A passive sonar multi-target trajectory reconstruction method and device

Abstract

The present application provides a method and device for reconstructing multi-target trajectories of a passive sonar, belonging to the technical field related to multi-target tracking. Sonar signal data of the passive sonar is acquired, and the acquired sonar signal data is filtered; a heuristic search algorithm is used for associating multi-target data to obtain sonar data of different targets after association; multi-target trajectory reconstruction is performed based on the sonar data of multiple different targets after association; by introducing an improved heuristic search algorithm, the jump of sonar data caused by target crossing can be avoided, ensuring the continuity and accuracy of the target trajectory and avoiding the failure of trajectory reconstruction caused by target jump; by utilizing the global optimization characteristics of the heuristic search algorithm, the problem of multi-target data association can be efficiently solved. By establishing a cost function of the target trajectory, the heuristic search algorithm can accurately associate the sonar data of different targets, thereby realizing the accurate reconstruction of multi-target trajectories.

Description

Technical Field

[0001] This application relates to the technical field of multi-target tracking, and particularly to a method and device for reconstructing multi-target trajectories of passive sonar. Background Art

[0002] Passive sonar technology is widely used in underwater target detection and tracking, and particularly plays an important role in military and marine scientific research. Passive sonar detects and locates targets by receiving acoustic wave signals emitted by the targets. The main purpose of its data filtering technology is to extract useful target signals from a complex underwater acoustic environment, suppress noise and interference, thereby improving the accuracy of target detection, location, and tracking.

[0003] However, in practical applications, due to the complexity of the underwater environment and the weakness of target signals, target signals are usually masked by background noise (such as ocean ambient noise, hydrophone self-noise) and interference signals (such as noise from other underwater devices, multi-path effects, etc.), resulting in discontinuous jumps in passive sonar signals, which leads to failure or large errors in target trajectory reconstruction.

[0004] To solve the above technical problems, the existing technology adopts a passive sonar target tracking method based on Kalman filtering. The passive sonar target tracking method based on Kalman filtering predicts and updates the target position by establishing a target motion model and using Kalman filtering. However, when dealing with target jumps, due to the limitations of model assumptions, this method cannot effectively handle sudden changes in target signals, resulting in failure of trajectory reconstruction. Summary of the Invention

[0005] The embodiments of this application provide a method and device for reconstructing multi-target trajectories of passive sonar, aiming to solve the problem of trajectory reconstruction failure caused by the jump of passive sonar detected targets due to target jumps when an underwater unmanned vehicle uses passive sonar to track marine targets, and use a heuristic search algorithm to realize the association of multi-target data and achieve the reconstruction of multi-target trajectories of passive sonar.

[0006] In a first aspect, this application provides a method for reconstructing multi-target trajectories of passive sonar, which is used for multi-target tracking and trajectory reconstruction of passive sonar. The method for reconstructing multi-target trajectories of passive sonar includes:

[0007] Obtain sonar signal data of passive sonar, and perform filtering processing on the obtained sonar signal data;

[0008] Perform association processing on multi-target data by using a heuristic search algorithm to obtain sonar data of different targets after association;

[0009] Reconstruct multi-target trajectories based on the sonar data of multiple different targets after association.

[0010] Optionally, the heuristic search algorithm is used to perform association processing on multi-target data to obtain sonar data of different targets after association, specifically including:

[0011] Set the search range of the heuristic search algorithm according to the number of targets detected for the first time and the number of detections;

[0012] Use the heuristic search algorithm to search for the motion trajectory of a single target. Take the first detection value of the i-th target as the current node, and calculate the cost value between all nodes at time k + 1 and the current node according to the target sonar measurement value of the current node and the sonar measurement angle of the i-th target at time k + 1;

[0013] Calculate the cost value for all targets at time k + 1 and update it to the cost function. Then, take the node with the minimum cost value as the current node for iteration, and repeat the update process of the cost function until the last node.

