A posture-aware deep-sea magnetic field correction method and system

Abstract

The present invention discloses a deep-sea magnetic field correction method and system for attitude perception, including: obtaining geographical location parameters and detection time parameters of the current detection location, where the geographical location parameters include longitude, latitude, and sea level depth, and further obtaining the theoretical magnetic field value of the current detection location according to the IGRF model; obtaining the actual detected magnetic field value of the three-component magnetometer at the same detection time at the current detection location, where the actual detected magnetic field value includes magnetic field components in three orthogonal directions, total magnetic field intensity, magnetic dip, and magnetic declination; constructing a deep-sea magnetic field error model including soft magnetic error, orthogonal error, and hard magnetic error, calculating the corresponding magnetic field error according to the theoretical magnetic field value and the actual detected magnetic field value, and using the principal component analysis method to extract magnetic field error characteristics to obtain error eigenvectors and eigenvalues; calculating the error contribution rate of each magnetic field error characteristic type, and performing principal component analysis according to the error contribution, and further calculating the magnetic field correction value.

Description

Technical Field

[0001] The present invention relates to the technical field of magnetic force detection and correction of three-component magnetometers, and particularly relates to a method and system for deep-sea magnetic correction with attitude perception. Background Art

[0002] Currently, when existing magnetometers perform magnetic field detection, they are all affected by factors including their own materials and site environments. Among them, the deep-sea scenario has a complex environment and numerous magnetic field influencing factors. Existing deep-sea magnetic detection methods mainly use three-component magnetometers for detection. The detection methods of three-component magnetometers mainly include using magnetoresistive sensors to measure the magnetic field intensities in each axial direction through three orthogonally arranged magnetoresistive elements, or using fluxgate technology to generate a periodic saturated magnetic field by coil excitation of an iron core and detecting the change in the second harmonic signal caused by the external magnetic field. However, the above existing magnetometers are affected by various types of magnetic field interferences, such as soft magnetic errors, orthogonal errors, and hard magnetic errors. Existing three-component magnetic correction methods mainly correct the hard magnetic errors of the material itself. There are still technical problems in separating the magnetic field interference factors for the correction methods of soft magnetic errors and orthogonal errors affected by the environmental magnetic field. In particular, there is no good existing technology for the analysis of the main magnetic field interference factors, resulting in poor correction effects of traditional three-component magnetometers and large problems in magnetic field detection during the deep-sea application of the three-component magnetometers. Summary of the Invention

[0003] One of the invention objectives of the present invention is to provide a method and system for deep-sea magnetic correction with attitude perception. The method and system apply the principal component analysis method to the deep-sea magnetic correction to analyze the influencing factors of magnetic force errors. The present invention uses the principal component analysis method in combination with the IGRF model (International Geomagnetic Reference Field model) to effectively separate the main types of magnetic field detection errors existing in the existing three-component magnetometers during the deep-sea exploration of AUVs (Autonomous Underwater Vehicles). According to the main error types, the measured data of the corresponding types of the three-component magnetometers are corrected, thereby improving the magnetic field detection accuracy of the three-component magnetometers in the deep-sea scenario.

[0004] Another object of the present invention is to provide a posture-aware deep-sea magnetic field correction method and system. The method and system pre-establish a deep-sea magnetic field error model, where the deep-sea magnetic field error model includes an environmental geomagnetic field matrix, a measured magnetic field matrix, a soft magnetic error matrix, an orthogonal error matrix, and a hard magnetic error matrix. Using the principal component analysis method and the deep-sea magnetic field model, perform eigen-decomposition on the three-component magnetometer in the deep-sea scenario for correlation coefficient matrices such as soft magnetic error, orthogonal error, and hard magnetic interference error. And the correlation coefficient matrix also decomposes the errors of magnetic field strength and magnetic field direction based on soft magnetic error, orthogonal error, and hard magnetic interference error. After eigen-decomposition, obtain the contribution rate of different error types to the magnetic field error according to the principal component analysis method, and perform principal component screening according to the error contribution rate.

