i.e., the cooperative detection buoy. ** .

[0050] In the rectangular coordinate system Oxyz, a collaborative detection buoy group (i * i ** Processing time k n j Individual measurement z ki ={(α) k(i)j ,β k(i)j ,e k(i)j |j=1,2,…,n j ,i=i * i ** The directional information α in} k(i)j Pitch information β k(i)jand energy information e k(i)j Cross-positioning technology was used to obtain the three-dimensional spatial location and energy information (x) of weak underwater targets. k(i *)j,y k(i*)j ,z k(i*)j ,e k(i*)j Based on the detection range of the collaborative detection buoy group and prior information on the movement of underwater weak targets, the time it takes for the underwater weak target to pass through the sonar buoy array is estimated to obtain the information of the segment to be detected. Optionally, the adaptive Hough detection processing module specifically includes:

[0051] The detection point acquisition unit is used to adaptively acquire the input detection information of the Hough detection processing module, project the detection segment information in three-dimensional space onto the horizontal and vertical plane information; and use the pre-detection tracking theory to sample and process the position information and energy information of the horizontal and vertical planes respectively to obtain the detection point set of the horizontal and vertical planes.

[0052] The detection segment information sampling unit is used to input detection information for the linear motion detection processing unit and the curvilinear motion detection processing unit. The dimension of the set of position points corresponding to each time step in the same segment information is N. The set of position points Z is arbitrarily selected over K consecutive time steps (K≥5). k For example, in the Cartesian coordinate system Oxyz, the set of points at time k is Z. k ={(x i*k ,y i*k ,z i*k ,e i*k (k = 1, ..., K).

[0053] Initialization and discretization processing unit, used to process the location point set Z k Project onto the horizontal plane xy to construct the data space A xy ={Z1,…,Z K} and parameter space B xy , for B xy Initialize, discretize, and construct the parameter accumulation matrix D. xy and energy accumulation matrix E xy , for D xy and E xy Initialization processing;

[0054] The linear motion detection and processing unit uses a set of linear Hough detection algorithms to process A. xy and B xy Perform Hough detection based on dual threshold accumulation of point count and energy to obtain the initial target point traces that satisfy the dual thresholds of point count and energy on the horizontal plane xy;

[0055] Optionally, the dual-threshold accumulation formula for points and energy is as follows:

[0056]

[0057] Where, (θ t ,ρ m ) represents B xy The coordinates of the center point of each parameter unit are shown in the text; "+=" indicates an accumulation operation.

[0058] The location point set Z k Projecting onto the vertical plane yz, constructing the data space A yz and parameter space B yz Similarly, underwater target points with high fitting degree on the horizontal plane yz are obtained;

[0059] If the Hough detection point for the line class does not exist, construct the curve parameter space C on the horizontal plane xy. xy and C xy Initialization, discretization, and construction of the parameter accumulation matrix F xy and energy accumulation matrix G xy and for F xy and G xy Initialization processing;

[0060] The curve parameter set acquisition unit is used to obtain data in the horizontal plane xy data space A. xy Three data points are randomly selected from [Z]. i Z j Z k (i,j,k=1,…,K), obtain the curve parameter set based on the set of equations for the random Hough transform of the curve;

[0061] The curve motion detection processing unit is used for processing data space A using a set of random Hough detection algorithms for curves. xy and parameter space C xy Perform random Hough detection based on dual-threshold accumulation of point count and energy to obtain initial target points that satisfy the dual accumulation thresholds on the horizontal plane xy.

[0062] The location point set Z k Projecting onto the vertical plane yz, constructing the data space A yz and parameter space B yz Similarly, the underwater target points on the horizontal plane yz are obtained.

[0063] Optionally, the detection point optimization processing module specifically includes:

[0064] If the detection points exist in the adaptive Hough detection processing module, point optimization processing is performed. Data points at any adjacent times k, k+1, and k+2 in the underwater target point points in the horizontal plane xy are selected, and the velocity, acceleration, heading, and rate of change of heading measured by the sonar buoy in the horizontal plane xy are calculated as constraints. Based on the prior information of underwater target motion, initial points that meet the constraints are merged, and initial points that do not meet the constraints are eliminated to improve the accuracy of the target points, that is, to obtain underwater target points with high fitting degree in the horizontal plane xy. Similarly, underwater target points with high fitting degree are obtained in the vertical plane yz. The same data point sets at the same time are merged to realize the spatial fusion of the same points in the two planes xy and yz, and to complete the passive detection of weak underwater targets.