[0014] Optionally, the calculation of the cost value between all nodes at time t2 and the current node according to the target sonar measurement value of the current node and the sonar measurement angle of the i-th target at time k + 1 includes:

[0015] According to the target sonar measurement value of the current node and the sonar measurement angle of the i-th target at time k + 1, use the formula Calculate the cost value between all nodes at time k + 1 and the current node , where represents the sonar measurement angle of the i-th target at time k + 1, represents the target sonar measurement value of the current node.

[0016] Optionally, the calculation of the cost value for all targets at time k + 1 and the update to the cost function, then taking the node with the minimum cost value as the current node for iteration and repeating the update process of the cost function until the last node includes:

[0017] According to the formula Calculate the cost value for all targets at time k + 1 and update it to the cost function, where is the cost value of the node corresponding to the i-th target at the current moment, is the cost value of the node corresponding to the i-th target at the next moment, is the cost value of the i-th target at time k + 1;

[0018] According to the formula Determine the node with the minimum cost value, then take the node with the minimum cost value as the current node for iteration, and repeat the update process of the cost function until the last node.

[0019] Optionally, the multi-target trajectory reconstruction based on the sonar data of multiple different targets after association includes:

[0020] Taking the set of paths with the minimum cost as the optimal path for this search, determining the sonar data corresponding to the optimal path as the target sonar data, and performing multi-target trajectory reconstruction based on the sonar data of multiple different targets.

[0021] Optionally, the multi-target trajectory reconstruction based on the sonar data of multiple different targets includes:

[0022] Performing multi-target trajectory reconstruction on the sonar data of multiple different targets using the interactive multi-model algorithm, introducing multiple models to characterize the working state of the target during the multi-target trajectory reconstruction, and using a weighted fusion mechanism for system state estimation.

[0023] Optionally, the multi-target trajectory reconstruction based on the sonar data of multiple different targets includes:

[0024] Setting the initial transition matrix P and the credibility coefficient , at the k−1 moment, each model has an optimal estimate and the estimated covariance , weighting and fusing the model estimation states and covariance matrices related to each model to obtain the optimal state and covariance matrix of the fused target at the k−1 moment, where the models used include the uniform motion model and the circular motion model;

[0025] Respectively performing filtering updates on the above-described uniform motion model and circular motion model, and obtaining the estimated target state and covariance matrix of each model after update. When each model has obtained the updated estimated target state, evaluate the goodness of the model's estimation of the current state according to the estimated target state;

[0026] According to the calculated likelihood value of each model at the k moment, the accuracy of the description of each model can be obtained, thereby realizing the update of the model credibility;

[0027] According to the updated confidence successively fuse the models to obtain the output target state and covariance matrix of the IMM filtering algorithm;

[0028] According to the output target state and covariance matrix of the IMM filtering algorithm, perform multi-target trajectory reconstruction on the sonar data of multiple different targets using the trajectory reconstruction system model.

[0029] Optionally, the multi-target trajectory reconstruction using the trajectory reconstruction system model based on the sonar data of multiple different targets includes:

[0030] Based on sonar data for multiple different targets, according to the motion equation and the observation equation perform multi-target cabinet reconstruction, where is the system state vector at time k, is the system state vector at time k−1, is the process noise at time k−1, is the observation vector at time k, is the observation equation operator at time k, is the measurement error at time k.

[0031] Optionally, the models adopted include a uniform motion model and a circular motion model, where the calculation formula of the uniform motion model is: and the calculation formula of the circular motion model is: , represents the two-dimensional coordinate position of the target in the NEG coordinate system, is its heading in the geodetic coordinate system, is the eastward velocity and northward velocity of the target in the geodetic coordinate system, is the angular velocity of the target, , , , respectively represent the process noise of the position and velocity at time k−1, and T is the navigation data sampling time interval.

[0032] In a second aspect, the present application also provides a passive sonar multi-target trajectory reconstruction device, and the passive sonar multi-target trajectory reconstruction device is configured to perform the above-mentioned passive sonar multi-target trajectory reconstruction method.

