阅读说明：本技术 一种矢量磁力梯度仪转向差校正方法 (Vector magnetic gradiometer steering difference correction method ) 是由 赵瑜 周波 朱文兵 周建斐 綦俊峰 刘雨聪 蒋弘威 于 2021-07-02 设计创作，主要内容包括：本发明涉及一种矢量磁力梯度仪转向差校正方法,适用于两套矢量磁力仪组成磁力梯度仪的分量梯度误差校正,通过建立梯度误差校正模型,引入三轴无磁基准转台进行精确的姿态旋转,按照规范的姿态变换来分离和求取单个模型参数,逐个完成误差校正模型内的参数解算,该方法可避免复杂繁琐的矩阵运算；然后引入姿态航向传感器来获取矢量磁力梯度仪的三种角度信息,通过各分量剩余的梯度误差与角度数据的关系式来求取姿态误差校正参数。完成上述校正过程后,各分量梯度误差由最大±380.0nT降低至±4.0nT,校正误差下降了98.95％,极大的降低了矢量磁力梯度仪的输出转向差。(The invention relates to a vector magnetic gradiometer steering difference correction method, which is suitable for component gradient error correction of a magnetic gradiometer consisting of two sets of vector gradiometers, wherein a gradient error correction model is established, a three-axis nonmagnetic reference rotary table is introduced for accurate attitude rotation, a single model parameter is separated and solved according to standard attitude transformation, and parameter calculation in the error correction model is completed one by one, so that the method can avoid complex and fussy matrix operation; and then, introducing an attitude heading sensor to acquire three kinds of angle information of the vector magnetic gradiometer, and solving an attitude error correction parameter through a relational expression of residual gradient errors of all components and angle data. After the correction process is completed, the gradient error of each component is reduced to +/-4.0 nT from +/-380.0 nT at the maximum, the correction error is reduced by 98.95 percent, and the output steering difference of the vector magnetic gradiometer is greatly reduced.)

1. A vector magnetic gradiometer steering difference correction method is characterized in that: firstly, two sets of vector magnetometers are combined into a vector magnetic gradiometer according to the same direction to establish a gradient error correction model, error parameter variables are determined, a three-axis nonmagnetic reference rotary table is introduced, so that the vector magnetic gradiometer can be gradually rotated to any specific angle in a vector space through posture transformation to obtain component gradient error values of the specific angle, and the error values are corrected one by one according to the characteristics of each error parameter of the gradient error correction model until all the error values are corrected; and secondly, acquiring angle information of the vector magnetic gradiometer by introducing an attitude heading sensor and rigidly fixing the attitude heading sensor and the vector magnetic gradiometer, then enabling the vector magnetic gradiometer to rotate randomly in space, synchronously acquiring gradient data and angle data of each component in the rotating process, and correcting the residual error amount by establishing a relational expression between the gradient data and the angle data of each component.

2. The vector magnetic gradiometer steering difference correction method of claim 1, wherein: the gradient error correction model comprises two axis alignment errors, a gradient zero point error and a difference error, and a model matrix is as follows:

wherein:

X1C, Y1C, Z1C, X2C, Y2C, Z2C respectively represent the original output of the X component, Y component, Z component of the first set of vector magnetometers, the original output of the X component, Y component, Z component of the second set of vector magnetometers;

DeltaX, DeltaY and DeltaZ respectively represent an X-axis gradient value, a Y-axis gradient value and a Z-axis gradient value;

ixbe, iybe and izbe respectively represent X-axis alignment error, Y-axis alignment error and Z-axis alignment error;

ixofs, iyofs and izofs respectively represent an X-axis gradient zero-point error, a Y-axis gradient zero-point error and a Z-axis gradient zero-point error;

ixocxy and ixocxz respectively represent a difference error of the Y axis to the X axis and a difference error of the Z axis to the X axis;

the iyocyx and the iyocyz respectively represent a difference error of an X axis to a Y axis and a difference error of a Z axis to the Y axis;

the izoczy and the izoczx respectively represent the difference error of the Y axis to the Z axis and the difference error of the X axis to the Z axis;

x1CC, Y1CC, and Z1CC represent the corrected outputs of the X, Y, and Z components of the first set of vector magnetometers, respectively.

3. The method for correcting the steering difference of the vector magnetic gradiometer according to claim 1, wherein in the posture conversion, in order to realize the calculation of the gradient error correction model, the vector magnetic gradiometer needs to be precisely rotated to a specific posture in space, and a set of three-axis nonmagnetic reference rotary table is designed and introduced, wherein the reference rotary table can be rotated at a full heading angle of 360 degrees, rotated at a full roll angle of 360 degrees, rotated at a full pitch angle of ± 90 degrees, and the reference scale is less than or equal to 0.1 degrees, so that the vector magnetic gradiometer can be precisely rotated and fixed at any required posture.

