Method for optimizing fluctuating magnetic field output drift of hemispherical resonator gyroscope

Abstract

A fluctuating magnetic field output drift optimization method for a hemispherical resonator gyroscope comprises the following steps: step 1, primary optimization: calculating a quaternion as a state value of Kalman filtering by using a Gauss-Newton iteration method through measured magnetic field intensity and acceleration, and obtaining a primary optimization estimated value of the quaternion according to the state value; step 2, conversion correction; and step 3, secondary optimization: taking the corrected quaternion as a state quantity, then taking the quaternion obtained by the first Kalman filtering as an observed quantity, carrying out second Kalman filtering, and completing secondary optimization of the quaternion, in the above method, obtaining an optimal estimated quaternion through Kalman filtering, and finally obtaining an optimal estimated quaternion through Kalman filtering. And then compensating the drift of the gyroscope by using a complementary filtering algorithm to obtain a corrected quaternion, and carrying out second optimization on the quaternion through Kalman filtering by using the obtained quaternion and the optimal estimated quaternion obtained by Kalman filtering, thereby reducing the output drift of the hemispherical resonator gyroscope through the optimization method.

Description

Technical Field

[0001] The invention relates to the technical field of gyroscopes, and in particular to a method for optimizing the output drift of a fluctuating magnetic field of a hemispherical resonant gyroscope. Background Art

[0002] Hemispherical Resonator Gyroscope (HRG) is a high-precision inertial navigation instrument that uses a resonant structure to measure angular velocity. In HRG, magnetic materials are usually used to form a resonant body. When a fluctuating magnetic field is introduced, it will affect the magnetization state of these magnetic materials. In addition, changes in the magnetic field due to the complex external environment will lead to changes in the magnetic moment in the resonant body, and further cause the output signal to drift.

[0003] In the prior art, there are many fusion algorithms for the problems of gyroscopes being susceptible to interference and gyroscope integral drift, such as Kalman filtering, unscented Kalman filtering (UKF), extended Kalman filtering (EKF), etc. These filtering methods require the establishment of accurate state equations and observation equations. There is also a filtering algorithm that is extended on the basis of complementary filtering, such as classical complementary filtering, complementary filtering algorithm based on gradient descent method, etc., but the accuracy of this filtering algorithm is not high. Even after calibration and filtering, the gyroscope still has a certain degree of output drift, causing the attitude azimuth to gradually deviate from the true value over time.

[0004] Therefore, there is an urgent need for an optimization method that can solve the output drift optimization caused by fluctuating magnetic field interference while solving the divergence of gyroscope azimuth error. Summary of the invention

[0005] The present invention aims to provide a method for optimizing the drift of a fluctuating magnetic field output of a hemispherical resonant gyroscope, which is used to solve the problem of interference of the fluctuating magnetic field while solving the divergence of the azimuth error of the gyroscope.

[0006] The present invention is achieved through the following technical solutions: A method for optimizing the drift of the fluctuating magnetic field output of a hemispherical resonant gyroscope comprises the following steps: step 1, primary optimization, using the quaternion calculated by the Gauss-Newton iteration method through the measured magnetic field intensity and acceleration as the state value of the Kalman filter, and then obtaining the primary optimization estimation value of the quaternion according to the state value; step 2, conversion correction, after the completion of step 1, the gravity acceleration and the geomagnetic component in the geographic coordinate system are converted to the carrier coordinate system through the conversion matrix between the carrier coordinate system and the geographic coordinate system, and then the magnetic field intensity and acceleration measured in the carrier coordinate system are subjected to the vector product operation, The sum of the vector products after proportional integral low-pass filtering is fused with the angular velocity measured by the gyroscope to obtain the angular velocity after compensation of the gyroscope, and the correction quaternion is obtained by using the complementary filtering algorithm and the fourth-order Runge-Kutta method; Step 3, secondary optimization, after the completion of step 2, the correction quaternion obtained by using the complementary filtering algorithm and the fourth-order Runge-Kutta method is used as the state quantity, and the first optimized estimate of the quaternion obtained by the first Kalman filtering in step 1 is used as the observation quantity for the second Kalman filtering, completing the secondary optimization of the quaternion, and then the attitude matrix is ​​established through the quaternion to obtain the optimized output.