[0065] According to specific embodiments provided by the present invention, the following technical effects are disclosed: The adaptive detection method and system for underwater maneuvering weak targets based on passive sonar buoy arrays provided by the present invention includes: automatically determining a cooperative detection sonar buoy group in a passive directional sonar buoy array using the maximum energy method and the minimum spacing method; obtaining the information of the segment to be detected in three-dimensional space using cooperative positioning technology; the buoys in the passive directional sonar buoy array being passive directional sonar buoys with azimuth and elevation angle measurement functions; projecting the information of the segment to be detected in three-dimensional space onto horizontal and vertical plane information; and utilizing detection... The forward tracking theory samples and processes the position and energy information of the horizontal and vertical planes respectively to obtain the detection point sets of the horizontal and vertical planes. The straight line Hough detection algorithm set is used to perform dual-threshold Hough detection on the detection point sets of the two planes respectively. If the initial point exists, the point optimization method is used to realize the spatial fusion of the same point of the two planes to obtain the three-dimensional motion track of the underwater target. If the straight line Hough detection algorithm set cannot complete the detection, the curve-type random Hough detection algorithm set is used to perform dual-threshold random Hough detection on the segmented information of the two planes respectively. This invention utilizes a passive directional sonar buoy array with azimuth and pitch measurement capabilities to collaboratively locate the three-dimensional spatial position information of underwater targets. The detection segment information in three-dimensional space is projected onto horizontal and vertical planes, and sampling and processing are performed using pre-detection tracking theory to obtain a set of detection points on both planes. A set of straight-line Hough detection algorithms is used to perform dual-threshold Hough detection on the detection segments on both planes. When straight-line detection fails, a set of curve-based random Hough detection algorithms is used to perform dual-threshold random Hough detection on the segments on both planes, thus achieving effective detection of underwater weak targets under unknown motion paths. Attached Figure Description

[0066] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0067] Figure 1 This is a schematic diagram of the adaptive detection method for underwater maneuvering weak targets based on passive sonar buoy array provided in Embodiment 1 of the present invention;

[0068] Figure 2 This is a flowchart of the implementation of the underwater maneuvering weak target adaptive detection method based on passive sonar buoy array provided in Embodiment 1 of the present invention;

[0069] Figure 3 This is a schematic diagram illustrating the construction principle of the collaborative detection sonar buoy group provided in Embodiment 1 of the present invention.

[0070] Figure 4 This is a schematic diagram of the collaborative positioning principle of the collaborative detection sonar buoy group provided in Embodiment 1 of the present invention;

[0071] Figure 5 Here is a flowchart of the Hough detection algorithm provided in Embodiment 1 of the present invention;

[0072] Figure 6 This is a diagram of the Hough detection method for straight lines provided in Embodiment 1 of the present invention;

[0073] Figure 7 This is a diagram of the curve-based random Hough detection method provided in Embodiment 1 of the present invention;

[0074] Figure 8 This is a block diagram of an underwater maneuvering weak target adaptive detection system based on a passive sonar buoy array, provided in Embodiment 2 of the present invention. Detailed Implementation

[0075] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0076] The main drawbacks of traditional passive directional sonar buoys for detecting weak underwater maneuvering targets include:

[0077] Existing passive directional sonar buoy cooperative positioning relies on manual selection of buoys, which reduces positioning accuracy and efficiency.

[0078] Existing passive directional sonar buoy arrays only utilize azimuth information for horizontal level detection, lacking depth level detection, which affects the completeness and reliability of passive underwater target detection.

[0079] Existing Hough detection methods for straight lines and random Hough detection methods for curves are only effective for detecting moving targets with known paths, while their performance is poor for detecting moving targets with unknown paths.

[0080] The intensity of marine environmental noise is much greater than that of the target sound source, often drowning out the target signal. Therefore, the measurement error of marine environmental noise has a significant impact on the final target detection accuracy.

[0081] To address the shortcomings of the prior art, the present invention aims to provide an adaptive detection method and system for underwater maneuvering weak targets based on a passive sonar buoy array, so as to reduce the adverse effects of human intervention on the cooperative positioning of sonar buoys and improve the applicability and reliability of maneuvering target detection tracks.