4. The vector magnetic gradiometer steering difference correction method of claim 1, wherein in the design of the nonmagnetic characteristic of the three-axis nonmagnetic reference turntable, in order to realize the calculation of the gradient error correction model, the turntable equipment is required not to introduce additional magnetic anomaly interference, the turntable used for posture change of the vector magnetic gradiometer is completely nonmagnetic, the main material is nonmagnetic aircraft aluminum cutting processing, the bearing is nonmagnetic ceramic material, and the accessory is nonmagnetic navy copper material, so as to ensure that the whole nonmagnetic turntable does not introduce additional magnetic anomaly interference to the correction work.

5. The method for correcting the steering difference of the vector magnetic gradiometer according to claim 1, wherein, in the calculation of model parameters, in order to realize the calculation of the gradient error correction model, the vector magnetic gradiometer is aligned to a specific orientation and angle through the turntable, the gradient error of a specific component is determined by only a single parameter in the model, and the parameter is obtained and calculated and substituted into the model; and aligning the vector magnetic gradiometer to other specific directions and angles through the turntable to obtain accurate values of other single parameters in the model, and operating one by one until the parameters in the model are calculated.

6. The method for correcting the steering error of the vector magnetic gradiometer according to claim 1, wherein in the aspect of attitude error correction, after the gradient error correction model is completed, a set of attitude heading sensor is introduced to obtain three angles of a heading angle, a roll angle and a pitch angle of the vector magnetic gradiometer, the remaining gradient errors of each component are synchronously obtained, the corresponding relation between the gradient errors of each component and the attitude angle information is obtained and eliminated, and the gradient steering error data after the attitude angle errors are eliminated is output.

Technical Field

The invention belongs to the field of error correction of magnetic sensors, and mainly relates to a method for correcting steering difference of a vector magnetic gradiometer.

Background

The vector magnetometer is formed by combining three magnetic component sensors which are perpendicular to each other in pairs, the three magnetic component sensors are used for respectively measuring the geomagnetic components of the geomagnetic field in the axial direction of the geomagnetic field, and the three magnetic component sensors are used for calculating according to three-dimensional coordinate axes to obtain a synthesized total field value, so that the magnetic field values measured by the vector magnetometer comprise the size and the direction of the components and the size of the total field value. The magnetic field measured by the scalar magnetometer is a total geomagnetic field, and only the magnitude and the direction of the total geomagnetic field are the same, so that the vector magnetometer has more comprehensive measured information compared with the scalar magnetometer.

Two sets of vector magnetometers are arranged according to the direction that the axial directions of all components are consistent, so that the directions of two X-axis components are consistent, the directions of Y-axis components are consistent, and the directions of Z-axis components are consistent, thereby forming the vector magnetometers, measuring the quantity of six magnetic components and three vector magnetic gradient data, synthesizing total field magnetic gradient data by one path, and compared with a single vector magnetometer or a scalar magnetometer, measuring magnetic information is richer, reflecting the magnetic field characteristics of a substance from a plurality of parameters, and explaining information is richer.

Magnetic information measured by the vector magnetic gradiometer is comprehensive, but errors influencing measurement accuracy are more, and useful information can be submerged in noise signals if effective processing is not carried out. The error factor includes two aspects: one is the output error of a single vector magnetometer, which comprises triaxial non-orthogonality error, zero offset, linearity non-uniformity and the like, the error has a mature error model, and the compensation of the error can be realized by acquiring magnetic field data under different postures and substituting the magnetic field data into the error model for correction; the other type is the gradient steering error of the vector magnetic gradiometer, and comprises a two-axis alignment error, a gradient zero error, a differential error and the like, wherein the error can cause the gradient value error output in the acquisition process of the vector magnetic gradiometer to be very large.

