Aeromagnetic compensation method and system for large unmanned aerial vehicle

Abstract

The invention discloses an aeromagnetic compensation method and system for a large unmanned aerial vehicle, and relates to the technical field of aeromagnetic exploration compensation, and the method comprises the steps: obtaining a compensation matrix through an inertial navigation system, a GPS and a three-axis fluxgate magnetometer of the large unmanned aerial vehicle based on a T-L model; and based on the compensation matrix, performing aeromagnetic compensation on the large unmanned aerial vehicle by increasing the maneuvering motion range of the large unmanned aerial vehicle and utilizing a compensation optimization algorithm. The large multi-load unmanned aerial vehicle is effectively compensated, and the actual use value of the large unmanned aerial vehicle is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of aeromagnetic compensation, and in particular to an aeromagnetic compensation method and system for large unmanned aerial vehicles. Background Art

[0002] In recent years, drones have increasingly been used as carriers for aeromagnetic surveys. With the increasing demand for aerial surveys, the need for larger payloads has led to the use of larger drones, making them more suitable for magnetic exploration missions. Airborne magnetic exploration is a physical method that uses an aircraft-mounted magnetometer to measure the strength or gradient of the Earth's magnetic field. The primary purpose of airborne magnetic exploration is to study underground geological structures and explore mineral resources. Airborne magnetometers and their associated auxiliary equipment are mounted on an aircraft and flown over the survey area at a predetermined altitude along a predetermined survey line to collect geomagnetic field data. This data is then analyzed to map the geomagnetic field distribution and, in turn, to infer the underground rock structure and mineral deposit locations. During airborne magnetic surveys, changes in the aircraft's flight direction and motion generate magnetic fields from the aircraft's magnetic components. These fields can interfere with the magnetometer, affecting measurement accuracy. To ensure accurate measurements of the Earth's magnetic field, aeromagnetic compensation is required to offset this interference. Compensation plays a crucial role in flight exploration; the process of eliminating these interfering magnetic fields is known as aeromagnetic compensation.

[0003] In the actual flight compensation process, many problems will be encountered. First of all, the accuracy of the three-axis fluxgate magnetometer is poor, and its anti-noise interference ability is poor. Therefore, interference will inevitably occur during the aircraft's turning process, affecting the accuracy of the three-axis fluxgate magnetometer. The increase in error will further lead to low accuracy in solving the compensation coefficient. Secondly, the compensation matrix obtained during the compensation process will have the problem of multicollinearity. Large maneuvers will reduce the problem of multicollinearity to a certain extent, while large maneuvers during the compensation process will increase the interference magnetic field.

[0004] In the existing technology, small drones are usually used to reduce interfering magnetic fields. However, the drone platforms usually used for flight are often large in size because they need to carry many instruments and equipment. When combined with the above-mentioned large maneuvering action plan, the interfering magnetic field that needs to be compensated is even larger. Summary of the Invention

[0005] In order to solve the above problems, the present invention provides an aeromagnetic compensation technology for large UAVs, aiming to compensate for the large interference noise of large UAVs and achieve better compensation effect.

[0006] To achieve the above technical objectives, the present application provides an aeromagnetic compensation method for large UAVs, comprising the following steps:

[0007] Based on the TL model, the compensation matrix is ​​obtained through the inertial navigation system, GPS and three-axis fluxgate magnetometer of the large UAV;

[0008] Based on the compensation matrix, the aeromagnetic compensation of large UAV is performed by increasing the maneuvering amplitude of the large UAV and using the compensation optimization algorithm.

[0009] Preferably, when obtaining the compensation matrix, the compensation matrix is ​​constructed using Euler angles based on the TL model through an inertial navigation system, GPS and a three-axis fluxgate magnetometer.

[0010] Preferably, when increasing the amplitude of the maneuver, a large-angle compensation flight strategy is adopted to reduce the multicollinearity problem in the compensation matrix.

[0011] Preferably, when using the compensation optimization algorithm, the compensation optimization algorithm is constructed by using the Lasso regression algorithm and the Newton iteration algorithm.