[0082] Example 1

[0083] This embodiment provides an adaptive detection method for underwater maneuvering weak targets based on a passive sonar buoy array. (See also...) Figure 1 and Figure 2 The method includes:

[0084] Step 1: In the passive directional sonar buoy array, the maximum energy method and the minimum spacing method are used to automatically determine the cooperative detection sonar buoy group. In three-dimensional space, the cooperative detection buoy group uses cooperative positioning technology to eliminate underwater noise and interference outside the cooperative detection area and obtain the information of the segment to be detected. The buoys in the passive directional sonar buoy array are passive directional sonar buoys with the function of measuring azimuth and pitch angle.

[0085] Step 2: Adaptively detect the position information of the horizontal and vertical planes using the pre-detection tracking theory and the Hough detection algorithm set, and obtain the traces that satisfy the double accumulation threshold from noise and interference.

[0086] Step 3: By fusing dual-plane point tracks using multi-level motion constraints, a high-precision target trajectory is obtained, thus achieving passive detection of weak underwater targets.

[0087] The process of obtaining the collaborative detection sonar buoy group in this embodiment is described below:

[0088] See Figure 3Assume that all sonars in the passive sonar buoy coverage array have identical performance and synchronized data. Within the detection range of the coverage array, the cooperative detection buoy group is determined by detecting the position of each sonar buoy and the acquired signal strength, based on the maximum signal strength and minimum spacing method. Assume the sonar buoy coverage array consists of N rows and M columns of passive sonar buoys, where the spacing between adjacent rows and adjacent columns is D and d, respectively. In the rectangular coordinate system Oxyz, the coordinates of the sonar buoy positions are (x...). si ,y si ,z si (i = 1, 2, ..., N × M), the set of measurement data of the i-th sonar buoy at time k is:

[0089] Z i ={z ki |k=1,2,…,K} (1)

[0090] In the formula: z ki ={(α) k(i)j ,β k(i)j ,e k(i)j |j=1,2,…,n j Let} be the dataset received by the i-th sonar buoy at time k. Where n j To measure the number, α k(i)j Let β be the azimuth angle of the j-th measurement. k(i)j Let e ​​be the pitch angle measured for the j-th time. k(i)j This represents the energy information for the j-th measurement.

[0091] Based on the azimuth and elevation angles measured by N×M passive sonar buoys at time k, their energy information e is characterized. k(i)j The set of energy values ​​received by N×M sonar buoys at time k is:

[0092] {e kij |i=1,2,…,N×M,j=1,2,…,n i} (2)

[0093] The maximum energy value is searched using the maximum value method, and the corresponding sonar buoy is i. * .

[0094]

[0095] In the formula: the decision threshold is N s / 2, the number of buoys in the sonar coverage array is N s = N×M, where the buoy accumulation matrix is ​​N.

[0096] According to formula (3), further search for the two buoys i1 with the second maximum energy value in the sonar coverage array. * and i2 *Then, the corresponding sonar buoys are searched using the minimum spacing method to obtain the cooperative detection buoy group (i * i1 * ) or (i * i2 * The formula for the minimum spacing is:

[0097] min{R(i1 * i * ),R(i2 * i * (4)

[0098] Among them, R(i1) * i * ) is the buoy i * and buoy i1 * The spacing, R(i2) * i * ) is the buoy i * and buoy i2 * The spacing.

[0099] See Figure 4 In the rectangular coordinate system Oxyz, a collaborative detection buoy group (i * i ** Processing time k n j Individual measurement z ki ={(α) k(i)j ,β k(i)j ,e k(i)j |j=1,2,…,n j ,i=i * i ** The directional information α in} k(i)j Pitch information β k(i)j and energy information e k(i)j Cross-positioning technology was used to obtain the three-dimensional spatial location and energy information (x) of weak underwater targets. k(i*)j ,y k(i*)j ,z k(i*)j ,e k(i*)j Based on the detection range of the collaborative detection buoy group and the prior information on the movement of underwater weak targets, the time it takes for the underwater weak target to pass through the sonar buoy array is estimated, and the information of the segment to be detected is obtained.