[0033] Advantageous technical effects of the present application:

[0034] A passive sonar multi-target trajectory reconstruction method and device provided by the present application acquire sonar signal data of a passive sonar and perform filtering processing on the acquired sonar signal data; use a heuristic search algorithm to perform association processing on multi-target data to obtain sonar data of different targets after association; perform multi-target trajectory reconstruction based on the sonar data of multiple different targets after association; by introducing an improved heuristic search algorithm, it can avoid the sonar data jump caused by target crossing, ensure the continuity and accuracy of the target trajectory, and avoid the failure of trajectory reconstruction caused by target jump; by utilizing the global optimization characteristics of the heuristic search algorithm, it can efficiently solve the multi-target data association problem, and by establishing a cost function of the target trajectory, the heuristic search algorithm can accurately associate the sonar data of different targets, thereby realizing the accurate reconstruction of the multi-target trajectory. Description of the Drawings

[0035] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0036] Figure 1 It is a schematic flowchart of a method for reconstructing multi-target trajectories of a passive sonar provided by the present application;

[0037] Figure 2 It is a schematic diagram of searching for an optimal path provided by the present application;

[0038] Figure 3 It is a block diagram of the structure of an electronic device provided by the present application. Detailed implementation manners

[0039] In order to make the objectives, technical solutions, and advantages of the present application clearer, the following will further describe the present application in detail with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0040] It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the various processes do not mean the order of execution, and the execution order of the various processes should be determined according to their functions and internal logics, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0041] It should be understood that in the present application, "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0042] The following will specifically describe the technical solutions of the present application with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0043] Refer to Figure 1 and Figure 2As shown in the figure, an embodiment of the present invention provides a method for reconstructing the trajectories of multiple passive sonar targets, which is used for tracking and reconstructing multiple passive sonar targets. The aim is to solve the problem of trajectory reconstruction failure caused by the detection of target jumps by passive sonar when an underwater unmanned vehicle uses passive sonar to track sea targets. The method uses a heuristic search algorithm to associate multi-target data and realizes the reconstruction of the trajectories of multiple passive sonar targets.

[0044] Existing passive sonar target tracking and trajectory reconstruction technologies mainly rely on traditional filtering methods such as Kalman filtering and particle filtering. To a certain extent, these methods can handle noise interference, but their performance significantly degrades in the case of target signal jumps. Specifically, these methods of the existing technology rely on the assumptions of the continuity and smoothness of target motion. When the target signal undergoes a sudden change (such as a rapid turn of the target or signal loss), the prediction and update mechanisms of the filter cannot effectively respond, resulting in trajectory reconstruction failure or a significant increase in errors. A passive sonar system usually can only obtain the azimuth information of the target. In a multi-target scenario, due to the crossing and overlapping of target signals, it is difficult for traditional methods to accurately associate the sonar data of different targets, leading to difficulties in multi-target trajectory prediction and reconstruction.

[0045] Aiming at the above-mentioned shortcomings of the existing technology, the purpose of the present invention is to provide a method for reconstructing the trajectories of multiple passive sonar targets. By introducing an improved heuristic search algorithm, it can avoid the sonar data jumps caused by target crossing, ensure the continuity and accuracy of the target trajectory, and avoid trajectory reconstruction failure caused by target jumps. Utilizing the global optimization characteristics of the heuristic search algorithm, it can efficiently solve the problem of multi-target data association. By establishing a cost function for the target trajectory, the heuristic search algorithm can accurately associate the sonar data of different targets, thereby realizing the precise reconstruction of the multi-target trajectory.

[0046] Reference Figure 1 As shown in the figure, an embodiment of the present invention provides a method for reconstructing the trajectories of multiple passive sonar targets, which is used for tracking and reconstructing multiple passive sonar targets. The method for reconstructing the trajectories of multiple passive sonar targets provided by the embodiment of the present invention includes the following steps:

[0047] Step 100: Obtain the sonar signal data of the passive sonar and perform filtering processing on the obtained sonar signal data.

[0048] Specifically, an AUV underwater vehicle is equipped with a passive sonar as an observation station to observe moving sea targets. After starting the passive sonar, the passive sonar data is read through the serial port. Filtering processing is performed on the sonar signal data of the passive sonar. A threshold is set to remove the self-noise and environmental noise of the passive sonar, and the azimuth values of all filtered targets are sorted according to the signal strength and stored.