The invention provides a correction method of a magnetic gradient tensor system of a mobile platform CN201910461081.8, which comprises the steps of firstly modeling the system error of a single sensor, realizing the establishment of an error compensation model of the system of the single sensor, then randomly changing the attitude acquisition data in a three-dimensional space and carrying out inversion to obtain error parameters; and then modeling the non-alignment error of the magnetic gradient tensor system, enabling an X axis, a Y axis and a Z axis to be perpendicular to the horizontal table in sequence, respectively rotating and measuring data in the horizontal direction, substituting the data into a model and calculating coefficients, wherein simulation results show that the correction precision of the system can be improved to a greater extent. The method realizes effective correction in a specific direction on theoretical simulation, but simulation or actual test is not carried out in all directions, error correction in a full-range cannot be met, parameters cannot be matched after the state is changed, in addition, the established model adopts a differential evolution algorithm, various error factors are mixed together, a large amount of operation is needed during parameter calculation, and the real-time performance is not facilitated. The invention patent CN202010181000.1 discloses a unified correction method for magnetic gradient tensors of various orders, which obtains rotation correction data of the magnetic gradient tensors of various orders by establishing a unified correction frame for the magnetic gradient tensors of various orders, takes tensor invariants as constraint criteria, and adopts LM algorithm to solve optimal correction parameters. According to the method, an n-order correction matrix is established, calibration factor errors, non-orthogonal errors and non-alignment errors are fused, the matrix is expanded to a second order, a third order, a fourth order and an n-order, a complex matrix equation is obtained for optimal solution, the model is too complex, a large amount of operations are needed, the real-time performance is not facilitated, and the non-alignment error parameter values causing the gradient errors cannot be obtained independently. The invention patent CN201911293829.4 discloses a full-tensor magnetic gradient data compensation optimization method and system, aiming at each magnetic gradient component of full-tensor magnetic gradient data, different experimental data under various working conditions are adopted, relaxation coefficients and decision variables are introduced, constraint conditions of a model are established, correction coefficients are obtained according to a mathematical optimization model and the constraint conditions, and platform interference in the data is eliminated. The method only aims at a single vector magnetometer, realizes gradient error correction on each component of the single vector magnetometer, and does not mention component gradient error correction after the two vector magnetometers are combined. The invention patent CN202010048648.1 discloses a detection method based on magnetic gradient tensor, which first calculates the initial values of position vector and magnetic moment vector by using STAR method, then calculates new magnetic gradient contraction gradient, new position vector and magnetic moment vector, and repeats the above processes until the difference value of the position vector satisfies the convergence condition. The method belongs to the field of target detection, and uses a magnetic gradient tensor as a detection tool to obtain the position and the magnetic moment of a detection target, and optimizes the position and the magnetic moment through an algorithm. This method does not mention an error correction method of the magnetic gradient.

Disclosure of Invention

The invention aims to overcome the defects in the prior art, and provides a vector magnetic gradiometer steering difference correction method which can correct the magnetic gradient output errors of two paths of vector gradiometers, realize magnetic field vector measurement after error correction and improve the output accuracy of magnetic component gradients and synthesized total field magnetic gradients.

The object of the present invention is achieved by the following technical means. A method for correcting the steering difference of vector magnetic gradiometer can correct the gradient error of each component of two sets of vector magnetic gradiometers, and the gradient error of each component comes from the alignment error of two axes of each component of the magnetic gradiometer, the zero-point error of gradient, the differential error and the like. Firstly, two sets of vector magnetometers are combined into a vector magnetic gradiometer according to the same direction to establish a gradient error correction model, error parameter variables are determined, a three-axis nonmagnetic reference rotary table is introduced, so that the vector magnetic gradiometer can be gradually rotated to any specific angle in a vector space through posture transformation to obtain component gradient error values of the specific angle, and the error values are corrected one by one according to the characteristics of each error parameter of the gradient error correction model until all the error values are corrected; and secondly, acquiring angle information of the vector magnetic gradiometer by introducing an attitude heading sensor and rigidly fixing the attitude heading sensor and the vector magnetic gradiometer, then enabling the vector magnetic gradiometer to rotate randomly in space, synchronously acquiring gradient data and angle data of each component in the rotating process, and correcting the residual error amount by establishing a relational expression between the gradient data and the angle data of each component.

The theoretical basis and the working principle of the invention are as follows:

the theoretical basis is as follows:

two sets of vector magnetometers are combined into a vector magnetic gradiometer according to the same direction, when the vector magnetic gradiometer is actually installed, two X component axial directions cannot be completely parallel, similarly, two Y component axial directions and Z component axial directions cannot be completely parallel, so that gradients delta X, delta Y and delta Z of the X component, the Y component and the Z component cannot be zero all the time in the same geomagnetic field environment, when the vector magnetic gradiometer changes the direction, the error value change of each gradient value is irregular, so that the factors of the errors comprise two-axis alignment errors, gradient zero point errors, differential errors and the like, the error factors are fused together and cannot be distinguished, when the vector magnetic gradiometer is aligned to a specific angle, the influence of a single error factor can play a role, the influence of other error factors can be reduced to the negligible degree, and based on the characteristics, a three-axis non-magnetic reference turntable is adopted, and adjusting the vector magnetic gradiometers to preset postures one by one, and then correcting specific error factors until all the error factors are corrected.