[0100] In this embodiment, see Figure 5 The adaptive Hough detection algorithm set detection method in step two specifically includes:

[0101] Step 1: Detect fragment information sampling. Project the information of the fragment to be detected in 3D space onto horizontal and vertical plane information. Use pre-detection tracking theory to sample and process the position and energy information of the horizontal and vertical planes respectively, obtaining the detection point trace sets of the horizontal and vertical planes. Assuming that the dimension of the position point set corresponding to each time step in the same segment information is N, arbitrarily select the position point set within K consecutive time steps (K≥5). For example, in the Cartesian coordinate system Oxyz, the position point set at time step k is Z. k ={(x ik ,y ik ,z ik )|i=1,…,N}(k=1,…,K).

[0102] Step 2: Project the set of location points onto the horizontal plane xy to construct the data space A. xy ={Z1,…,Z K} and parameter space B xy and to B xy Initialization and discretization processing.

[0103] Step 3: Construct the parameter accumulation matrix D xy and energy accumulation matrix E xy and D xy and E xy Initialization process.

[0104] Step 4: See Figure 6 The Hough detection algorithm set of the straight line class is used to detect A. xy and B xy HT detection based on dual threshold accumulation of point count and energy is performed, and the initial target point traces that satisfy the dual thresholds of point count and energy are calculated on the horizontal plane xy.

[0105] Step 5: Similarly, project the set of location points onto the vertical plane yz to construct the data space A. yz and parameter space B yz Acquire underwater target points on the vertical plane yz.

[0106] Step 6: If the Hough detection point for the line class does not exist, construct the curve parameter space C on the horizontal xy plane. xy and C xy Initialization and discretization processing.

[0107] Step 7: Construct the parameter accumulation matrix F xy and energy accumulation matrix G xy and for F xy and G xy Initialization process.

[0108] Step 8: In the horizontal plane xy data space Axy Three data points are randomly selected from [Z]. i Z j Z k ](i,j,k=1,…,K), calculate the curve parameter set based on the set of equations for the random Hough transform of the curve.

[0109] Step 9: See Figure 7 A set of curve-based random Hough detection algorithms is used to analyze the data space A. xy and parameter space C xy Random Hough detection based on dual-threshold accumulation of points and energy is performed, and initial target points satisfying the dual accumulation thresholds are calculated on the horizontal plane xy.

[0110] Step 10: Similarly, project the set of location points onto the vertical plane yz to construct the data space A. yz and parameter space B yz Acquire underwater target points on the vertical plane yz.

[0111] Step 11: If the detection is unsuccessful, re-segment the sampling process and repeat the projection and detection process to obtain the initial detection points.

[0112] In this embodiment, the dot optimization processing method in step three specifically includes:

[0113] If the detected points exist in step two, point optimization processing is performed. Data points at any adjacent times k, k+1, and k+2 in the underwater target point points in the horizontal plane xy are selected, and the velocity, acceleration, heading, and rate of change of heading measured by the sonar buoy in the horizontal plane xy are calculated as constraints. Based on the prior information of underwater target motion, the initial points that meet the constraints are merged, and the initial points that do not meet the constraints are removed to improve the accuracy of the target points, that is, to obtain underwater target points with high fitting degree in the horizontal plane xy.

[0114] Similarly, underwater target points with high fitting degree are obtained on the vertical plane yz.

[0115] Merging the same data point sets at the same time, the spatial fusion of the same points in the dual-plane xy and yz planes is achieved, thus completing the passive detection of weak underwater targets. Compared with the prior art, the beneficial effects of the underwater maneuvering weak target adaptive detection method based on passive sonar buoy array described in this invention are as follows: (1) This invention utilizes the construction of a passive directional sonar buoy detection group to achieve automatic collaborative positioning, reducing the uncertainty of human operation, improving positioning reliability and increasing positioning efficiency.

[0116] (2) This invention directly utilizes the azimuth and pitch information of existing passive directional sonar buoy groups to perform horizontal plane detection and depth plane detection, thereby improving the completeness and reliability of passive underwater target detection.

[0117] (3) This invention utilizes existing Hough detection and pre-detection tracking theories to reduce the impact of factors such as marine environmental noise, underwater and surface interference targets c, and passive sonar detection threshold settings, which can effectively reduce the probability of false alarms and missed detections of sonar buoys.

[0118] (4) By constructing a set of Hough detection algorithms for straight lines and a set of random Hough detection algorithms for curves, this invention achieves good detection accuracy for maneuvering targets with unknown trajectories.