[0049] Step 200: Use a heuristic search algorithm to perform association processing on multi-target data to obtain sonar data of different targets after association.

[0050] When there are multiple targets in the environment, simply outputting the azimuth of the target with the strongest signal may mask the signal of the real target due to the movement of the AUV underwater vehicle, resulting in target crossing. Therefore, it is necessary to set the search range of the heuristic search algorithm according to the number of targets detected for the first time and the number of detections; use the heuristic search algorithm to search for the movement trajectory of a single target, take the first detection value of the i-th target as the current node, and calculate the cost value between all nodes at time k + 1 and the current node according to the target sonar measurement value of the current node and the sonar measurement angle of the i-th target at time k + 1; calculate the cost value for all targets at time k + 1 and update it to the cost function, and then take the node with the smallest cost value as the current node for iteration, repeating the update process of the cost function until the last node, and then complete the association processing of multi-target data based on the target detection information to obtain the sonar data of different targets after association.

[0051] Further, according to the target sonar measurement value of the current node and the sonar measurement angle of the i-th target at time k + 1, use the formula to calculate the cost value between all nodes at time k + 1 and the current node , where represents the sonar measurement angle of the i-th target at time k + 1, represents the target sonar measurement value of the current node. According to the formula calculate the cost value for all targets at time k + 1 and update it to the cost function, where is the cost value of the node corresponding to the i-th target at the current moment, is the cost value of the node corresponding to the i-th target at the next moment, is the cost value of the i-th target at time k + 1; according to the formula determine the node with the smallest cost value, and then take the node with the smallest cost value as the current node for iteration, repeating the update process of the cost function until the last node.

[0052] Refer to Figure 2 As shown, first set the search range of the heuristic search algorithm. The number of targets detected for the first time is the number of trajectory reconstructions (because the targets detected each time are sorted according to signal strength and are not associated with the detections in the previous frame. Subsequently, data association will be performed on the detection sequences of each target). Assume that there are N targets detected for the first time and a total of M detections are performed. Then the search range of the heuristic search algorithm is N×M.

[0053] After that, calculate the cost function. Use the heuristic search algorithm to search for the motion trajectory of a single target. Take the first detection value of the i-th target as the initial node, that is, the first target at time k0. Let k = k0 and calculate the cost values of all nodes at time k + 1 and the current node according to the following formula.

[0054]

[0055] Among them, represents the sonar measurement angle of the i-th target at time k + 1, represents the sonar measurement value of the target at the current node.

[0056] Calculate the cost values for all targets at time k + 1 and update them into the cost function. The calculation formula of the cost function is as follows:

[0057]

[0058] Among them, is the cost value of the node corresponding to the i-th target at the current moment, is the cost value of the node corresponding to the i-th target at the next moment, is the cost value of the i-th target at time k + 1;

[0059] According to the formula Determine the node with the minimum cost value. After that, use the node with the minimum cost value as the current node for iteration, and repeat the update process of the cost function until the last node to achieve the selection of the optimal path.

[0060] Step 300: Perform multi-target trajectory reconstruction based on the sonar data of multiple different targets after association.

[0061] In the embodiment of the present invention, a set of paths with the minimum cost is used as the optimal path for this search, and the sonar data corresponding to the optimal path is determined as the target sonar data. Multi-target trajectory reconstruction is performed based on the sonar data of multiple different targets. The interactive multiple model algorithm is used to perform multi-target trajectory reconstruction based on the sonar data of multiple different targets. During the multi-target trajectory reconstruction process, multiple models are introduced to characterize the working state of the target, and a weighted fusion mechanism is used for system state estimation.