After the work operation is finished, a set of attitude heading sensor and the vector magnetic gradiometer are generally and rigidly fixed to obtain three kinds of angle information of a roll angle, a pitch angle, a heading angle and the like of the vector magnetic gradiometer, then the vector magnetic gradiometer rotates randomly in space, gradient data and angle data of each component are synchronously obtained in the rotating process, and the influence of gradient errors of each component caused by the angle is eliminated by establishing a relational expression between the gradient data and the angle data of each component.

The working principle is as follows:

based on the theoretical basis, in order to realize the correction method of the steering difference of the vector magnetic gradiometer, firstly, a certain shaft of the vector magnetic gradiometer is adjusted to a specific direction of a geomagnetic field, usually the vertical direction or the horizontal direction of the geomagnetic field, three component gradient data are recorded through a set of three-shaft non-magnetic reference rotary table, then the shaft is rotated by 180 degrees, the changed three component gradient data are recorded, and a specific error parameter is adjusted according to the recorded change quantity, so that the change quantity is reduced to the minimum. After the operation is finished, the other axis is adjusted to the specific direction of the geomagnetic field, and the process is repeated until all error parameters are adjusted. And then, magnetic gradient data and angle data in the attitude transformation process are synchronously acquired through an attitude heading sensor, a relational expression between each component gradient data and the angle data is established, the component gradient error influence related to the attitude angle is eliminated, and finally the corrected magnetic gradient output is obtained.

The scheme for solving the technical problem is to provide and realize a vector magnetic gradiometer steering difference correction method, and the scheme is as follows:

firstly, establishing a correction model: when the vector magnetic gradiometer steering error is corrected, firstly, the factors causing the steering error need to be found out in principle, and a steering error correction model is established, in the scheme, the factors causing the steering error are confirmed to be two-axis alignment errors, gradient zero point errors and differential errors, and the specific correction model is as follows:

wherein:

X1C, Y1C, Z1C, X2C, Y2C, Z2C respectively represent the original output of the X component, Y component, Z component of the first set of vector magnetometers, the original output of the X component, Y component, Z component of the second set of vector magnetometers;

DeltaX, DeltaY and DeltaZ respectively represent an X-axis gradient value, a Y-axis gradient value and a Z-axis gradient value;

ixbe, iybe and izbe respectively represent X-axis alignment error, Y-axis alignment error and Z-axis alignment error;

ixofs, iyofs and izofs respectively represent an X-axis gradient zero-point error, a Y-axis gradient zero-point error and a Z-axis gradient zero-point error;

ixocxy and ixocxz respectively represent a difference error of the Y axis to the X axis and a difference error of the Z axis to the X axis;

the iyocyx and the iyocyz respectively represent a difference error of an X axis to a Y axis and a difference error of a Z axis to the Y axis;

the izoczy and the izoczx respectively represent the difference error of the Y axis to the Z axis and the difference error of the X axis to the Z axis;

x1CC, Y1CC, and Z1CC represent the corrected outputs of the X, Y, and Z components of the first set of vector magnetometers, respectively.

Introducing a reference turntable: in order to realize the correction work of the model, the vector magnetic gradiometer needs to be rotated to a specific direction and posture as required, in order to obtain accurate posture information, a set of three-axis nonmagnetic reference turntable equipment is introduced, the three-axis nonmagnetic reference turntable equipment can be freely rotated and fixed around three axial directions, the three axial directions are provided with reference scales, the scale precision is less than or equal to 0.1 degree, and the vector magnetic gradiometer can be ensured to be accurately rotated to any vector direction and posture.

Determining a correction method: in order to realize the correction work of the model, the vector magnetic gradiometer is aligned to a specific angle through the spatial rotation of the reference turntable, so that a single parameter of the model plays a determining role in the difference value of the steering difference, the influence of other parameters on the difference value of the steering difference is ignored, the corresponding relation between the single parameter of the model and the steering difference is obtained, and the parameter value is calculated and substituted into the model. And then rotating the vector magnetic gradiometer to other specific angles by adopting the same method, acquiring the corresponding relation between other single parameters and the steering difference in the model, calculating and substituting the parameter values, and completing one by one until all the parameters are calculated.

Correcting the attitude error: in order to realize the correction work of the model, an attitude heading sensor is introduced and is used for acquiring steering difference information related to an attitude angle, the attitude angle information comprises three angles such as a heading angle, a roll angle and a pitch angle, and then a gradient error caused by the attitude angle is eliminated according to a corresponding relation between the steering difference and the attitude angle.

