Multi-target positioning method and system based on magnetic measurement

Abstract

The invention provides a multi-target positioning method and system based on magnetic measurement. A magnetic gradient tensor system collects magnetic gradient tensor elements above targets and normalized magnetic source intensity NSS generated by multiple targets. A coordinate system is established, a solution space is set, the solution space is subdivided, each grid point in the solution space has a magnetic dipole, the normalized magnetic source intensity obtained in the acquisition step is vectorized, and a linear equation set between the normalized magnetic source intensity and a grid magnetic dipole magnetic moment module is established by combining data of a plurality of acquisition points. The magnetic moment module value of the magnetic dipole corresponding to the grid point where the target is located is not 0, the magnetic moment module values of other magnetic dipoles are equal to 0, the size and the position of the non-0 magnetic moment in the solution space are estimated through an L1 norm minimization method, the number of the non-0 magnetic moment is equal to the number of the target, and the module value of the non-0 magnetic moment corresponds to the magnetic moment vector module of the target. The multi-target positioning method does not need to estimate the number of targets in advance.

Description

Technical Field

[0001] The present invention belongs to the technical field of target detection, and particularly relates to a multi-target positioning method and system based on magnetic measurement. Background Art

[0002] Magnetic measurement is an important non-destructive target detection method and has important applications in fields such as underground pipelines, underground / underwater metal targets, and resource exploration. Affected by the induced magnetization of the geomagnetic field, ferromagnetic targets will generate an induced magnetic field, resulting in the distortion of the low magnetic field near the targets. By collecting the magnetic field above the targets, information such as the position and attitude of the targets can be inverted to achieve target detection and recognition.

[0003] Existing magnetic target positioning methods generally assume a single target. When there are multiple targets with adjacent positions, the magnetic anomaly fields generated by them are aliased. Based on traditional magnetic target positioning methods, there are problems: (1) The number of targets cannot be determined; (2) The inverted target position is different from the true position, and the positioning error is large. For the problem of multi-target detection, generally, the number of targets is required as prior information, and then the three-dimensional positions of all targets are estimated. Summary of the Invention

[0004] The present invention provides a multi-target positioning method and system based on magnetic measurement to solve the problems raised in the above background art.

[0005] To solve the above technical problems, the first aspect of the present invention discloses a multi-target positioning method based on magnetic measurement, including the following steps:

[0006] a. Setting and meshing step: Establish a coordinate system to set the solution space, and mesh the solution space;

[0007] b. Acquisition step: Collect the magnetic gradient tensor elements above the targets and the normalized magnetic source strength NSS generated by multiple targets;

[0008] c. Data processing step: Based on the solution space, construct a point magnetic dipole model, input the collected magnetic gradient tensor elements and magnetic source strength NSS as input data into the point magnetic dipole model, and output the magnitude and position of the non-zero magnetic moment to achieve target positioning.

[0009] Furthermore, the point magnetic dipole model includes the solution of multi-magnetic dipole parameters based on the minimization of the L1 norm:

[0010]

[0011] where, ‖·‖1 represents the L1 norm, is the vector of the magnetic moment modulus parameters of the magnetic dipole.

[0012] Further, the solution algorithms for L1 norm minimization include the basis pursuit algorithm, interior point method, simplex method, or orthogonal matching pursuit algorithm, and the primal-dual interior point method.

[0013] Further, in the data processing step, the solution space is partitioned, and it is required that the distance between the magnetic field sensor and the target is greater than 2.5 times the maximum grid size.

[0014] Further, in the setting and partitioning step, the solution space is a cuboid.

[0015] Further, in the acquisition step, an observation horizontal plane is selected above the target, and the magnetic gradient tensor elements B xx 、B xy 、B xz 、B yx 、B yz are uniformly acquired at the observation plane using the magnetic gradient tensor system.

[0016] Further, the data processing step includes vectorizing the normalized magnetic source intensity collected on the observation plane, combining the data of multiple collection points on the observation plane, and establishing a linear equation system:

[0017]

[0018] s = Ψτ

[0019] where Ψ is a P×Q matrix, and r ij (i = 1, 2, …, P, j = 1, 2, …, Q) is the distance between the i-th observation point on the observation plane and the j-th magnetic dipole in the solution space. By solving the linear equation system, the magnetic moment modulus distribution τ of the magnetic dipoles in the solution space can be determined;

[0020] Preferably, the total number of sampling points is P = lenx * leny, where lenx is the number of sampling points in the x direction and leny is the number of sampling points in the y direction.