[0007] It should be noted that, in view of the problem of output drift caused by fluctuating magnetic fields in existing hemispherical resonant gyroscopes, a method for optimizing the output drift of fluctuating magnetic fields of hemispherical resonant gyroscopes is proposed, which is used to solve the interference problem of fluctuating magnetic fields while solving the azimuth error divergence of the gyroscope. In general, the hemispherical resonant gyroscope is subjected to Kalman filtering to obtain the optimal estimated quaternion, and then the complementary filtering algorithm is used to compensate for the drift of the gyroscope to obtain the corrected quaternion. The quaternion obtained this time and the optimal estimated quaternion obtained by Kalman filtering are used again by Kalman filtering to optimize the quaternion for the second time, and then the output drift of the hemispherical resonant gyroscope is reduced by the optimization method.

[0008] Compared with the prior art, the present invention has the following advantages and beneficial effects: 1. The present invention obtains the optimal estimated quaternion of the hemispherical resonant gyroscope through Kalman filtering, and then uses a complementary filtering algorithm to compensate for the drift of the gyroscope to obtain a corrected quaternion. The quaternion obtained this time and the optimal estimated quaternion obtained by Kalman filtering are used again through Kalman filtering to optimize the quaternion for the second time, thereby achieving the reduction of the output drift of the hemispherical resonant gyroscope through the optimization method; 2. The Kalman filter of the present invention calculates a gain matrix, which reflects the influence of the observation on the state estimation. The size of the gain matrix depends on the variance of the observation noise and the size of the state prediction error covariance. Through this gain matrix, the filter can reasonably fuse the observation data and the prediction data to obtain a more accurate state estimation; 3. The conversion process of the present invention can accurately convert the gravitational acceleration and geomagnetic components in the geographic coordinate system to the carrier coordinate system, which enables the measurement data of the gyroscope to be directly used for navigation and positioning calculations without the need for additional coordinate conversion steps. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of the present invention, and do not constitute a limitation of the embodiments of the present invention. In the drawings: Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0010] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and drawings. The schematic implementation modes and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention. It should be noted that the present invention is already in the actual development and use stage.

[0011] In order to make the technical solution of the present invention clearer, the professional terms involved in the present invention are explained here; The term "Gauss-Newton iteration method" refers to an iterative method for solving nonlinear least squares problems. In the present invention, it is specifically used to solve quaternions through iterative calculations based on measured magnetic field strength and acceleration data. Quaternions are a mathematical tool for representing rotations in three-dimensional space. The Gauss-Newton iteration method is used in this process to optimize the value of the quaternion so that it better conforms to the actual measurement data. The term "Kalman filter" refers to a recursive filter that estimates the state of a dynamic system from noisy measurements. In the present invention, Kalman filtering is also used to process gyroscope and accelerometer data to estimate and optimize quaternions, thereby reducing the effects of measurement noise and system errors. Through Kalman filtering, a more accurate and stable state estimate can be obtained; The term "quaternion" refers to a mathematical tool for representing and calculating rotation in three-dimensional space, which consists of a real part and three imaginary parts. In the present invention, quaternion is used to describe and implement the posture change of the carrier in three-dimensional space, which can avoid the universal lock problem and is more suitable for representing continuous three-dimensional rotation than Euler angle; The term "geomagnetic component" refers to the vector components of the earth's magnetic field in different directions. In the present invention, the geomagnetic component refers to the north component, east component and vertical component of the earth's magnetic field measured in a geographic coordinate system, and these components are used to determine the direction and attitude; The term "carrier coordinate system" refers to a coordinate system fixed to the object under study. In the present invention, the carrier coordinate system refers to a coordinate system fixed to the hemispherical resonant gyroscope, which is used to measure and represent the movement and output change of the gyroscope relative to the carrier itself; The term "fourth-order Runge-Kutta method" refers to a numerical integration method for finding approximate solutions to ordinary differential equations. In the present invention, the fourth-order Runge-Kutta method is used to calculate a more accurate output based on the angular velocity measured by the gyroscope and the correction quaternion obtained by the complementary filtering algorithm; The term "complementary filtering algorithm" refers to an algorithm that combines sensor data of different frequency characteristics. In the present invention, the complementary filtering algorithm is used to combine the data of the gyroscope and the magnetometer to compensate for the drift of the gyroscope and correct the interference of the magnetometer to obtain a more accurate output estimate; The term "attitude matrix" refers to a matrix used to represent the direction of an object in three-dimensional space. In the present invention, the attitude matrix is ​​a 3x3 rotation matrix converted from quaternions and is used to describe the direction and orientation of a carrier in three-dimensional space.