The innovation points of the invention are as follows:

creativity: the method for correcting the steering difference of the vector magnetic gradiometer is provided and realized, and an error correction model is established by determining the error source of the steering difference; the vector magnetic gradiometer is only related to a single error variable under a specific posture, and a correction method for separating and calculating the error variables one by one is determined, so that the correction work is simple and easy to operate, and a complicated correction matrix does not need to be established and complicated operation is not needed;

easy operability: in the process of correcting the steering difference of the vector magnetic gradiometer, the vector magnetic gradiometer needs to be rotated and fixed at a specific angle in a vector space, a set of three-axis nonmagnetic reference rotary table is introduced to realize the easy operability of the work, the reference rotary table can rotate at a full course angle of 360 degrees, rotate at a full roll angle of 360 degrees and rotate at a full pitch angle of +/-90 degrees, the reference scale is less than or equal to 0.1 degree, the vector magnetic gradiometer can be rotated and fixed at any required posture, and compared with manual operation, the rotary table operation is not only accurate and efficient, but also simple and standard;

③ no magnetism: in the process of correcting the steering difference of the vector magnetic gradiometer, the external environment is required not to introduce magnetic abnormal interference, the three-axis non-magnetic reference rotary table is made of a non-magnetic material, and the non-magnetic characteristic of the material ensures that the whole rotary table does not bring extra magnetic abnormal interference to the steering difference correction, so that the non-magnetism of the correction work is ensured;

calculating model parameters: the parameters in the error correction model mainly comprise two-axis alignment errors, gradient zero point errors and differential errors, when the vector magnetic gradiometer rotates to a specific attitude in space, the errors only depend on a single parameter, and the influence of other parameters is ignored. Based on the characteristics, when the model parameters are calculated, the vector magnetic gradiometer is rotated to a specific posture through the rotary table and is fixed, the relation between the gradient error and the corresponding single parameter is obtained and calculated, the specific value of the parameter is obtained and substituted into the model, and then other parameter values are obtained one by one according to the same method until all the parameters are obtained. According to the method, complex and fussy matrix operation is not needed, parameters in the model can be calculated one by one, the parameters have universality, only one time of calculation is needed, and the engineering application has practicability;

correcting the attitude error: and acquiring a course angle, a roll angle and a pitch angle of the vector magnetic gradiometer by introducing a set of attitude and course sensors, acquiring a relational expression of the residual gradient error and the attitude angle after the error correction model parameter is solved, and removing the gradient error related to the attitude angle.

The invention has the beneficial effects that: the method is simple to operate, does not need to establish a complex correction model, can correct each error amount one by one, is simple to calculate, has universality in single operation, does not need to be corrected again during cross-latitude operation, and greatly improves the output precision of each component gradient value. Before the steering difference correction is carried out, when the posture of the magnetic gradiometer is changed, the gradient difference value of each component reaches +/-380.0 nT, after the steering difference correction is finished, the gradient difference value of each component is reduced to be within +/-4.0 nT, the correction error is reduced by 98.95 percent, and the output precision of the vector magnetic gradiometer is greatly improved.

Drawings

FIG. 1 is a schematic diagram of the arrangement of the magnetic components inside a vector magnetic gradiometer.

FIG. 2 is a matrix expression of an error correction model of a vector magnetic gradiometer.

Fig. 3 illustrates step 1 of the calibration model parameter obtaining embodiment.

Fig. 4 illustrates the correction model parameter obtaining embodiment step 2.

Fig. 5 illustrates step 3 of the calibration model parameter obtaining embodiment.

Fig. 6 shows the correction model parameter obtaining embodiment step 4.

Fig. 7 shows the correction model parameter obtaining embodiment step 5.

Fig. 8 shows the correction model parameter obtaining embodiment step 6.

Fig. 9 is a diagram of an embodiment of attitude error correction.

Fig. 10 is an attitude error correction parameter expression.