[0021] The second aspect of the present invention discloses a multi-target positioning system based on magnetic measurement, including:

[0022] a. An acquisition module for acquiring the magnetic gradient tensor elements above the target and the normalized magnetic source intensity NSS generated by multiple targets;

[0023] b. A data processing module for determining the normalized magnetic source intensity NSS generated by multiple targets according to the acquired magnetic gradient tensor field. When the distance between the magnetic field sensor and the target is greater than 2.5 times the maximum size of the known target, it is simplified into a point-like magnetic dipole model. When the number of non-zero magnetic dipoles n in the solution space, the position of the i-th magnetic dipole is (x i , y i,z i ), with the magnetic moment modulus being m i , the total normalized magnetic source intensity generated by n magnetic dipoles is And establish a coordinate system, set up the solution space for meshing, vectorize the normalized magnetic source intensity collected on the observation plane s = [NSS1, NSS2,..., NSS P T , establish a linear equation system between the normalized magnetic source intensity and the magnetic moment modulus of the grid magnetic dipoles:

[0024]

[0025] s = Ψτ

[0026] where Ψ is a P×Q matrix, r ij (i = 1, 2,..., P, j = 1, 2,..., Q) is the distance between the i-th observation point on the observation plane and the j-th magnetic dipole in the solution space. By solving the linear equation system, determine the magnetic moment modulus distribution τ of the magnetic dipoles in the solution space, and use the L1 norm minimization method to solve the magnitude and position of the non-zero magnetic moments in the solution space.

[0027] Furthermore, the acquisition module includes a sensor array for acquiring five magnetic gradient tensor elements B xx , B xy , B xz , B yx , B yz and the normalized magnetic source intensity NSS generated by multiple targets. The five magnetic gradient tensor elements satisfy the formula where B ij (i, j = x, y, z) is the magnetic gradient tensor element generated by the dipole, B i is the i component of the magnetic induction intensity, r j is the j component of the distance vector between the observation point where the magnetic field sensor is located and the target position. The normalized magnetic source intensity NSS generated by multiple targets satisfies the formula where n is the number of non-zero magnetic dipoles in the solution space, (x i , y i , z i ) is the position of the i-th magnetic dipole, and the magnetic moment modulus is m i .

[0028] Furthermore, the system can achieve multi-target positioning without pre-estimating the number of targets. The system also includes a user interface module for displaying the position and magnitude of the non-zero magnetic moments, as well as the number of targets;

[0029] Preferably, the system also includes a calibration module for calibrating the sensors of the acquisition module to improve the measurement accuracy; ​

[0030] Preferably, the system further includes a data storage module for storing information such as the collected magnetic gradient tensor field data, the collected normalized magnetic source strength NSS, and the positions and magnitudes of the non-zero magnetic moments obtained by solving.

[0031] Compared with the prior art, the present invention provides a multi-target positioning method based on magnetic measurement, having the following beneficial effects:

[0032] (1) Innovatively propose a multi-target positioning method based on magnetic measurement. The magnetic gradient tensor system collects the magnetic gradient tensor elements above the target to determine the normalized magnetic source strength NSS generated by the target. A coordinate system is established to set up the solution space, and the solution space is divided. There is a magnetic dipole at each grid point in the solution space. Therefore, the normalized magnetic source strength obtained in the collection step is vectorized, and a linear equation system between the normalized magnetic source strength and the magnetic moment modulus of the grid magnetic dipole is established by combining the data of multiple collection points;

[0033] (2) Based on the fact that the distribution of non-zero magnetic dipoles in the solution space is sparse, only the magnetic moment modulus value of the magnetic dipole corresponding to the grid point where the target is located is non-zero, and the magnetic moment modulus values of other magnetic dipoles are equal to 0. The L1 norm minimization method is used to estimate the magnitudes and positions of non-zero magnetic moments in the solution space. The number of non-zero magnetic moments is equal to the number of targets, and the modulus value of the non-zero magnetic moment corresponds to the magnetic moment vector modulus of the target. This multi-target positioning method does not require pre-estimation of the number of targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present invention will be further described below with reference to the drawings and embodiments.