[0005] Another object of the present invention is to provide a posture-aware deep-sea magnetic field correction method and system. The method and system are also provided with a high-precision crystal gyroscope. The present invention uses the high-precision crystal gyroscope to calculate the posture angle of the corresponding three-component magnetometer in the deep-sea scenario, and constructs a relevant angle transformation matrix of the three-component magnetometer according to the posture angle. The angle transformation matrix is used for eigen-decomposition of the correlation coefficient matrix, so as to know the influence of the three-component magnetometer on the detected magnetic field error at the rotation angle.

[0006] Another object of the present invention is to provide a posture-aware deep-sea magnetic field correction method and system. The method and system also adopt a quaternion update method and a filter to update the posture angle of the three-component magnetometer by the high-precision crystal gyroscope in real time in the deep-sea scenario. Using the quaternion update method to calculate the posture angle of the three-component magnetometer in the deep-sea scenario can effectively improve the detection accuracy of the posture angle, especially can improve the detection accuracy of the posture angle of an AUV (Autonomous Underwater Vehicle) in a dynamic scenario, thereby reducing the error brought by the detection accuracy of the AUV (Autonomous Underwater Vehicle) in the motion state where the gyroscope is installed itself.

[0007] In order to achieve at least one of the above-mentioned objects of the invention, the present invention further provides a posture-aware deep-sea magnetic field correction method, the method comprising:

[0008] Obtain the geographical location parameters and detection time parameters of the current detection location, where the geographical location parameters include longitude, latitude, and sea level depth, and input the geographical location parameters and time parameters into the IGRF model to obtain the theoretical magnetic field value of the current detection location;

[0009] Obtain the actual detected magnetic field value of the three-component magnetometer at the same detection time at the current detection location, where the actual detected magnetic field value includes magnetic field components in three orthogonal directions, total magnetic field strength, magnetic dip angle, and magnetic declination;

[0010] Construct a deep - sea magnetic field error model including soft magnetic error, orthogonal error and hard magnetic error, calculate the corresponding magnetic field error according to the theoretical magnetic field value and the actually detected magnetic field value, and use the magnetic field error model and the principal component analysis method to extract the magnetic field error characteristics, obtaining the error eigenvector and eigenvalue;

[0011] Calculate the error contribution rate of each magnetic field error characteristic type according to the error eigenvector and eigenvalue, perform principal component analysis according to the error contribution of each magnetic field error characteristic type, and calculate the magnetic field correction value according to the principal component analysis result.

[0012] According to one preferred embodiment of the present invention, the method for constructing the deep - sea magnetic field error model includes: constructing the deep - sea magnetic field error model using the following formula: B sensor =A soft *C sn *B real +B hard , where B sensor is the measured value of the three - component magnetometer, A soft is the soft magnetic error matrix, C sn is the orthogonal error matrix, B real is the environmental geomagnetic field intensity, B hard is the hard magnetic error matrix. Further, establish a linear correction model: B 校正 =K*B sensor +B hard , where K is the correction coefficient matrix, B 校正 is the correction matrix. Solve the correction coefficient matrix K and the hard magnetic error matrix B sensor by minimizing the residual between the measured value B hard and the theoretical value, and use the linear correction model to output the corrected magnetic field intensity and corrected magnetic field direction including different magnetic field components.

[0013] According to another preferred embodiment of the present invention, the principal component analysis method includes: using an AUV underwater robot to perform magnetic force detection at different positions with different latitudes, longitudes and sea - level depths in the deep sea, obtaining the magnetic field components in three orthogonal directions, the total magnetic field intensity, the magnetic dip angle and the magnetic declination at the corresponding positions at the corresponding detection times, and using the IGRF model to input the geographical parameters and detection time parameters of the different positions, outputting multiple error values including the magnetic field components in three orthogonal directions, the total magnetic field intensity, the magnetic dip angle and the magnetic declination, performing standardization processing on each type of error value, and calculating the correlation coefficient of each error type with respect to the error values of the magnetic field components in three orthogonal directions, the total magnetic field intensity, the magnetic dip angle and the magnetic declination using the correlation coefficient matrix.