[0062] A passive sonar multi-target trajectory reconstruction method provided by the embodiment of the present invention. Among them, the process of performing multi-target trajectory reconstruction based on the sonar data of multiple different targets is as follows:

[0063] Set the initial transition matrix P and the credibility coefficient , at time k - 1, each model has an optimal estimate and the estimation covariance , the model estimation status and covariance matrix related to each model are weighted and fused to obtain the optimal status and covariance matrix of the target at the k−1 moment. The models adopted include a uniform motion model and a circular motion model. The calculation formula of the uniform motion model is: , the calculation formula of the circular motion model is: , represents the two-dimensional coordinate position of the target in the NEG coordinate system, is its heading in the geodetic coordinate system, are the eastward and northward velocities of the target in the geodetic coordinate system, is the angular velocity of the target, , , , respectively represent the process noise of the position and velocity at the k−1 moment, and T is the navigation data sampling time interval.

[0064] The uniform motion model and the circular motion model described above are each filtered and updated to obtain the estimated target status and covariance matrix after the update of each model. When each model has obtained the updated estimated target status, the goodness of the model's estimation of the current status is evaluated according to the estimated target status;

[0065] According to the likelihood value of each model calculated at the k moment, the accuracy of the description of each model can be obtained, thereby realizing the update of the model credibility;

[0066] According to the updated confidence the models are fused in sequence to obtain the output target status and covariance matrix of the IMM filtering algorithm;

[0067] According to the output target status and covariance matrix of the IMM filtering algorithm, a multi-target trajectory reconstruction is performed using a trajectory reconstruction system model based on the sonar data of multiple different targets. Based on the sonar data of multiple different targets, according to the motion equation and the observation equation a multi-target cabinet reconstruction is performed, where is the system state vector at the k moment, is the system state vector at the k−1 moment, is the process noise at the k−1 moment, is the observation vector at the k moment, is the observation equation operator at the k moment, is the measurement error at the k moment.

[0068] The passive sonar multi-target trajectory reconstruction method provided by the embodiments of the present invention takes the set of paths with the minimum cost as the optimal path for this search, that is, the path of the target motion this time, and performs trajectory reconstruction according to this set of angle values. The trajectory reconstruction is implemented using the IMM-UKF algorithm. The Interacting Multiple Model (IMM) algorithm, as a soft switching strategy, introduces two or more models to characterize the possible working states of the target and uses an efficient weighted fusion mechanism for system state estimation. The interacting multiple model algorithm effectively alleviates the problem of target tracking failure when the target makes a turn.

[0069] Specifically, the system state vector (state value) is a multi-dimensional vector describing the target state, and its state value at time k can be expressed as:

[0070]

[0071] Among them, represents the two-dimensional coordinate position of the target in the NEG coordinate system, is its heading in the geodetic coordinate system, are the eastward and northward velocities of the target in the geodetic coordinate system, is the angular velocity of the target. Since this system is a discrete-time system, let T be the navigation data sampling time interval, then the state transition model (motion equation) of the target is:

[0072]

[0073] Among them, is the process noise at time k−1, f(⋅) represents the state transition operator, is the state value at time k−1. The process noise follows a zero-mean Gaussian distribution, that is, w∼N(0,W), and W is the process noise covariance matrix, which is expressed as: .

[0074] The observation value at time K can be described as: , where represents the angular relationship between the target and the AUV, which can be obtained from the AUV heading and passive sonar observations. The relationship between the observation value and the state value (observation equation) is shown in the following formula: , where is the observation noise, h(⋅) is the observation equation operator, and the observation equation operator is expressed as: , where and represent the position coordinates of the AUV respectively. The target motion trajectory is reconstructed according to the above motion equation and observation equation.

[0075] The models adopted in the embodiments of the present invention include a uniform motion model and a circular motion model. Among them, assuming that the target has r motion states, we can assume that it has r state transition equations. Construct the target state vector , and the adopted uniform motion model and circular motion model respectively represent the linear motion and turning motion models of the target.

[0076] The calculation formula of the uniform motion model is:

[0077]

[0078] The calculation formula of the circular motion model is:

[0079]

[0080] Among them, represents the two-dimensional coordinate position of the target in the NEG coordinate system, is its heading in the geodetic coordinate system, is the eastward speed and northward speed of the target in the geodetic coordinate system, is the angular velocity of the target, , , , respectively represent the process noise of the position and speed at the k−1 moment, and T is the navigation data sampling time interval.