FIG. 11 is a comparison graph of gradient error of each component before and after correction, which is a total of four graphs from top to bottom, wherein the first graph is a comparison curve before and after correction of the component gradient DeltaX under different postures, the blue graph is a gradient error value of the DeltaX before correction, the error amount is-220 nT-75 nT, the green graph is a gradient error value of the DeltaX after correction, and the error amount is-3.0 nT-2.2 nT; the second curve graph is a comparison curve before and after correction of the component gradient delta Y under different postures, the blue curve is a delta Y gradient error value before correction, the error amount is-370 nT-380 nT, the green curve is a delta Y gradient error value after correction, and the error amount is-4.0 nT-3.0 nT; the third graph is a comparison curve before and after correction of the component gradient delta Z under different postures, the blue curve is a delta Z gradient error value before correction, the error amount is-115 nT-70 nT, the green curve is a delta Z gradient error value after correction, and the error amount is-2.0 nT-3.0 nT; the fourth curve chart is a comparison curve before and after correction of the synthetic total field magnetic gradient under different postures, the blue curve is a synthetic total field magnetic gradient error value before correction, the error amount is-200 nT-220 nT, the green curve is a synthetic total field magnetic gradient error value after correction, and the error amount is-1.8 nT-4.0 nT. The corrected gradient error is reduced from a few hundred nT level to a few nT level, which shows that the correction method has obvious effect.

The reference numbers in the drawings of the specification are as follows:

1. the X1 axis of vector magnetometer 1; 2. the Y1 axis of vector magnetometer 1; 3. the Z1 axis of vector magnetometer 1; 4. the X2 axis of the vector magnetometer 2; 5. the Y2 axis of the vector magnetometer 2; 6. the Z2 axis of vector magnetometer 2; 7. a geomagnetic field direction (B); 8. a vector magnetic gradiometer; 9. an attitude and heading sensor; 10. a Roll shaft Roll; 11. a Pitch axis Pitch; 12. a course axis Yaw; 13. a rigid support frame.

Detailed Description

The invention will be described in detail below with reference to the following drawings: the specific implementation mode comprises two parts: a correction model parameter acquisition implementation; an attitude error correction implementation.

The correction model parameter obtaining embodiment comprises the following steps:

the schematic layout of each component in the vector magnetic gradiometer (8) is shown in fig. 1, and the vector magnetic gradiometer internally comprises two sets of vector gradiometers: the vector magnetometer comprises a vector magnetometer 1 and a vector magnetometer 2 which are rigidly fixed, wherein an X1 axis (1) of the vector magnetometer 1 and an X2 axis (4) of the vector magnetometer 2 are arranged in the same direction, a Y1 axis (2) of the vector magnetometer 1 and a Y2 axis (5) of the vector magnetometer 2 are arranged in the same direction, and a Z1 axis (3) of the vector magnetometer 1 and a Z2 axis (6) of the vector magnetometer 2 are arranged in the same direction. The difference in readings between X1 and X2 is denoted as Δ X, the difference in readings between Y1 and Y2 is denoted as Δ Y, and the difference in readings between Z1 and Z2 is denoted as Δ Z.

The matrix expression of the vector magnetic gradiometer error correction model is shown in fig. 2, in which:

X1C, Y1C, Z1C, X2C, Y2C, Z2C respectively represent the original output of the X component, Y component, Z component of the first set of vector magnetometers, the original output of the X component, Y component, Z component of the second set of vector magnetometers;

DeltaX, DeltaY and DeltaZ respectively represent an X-axis gradient value, a Y-axis gradient value and a Z-axis gradient value;

ixbe, iybe and izbe respectively represent X-axis alignment error, Y-axis alignment error and Z-axis alignment error;

ixofs, iyofs and izofs respectively represent an X-axis gradient zero-point error, a Y-axis gradient zero-point error and a Z-axis gradient zero-point error;

ixocxy and ixocxz respectively represent a difference error of the Y axis to the X axis and a difference error of the Z axis to the X axis;

the iyocyx and the iyocyz respectively represent a difference error of an X axis to a Y axis and a difference error of a Z axis to the Y axis;

the izoczy and the izoczx respectively represent the difference error of the Y axis to the Z axis and the difference error of the X axis to the Z axis;

x1CC, Y1CC, and Z1CC represent the corrected outputs of the X, Y, and Z components of the first set of vector magnetometers, respectively.

Calibration model parameter acquisition embodiment step 1 as shown in fig. 3, the turntable was adjusted to change the attitude of the vector magnetic gradiometer (8), the output of Z1(3) was observed, the output of Z1(3) was made zero, and the direction was perpendicular to the geomagnetic field direction (B) (7); observing the output of the X1(1), so that the output of the X1(1) is zero and is vertical to the direction (B) (7) of the geomagnetic field; the output of Y1(2) was observed, and the forward direction of the output of Y1(2) was maximized, and the forward direction was parallel to the geomagnetic field direction (B) (7), and at this time, the vector magnetic gradiometer (8) was fixed, and the Δ X and Δ Y readings were recorded. And adjusting the turntable to rotate the vector magnetic gradiometer (8) by 180 degrees around the Z1 axis, and recording the reading values of delta X and delta Y after rotation.