[0014] According to another preferred embodiment of the present invention, the eigenvalues and eigenvectors of the correlation coefficient matrix are further calculated, wherein the eigenvectors are orthogonal eigenvectors of the eigenvalues corresponding to the error types, the eigenvalues are error correlation values calculated corresponding to the error types, and the contribution rate of each principal component is calculated based on the eigenvectors and eigenvalues, and the total error contribution rate of different principal components is calculated. Based on the total error contribution rate, the correction of different error types required to meet the principal component error contribution rate requirement is calculated through the linear correction model B 校正 =K*B sensor +B hard Calculate the final correction value of each error type.

[0015] According to another preferred embodiment of the present invention, the calculation method of the principal component error contribution rate includes: defining the eigenvalue corresponding to the error type j as λ j , then the error contribution rate of this eigenvalue type ; where m is the total error type, k represents the corresponding error type identifier. After obtaining the error contribution rate of the corresponding eigenvalue type, the total error contribution rate Z P is further calculated.

[0016] According to another preferred embodiment of the present invention, the calculation method of the total error contribution rate Z p includes: sorting the corresponding error type eigenvalues λ j from large to small, and selecting p error types from large to small and calculating the total error contribution rate Z p according to the following formula: , set the total error contribution rate threshold Z s . If the calculated total error contribution rate Z p < the total error contribution rate threshold Z s , then p = p + 1, and substitute p + 1 back into the total error contribution rate Z p for calculation until the calculated total error contribution rate Z p ≥ the total error contribution rate threshold Z s .

[0017] According to another preferred embodiment of the present invention, for the rotation error of the magnetic field direction, a configuration including a high-precision gyroscope and an acceleration sensor is used to collect the rotation angle of the three-component magnetometer in real time. Among them, the four-element method is used for the real-time update of the collected rotation angle of the three-component magnetometer. The four-element method includes: constructing a four-element differential equation describing the attitude of the corresponding component of the three-component magnetometer: , where q is the four-element and , ω is the angular velocity, x, y, z respectively represent different component elements, ω x , ω yand ω z are the angular velocities of the corresponding components respectively, is the quaternion multiplication.

[0018] According to another preferred embodiment of the present invention, the quaternion update method further includes: updating the quaternion according to the following fourth-order Runge-Kutta formula:

[0019] ;

[0020] ;

[0021] ;

[0022] ;

[0023] where k1, k2, k3 and k4 are the slope coefficients respectively, q t is the quaternion at the corresponding detection time t, h is the detection time step, and ω is the angular velocity; and further update the quaternion at the next detection time t + 1 according to the following formula: q t+1 = q t + h( k 1 + 2 k 2 + 2 k 3 + k 4) / 6, where the angular velocity including the corresponding components can be obtained through the updated quaternion q t+1 and the rotation matrix of the corresponding component can be obtained by integrating and solving in combination with the corresponding detection time step.

[0024] In order to achieve at least one of the above-mentioned invention purposes, the present invention further provides an attitude perception deep-sea magnetic correction system, and the system executes the above-mentioned attitude perception deep-sea magnetic correction method.

[0025] The present invention further provides a computer-readable storage medium, and the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the above-mentioned attitude perception deep-sea magnetic correction method. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It shows a schematic flowchart of an attitude perception deep-sea magnetic correction method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations. The basic principles defined in the following description of the present invention can be applied to other implementation schemes, variation schemes, improvement schemes, equivalent schemes, and other technical schemes without departing from the spirit and scope of the present invention.

[0028] It can be understood that the term "a" should be understood as "at least one" or "one or more". That is, in one embodiment, the number of an element can be one, while in other embodiments, the number of the element can be multiple. The term "a" cannot be understood as a limitation on the quantity.