[0081] The process of multi-target trajectory reconstruction using the IMM algorithm based on the sonar data of multiple different targets and the trajectory reconstruction system model in the embodiments of the present invention is as follows:

[0082] Set the initial transfer matrix P and the credibility coefficient , and the conversion between models is defined as follows through the Markov probability transfer matrix:

[0083]

[0084]

[0085] Among them, the matrix element represents the probability that the target motion model transfers from the i-th state to the j-th state.

[0086] At the k-1 moment, each model has an optimal estimate and the estimation covariance . The estimated states and covariance matrices of the models related to each model are weighted and fused to obtain the optimal state and covariance matrix of the fused target at the k-1 moment:

[0087]

[0088]

[0089] Among them, is the correlation coefficient of fusion, and its calculation formula is .

[0090] Perform filtering updates on the uniform motion model and the circular motion model described above respectively, and obtain the estimated target state and covariance matrix .

[0091] When each model estimates the state of the target, the model will be evaluated for its goodness of estimating the current state according to the estimated target state, and the calculation is as follows:

[0092]

[0093] Among them, and are the measurement error and measurement error covariance matrix, and the calculation formula is as follows:

[0094]

[0095]

[0096] According to the likelihood value of each model at time k calculated, the accuracy of each model's description can be obtained, thereby realizing the update of the model credibility :

[0097]

[0098] Among them, is the credibility after fusion, and c is the denominator coefficient for normalizing , and the calculation is as follows:

[0099]

[0100]

[0101] According to the updated confidence the models can be fused in sequence to obtain the output target state and covariance matrix of the IMM filtering algorithm:

[0102]

[0103]

[0104] Continue to search for the next target trajectory and perform reconstruction until the state estimation for all time instants is completed, until all multi-target trajectory reconstructions of sonar data based on multiple different targets are completed according to the target state and covariance matrix output by the IMM filtering algorithm.

[0105] A passive sonar multi-target trajectory reconstruction method and device provided by this application obtain sonar signal data of a passive sonar and perform filtering processing on the obtained sonar signal data; perform multi-target data association processing using a heuristic search algorithm to obtain sonar data of different targets after association; perform multi-target trajectory reconstruction based on the sonar data of multiple different targets after association; by introducing an improved heuristic search algorithm, it can avoid the sonar data jump caused by target crossing, ensure the continuity and accuracy of the target trajectory, and avoid the failure of trajectory reconstruction caused by target jump; by utilizing the global optimization characteristics of the heuristic search algorithm, it can efficiently solve the multi-target data association problem. By establishing a cost function for the target trajectory, the heuristic search algorithm can accurately associate the sonar data of different targets, thereby achieving the precise reconstruction of the multi-target trajectory.

[0106] Based on the same inventive concept, an embodiment of this application also provides a passive sonar multi-target trajectory reconstruction device that can execute the processing flow provided by an embodiment of a passive sonar multi-target trajectory reconstruction method.

[0107] A passive sonar multi-target trajectory reconstruction device provided by an embodiment of this application can be used to execute a passive sonar multi-target trajectory reconstruction method in any of the above embodiments. Its implementation principle and technical effects are similar and will not be elaborated here. The device provided by the embodiment of this application can specifically be used to execute the solution provided by the corresponding method embodiment. The specific functions and achievable technical effects will not be elaborated here.

[0108] It should be noted that it should be understood that the division of each module of the above device is only a logical function division. In actual implementation, it can be fully or partially integrated into a physical entity, or physically separated. And these modules can all be implemented in the form of software called by a processing element; they can also all be implemented in the form of hardware; or some modules can be implemented in the form of software called by a processing element, and some modules can be implemented in the form of hardware. In addition, all or part of these modules can be integrated together or independently implemented. The processing element mentioned here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed by the hardware integrated logic circuit or software-form instructions in the processor element.