[0012] Embodiment 1: As attached Figure 1 As shown, a method for optimizing the drift of a fluctuating magnetic field output of a hemispherical resonant gyroscope comprises the following steps: Step 1, primary optimization, using the measured magnetic field strength and acceleration to solve the quaternion using the Gauss-Newton iteration method as the state value of the Kalman filter, and then obtaining a primary optimization estimate of the quaternion based on the state value; Step 2, conversion correction, after completing step 1, the gravity acceleration and geomagnetic component in the geographic coordinate system are converted to the carrier coordinate system through the conversion matrix between the carrier coordinate system and the geographic coordinate system, and then the magnetic field strength and acceleration measured in the carrier coordinate system are vector-producted, and the sum of the vector products is subjected to proportional integral low-pass filtering and the result is fused with the angular velocity measured by the gyroscope to obtain the angular velocity after compensation of the gyroscope, and the correction quaternion is obtained by using the complementary filtering algorithm and the fourth-order Runge-Kutta method; Step 3, secondary optimization. After step 2 is completed, the corrected quaternion obtained by the complementary filtering algorithm and the fourth-order Runge-Kutta method is used as the state quantity, and the first optimized estimate of the quaternion obtained by the first Kalman filtering in step 1 is used as the observation quantity for the second Kalman filtering to complete the secondary optimization of the quaternion, and then the attitude matrix is ​​established through the quaternion to obtain the optimized output.

[0013] It should be noted that the filtering method in the prior art not only needs to establish accurate state equations and observation equations, but also needs to be expanded on the basis of complementary filtering, such as classical complementary filtering, complementary filtering algorithm based on gradient descent method, etc., and the accuracy of the above algorithms is not high. Even after calibration and filtering, the gyroscope still has a certain degree of output drift, causing the output of the gyroscope to gradually deviate from the true value over time. Based on the above problems, a method for optimizing the output drift of a hemispherical resonant gyroscope fluctuating magnetic field is proposed. The optimal estimated quaternion is obtained by passing the hemispherical resonant gyroscope through Kalman filtering, and then the drift of the gyroscope is compensated by using a complementary filtering algorithm to obtain a corrected quaternion. The quaternion obtained this time and the optimal estimated quaternion obtained by Kalman filtering are used again by Kalman filtering to optimize the quaternion for the second time, and then the output drift of the hemispherical resonant gyroscope is reduced by the optimization method.

[0014] It should also be noted that the essence of the geomagnetic component measured in this embodiment is the magnetic field intensity; in addition, the complementary filtering algorithm adopted mainly adjusts the frequency characteristics through the proportional link, and integrates the high-frequency characteristics of the gyroscope and the low-frequency characteristics of the matching compass module to reduce errors. The fourth-order Runge-Kutta method is a numerical integration method for solving ordinary differential equations. More specifically, it is used to perform attitude calculation based on the compensated angular velocity. Through the combination of the two methods, the technical solution in this embodiment can effectively solve the integral drift problem and the magnetic field interference problem of the gyroscope, while maintaining the high accuracy and stability of attitude estimation.

[0015] More specifically, the Gauss-Newton iteration method is used to solve the quaternion, which is a mathematical tool used to represent rotation in three-dimensional space. It consists of a real part and three imaginary parts. Compared with the Euler angle, the quaternion has better numerical stability, especially when dealing with large-angle rotations, and can avoid the gimbal lock problem.