[0029] Please combine Figure 1 , the present invention provides a deep-sea magnetic correction method and system for attitude perception. The method includes the following steps: First, it is necessary to determine the deep-sea magnetic detection area, and an autonomous underwater vehicle (AUV) is configured in the deep-sea magnetic detection area. The AUV is configured with, including but not limited to, a three-component magnetometer, a high-precision crystal gyroscope, etc. The high-precision crystal gyroscope and the acceleration sensor can be integrally installed in the three-component magnetometer to form a sealed structure for real-time collection of the magnetic detection data and the rotational attitude data of the AUV in the corresponding deep-sea detection area. The AUV is also configured with position sensors including but not limited to GPS, etc., for detecting geographical location parameters such as the longitude and latitude of the AUV on the ocean plane, and a pressure sensor including but not limited to can be configured in the AUV, and the water pressure value detected by the pressure sensor is used for depth conversion to obtain the sea level depth where the AUV is located. After the present invention obtains the geographical location parameters of the AUV including longitude, latitude, and sea level depth, the geographical location parameters and the detection time parameters are further input into the IGRF model (International Geomagnetic Reference Field model) for theoretical magnetic field value calculation. It should be noted that the IGRF model is a model of the standard theoretical value of the global magnetic field, which varies with different detection times. In the present invention, the latest version of the IGRF model needs to be used as the reference magnetic field model for the corresponding deep-sea location to obtain the magnetic field theoretical value. Calculate the error value according to the theoretical value and the actual detection value, and establish a deep-sea magnetic field error model and a linear error correction model. Use the error contribution rate obtained by the deep-sea magnetic field error model and the principal component analysis method and the linear error correction model to correct the measured value of the corresponding type of magnetic field, so as to improve the error correction effect of the three-component magnetometer.

[0030] Specifically, the method for calculating the corresponding magnetic field theoretical value using the IGRF model in the present invention includes the following steps: First, the geographical location parameters and time parameters of the underwater robot AUV need to be obtained. Among them, the geographical location parameters are converted, and the longitude and latitude in the geographical location parameters are converted into co-latitude θ and geocentric distance r. Among them, the altitude h is converted into the difference r - h between the geocentric distance r and the altitude h, which is used as the elevation data for the corresponding IGRF model. The co-latitude θ, geocentric distance r, elevation r - h, and detection time series t are input into the latest version of the IGRF model. The scalar potential V is calculated using the spherical harmonic function of the IGRF model, and time difference calculation and magnetic field component calculation are performed. Finally, the IGRF model outputs the total magnetic field intensity F, horizontal magnetic field intensity (H), northward magnetic field intensity (X), eastward magnetic field intensity (Y), vertical magnetic field intensity (Z), magnetic declination (D), and magnetic inclination (I). Among them, the total magnetic field intensity F is the vector sum of the three orthogonal components of the northward magnetic field intensity (X), eastward magnetic field intensity (Y), and vertical magnetic field intensity (Z), that is, the total magnetic field intensity F = , and the total magnetic field direction angle is the direction of the vector sum of the corresponding three orthogonal components.

[0031] Further, the formula for the deep-sea magnetic field error model constructed in the present invention is as follows: B sensor =A soft *C sn *B real +B hard , where B sensor is the measured value of the three-component magnetometer, A soft is the soft magnetic error matrix, C sn is the orthogonal error matrix, B real is the environmental geomagnetic field intensity, and B hard is the hard magnetic error matrix. Among them, the types of magnetic field errors can be divided into, including but not limited to, soft magnetic errors of the horizontal magnetic field, soft magnetic errors of the northward magnetic field, total magnetic field soft magnetic errors, as well as orthogonal errors of magnetic inclination, orthogonal errors of magnetic declination, and hard magnetic errors of the total magnetic field, etc. Since the magnetic field lines of the ideal geomagnetic field model are spherical extension structures, but due to various types of errors including detection materials, mechanical mechanisms, ocean environments, and the composition of seabed rock plates, the specific values of the above error types can be obtained through least squares calculation by fitting an ellipsoid model, and the corresponding fitting calculation can be performed through the ellipsoid model provided by the GitHub open-source library. The fitting calculation methods include but not limited to the quadratic surface equation calculation of the output data of the magnetic field sensor and the geometric parameter extraction calculation of the ellipsoid surface. Since the ellipsoid surface fitting calculation based on the magnetic field model is a prior art, the present invention will not elaborate on it in detail. The present invention further sets a linear magnetic field correction model, and the formula for the linear magnetic field correction model is as follows: B校正 = K * B sensor + B hard , where K is the calibration coefficient matrix and B 校正 is the calibration matrix. The calibration coefficient matrix K and the hard magnetic error matrix B are solved by minimizing the residuals between the measured value B sensor and the theoretical value. The calibrated magnetic field strength and the calibrated magnetic field direction including different magnetic field components are output using the linear calibration model. Among them, minimizing the residuals between the measured value B hard and the model theoretical value is given by the formula sensor . .