[0035] Figure 1 is the magnetic anomaly field generated by the position adjacent to the target being aliased;

[0036] Figure 2 is the position distribution of multiple targets in the solution space;

[0037] Figure 3 is the observed plane magnetic gradient tensor field (unit: nT / m);

[0038] Figure 4 is the observed plane normalized magnetic source strength (unit: nT / m);

[0039] Figure 5 is the solution τ of the linear equation system. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] In an exemplary embodiment of the present invention,

[0041] First, the magnetic anomaly field generated by ferromagnetic targets is introduced. When the distance between the magnetic field sensor and the target is greater than 2.5 times the maximum size of the target, the target can be simplified into a point magnetic dipole model. The magnetic dipole model is characterized by a three-dimensional magnetic moment vector and a three-dimensional position vector. The magnetic anomaly field generated by the target is:

[0042]

[0043] where μ0 = 4π×10 -7 H / m is the magnetic permeability of air, is the three-dimensional magnetic moment vector, is the distance vector from the observation point where the magnetic field sensor is located to the target position, and r is the modulus of the distance vector. Thus, the magnetic gradient tensor field generated by the target can be determined In a rectangular coordinate system, G is a second-order tensor:

[0044]

[0045] The magnetic gradient tensor matrix G is a symmetric matrix and satisfies that the trace of the matrix is 0. Therefore, only 5 of the magnetic gradient tensor elements are independent, including two diagonal elements and three non-diagonal elements. The magnetic gradient tensor element B ij (i, j = x, y, z) is calculated as follows:

[0046]

[0047] where only when i and j are the same, δ ij = 1, and in other cases δ ij = 0. r i is the i-component of the distance vector , and m i is the i-component of the magnetic moment vector . The normalized source strength NSS (Normalized Source Strength) μ of the target is a scalar independent of the direction of the target magnetic moment vector, and it is only related to the modulus m of the magnetic moment vector and the distance r. At the same time, the normalized source strength μ of the magnetic dipole can be determined by the three eigenvalues λ1, λ2, and λ3 of the magnetic gradient tensor matrix G. The calculation formula is as follows:

[0048]

[0049] Therefore, by collecting the magnetic gradient tensor field above the target, determining the magnetic gradient tensor matrix G, and calculating its eigenvalues, the normalized source strength NSS of the target can be determined.

[0050] Secondly, the multi-target positioning method is introduced. As Figure 1As shown in the figure, there are two ferromagnetic targets 1 and 2 with close positions in the underground space. Under the action of geomagnetic induction magnetization, the targets will generate magnetic anomaly fields. Curve 1 corresponds to the magnetic anomaly field generated when only target 1 exists, curve 2 corresponds to the magnetic anomaly field generated when only target 2 exists, and the uppermost curve is the real magnetic anomaly field collected by a magnetic field sensor along a horizontal side line, which is the result of the superposition of curve 1 and curve 2. It can be seen that the magnetic anomaly fields generated by two adjacent targets are aliased, and the number of targets cannot be judged from the measured magnetic anomaly field curve.

[0051] A rectangular coordinate system 0-xyz is established, with the ground as the xoy plane. Thus, the space where the target is located is the space where z < 0. A rectangular solution space is set to ensure that all ferromagnetic targets are in the solution space, as Figure 2 shown. The solution space is evenly divided, and the number of grid divisions is Q = nx × ny × nz. It is assumed that there is a magnetic dipole at each grid point in the solution space. The magnetic moment of the magnetic dipole at the grid point without a target is 0, and the modulus of the magnetic moment of the magnetic dipole corresponding to the grid point where the target is located is non-zero. When the number of non-zero magnetic dipoles in the solution space is n, the position of the i-th magnetic dipole is (x i , y i , z i ), and the modulus of the magnetic moment is m i . The total normalized magnetic source intensity generated by n magnetic dipoles is:

[0052]

[0053] An observation horizontal plane is selected above the target, and the magnetic gradient tensor elements B xx , B xy , B xz , B yx , B yz at the observation plane are uniformly collected by a magnetic gradient tensor system, and a magnetic gradient tensor matrix is established. The total number of sampling points on the observation plane is P = lenx * leny, where lenx is the number of sampling points along the x direction and leny is the number of sampling points along the y direction. According to formula (4), the total normalized magnetic source intensity NSS on the observation plane can be collected, and the normalized magnetic source intensity collected on the observation plane is vectorized as s = [NSS1, NSS2,..., NSS P T . Combining the data of multiple collection points on the observation plane, a linear equation system is established:

[0054]

[0055] s = Ψτ

[0056] Among them, Ψ is a P×Q matrix, r ij ​(i = 1, 2, …, P, j = 1, 2, …, Q) is the distance between the i-th observation point on the observation plane and the j-th magnetic dipole in the solution space. By solving the linear equations, the distribution τ of the magnetic moment modulus of the magnetic dipoles in the solution space can be determined.