[0012] Preferably, when performing aeromagnetic compensation on a large UAV, 36 compensation coefficients are obtained based on the compensation matrix and the magnetic field data obtained by the optically pumped magnetometer of the large UAV are solved;

[0013] According to the compensation coefficient and compensation matrix, the interference magnetic field is obtained;

[0014] The magnetic field data is used to subtract the interfering magnetic field to obtain the compensated magnetic field data.

[0015] Preferably, when performing aeromagnetic compensation, the standard deviation is used to evaluate the data before and after compensation.

[0016] The present invention also discloses an aeromagnetic compensation system for large UAVs, which is used to implement the above-mentioned aeromagnetic compensation method for large UAVs, including:

[0017] A compensation matrix building module is used to obtain the compensation matrix based on the TL model through the inertial navigation system, GPS and three-axis fluxgate magnetometer of the large UAV;

[0018] The aeromagnetic compensation module is used to perform aeromagnetic compensation on large UAVs based on the compensation matrix, by increasing the maneuvering amplitude of the large UAV and using the compensation optimization algorithm.

[0019] Preferably, the compensation matrix construction module is further configured to construct the compensation matrix using Euler angles.

[0020] Preferably, the aeromagnetic compensation module is further used to adopt a large-angle compensation flight strategy to reduce the multicollinearity problem in the compensation matrix; and to construct a compensation optimization algorithm through the Lasso regression algorithm and the Newton iteration algorithm.

[0021] Preferably, the aeromagnetic compensation module is further configured to solve, based on the compensation matrix, the magnetic field data acquired by the optically pumped magnetometer of the large unmanned aerial vehicle to obtain 36 compensation coefficients;

[0022] The interference magnetic field is obtained according to the compensation coefficient and the compensation matrix; the interference magnetic field is subtracted from the magnetic field data to obtain the compensated magnetic field data, wherein the data before and after compensation are evaluated using the standard deviation.

[0023] The present invention discloses the following technical effects:

[0024] The present invention effectively compensates for large multi-payload UAVs and improves the practical use value of large UAVs. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0026] Figure 1 It is the geographical and aircraft coordinate diagram of the present invention;

[0027] Figure 2 is a flow chart of aeromagnetic compensation according to the present invention;

[0028] Figure 3 This is a flow chart of the Lasso regularized Newton iterative algorithm described in the present invention;

[0029] Figure 4 The flight circle of the present invention includes a compensation flight circle A and a verification flight circle B;

[0030] Figure 5 The present invention is to realize self-compensation of flight circle A by LRNM;

[0031] Figure 6 The present invention provides a method for compensating flight circle A by using LRNM to compensate for verification flight circle B.

[0032] Figure 7 is the compensation result of the three methods described in the present invention for verifying flight circle B;

[0033] Figure 8 It is the level flight C trajectory of the present invention;

[0034] Figure 9 This is the compensation result of the level flight segment C using three methods described in the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application.

[0036] like Figures 1-9 As shown, the present invention provides an aeromagnetic compensation technology for large unmanned aerial vehicles. According to the traditional TL model, the shortcomings of large noise and large error of the three-axis fluxgate magnetometer are improved. The inertial navigation system, GPS and three-axis fluxgate magnetometer are selected to jointly obtain the compensation matrix, and the amplitude of the compensation maneuver is increased, which will reduce the multicollinearity problem in the compensation matrix to a certain extent. In combination with the shortcomings of the large interference field of large unmanned aerial vehicles, a new compensation optimization algorithm, Lasso Regression Newton Iteration Method (LRNM) is proposed. By comparing with the traditional solution methods of least squares method (LS) and singular value decomposition method (SVD), the present invention achieves better compensation effect. Combined with the above improved model and large maneuver compensation scheme, it can compensate for the large interference noise of large unmanned aerial vehicles. The aeromagnetic compensation technology provided by the present invention specifically includes the following contents:

[0037] When using the traditional TL model for aeromagnetic compensation, the required cosine values ​​are obtained with the help of a three-axis fluxgate vector magnetometer. However, due to its own reasons, this instrument has poor accuracy and low noise immunity. During the flight, the aircraft will inevitably perform maneuvers, and these maneuvers will introduce a lot of noise. This noise is difficult to be filtered by simple filters. If the three-axis fluxgate vector magnetometer is used alone to obtain the cosine value of the compensation matrix, the noise will also be introduced, resulting in inaccurate calculation of the compensation coefficient by the equation. Therefore, the best method is to avoid using the three-axis fluxgate vector magnetometer as the only instrument to obtain data to obtain the cosine value. The present invention provides a method that uses an inertial navigation system and GPS to assist the three-axis fluxgate vector magnetometer to obtain the required cosine value in the compensation matrix by obtaining the attitude angle.