[0032] It is worth mentioning that in the present invention, the principal component analysis method is used to perform principal component analysis on different error types with respect to the principal components including but not limited to the total magnetic field error. The specific steps are as follows: An AUV underwater robot is used to perform magnetic force detection at different positions with different latitudes, longitudes, and sea surface depths in the deep sea, and the magnetic field components in three orthogonal directions, the total magnetic field strength, the magnetic dip angle, and the magnetic declination at the corresponding positions at the corresponding detection times are obtained. The IGRF model is used to input the geographical parameters and detection time parameters of the different positions, and a plurality of error values including the magnetic field components in three orthogonal directions, the total magnetic field strength, the magnetic dip angle, and the magnetic declination are output. Standardization processing is performed on the error values of each type. The standardization processing method includes: determining the magnetic field index object i, where the above magnetic field error index object i includes the total magnetic field strength F, the northward magnetic field strength (X), the eastward magnetic field strength (Y), the vertical magnetic field strength (Z), the magnetic declination (D), and the magnetic dip angle (I). Further, the magnetic field error index type j is determined, where the magnetic field error index type j includes but not limited to soft magnetic error, orthogonal error, hard magnetic error, and direction error, etc. The soft magnetic error includes scale error, sensitivity error, temperature drift error, and hysteresis error caused by sensor scaling, etc. The above error types are only examples in the present invention, and specific more detailed classifications can be selected according to actual needs, which will not be elaborated in detail in the present invention. The correlation coefficient matrix is used to calculate the correlation coefficient of each error type with respect to the error values of the magnetic field components in the three orthogonal directions, the total magnetic field strength, the magnetic dip angle, and the magnetic declination.

[0033] In the present invention, standardization processing needs to be performed on the error data. The standardization processing method includes: defining the standardized magnetic field error value as B ij， where i represents the magnetic field error index object and j represents the magnetic field error index type. The standardized magnetic field error value is , where X ij represents the original data, υ j represents the mean of the j-th type of index, and σ jis the standard deviation of the j-th type of index. After calculating the standardized magnetic field error value B ij , the correlation coefficient r ij is further calculated as follows: , where Cov represents the covariance function, r represents the correlation coefficient of the standardized error matrix, and var represents the variance. Based on the correlation coefficient r ij , the corresponding correlation coefficient matrix R is constructed. In another preferred embodiment of the present invention, if the calculation of the standardized magnetic field error value B ij is not performed in order to retain the original information, the covariance matrix can be directly used for calculation, and the present invention will not elaborate on this in detail.

[0034] The eigenvalues and eigenvectors of the correlation coefficient matrix R are further calculated. The eigenvectors are the orthogonal eigenvectors corresponding to the eigenvalues of the corresponding error types, and the eigenvalues are the error correlation values calculated for the corresponding error types. Based on the eigenvectors and eigenvalues, the contribution rate of each principal component is calculated, and the total error contribution rate of different principal components is calculated. Based on the total error contribution rate, the correction of different error types required to meet the principal component error contribution rate requirement is calculated. Through the linear correction model B 校正 =K*B sensor +B hard , the final correction value of each error type is calculated.

[0035] The calculation method of the principal component error contribution rate includes: defining the eigenvalue corresponding to the error type j as λ j , then the error contribution rate of this eigenvalue type ; where m is the total error type, and k represents the corresponding error type identifier. After obtaining the error contribution rate of the corresponding eigenvalue type, the total error contribution rate Z P is further calculated.

[0036] The calculation method of the total error contribution rate Z p includes: sorting the corresponding error type eigenvalues λ j from large to small, and selecting p error types from large to small and calculating the total error contribution rate Z p according to the following formula: . Set the total error contribution rate threshold Z s . If the calculated total error contribution rate Z p <the total error contribution rate threshold Z s , then p = p + 1, and substitute p + 1 back into the total error contribution rate Z p for calculation until the calculated total error contribution rate Z p ≥the total error contribution rate threshold Z s .