[0057] Finally, a method for solving the parameters of multiple magnetic dipoles based on the minimization of the L1 norm is presented. The solution τ of the linear equations corresponds to the magnetic moment modulus of Q magnetic dipoles in the solution space. Only the magnetic moment of the magnetic dipole at the position of the target is non-zero, and the magnetic moments of the magnetic dipoles at the remaining grid positions are equal to 0. Therefore, the system of equations (6) is a sparse solution problem, and the method based on the minimization of the L1 norm is used to determine τ.

[0058]

[0059] where, ‖·‖1 represents the L1 norm, is the parameter vector of the magnetic moment modulus of the magnetic dipole. There are various solution algorithms for the minimization of the L1 norm, such as the basis pursuit algorithm, the interior point method, the simplex method, the orthogonal matching pursuit algorithm, etc. Here, the primal-dual interior point method is used to solve τ. The number of targets can be determined according to the number of non-zero solutions, and the three-dimensional positions of all targets can be determined by the positions of the non-zero solutions.

[0060] The positioning results of the above embodiments are detected by simulation analysis: The simulation experiment settings are as follows: The geomagnetic field intensity is 55000 nT, the geomagnetic inclination is 57.58°, and the geomagnetic declination is -5.14°. Two ferromagnetic targets are buried in the underground space, and the position of target 1 is (9, 10, -2) m, and the magnetic dipole moment is (0, 4, 0) A·m 2 , and the position of target 2 is (9.5, 10, -2) m, and the magnetic dipole moment is (1, 2, -5) A·m 2 .

[0061] The magnetic gradient tensor elements above the targets are collected. The observation plane is a horizontal plane of 20 m × 20 m, with a height of 1 m from the ground. The sampling interval in the x and y directions is 1 m, and the number of sampling points is P = 441. The distribution of the collected magnetic gradient elements is as Figure 3 shown. The normalized magnetic source strength NSS on the observation plane is as Figure 4 shown. It can be seen from the figure that the magnetic anomaly fields generated by the two ferromagnetic targets are aliased, and the number of targets cannot be directly judged.

[0062] Based on the multi-target positioning method proposed in this patent, a solution space of 20 m × 20 m × 3 m is set. The solution space is evenly divided, and the length of the divided grid is 0.5 m, with a total of 11767 grid nodes. According to the distribution of the normalized magnetic source strength NSS on the observation plane, the primal-dual interior point method is used to solve the distribution result τ of the magnetic moment modulus values of the magnetic dipoles in the underground space, as Figure 5As shown. Among them, there are two non-zero elements, indicating that the number of targets in the underground space is 2. According to the positions of the non-zero elements, the three-dimensional positions of the targets are obtained, and the results are (9, 10, -2) m and (9.5, 10, -2) m. Therefore, in the case of no noise, the multi-target localization result proposed in this patent is consistent with the theoretical localization result of the targets.

[0063] In an exemplary embodiment of the present invention, this embodiment introduces a multi-target localization system based on magnetic measurement, which can efficiently and accurately achieve the localization of multiple targets. The system mainly includes an acquisition module, a data processing module, a user interface module, a calibration module, and a data storage module. The functions of each module and their applications in the system will be described in detail below.