[0038] Definitions of geographic coordinate system and aircraft coordinate system:

[0039] like Figure 1As shown, the geographic coordinate system is based on an abstract model of the Earth—an ellipsoid. The axis around which this ellipsoid rotates is called the Earth's axis, with the North Pole at its north end and the South Pole at its south end. The aircraft can be considered a point close to the Earth's surface, with its X-axis pointing north, its Y-axis pointing east, and its Z-axis perpendicular to the ground, pointing toward the Earth's center.

[0040] The aircraft coordinate system is centered at O, with the X-axis parallel to the fuselage and pointing toward the nose. The Y-axis is perpendicular to the fuselage and points toward the wings. The Z-axis is perpendicular to the belly and points toward the ground, following the left-hand rule. As the aircraft maneuvers in mid-air, the coordinate system will create a certain angle with the original assumed coordinate system, resulting in attitude angles. The roll angle, denoted by θ, is defined as the aircraft's coordinates maneuvering around the X-axis. The pitch angle, denoted by Ψ, is defined as the aircraft's coordinates maneuvering around the Y-axis. The yaw angle, denoted by Φ, is defined as the aircraft's coordinates maneuvering around the Z-axis. These three angles represent the flight attitude and can be used to convert geographic coordinates to aircraft coordinates.

[0041] The transformation matrix between geographic coordinates and aircraft coordinates is as follows:

[0042] The coordinate change matrix of the horizontal scrolling action:

[0043] Coordinate change matrix of pitching action:

[0044] The coordinate change matrix of the yaw action:

[0045] Then the matrix obtained by multiplying these three equations in the order of roll, pitch, and yaw is the complete coordinate change matrix:

[0046]

[0047] After the coordinates are changed, the overall equation of the three-axis magnetic components can be written as:

[0048]

[0049] Among them, H X , H Y , H Z is the component of the geomagnetic field on the three axes in the aircraft coordinate system, H x , H y , H zIt is the component of the geomagnetic field on three axes in the geographic coordinate system. It can be seen from this equation that as long as the components of the initial geomagnetic field in the geographic coordinate system and the real-time attitude change angle obtained by the inertial navigation system and GPS during the movement of the aircraft in the air are obtained, the present invention can obtain the three-axis magnetic components in the real-time aircraft coordinate system by this equation. The components of the initial geomagnetic field in the geographic coordinate system can be obtained by querying the International Geomagnetic Reference Field (IGRF). IGRF is a very important geomagnetic query tool. In this model, you only need to input longitude, latitude, and altitude to obtain the total geomagnetic field intensity F, horizontal intensity H, vertical intensity Z, X and Y are the north and east components of H, and the magnetic declination D and magnetic inclination I. These data can be used to calculate H x , H y , H z The value of .

[0050] Through the above coordinate change method, a method can be obtained to obtain the three-axis magnetic component without the need for a three-axis fluxgate sensor. Combined with the ground-to-flying coordinate transformation, the present invention can obtain a new TL equation:

[0051]

[0052] H g =J×F

[0053] Where H g is the aircraft interference field of the new TL model, and J represents a vector consisting of Among them, j i represents the three direction cosines, It represents the derivative of the three direction cosines with respect to time, and F represents the unknown coefficients to be solved, which are 18, F = [f1, f2, f3 ... f18]; here the present invention uses e i Represents the direction cosines cosα, cosβ, and cosγ in the equation. Represents the time derivatives of the direction cosines cosα, cosβ, and cosγ in the equation, x i represents the coefficient of the fixed field, v ij represents the coefficient of the induction field, b ij Represents the coefficient of the eddy current field. The direction cosine vectors e1, e2, and e3 are obtained by the following formula:

[0054]

[0055] Here, H x , H y , H z It can be obtained by querying IGRF.