[0119] (5) The present invention can realize the trajectory detection of weakly maneuvering targets in three-dimensional underwater space.

[0120] Example 2

[0121] This embodiment provides an adaptive detection system for underwater maneuvering weak targets using a passive sonar buoy array. (See also...) Figure 8 The system includes:

[0122] The passive buoy array preprocessing module T1 automatically constructs a collaborative detection buoy group using the maximum energy method and the minimum spacing method. It uses the collaborative detection buoy group to locate underwater acoustic signals and eliminates underwater noise and interference outside the collaborative detection area to obtain information on the segment to be detected. The buoys in the passive directional sonar buoy array are passive directional sonar buoys with the function of measuring azimuth and pitch angles.

[0123] The adaptive Hough detection processing module T2 uses the pre-detection tracking theory and the Hough detection algorithm set to adaptively detect the position information of the horizontal and vertical planes, and obtains the traces that meet the double accumulation threshold from noise and interference.

[0124] The detection point optimization processing module T3 uses multi-level motion constraints to fuse dual-plane point traces, obtaining high-precision target real tracks, thus realizing passive detection of weak underwater targets.

[0125] In this embodiment, the passive buoy array preprocessing module T1 specifically includes:

[0126] Assume the sonar buoy coverage array consists of N rows and M columns of passive sonar buoys, where the spacing between adjacent rows and adjacent columns is D and d, respectively; in the rectangular coordinate system Oxyz, the coordinates of the sonar buoy positions are (x... si ,y si ,z si (i = 1, 2, ..., N × M), the set of measurement data of the i-th sonar buoy at time k is:

[0127] Z i ={z ki |k=1,2,…,K}

[0128] The dataset received by the i-th sonar buoy at time k is...

[0129] z ki ={(α) k(i)j ,β k(i)j ,e k(i)j |j=1,2,…,n j}

[0130] Where, n j To measure the number, α k(i)j Let β be the azimuth angle of the j-th measurement. k(i)j Let e ​​be the pitch angle measured for the j-th time. k(i)j This represents the energy information for the j-th measurement.

[0131] The set of energy values ​​received by N×M sonar buoys at time k

[0132] {e kij |i=1,2,…,N×M,j=1,2,…,n i}

[0133] The sound intensity acquisition unit is used to acquire the sonar buoy i corresponding to the maximum sound intensity at time k. *

[0134]

[0135] Wherein, the decision threshold is N s / 2,N s =N×M is the number of buoys in the sonar coverage array, and N is the buoy accumulation matrix.

[0136] Further search for the two buoys i1 with the second-highest energy values ​​in the sonar coverage array * and i2 * Then, the corresponding sonar buoys are searched using the minimum spacing method to obtain the cooperative detection buoy group (i * i1 * ) or (i * i2 * The formula for the minimum spacing is:

[0137] min{R(i1 * i * ),R(i2 * i * )}

[0138] Among them, R(i1) * i * ) is the buoy i* and buoy i1 * The spacing, R(i2) * i * ) is the buoy i * and buoy i2 * The spacing.

[0139] In the rectangular coordinate system Oxyz, a collaborative detection buoy group (i * i ** Processing time k n j Individual measurement z ki ={(α) k(i)j ,β k(i)j ,e k(i)j |j=1,2,…,n j ,i=i * i ** The directional information α in} k(i)j Pitch information β k(i)j and energy information e k(i)j Cross-positioning technology was used to obtain the three-dimensional spatial location and energy information (x) of weak underwater targets. k(i*)j ,y k(i*)j ,z k(i*)j ,e k(i*)j Based on the detection range of the collaborative detection buoy group and prior information on the movement of underwater weak targets, the time it takes for the underwater weak target to pass through the sonar buoy array is estimated, thus obtaining the information of the segment to be detected. In this embodiment, the adaptive Hough detection processing module T2 specifically includes:

[0140] The detection point acquisition unit is used to adaptively acquire the input detection information of the Hough detection processing module, project the detection segment information in three-dimensional space onto the horizontal and vertical plane information; and use the pre-detection tracking theory to sample and process the position information and energy information of the horizontal and vertical planes respectively to obtain the detection point set of the horizontal and vertical planes.