[0016] The Kalman filter used in the present invention is a recursive filter that can estimate the state of a dynamic system from measurements containing noise. In an optimization, the Kalman filter uses the quaternion calculated by the Gauss-Newton iterative method as the state value, combines the dynamic model and observation model of the system, and estimates the true state of the system through two steps of prediction and update. Among them, the state equation describes the evolution of the system state over time, and the observation equation describes how to obtain observation data from the system state.

[0017] In addition, the Kalman filter predicts the state at the next moment based on the system's state transfer matrix and control input. The state transfer matrix describes the transfer relationship of the system state from one moment to another. In the update step, the filter uses the statistical characteristics of the observed data and observation noise to correct the predicted state, thereby obtaining a more accurate state estimate.

[0018] For the conversion correction process: The purpose of the conversion correction step is to convert the gravity acceleration and geomagnetic components in the geographic coordinate system to the carrier coordinate system so as to more accurately reflect the actual working environment of the gyroscope.

[0019] First, determine the transformation matrix between the geographic coordinate system and the carrier coordinate system. The transformation matrix is ​​determined by the carrier's posture and can be represented by quaternions. Quaternions can easily convert between coordinate systems, thereby avoiding the gimbal lock problem in three-dimensional rotation.

[0020] Next, the gravitational acceleration and geomagnetic components in the geographic coordinate system are converted to the carrier coordinate system through the transformation matrix to obtain more accurate gravity and geomagnetic measurements. Subsequently, the magnetic field strength and acceleration measured in the carrier coordinate system need to be vector-producted. The result of the vector product is a new vector that is perpendicular to the original two vectors and its size is proportional to the area of ​​the parallelogram formed by the two vectors.

[0021] Finally, the result of the vector product is fused with the angular velocity measured by the gyroscope. Through the above process, the output of the gyroscope compensation can be obtained, and this output result is corrected and can more accurately reflect the actual motion state of the gyroscope; then, the compensated angular velocity is numerically integrated using the fourth-order Runge-Kutta method to obtain the corrected quaternion, and then the accuracy of the integration is improved by considering four different slopes in each integration step.

[0022] For the second optimization: The secondary optimization process is the last step in the entire optimization process. Its purpose is to further improve the accuracy of the quaternion, thereby reducing the output drift of the gyroscope. This step also uses the Kalman filter algorithm, but unlike the primary optimization, it uses the corrected quaternion obtained by the complementary filtering algorithm as the state quantity, and uses the quaternion obtained by the first Kalman filter as the observation quantity.

[0023] As time accumulates, its measurements will produce errors due to drift, while the magnetometer can provide a stable reference to the earth's magnetic field, but it is not able to respond to fast-changing dynamics. The complementary filter algorithm can provide a more accurate angular position estimate by combining the advantages of these two sensors. In the secondary optimization, the Kalman filter again uses the state transfer matrix and the observation model to predict and update the quaternion. Since the state quantity at this time is based on the corrected quaternion obtained by the complementary filter algorithm, it already includes compensation for the gyroscope drift. At the same time, the observation quantity is based on the quaternion obtained by the first Kalman filter, which provides additional constraints for the filter and helps to further reduce the error.

[0024] In the update step, the Kalman filter calculates a gain matrix, which reflects the influence of the observation on the state estimation. The size of the gain matrix depends on the variance of the observation noise and the size of the state prediction error covariance. Through this gain matrix, the filter can reasonably fuse the observation data and the prediction data to obtain a more accurate state estimate. Finally, the attitude matrix is ​​established through quaternions to obtain the optimized output. This output is optimized twice, which not only takes into account the dynamic characteristics of the system, but also the error characteristics of the sensor. Therefore, it can more accurately reflect the actual motion state of the gyroscope, thereby significantly reducing the output drift.

[0025] It should be noted that the magnetic field strength and acceleration measured in step 1 are update quantities with time as reference, and the state equation of the Kalman filter satisfies: ;in, Indicates the moment; Indicates time The state vector of Represents the state transfer matrix, describing the transition from time To time state transfer; express The state vector of represents the gain matrix of the observation vector; Indicates time The observation vector of represents the observation noise; represents the control matrix; represents process noise.