[0056] The modified TL model recommended by the present invention combines the traditional TL model with the TL equation processed after coordinate transformation, which can be written as:

[0057] H=H d +H g

[0058] H d =H p +H i +H r

[0059] Where H d The aircraft interference field in the traditional TL model consists of fixed magnetic field, induced magnetic field and eddy current field. p is the aircraft's fixed magnetic field, H i is the induced magnetic field, H r It is the eddy current field;

[0060] The entire aeromagnetic compensation flow chart is as follows: Figure 2 As shown:

[0061] (1) Data obtained by optically pumped magnetometer, three-axis fluxgate vector magnetometer, inertial navigation system, and GPS during flight;

[0062] (2) All these data are filtered to remove the influence of geomagnetic gradient;

[0063] (3) The data obtained by the three-axis fluxgate, inertial navigation system, and GPS can be calculated to obtain a matrix composed of new cosine vectors;

[0064] (4) By solving equations with the data obtained from the optical pump, 36 compensation coefficients are obtained;

[0065] (5) Obtain new data through new flight circles;

[0066] (6) Repeat steps (2) and (3) for these data;

[0067] (7) Multiplying the obtained compensation coefficient with the new compensation matrix to obtain the interference magnetic field;

[0068] (8) Subtract the interfering magnetic field to obtain compensated magnetic field data.

[0069] In aeromagnetic compensation, an evaluation standard is required for the data before and after compensation. The present invention uses the standard deviation (STD) to evaluate the data before and after compensation. The ratio of the standard deviation before and after compensation is defined as the improvement ratio (IR). The size of the improvement ratio is used to evaluate the effect of the entire compensation process. The specific formula is: Among them, H b , Ha Represent the data before and after compensation respectively.

[0070] Optimization algorithm:

[0071] The present invention proposes a fusion algorithm. This method combines the optimization algorithms of Lasso regression and Newton iteration. The optimization algorithm obtained by integrating the advantages of the two is called Lasso Regularized Newton Iteration Method (LRNM) in the present invention. The amount of data obtained during the aeromagnetic compensation process is very large, and multicollinearity problems are inevitable, resulting in model instability, which is the so-called "ill-conditioned equation". Lasso regression reduces the model coefficients by adding a regularization term (L1 norm penalty term) to the loss function, and even reduces some coefficients to zero. Doing so not only helps prevent overfitting, but also enables feature selection, thereby simplifying the model and improving its generalization ability.

[0072] In addition, the regularization property of Lasso regression enables it to handle complex data sets and maintain stability and interpretability even when there is a high correlation between independent variables. Lasso regression objective function:

[0073] min||y-Xβ||^2+λ||β||1

[0074] Among them, ||y-Xβ||^2 is the residual sum of squares, which measures the error between the predicted value and the true value, ||β||1 is the sum of the absolute values ​​of all elements in the β vector, which is the L1 regularization term, and λ is the regularization parameter that controls the penalty strength.

[0075] Lasso regression is insensitive to noise and outliers because it uses L1 regularization. The basic idea of ​​the Newton iteration method is to use the gradient information and second-order derivative of the current iteration point to make a quadratic approximation of the target function, and then take the minimum point of this quadratic function as the new iteration point. This process is repeated until the minimum point of the function is found. Mathematically, this can be achieved by solving the first-order derivative (gradient) of the function and setting it to zero, thereby finding the stationary point.

[0076] The method recommended by this invention is to combine the two and solve an overdetermined matrix of the original equation. The main steps are as follows: Figure 3 As shown; in the figure, X is the input matrix, y is the target vector, and λ is the regularization parameter.

[0077] The Newton iteration method boasts strong data processing capabilities, fast quadratic convergence, and robust noise immunity. Lasso regularization prevents the optimization algorithm from falling into local minima. The combined advantages of these two methods make the algorithm more stable and better able to handle ill-conditioned equations. Compared to traditional methods such as SVD and LS, this method represents significant improvement. The following comparison of traditional methods with the proposed method using actual flight data demonstrates its superior accuracy in determining compensation coefficients and its excellent performance in achieving aeromagnetic compensation.