[0064] The module includes:

[0065] a. Acquisition module. The acquisition module is the core input part of the system, used to acquire the magnetic gradient tensor elements above the target and the normalized magnetic source strength NSS generated by multiple targets; it is used to acquire five magnetic gradient tensor elements B xx 、B xy 、B xz 、B yx 、B yz and a sensor array for the normalized magnetic source strength NSS generated by multiple targets. The five magnetic gradient tensor elements satisfy the formula where B ij (i,j = x,y,z) is the magnetic gradient tensor element generated by the dipole, B i is the i component of the magnetic induction intensity, r j is the j component of the distance vector between the observation point where the magnetic field sensor is located and the target position. The normalized magnetic source strength NSS generated by multiple targets satisfies the formula where n is the number of non-zero magnetic dipoles in the solution space, (x i ,y i ,z i ) is the position of the i-th magnetic dipole, and the magnetic moment modulus is m i . In actual operation, the sensor array is arranged on the observation plane, and the required magnetic gradient tensor elements and NSS values are obtained through precise measurement, providing basic data for subsequent data processing.

[0066] b. Data processing module. The data processing module is the core processing part of the system, used to determine the normalized magnetic source strength NSS generated by multiple targets according to the acquired magnetic gradient tensor field. When the distance between the magnetic field sensor and the target is greater than 2.5 times the known maximum size of the target, it is simplified into a point-like magnetic dipole model. When solving for the number n of non-zero magnetic dipoles in the space, the position of the i-th magnetic dipole is (x i ,yi , z i ), with the magnetic moment modulus being m i , and the total normalized magnetic source intensity generated by n magnetic dipoles is And establish a coordinate system, set up the solution space for meshing, and vectorize the normalized magnetic source intensity collected on the observation plane s = [NSS1, NSS2,..., NSS P T , and establish a linear equation system between the normalized magnetic source intensity and the magnetic moment modulus of the grid magnetic dipoles:

[0067]

[0068] s = Ψτ

[0069] where Ψ is a P×Q matrix, r ij (i = 1, 2,..., P, j = 1, 2,..., Q) is the distance between the i-th observation point on the observation plane and the j-th magnetic dipole in the solution space. By solving the linear equation system, determine the magnetic moment modulus distribution τ of the magnetic dipoles in the solution space, and use the L1 norm minimization method to solve the magnitude and position of the non-zero magnetic moments in the solution space.

[0070] c. User interface module, used to display the position and magnitude of the non-zero magnetic moments, as well as the number of targets. Through an intuitive graphical interface, users can conveniently view the positioning results and perform subsequent operations as needed.

[0071] d. Calibration module, used to calibrate the sensors of the acquisition module to improve the measurement accuracy; in practical applications, the sensors may have deviations due to environmental factors or long-term use. Through the calibration module, the accuracy of the sensors can be ensured, thereby improving the positioning accuracy of the entire system.

[0072] e. Data storage module, used to store information such as the collected magnetic gradient tensor field data, the collected normalized magnetic source intensity NSS, and the position and magnitude of the solved non-zero magnetic moments. Through the data storage module, users can conveniently query historical data and perform further analysis and processing.

[0073] In the system application example, assume that in a certain underground detection task, it is necessary to determine the positions and magnitudes of multiple unknown metal objects. At this time, the multi-target positioning system based on magnetic measurement described in this embodiment can be adopted.

[0074] First, arrange a sensor array on the observation plane through the acquisition module and collect the required magnetic gradient tensor elements and NSS values. Then, use the data processing module to process the collected data and solve the positions and magnetic moment moduli of the non-zero magnetic dipoles. Finally, display the positioning results through the user interface module and perform subsequent operations as needed.​

[0075] In practical applications, the system has shown good stability and accuracy, and successfully achieved multi-target positioning. At the same time, with the support of the calibration module and the data storage module, the measurement accuracy and data processing ability of the system have been further improved.

[0076] In summary, the multi-target positioning system based on magnetic measurement described in this embodiment has broad application prospects and important practical significance.