[0141] The detection segment information sampling unit is used to input detection information for the linear motion detection processing unit and the curvilinear motion detection processing unit. The dimension of the set of position points corresponding to each time step in the same segment information is N. The set of position points Z is arbitrarily selected over K consecutive time steps (K≥5). k For example, in the Cartesian coordinate system Oxyz, the set of points at time k is Z. k ={(x i*k ,y i*k ,z i*k ,e i*k (k = 1, ..., K).

[0142] Initialization and discretization processing unit, used to process the location point set Z kProject onto the horizontal plane xy to construct the data space A xy ={Z1,…,Z K} and parameter space B xy , for B xy Initialize, discretize, and construct the parameter accumulation matrix D. xy and energy accumulation matrix E xy , for D xy and E xy Initialization process.

[0143] The linear motion detection and processing unit uses a set of linear Hough detection algorithms to process A. xy and B xy Hough detection based on dual threshold accumulation of point count and energy is performed to obtain the initial target point traces that satisfy the dual thresholds of point count and energy on the horizontal plane xy.

[0144] Optionally, the dual-threshold accumulation formula for points and energy is as follows:

[0145]

[0146] Where, (θ t ,ρ m ) represents B xy The coordinates of the center point of each parameter unit are shown in the figure. "+=" indicates the cumulative operation.

[0147] Similarly, the set of location points Z k Projecting onto the vertical plane yz, constructing the data space A yz and parameter space B yz To obtain underwater target points with high fitting on the horizontal plane yz.

[0148] If the Hough detection point for the line class does not exist, construct the curve parameter space C on the horizontal plane xy. xy and C xy Initialization, discretization, and construction of the parameter accumulation matrix F xy and energy accumulation matrix G xy and for F xy and G xy Initialization processing;

[0149] The curve parameter set acquisition unit is used to obtain data in the horizontal plane xy data space A. xy Three data points are randomly selected from [Z]. i Z j Z k (i,j,k=1,…,K), obtain the curve parameter set based on the set of equations for the random Hough transform of the curve;

[0150] The curve motion detection processing unit is used for processing data space A using a set of random Hough detection algorithms for curves. xy and parameter space C xy Perform random Hough detection based on dual-threshold accumulation of point count and energy to obtain initial target points that satisfy the dual accumulation thresholds on the horizontal plane xy.

[0151] Similarly, the set of location points Z k Projecting onto the vertical plane yz, constructing the data space A yz and parameter space B yz To obtain the underwater target points on the horizontal plane yz.

[0152] In this embodiment 2, the detection point optimization processing module T3 specifically includes:

[0153] If the detection points exist in the adaptive Hough detection processing module, point optimization processing is performed. Data points at any adjacent times k, k+1, and k+2 in the underwater target point points in the horizontal plane xy are selected, and the velocity, acceleration, heading, and rate of change of heading measured by the sonar buoy in the horizontal plane xy are calculated as constraints. Based on the prior information of underwater target motion, initial points that meet the constraints are merged, and initial points that do not meet the constraints are eliminated to improve the accuracy of the target points, that is, to obtain underwater target points with high fitting degree in the horizontal plane xy. Similarly, underwater target points with high fitting degree are obtained in the vertical plane yz. The same data point sets at the same time are merged to realize the spatial fusion of the same points in the two planes xy and yz, and to complete the passive detection of weak underwater targets.

[0154] The system disclosed in the embodiments is described in a relatively simple manner because it corresponds to the method disclosed in the embodiments. For relevant details, please refer to the method section.