The first step is that the median value of the two values is obtained according to the two recorded values of the delta X, and the parameter value of the difference error ixocxy is calculated to enable the value of the delta X to be equal to the median value of the two values; secondly, calculating a parameter value of a gradient zero error ixofs to enable a value of delta X to be equal to zero; and thirdly, acquiring a median value of the two recorded values according to the two recorded values of the delta Y, and calculating a parameter value of the alignment error iybe to enable the value of the delta Y to be equal to the median value of the two recorded values.

In the previous step, as shown in fig. 4, in step 2 of the corrected model parameter acquisition embodiment, the turntable is adjusted to change the posture of the vector magnetic gradiometer (8), the output of Z1(3) is observed, the output of Z1(3) is made to be maximum in the reverse direction, and the direction of the geomagnetic field is made to be parallel to the reverse direction (B) (7); observing the output of the X1(1), so that the output of the X1(1) is zero and is vertical to the direction (B) (7) of the geomagnetic field; the output of Y1(2) was observed to be zero at Y1(2) perpendicular to the direction of the geomagnetic field (B) (7), at which time the vector magnetic gradiometer (8) was held stationary and the Δ X and Δ Z readings were recorded. And adjusting the turntable to rotate the vector magnetic gradiometer (8) by 180 degrees around the Y1 axis, and recording the reading values of delta X and delta Z after rotation.

The first step is that the median value of the two values is obtained according to the two recorded values of the delta X, and the parameter value of the difference error ixocxz is calculated to enable the value of the delta X to be equal to the median value of the two values; secondly, calculating a parameter value of a gradient zero error ixofs to enable a value of delta X to be equal to zero; and thirdly, acquiring the median value of the two recorded values according to the two recorded values of the delta Z, and calculating the parameter value of the alignment error izbe to enable the value of the delta Z to be equal to the median value of the two recorded values.

In the previous step, step 3 of the corrected model parameter acquisition embodiment is, as shown in fig. 5, adjusting the turntable to change the posture of the vector magnetic gradiometer (8), observing the output of Z1(3), and making the output of Z1(3) zero, perpendicular to the geomagnetic field direction (B) (7); observing the output of the X1(1), and enabling the X1(1) to output the maximum positive direction which is parallel to the geomagnetic field direction (B) (7); the output of Y1(2) was observed to be zero at Y1(2) perpendicular to the direction of the geomagnetic field (B) (7), at which time the vector magnetic gradiometer (8) was held stationary and the Δ X and Δ Y readings were recorded. And adjusting the turntable to rotate the vector magnetic gradiometer (8) by 180 degrees around the Z1 axis, and recording the reading values of delta X and delta Y after rotation.

The first step is that the median value of the two values is obtained according to the two recorded values of the delta Y, and the parameter value of the differential error iyocyx is calculated to enable the value of the delta Y to be equal to the median value of the two values; secondly, calculating a parameter value of a gradient zero error iyofs to enable a value of delta Y to be equal to zero; and thirdly, acquiring a median value of the two recorded values according to the two recorded values of the delta X, and calculating a parameter value of the alignment error ixbe so that the value of the delta X is equal to the median value of the two recorded values.

In the previous step, the calibration model parameter acquisition embodiment step 4, as shown in fig. 6, adjusts the turntable to change the attitude of the vector magnetic gradiometer (8), observes the output of Z1(3), maximizes the output of Z1(3) in the opposite direction, and reverses the direction of the parallel geomagnetic field (B) (7); observing the output of the X1(1), so that the output of the X1(1) is zero and is vertical to the direction (B) (7) of the geomagnetic field; the output of Y1(2) was observed to be zero at Y1(2) perpendicular to the direction of the geomagnetic field (B) (7), at which time the vector magnetic gradiometer (8) was held stationary and the readings of Δ Y and Δ Z were recorded. And adjusting the turntable to rotate the vector magnetic gradiometer (8) by 180 degrees around an X1 axis, and recording the reading values of delta Y and delta Z after rotation.

The first step is that the median value of the two values is obtained according to the two recorded values of the delta Y, and the parameter value of the differential error iyocyz is calculated to enable the value of the delta Y to be equal to the median value of the two values; secondly, calculating a parameter value of a gradient zero error iyofs to enable a value of delta Y to be equal to zero; and thirdly, acquiring the median value of the two recorded values according to the two recorded values of the delta Z, and calculating the parameter value of the alignment error izbe to enable the value of the delta Z to be equal to the median value of the two recorded values.