[0037] It should be noted that in the present invention, the rotation error for the direction also needs to be calculated. The configuration includes a high-precision gyroscope and an acceleration sensor, which are used to collect the rotation angles of the three-component magnetometer in real time. Among them, the quaternion method is used for real-time updating of the collected rotation angles of the three-component magnetometer. The quaternion method includes: constructing a quaternion differential equation describing the attitude of the corresponding components of the three-component magnetometer: , where q is the quaternion and , ω is the angular velocity, x, y, and z respectively represent different component elements, ω x , ω y and ω z are respectively the angular velocities of the corresponding components, is the quaternion multiplication. The quaternion update method further includes: updating the quaternion according to the following fourth-order Runge-Kutta formula:

[0038] ;

[0039] ;

[0040] ;

[0041] ;

[0042] where k1, k2, k3, and k4 are respectively the slope coefficients, q t is the quaternion at the corresponding detection time t, h is the detection time step, and ω is the angular velocity; and further update the quaternion at the next detection time t + 1 according to the following formula: q t+1 = q t +h( k 1 + 2 k 2 + 2 k 3 + k 4) / 6, where the angular velocity including the corresponding components can be obtained through the updated quaternion q t+1 . After integrating and solving in combination with the corresponding detection time step, the rotation matrix of the corresponding component can be obtained. It should be noted that the rotation matrix can be used for further conversion expressions including but not limited to magnetic declination error, magnetic inclination error, etc. That is to say, the rotation matrix can be added to the correction coefficient matrix K for linear correction.

[0043] Embodiments disclosed by the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. Embodiments disclosed by the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), the above-mentioned functions defined in the methods of the present application are performed. It should be noted that the above-mentioned computer-readable medium in the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection having one or more wire segments, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device. And in the present application, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and the computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted by any appropriate medium, including but not limited to: wireless segments, wire segments, optical cables, RF, etc., or any suitable combination of the above.

[0044] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks may occur in a different order than that noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0045] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and illustrated in the embodiments. Without departing from the said principles, the embodiments of the present invention may have any variations or modifications.

Claims

1. A deep-sea magnetic correction method for attitude perception, characterized in that: The method comprises: Obtaining geographic location parameters and detection time parameters of the current detection location, wherein the geographic location parameters include longitude and latitude and sea level depth, and inputting the geographic location parameters and time parameters into the IGRF model to obtain the theoretical magnetic field value of the current detection location; Obtaining the actual detected magnetic field value of the three-component magnetometer at the current detection location, wherein the actual detected magnetic field value includes magnetic field components in three orthogonal directions, the total magnetic field intensity, the magnetic inclination and the magnetic declination; A deep-sea magnetic field error model including soft magnetic error, orthogonal error and hard magnetic error is constructed, the corresponding magnetic field error is calculated according to the theoretical magnetic field value and the actual detected magnetic field value, and the magnetic field error feature is extracted using the magnetic field error model and principal component analysis method to obtain the error feature vector and feature value; The error contribution rate of each magnetic field error feature type is calculated according to the error eigenvector and eigenvalue, a principal component analysis is performed according to the error contribution of each magnetic field error feature type, and a magnetic field correction value is calculated according to the principal component analysis result.

2. A deep-sea magnetic correction method for attitude perception according to claim 1, characterized in that: The deep-sea magnetic field error model construction method comprises: constructing the deep-sea magnetic field error model using the following formula: sensor =A soft *C sn *B real +B hard , where B sensor is the measured value of the three-component magnetometer, A soft is the soft magnetic error matrix, C sn is the orthogonal error matrix, B real is the ambient geomagnetic field strength, B hard is the hard magnetic error matrix, and a linear correction model is further established: 校正 =K*B sensor +B hard , where K is the correction coefficient matrix, B 校正 is the correction matrix, by minimizing the measured value B sensor The correction coefficient matrix K and the hard magnetic error matrix B are solved by the residual of the theoretical value. hard , the linear correction model is used to output the correction magnetic field intensity and correction magnetic field direction including different magnetic field components.