[0078] Verification process:

[0079] To validate the model and solution proposed in this invention, a flight experiment was conducted in a certain sea area. The aircraft used was a state-of-the-art large-scale unmanned aerial vehicle (UAV), primarily equipped with an optical pump, a three-axis fluxgate, an inertial navigation system, and a GPS. The optical pump provided scalar total magnetic field data during flight, the three-axis fluxgate provided magnetic field data along the X, Y, and Z axes in the aircraft's coordinate system, and the inertial navigation system provided Euler angles during maneuvers, including pitch, roll, and yaw. The GPS provided flight altitude, flight speed, and real-time longitude and latitude during flight. The data sampling rate was 10 Hz, and the three-axis fluxgate data and the data acquired by the optical pump were filtered using a 0.05-0.6 Hz bandpass filter to minimize the impact of geomagnetic gradients.

[0080] like Figure 4 As shown, the present invention conducted two sets of compensation flight experiments, A and B. As can be seen from the figure, the compensation flight circle conducted by the present invention is very large, with a longitude span of approximately 0.15 degrees and a latitude span of approximately 0.15 degrees. Both flights were counterclockwise, flying from north to south, west to east, south to north, and east to west, respectively, with roll, pitch, and yaw maneuvers performed in each direction. The maneuvering amplitudes set in the present invention's experiments were ±10° for yaw, ±6° for pitch, and ±15° for roll. This design was intended to reduce multicollinearity in the compensation matrix to a certain extent and to better compensate for different columns in the matrix.

[0081] The overall flight altitude is around 3200m. The present invention uses compensation flight circle A as the compensation flight for calculating the compensation coefficient, and verification flight circle B as the verification flight for verifying the compensation effect of the compensation coefficient. Before compensation, the data obtained by the optically pumped magnetometer are processed by a bandpass filter (0.05-0.6Hz). Figure 5It can be seen that the data standard deviation of the compensation flight circle A is 2.211nT, and the noise is relatively large. Due to the large span of the flight circle, it is inevitably disturbed by the geomagnetic field gradient during flight. At the same time, in order to reduce multicollinearity to a certain extent, a large compensation maneuver is adopted, which leads to a large noise in the compensation process. In addition, due to the large aircraft platform and the large number of instruments, the interference magnetic field value to be compensated is very large, which makes compensation difficult. However, combined with the compensation model proposed in this paper and the optimization algorithm LRNM adopted, the final compensation effect is as follows Figure 5-Figure 6 As shown in Table 1, the compensation coefficient obtained for compensation flight circle A achieves an improvement of 20.46 when used to compensate for itself. The same compensation coefficient, when used to compensate for verification flight circle B, achieves an improvement of 17.67. To further demonstrate the effectiveness of the LRNM proposed in this invention, the LRNM is compared with the traditional LS and SVD methods. Compared to the SVD method, the LRNM achieves a 27% improvement, and compared to the LS method, the LRNM achieves a 34% improvement. This demonstrates the significant advantages of the LRNM optimization algorithm proposed in this invention. Furthermore, Table 2 shows that there is little difference between the LS and SVD methods when applied to aeromagnetic compensation data from large interfering magnetic fields. This demonstrates that the proposed method performs well in aeromagnetic compensation of large UAVs and offers significant compensation results.

[0082] Table 1 Compensation results of different methods for compensation flight circle A and verification flight circle B

[0083]

[0084] In order to verify the compensation effect of the compensation coefficient in level flight, a level flight test was finally carried out, such as Figure 8 In the level flight section C shown, no maneuvers are performed during the level flight. All interference noise comes from the large UAV platform and other instruments and equipment on board. The compensation coefficient is still obtained from the data of the correction flight circle A. The compensation result is as follows Figure 9 As shown in the figure, the LRNM method proposed in this invention achieves the best compensation effect. As shown in Table 2, the standard deviation of the noise before compensation is 0.249nT. After compensation, the LRNM method proposed in this invention improves the ratio by 29% compared to the SVD method and by 35% compared to the LS method. This shows that the optimization algorithm LRNM proposed in this invention has obvious advantages.