[0155] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An adaptive detection method for underwater maneuvering weak targets using a passive sonar buoy array, characterized in that, include: Step 1: The collaborative detection buoy group is automatically constructed using the maximum energy method and the minimum spacing method. The collaborative detection buoy group is used to locate the underwater acoustic signal and eliminate underwater noise and interference outside the collaborative detection area to obtain the information of the segment to be detected. The buoys in the passive directional sonar buoy array are passive directional sonar buoys with the function of measuring azimuth and pitch angles. Step 2: Adaptively detect the position information of the horizontal and vertical planes using the pre-detection tracking theory and the Hough detection algorithm set, and obtain the points that satisfy the double accumulation threshold from noise and interference; Step 3: By fusing dual-plane point tracks using multi-level motion constraints, a high-precision target trajectory is obtained, thus achieving passive detection of weak underwater targets. Step two specifically includes: The information of the segment to be detected in three-dimensional space is projected onto the horizontal plane information and the vertical plane information; the position and energy information of the horizontal plane and the vertical plane are sampled and processed respectively using the pre-detection tracking theory to obtain the set of points to be detected in the horizontal plane and the vertical plane. The Hough detection algorithm set is used to adaptively detect the unknown motion trajectory of weak underwater targets. First, the straight line Hough detection algorithm set is used to detect the target point set in the horizontal and vertical planes respectively. If no approximately straight line point is detected, the curve-type random Hough detection algorithm set is further used. If the detection is unsuccessful, the sampling process is re-segmented and the projection and detection process is repeated to obtain the initial detection point. Step two, the adaptive Hough detection algorithm set detection, specifically includes: In the Cartesian coordinate system middle, The set of location points at time is ,in, Set of location points Projected onto a horizontal plane Build a data space and parameter space and to Initialization and discretization; construction of the parameter accumulation matrix and energy accumulation matrix and to and Initialization processing; using a set of straight-line Hough detection algorithms. and Hough detection based on dual threshold accumulation of points and energy is performed in the horizontal plane. The initial target point trace that satisfies both the point count and energy thresholds is calculated; similarly, the location point set is... Projected onto the vertical plane Build a data space and parameter space On the horizontal plane To obtain highly fitted underwater target points; If the Hough detection points for straight lines do not exist, in the horizontal plane Construct curve parameter space and to Initialization and discretization; construction of the parameter accumulation matrix and energy accumulation matrix and to and Initialization processing; on the horizontal plane Data space Three data points were randomly selected. ,in, Based on the set of random Hough transform equations for curves, the set of curve parameters is calculated; a set of random Hough detection algorithms for curves is used to analyze the data space. and parameter space Perform random Hough detection based on dual threshold accumulation of point count and energy in the horizontal plane. The initial target point trace that satisfies the double accumulation threshold is calculated above; similarly, in the horizontal plane... To acquire underwater target points.

2. The adaptive detection method for underwater maneuvering weak targets using a passive sonar buoy array according to claim 1, characterized in that, Step one specifically includes: within the detection range of the coverage array, determining the cooperative detection buoy group by detecting the position of each sonar buoy and the acquired signal strength, based on the maximum signal strength and minimum spacing method. ,in, and These represent two buoys in a coordinated buoy detection group; in a rectangular coordinate system In the middle, a collaborative detection buoy group is adopted. deal with time Individual measurement Location information Pitch information and energy information Cross-positioning technology was used to obtain the three-dimensional spatial location and energy information of weak underwater targets. Based on the detection range of the collaborative detection buoy group and the prior information on the movement of underwater weak targets, the time it takes for the underwater weak target to pass through the sonar buoy array is estimated, and the information of the segment to be detected is obtained.

3. The adaptive detection method for underwater maneuvering weak targets using a passive sonar buoy array according to claim 2, characterized in that, Collaborative detection buoy group The calculation method is as follows: assuming the sonar buoy coverage array is composed of... OK Composed of a series of passive sonar buoys, time The set of energy values ​​received by each sonar buoy is At this point, the maximum energy value corresponds to the cooperative detection sonar buoy. for Among them: the judgment threshold is , The number of buoys in the sonar coverage array. Accumulate a matrix for the buoy; Based on the maximum energy value formula, further search was conducted on two buoys with the second-highest energy values ​​within the sonar coverage array. and Calculate the buoys separately With buoys and buoys Spacing , Then, according to the minimum spacing formula Searching for the corresponding sonar buoy is called collaborative detection buoy. .

4. The adaptive detection method for underwater maneuvering weak targets using a passive sonar buoy array according to claim 1, characterized in that, Step three specifically includes: if the detected dots exist in step two, perform dot optimization processing and select a horizontal plane. Any adjacent points in the underwater target trace , , The data points at each time point are used to calculate the measurements of the sonar buoy on the horizontal plane. The velocity, acceleration, heading, and rate of change of heading are used as constraints. Based on prior information about the underwater target's motion, initial points that meet the constraints are merged, while those that do not are eliminated, thus improving the accuracy of the target points. This is done on the horizontal plane. High-fit underwater target points can be obtained on the vertical plane; similarly, on the vertical plane... High-fit underwater target points are obtained; data points with the same time point are merged to achieve a dual-plane configuration. and Spatial fusion of identical points enables passive detection of weak underwater targets.

Images