3. A deep-sea magnetic correction method for attitude perception according to claim 2, characterized in that: The principal component analysis method includes: using an AUV underwater robot to perform magnetic detection at different positions of different longitudes and latitudes and sea level depths in the deep sea, obtaining magnetic field components, total magnetic field intensity, magnetic inclination and magnetic declination in three orthogonal directions corresponding to different positions at corresponding detection times, and using the IGRF model to input geographical parameters and detection time parameters of the different positions, output multiple error values ​​including magnetic field components, total magnetic field intensity, magnetic inclination and magnetic declination in three orthogonal directions, standardizing the error values ​​of each type, and using a correlation coefficient matrix to calculate the correlation coefficient of each error type to the error values ​​of the magnetic field components, total magnetic field intensity, magnetic inclination and magnetic declination in the three orthogonal directions.

4. A deep-sea magnetic correction method for attitude perception according to claim 3, characterized in that: Further calculate the eigenvalues ​​and eigenvectors of the correlation coefficient matrix, wherein the eigenvectors are orthogonal eigenvectors of the eigenvalues ​​of the corresponding error types, the eigenvalues ​​are error correlation values ​​calculated for the corresponding error types, and the contribution rate of each principal component is calculated based on the eigenvectors and eigenvalues, and the total error contribution rate of different principal components is calculated, and the correction of different error types that need to meet the principal component error contribution rate requirements is calculated based on the total error contribution rate, and the linear correction model B is used. 校正 =K*B sensor +B hard Calculate the final correction value for each error type.

5. A deep-sea magnetic correction method for attitude perception according to claim 4, characterized in that: The calculation method of the principal component error contribution rate includes: defining the characteristic value corresponding to the error type j as λ j , then the error contribution rate of this eigenvalue type is ; Where m is the total error type, k represents the corresponding error type identifier, after obtaining the error contribution rate of the corresponding eigenvalue type, further calculate the total error contribution rate Z P .

6. A deep-sea magnetic correction method for attitude perception according to claim 5, characterized in that: The total error contribution rate Z p The calculation method includes: converting the corresponding error type characteristic value λ j Sort from large to small, and select p error types from large to small according to the following formula to calculate the total error contribution rate Z p calculate: , set the total error contribution rate threshold Z s， If the calculated total error contribution rate Z p <Total error contribution rate threshold Z s , then p=p+1, and p+1 is substituted into the total error contribution rate Z p Calculate until the total error contribution rate Z is calculated p ≥Total error contribution rate threshold Z s .

7. The deep-sea magnetic correction method for attitude perception according to claim 1, characterized in that: For the rotation error of the magnetic field direction, the configuration includes a high-precision gyroscope for real-time acquisition of the rotation angle of the three-component magnetometer, wherein the four-element method is used to perform real-time update of the acquired rotation angle of the three-component magnetometer, wherein the four-element method includes: constructing a four-element differential equation describing the attitude of the corresponding components of the three-component magnetometer: , where q is a four-element , ω is the angular velocity, x, y, z represent different component elements, ω x ,ω y and ω z are the angular velocities of the corresponding components, is four-element multiplication.

8. The deep-sea magnetic correction method for attitude perception according to claim 7, characterized in that: The four-element updating method further includes: updating the four elements according to the following fourth-order Runge-Kutta method formula: ； ； ； ； Where k1, k2, k3 and k4 are slope coefficients, q t is the four elements corresponding to the detection time t, h is the detection time step, and ω is the angular velocity; and the four elements at the next detection time t+1 are further updated according to the following formula: q t+1 = q t +h( k 1+2 k 2+2 k 3+ k 4) / 6, where the updated four elements are q t+1 The angular velocity including the corresponding component can be obtained, and the rotation matrix of the corresponding component can be obtained by integrating and solving the corresponding detection time step.

9. A deep-sea magnetic correction system with attitude perception, characterized in that: The system executes a posture-aware deep-sea magnetic correction method as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement a posture-aware deep-sea magnetic correction method as described in any one of claims 1-8.

Images