[0085] Table 2 Compensation results during level flight

[0086] method Before compensation (STD) After compensation (STD) Improvement ratio IR LRNM 0.249nT 0.0158nT 15.73 LS 0.249nT 0.0214nT 11.64 SVD 0.249nT 0.0205nT 12.15

[0087] It can be seen from the above embodiments that the LRNM method, a new optimization solution algorithm proposed in the present invention, has improved compensation effects compared with the traditional LM and SVD methods. Therefore, it is proved that this method can effectively solve the magnetic compensation problem of large multi-payload UAVs, which is of great significance to the field of voyage compensation.

[0088] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0089] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0090] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. An aeromagnetic compensation method for large unmanned aerial vehicles, characterized in that: The following steps are involved: Based on the TL model, the compensation matrix is ​​obtained by using the inertial navigation system, GPS and three-axis fluxgate magnetometer of the large UAV; Based on the compensation matrix, the large UAV is subjected to aeromagnetic compensation by increasing the maneuvering amplitude of the large UAV and utilizing a compensation optimization algorithm.

2. The aeromagnetic compensation method for large unmanned aerial vehicles according to claim 1, characterized in that: When obtaining the compensation matrix, the compensation matrix is ​​constructed using Euler angles based on the TL model through an inertial navigation system, GPS, and a three-axis fluxgate magnetometer.

3. The aeromagnetic compensation method for large unmanned aerial vehicles according to claim 2, characterized in that: When increasing the amplitude of the maneuver, a large angle compensation flight strategy is adopted to reduce the multicollinearity problem in the compensation matrix.

4. The aeromagnetic compensation method for large unmanned aerial vehicles according to claim 3, characterized in that: When the compensation optimization algorithm is used, the compensation optimization algorithm is constructed by using the Lasso regression algorithm and the Newton iteration algorithm.

5. The aeromagnetic compensation method for large unmanned aerial vehicles according to claim 4, characterized in that: When performing aeromagnetic compensation on a large UAV, the compensation matrix is ​​used to solve the magnetic field data obtained by the optically pumped magnetometer of the large UAV to obtain 36 compensation coefficients; Obtaining an interfering magnetic field according to the compensation coefficient and the compensation matrix; The magnetic field data is used to subtract the interfering magnetic field to obtain compensated magnetic field data.

6. The aeromagnetic compensation method for large unmanned aerial vehicles according to claim 5, characterized in that: When performing aeromagnetic compensation, the standard deviation is used to evaluate the data before and after compensation.

7. An aeromagnetic compensation system for large unmanned aerial vehicles, characterized in that: include: A compensation matrix building module is used to obtain the compensation matrix based on the TL model through the inertial navigation system, GPS and three-axis fluxgate magnetometer of the large UAV; The aeromagnetic compensation module is used to perform aeromagnetic compensation on the large UAV based on the compensation matrix by increasing the maneuvering amplitude of the large UAV and using a compensation optimization algorithm.

8. The aeromagnetic compensation system for large unmanned aerial vehicles according to claim 7, characterized in that: The compensation matrix construction module is further used to construct the compensation matrix using Euler angles.

9. The aeromagnetic compensation system for large unmanned aerial vehicles according to claim 8, characterized in that: The aeromagnetic compensation module is also used to adopt a large-angle compensation flight strategy to reduce the multicollinearity problem in the compensation matrix; and to construct the compensation optimization algorithm through the Lasso regression algorithm and the Newton iteration algorithm.

10. The aeromagnetic compensation system for large unmanned aerial vehicles according to claim 9, characterized in that: The aeromagnetic compensation module is further configured to solve, based on the compensation matrix, the magnetic field data acquired by the optically pumped magnetometer of the large UAV to obtain 36 compensation coefficients; An interfering magnetic field is obtained according to the compensation coefficient and the compensation matrix; and the interfering magnetic field is subtracted from the magnetic field data to obtain compensated magnetic field data, wherein the data before and after compensation are evaluated using a standard deviation.

Images