[0109] Figure 3 It is a schematic structural diagram of an electronic device provided by an embodiment of this application. As Figure 3As shown, the electronic device may include: a processor 21, a memory 22, and computer program instructions stored on the memory 22 and executable on the processor 21. When the processor 21 executes the computer program instructions, a passive sonar multi-target trajectory reconstruction method provided in any of the foregoing embodiments is implemented.

[0110] Optionally, the various components of the electronic device may be connected via a system bus.

[0111] The memory 22 may be a separate storage unit or an integrated storage unit in the processor. The number of processors is one or more.

[0112] Optionally, the electronic device may further include a communication interface for interacting with other devices.

[0113] It should be understood that the processor 21 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the present application may be directly embodied as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0114] The system bus may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The system bus may be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity, only a thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus. The memory may include a random access memory (RAM), and may also include a non-volatile memory (NVM), such as at least one disk memory.

[0115] All or part of the steps of the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a readable memory. When the program is executed, it performs the steps including the above method embodiments; and the foregoing memory (storage medium) includes: read-only memory (ROM), RAM, flash memory, hard disk, solid state drive, magnetic tape, floppy disk, optical disc, and any combination thereof.

[0116] The electronic device provided in the embodiments of the present application can be used to execute a passive sonar multi-target trajectory reconstruction method provided in any of the above method embodiments, and its implementation principle and technical effects are similar, which will not be elaborated here.

[0117] The embodiments of the present application provide a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions run on a computer, the computer is enabled to execute the above passive sonar multi-target trajectory reconstruction method.

[0118] For the above computer-readable storage medium, the above-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic memory, flash memory, magnetic disk or optical disc. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.

[0119] Optionally, the readable storage medium is coupled to the processor, so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in the device.

[0120] The embodiments of the present application also provide a computer program product, which includes a computer program. The computer program is stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium, and when the at least one processor executes the computer program, the above passive sonar multi-target trajectory reconstruction method can be implemented.

[0121] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A passive sonar multi-target trajectory reconstruction method for passive sonar multi-target tracking and trajectory reconstruction, characterized in that: The passive sonar multi-target trajectory reconstruction method comprises: Acquire sonar signal data of passive sonar, and perform filtering processing on the acquired sonar signal data; A heuristic search algorithm is used to associate multi-target data to obtain sonar data of different targets after association. Reconstruct multi-target trajectories based on the sonar data of multiple different targets after association; The heuristic search algorithm is used to perform association processing on the multi-target data to obtain the sonar data of different targets after association, specifically including: The search range of the heuristic search algorithm is set according to the number of targets detected for the first time and the number of detections; A heuristic search algorithm is used to search the motion trajectory of a single target. The first detection value of the i-th target is taken as the current node. The cost value of all nodes and the current node at time k+1 is calculated based on the target sonar measurement value of the current node and the sonar measurement angle of the i-th target at time k+1. Calculate the cost value for all targets at time k+1 and update it into the cost function. Then, take the node with the smallest cost value as the current node for iteration and repeat the cost function update process until the last node. The cost values ​​of all nodes and the current node at time t2 are calculated based on the target sonar measurement value of the current node and the sonar measurement angle of the i-th target at time k+1, including: According to the target sonar measurement value of the current node and the sonar measurement angle of the i-th target at time k+1, the formula C(k+1) is used i =|θ(k+1) i -θ(current)|Calculate the cost value C(k+1) of all nodes and the current node at time k+1 i , where θ(k+1) i represents the sonar measurement angle of the i-th target at time k+1, and θ(current) represents the target sonar measurement value of the current node.

2. The passive sonar multi-target trajectory reconstruction method according to claim 1, characterized in that: The cost value is calculated for all targets at time k+1 and updated into the cost function, and then the node with the smallest cost value is used as the current node for iteration, and the cost function update process is repeated until the last node, including: According to the formula f(k+1) i =f(k)+C(k+1) i Calculate the cost value of all targets at time k+1 and update it to the cost function, where f(k) is the cost value of the node corresponding to the i-th target at the current time, and f(k+1) i is the cost value of the node corresponding to the i-th target at the next moment, C(k+1) i is the cost of the i-th target at time k+1; According to the formula f(k+1)=min(f(k+1) i ) Determine the node with the smallest cost value, then iterate the node with the smallest cost value as the current node, and repeat the cost function update process until the last node.