[0077] Further, when a multi-target positioning system based on magnetic measurement calibrates the sensors of the acquisition module through a calibration module, it includes: obtaining the output information of the sensors, and performing cross-verification based on the output information to determine whether the sensors need to be calibrated, so as to obtain a preliminary calibration analysis result; performing preliminary self-calibration according to the preliminary calibration analysis result. When the preliminary calibration analysis result indicates that the sensors need to be calibrated, software self-calibration is performed for the sensors respectively, and then the output information is re-obtained based on the sensors after software self-calibration, and cross-verification is performed according to the output information to determine whether the sensors need to be calibrated, so as to obtain a secondary calibration analysis result; performing secondary self-calibration according to the secondary calibration analysis result. When the secondary calibration analysis result indicates that the sensors do not need to be calibrated, the sensor calibration is completed. When the secondary calibration analysis result indicates that the sensors need to be calibrated, hardware self-calibration is performed for the sensors respectively, and then the output information is re-obtained based on the sensors after hardware self-calibration, and cross-verification is performed according to the output information to determine whether the sensors need to be calibrated, so as to obtain a re-calibration analysis result; performing an anomaly reminder according to the re-calibration analysis result. When the re-calibration analysis result indicates that the sensors do not need to be calibrated, the sensor calibration is completed. When the re-calibration analysis result indicates that the sensors need to be calibrated, an anomaly reminder is given for the sensors. The above technical solution realizes the calibration of the sensors through the calibration module, reduces the error of the sensor-induced acquisition data, and improves the accuracy of the magnetic gradient tensor elements above the target and the normalized magnetic source strength NSS generated by multiple targets collected by the acquisition module, so that the data processing module can analyze based on the more accurate acquisition data of the magnetic gradient tensor elements above the target and the normalized magnetic source strength NSS generated by multiple targets, providing guarantee for the analysis and processing of the data processing module, effectively improving the accuracy of the multi-target positioning system based on magnetic measurement, and ensuring the accuracy of the magnitude and position of the non-0 magnetic moment in space. By gradually performing software self-calibration and hardware self-calibration for calibration, the sensors can complete the calibration based on self-calibration without attracting relevant personnel for manual processing, reducing unnecessary troubles, and at the same time, it can also reduce the time occupied by sensor calibration, improving the efficiency of the calibration module for calibrating the sensors of the acquisition module. Moreover, the calibration effect can be clarified by obtaining the re-calibration analysis result, ensuring that the calibrated sensors have high accuracy, and in case of necessity, relevant personnel are assisted to calibrate through anomaly reminder, ensuring the calibration effect of the calibration module for the sensors of the acquisition module.

[0078] When the data storage module stores the collected magnetic gradient tensor field data, the collected normalized magnetic source strength NSS, and the position and magnitude information of the non-zero magnetic moment obtained by solving, validity checks are respectively performed on the collected magnetic gradient tensor field data, the collected normalized magnetic source strength NSS, and the position and magnitude information of the non-zero magnetic moment obtained by solving, and validity check results are obtained. When the validity check results pass, encryption processing is respectively performed on the collected magnetic gradient tensor field data, the collected normalized magnetic source strength NSS, the position and magnitude information of the non-zero magnetic moment obtained by solving, etc., to obtain the first encrypted information, the second encrypted information, and the third encrypted information. Then, data attribute matching is performed on the first encrypted information, the second encrypted information, and the third encrypted information to obtain a data matching result, and data information storage is performed according to the data matching result. By performing the validity check as described above, it is avoided that the data storage module stores invalid data information when storing the collected magnetic gradient tensor field data, the collected normalized magnetic source strength NSS, and the position and magnitude information of the non-zero magnetic moment obtained by solving, reducing the space occupied by invalid information in the storage module. Moreover, through encryption processing, the security of the data storage module is provided, preventing the information in the data storage module from being easily damaged. At the same time, by performing data attribute matching on the first encrypted information, the second encrypted information, and the third encrypted information, the orderliness of the stored information is ensured, avoiding data chaos that may cause the data storage module to malfunction. As a result, the data storage module can better store data information, providing convenience for information retrieval from the data storage module.

[0079] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. If these modifications and variations of the present invention fall within the scope of the claims of the present invention and its equivalent technologies, the present invention also intends to include these changes and modifications.

Claims

1. A multi-target positioning method based on magnetic measurement, comprising the following steps: a. Setting and dissection step: Establish a coordinate system to set the solution space, and dissect the solution space; b. Acquisition step: Acquire the magnetic gradient tensor elements above the target and the normalized magnetic source strength NSS generated by multiple targets; c. Data processing step: Construct a point magnetic dipole model based on the solution space, input the acquired magnetic gradient tensor elements and magnetic source strength NSS as input data into the point magnetic dipole model, and output the magnitude and position of the non-zero magnetic moment to achieve target positioning.

2. The multi-target positioning method based on magnetic measurement according to claim 1, characterized in that The point magnetic dipole model includes the solution of multi-magnetic dipole parameters based on the minimization of the L1 norm: where, ||·||1 represents the L1 norm, is the magnetic moment modulus parameter vector of the magnetic dipole.