In the previous step, step 5 of the corrected model parameter acquisition embodiment is, as shown in fig. 7, adjusting the turntable to change the posture of the vector magnetic gradiometer (8), observing the output of Z1(3), and making the output of Z1(3) zero, perpendicular to the geomagnetic field direction (B) (7); observing the output of the X1(1), so that the output of the X1(1) is the maximum reverse, and the direction (B) (7) of the geomagnetic field is parallel reversely; the output of Y1(2) was observed to be zero at Y1(2) perpendicular to the direction of the geomagnetic field (B) (7), at which time the vector magnetic gradiometer (8) was held stationary and the Δ Z and Δ X readings were recorded. And adjusting the turntable to rotate the vector magnetic gradiometer (8) by 180 degrees around the Y1 axis, and recording the reading values of delta Z and delta X after rotation.

The first step is that the median value of the two values is obtained according to the two recorded values of the delta Z, and the parameter value of the difference error izoczx is calculated to enable the value of the delta Z to be equal to the median value of the two values; secondly, calculating a parameter value of a gradient zero point error izofs to enable a value of delta Z to be equal to zero; and thirdly, acquiring a median value of the two recorded values according to the two recorded values of the delta X, and calculating a parameter value of the alignment error ixbe so that the value of the delta X is equal to the median value of the two recorded values.

In the previous step, as shown in fig. 8, in the corrected model parameter acquisition embodiment step 6, the turntable is adjusted to change the posture of the vector magnetic gradiometer (8), the output of Z1(3) is observed, the output of Z1(3) is zero, and the direction is perpendicular to the geomagnetic field direction (B) (7); observing the output of the X1(1), so that the output of the X1(1) is zero and is vertical to the direction (B) (7) of the geomagnetic field; the output of Y1(2) was observed to be the maximum in the opposite direction of the output of Y1(2), and the direction of the earth magnetic field (B) (7) was reversed, and at this time, the vector magnetic gradiometer (8) was fixed and the Δ Z and Δ Y readings were recorded. And adjusting the turntable to rotate the vector magnetic gradiometer (8) by 180 degrees around an X1 axis, and recording the reading values of delta Z and delta Y after rotation.

The first step is that the median value of the two values is obtained according to the two recorded values of the delta Z, and the parameter value of the difference error izoczy is calculated to enable the value of the delta Z to be equal to the median value of the two values; secondly, calculating a parameter value of a gradient zero point error izofs to enable a value of delta Z to be equal to zero; and thirdly, acquiring a median value of the two recorded values according to the two recorded values of the delta Y, and calculating a parameter value of the alignment error iybe to enable the value of the delta Y to be equal to the median value of the two recorded values.

So far, the calculation of all the parameters is completed through the six steps.

As shown in fig. 9, the attitude heading sensor 9 and the vector magnetic gradiometer 8 are fixed to the rigid support frame 13, and the attitude of the rigid support frame 13 is changed to synchronously change the attitude of the vector magnetic gradiometer 8 and the attitude heading sensor 9. When the attitude error correction is implemented, according to the reading of the heading axis Yaw (12) of the attitude heading sensor 9, the heading angles are respectively aligned to eight azimuth angles of 0 degrees, 45 degrees, 90 degrees, 135 degrees, 180 degrees, 225 degrees, 270 degrees and 315 degrees, the angles do not need to be aligned strictly, in each heading angle, the posture of the rigid support frame 13 is changed, the Roll axis Roll (10) of the attitude heading sensor 9 rotates for a circle, the Pitch axis Pitch (11) rotates for a circle, in the rotating process, delta X, delta Y, delta Z, the heading angle Yaw, the Pitch angle Pitch and the Roll angle Roll are synchronously recorded, then the next heading angle is rotated, and the operations are repeated until all the heading angles are rotated.

The relational expressions of the delta X, the delta Y, the delta Z, the course angle Yaw, the Pitch angle Pitch and the Roll angle Roll are shown in FIG. 10, wherein:

DeltaX, DeltaY and DeltaZ respectively represent an X component gradient value, a Y component gradient value and a Z component gradient value before attitude error correction;

yaw, Pitch and Roll respectively represent output values of a course angle, a Pitch angle and a Roll angle;

a1, b1, c1, a2, b2, c2, a3, b3 and c3 are nine parameters of a3 × 3 relation matrix;

Δ Xa, Δ Ya, and Δ Za represent the X-component gradient value, Y-component gradient value, and Z-component gradient value after the attitude error correction, respectively.

And acquiring nine parameters of the relation matrix by a least square method according to the recorded data.

By the above steps, the attitude error correction implementation is completed.

It should be understood that equivalent substitutions and changes to the technical solution and the inventive concept of the present invention should be made by those skilled in the art to the protection scope of the appended claims.