3. The passive sonar multi-target trajectory reconstruction method according to claim 2 is characterized in that: The multi-target trajectory reconstruction based on the sonar data of the multiple different targets after association includes: The set of paths with the lowest cost is taken as the optimal path for this search, the sonar data corresponding to the optimal path is determined as the target sonar data, and multi-target trajectories are reconstructed based on the sonar data of multiple different targets.

4. The passive sonar multi-target trajectory reconstruction method according to claim 3 is characterized in that: The multi-target trajectory reconstruction based on the sonar data of multiple different targets includes: An interactive multi-model algorithm is used to reconstruct multi-target trajectories based on the sonar data of multiple different targets. Multiple models are introduced in the process of multi-target trajectory reconstruction to characterize the working status of the target, and a weighted fusion mechanism is used to estimate the system state.

5. The passive sonar multi-target trajectory reconstruction method according to claim 4 is characterized in that: The multi-target trajectory reconstruction based on the sonar data of multiple different targets includes: Set the initial transfer matrix P and the credibility coefficient μ0. At time k-1, each model has an optimal estimate and the estimated covariance The model estimation state and covariance matrix related to each model are weighted fused to obtain the optimal state and covariance matrix of the fused target at time k-1, wherein the models used include uniform motion model and circular motion model; Perform filtering updates on the uniform motion model and circular motion model described above, respectively, and calculate the estimated target state after each model is updated. and the covariance matrix When each model obtains an updated estimated target state, the model's estimation of the current state is evaluated based on the estimated target state; According to the calculated likelihood value of each model at time k, the accuracy of the description of each model can be obtained, thereby realizing the update of the model credibility; According to the updated confidence μ, the models are fused in sequence to obtain the output target state and covariance matrix of the IMM filtering algorithm; According to the output target state and covariance matrix of the IMM filtering algorithm, the trajectory reconstruction system model is used to reconstruct multi-target trajectories based on the sonar data of multiple different targets.

6. The passive sonar multi-target trajectory reconstruction method according to claim 5, characterized in that: The method of reconstructing multiple target trajectories using a trajectory reconstruction system model based on sonar data of multiple different targets includes: Based on the sonar data of multiple different targets, according to the motion equation X k =f(X k-1 ,w k-1 ) and the observation equation Z k =h(X k )+r k Perform multi-target cabinet reconstruction, where X k is the system state vector at time k, X k-1 is the system state vector at time k-1, w k-1 is the process noise at time k-1, Z k is the observation vector at time k, h(X k ) is the observation equation operator at time k, r k is the measurement error at time k.

7. The passive sonar multi-target trajectory reconstruction method according to claim 6, characterized in that: The models used include a uniform motion model and a circular motion model, wherein the calculation formula of the uniform motion model is: The calculation formula of the circular motion model is: (x k ,y k ) represents the two-dimensional coordinate position of the target at time k in the NEG coordinate system, (x k-1 ,y k-1 ) represents the two-dimensional coordinate position of the target at time k-1 in the NEG coordinate system, is its heading at time k in the geodetic coordinate system, is its heading at time k-1 in the geodetic coordinate system, (v x,k ,v y,k ) are the eastward and northward velocities of the target at time k in the geodetic coordinate system, (v x,k-1 ,v y,k-1 ) are the eastward and northward velocities of the target at time k-1 in the geodetic coordinate system, ω k is the angular velocity of the target at time k, ω k-1 is the angular velocity of the target at time k-1, w x,k-1 、w y,k-1 , They represent the process noise of position and velocity at time k-1 respectively, T is the navigation data sampling time interval, Represents the process noise of the heading angle at time k-1.

8. A passive sonar multi-target trajectory reconstruction device, characterized in that: The passive sonar multi-target trajectory reconstruction device is configured to execute the passive sonar multi-target trajectory reconstruction method according to any one of claims 1 to 7.

Images