3. The multi-target positioning method based on magnetic measurement according to claim 2, wherein The solution algorithms for the minimization of the L1 norm include the basis pursuit algorithm, the interior point method, the simplex method, or the orthogonal matching pursuit algorithm and the primal-dual interior point method.

4. The multi-target positioning method based on magnetic measurement according to claim 1, characterized in that In the data processing step, for the dissection of the solution space, it is required to control the distance between the magnetic field sensor and the target to be greater than 2.5 times the maximum grid size.

5. The multi-target positioning method based on magnetic measurement according to claim 1, characterized in that, In the setting and dissection step, the solution space is a cuboid.

6. The multi-target positioning method based on magnetic measurement according to claim 1, wherein, In the acquisition step, an observation horizontal plane is selected above the target, and the magnetic gradient tensor elements B xx , B xy , B xa , B yx , B yz are uniformly acquired at the observation plane by using a magnetic gradient tensor system.

7. The multi-target positioning method based on magnetic measurement according to claim 1, wherein The data processing step includes vectorizing the normalized magnetic source strength collected on the observation plane, combining the data of multiple collection points on the observation plane, and establishing a linear equation system: s = Ψτ where Ψ is a P×Q matrix, and r ij (i = 1, 2, …, P, j = 1, 2, …, Q) is the distance between the i-th observation point on the observation plane and the j-th magnetic dipole in the solution space. By solving the linear equations, the distribution of the magnetic moment modulus τ of the magnetic dipoles in the solution space can be determined. The total number of sampling points is P = lenx * leny, where lenx is the number of sampling points in the x direction and leny is the number of sampling points in the y direction.

8. A multi-target positioning system based on magnetic measurement, characterized in that including: a. An acquisition module for acquiring the magnetic gradient tensor elements above the target and the normalized magnetic source strength NSS generated by multiple targets; b. Data processing module, which is used to determine the normalized magnetic source strength NSS generated by multiple targets according to the collected magnetic gradient tensor field. When the distance between the magnetic field sensor and the target is greater than 2.5 times the known maximum size of the target, it is simplified into a point-like magnetic dipole model. When solving for the number n of non-zero magnetic dipoles in the space, the position of the i-th magnetic dipole is (x i , y i , z i ), the magnitude of the magnetic moment is m i , and the total normalized magnetic source strength generated by n magnetic dipoles is And establish a coordinate system, set up the solution space for meshing, vectorize the normalized magnetic source strength collected on the observation plane s = [NSS1, NSS2,..., NSS P T , and establish a linear equation system between the normalized magnetic source strength and the magnitude of the magnetic moment of the grid magnetic dipole:​ s = Ψτ where Ψ is a P×Q matrix, r ij (i = 1, 2, …, P, j = 1, 2, …, Q) is the distance between the i-th observation point on the observation plane and the j-th magnetic dipole in the solution space. By solving the linear equations, the distribution τ of the magnetic moment modulus of the magnetic dipoles in the solution space is determined, and the magnitude and position of the non-zero magnetic moments in the solution space are solved using the L1 norm minimization method.

9. The system according to claim 8, wherein The acquisition module includes a sensor array for acquiring five magnetic gradient tensor elements B xx , B xy , B xz , B yx , B yz and the normalized magnetic source strength NSS generated by multiple targets. The five magnetic gradient tensor elements satisfy the formula where B ij (i, j = x, y, z) is the magnetic gradient tensor element generated by the dipole, B i is the i-component of the magnetic induction intensity, r j is the j-component of the distance vector between the observation point where the magnetic field sensor is located and the target position. The normalized magnetic source strength NSS generated by multiple targets satisfies the formula where n is the number of non-zero magnetic dipoles in the solution space, (x i , y i , z i ) is the position of the i-th magnetic dipole, and the magnetic moment modulus is m i .

10. The system according to claim 8 or 9, characterized in that The system can achieve multi-target positioning without pre-estimating the number of targets. The system further includes a user interface module for displaying the position and magnitude of the non-zero magnetic moment, as well as the number of targets; the system further includes a calibration module for calibrating the sensors of the acquisition module to improve the measurement accuracy; The system further includes a data storage module for storing information such as the collected magnetic gradient tensor field data, the acquired normalized magnetic source strength NSS, the position and magnitude of the solved non-zero magnetic moment, etc.

Images