[0026] It should also be noted that, first, the magnetic field strength and acceleration data are collected in real time through high-precision sensors; second, the collected raw data usually contains noise and interference, and is pre-processed through filters (such as low-pass filters) to reduce the impact of high-frequency noise; finally, in each sampling cycle, the magnetic field strength and acceleration data are updated to reflect the latest measurement results. By using the update amount based on the time reference, it can be ensured that the Kalman filter always performs state estimation based on the latest measurement data, which helps to improve the accuracy of state estimation and thus reduce the output drift of the gyroscope. The above means solve the problem of reduced measurement accuracy of the gyroscope in a dynamic environment. In a dynamic environment, the magnetic field strength and acceleration may change rapidly. If the filter cannot update these measurements in time, it may cause a lag in state estimation, which in turn affects the measurement accuracy of the gyroscope.

[0027] For the state equation: The state equation of the Kalman filter can realize the accurate estimation of the gyroscope state. Even in the presence of noise and interference, a more accurate and stable state estimate can be obtained through Kalman filtering, thereby reducing the impact of measurement noise and system errors. It also solves the problem of accurately estimating the gyroscope state in a dynamic environment. In a dynamic environment, due to the influence of external interference and system noise, the gyroscope's measurement data may contain large errors. The Kalman filter can effectively filter out these errors by combining the system's dynamic model and observation model, providing a more accurate state estimate.

[0028] It should be noted that the state transfer matrix of the state value in step 1 satisfies: ; The measurement process in step 1 is is the sampling period and satisfies: , represents the angular error of the gyroscope on the X-axis, represents the angular velocity around the X axis; , represents the angular error of the gyroscope on the Y axis, represents the angular velocity around the Y axis; , represents the angular error of the gyroscope on the Z axis, Represents the angular velocity around the Z axis; , represents the initial output angle error of the gyroscope.

[0029] It should also be noted that the physical model of the gyroscope's behavior in a fluctuating magnetic field is established by taking into account the gyroscope's geometric structure, material properties, and dynamic changes in the magnetic field. Within the framework of the Kalman filter, the state transfer matrix is ​​used to predict the state of the system at a certain moment in the future. Based on the state information at the current moment, in the prediction step of the Kalman filter, the state transfer matrix is ​​used to predict the state at the current moment based on the state estimate and control input at the previous moment; the state transfer matrix is ​​also used to predict the error covariance matrix; in the update step of the Kalman filter, the information of the state transfer matrix is ​​used to calculate the Kalman gain; by optimizing the state transfer matrix, the performance of the Kalman filter can be improved, thereby reducing the output drift of the hemispherical resonant gyroscope and improving its measurement accuracy.

[0030] It should be noted that in step 2, the conversion process of the gravitational acceleration and geomagnetic component in the geographic coordinate system to the carrier coordinate system satisfies: ; ; Represents the measured value of gravitational acceleration along the x direction; Represents the measured value of gravity acceleration along the y direction; Represents the measured value of gravity acceleration along the z direction; represents the transformation matrix; represents the acceleration due to gravity; Represents the pose matrix.

[0031] It should also be noted that through the above conversion process, the gravity acceleration and geomagnetic component in the geographic coordinate system can be accurately converted to the carrier coordinate system. This allows the gyroscope's measurement data to be directly used for navigation and positioning calculations without the need for additional coordinate conversion steps.

[0032] It should be noted that the calculation process of the vector product in step 2 satisfies: ; ;in, Respectively represent the geomagnetic component in the x direction, the geomagnetic component in the y direction, and the geomagnetic component in the z direction; They represent the measured geomagnetic component in the x direction, the geomagnetic component in the y direction, and the geomagnetic component in the z direction respectively.

[0033] It should also be noted that the geomagnetic components in the geographic coordinate system are obtained through the geomagnetic sensor, which usually includes the north component, the east component and the vertical component; the magnetic field strength in the carrier coordinate system is measured by the magnetic field sensor, and these measurements may be affected by many factors, including changes in the ambient magnetic field and the noise of the sensor itself; the geomagnetic components in the geographic coordinate system are converted to the carrier coordinate system; after obtaining the geomagnetic components and magnetic field strength in the carrier coordinate system, the vector product of the two vectors can be calculated. This step can be implemented using a standard vector operation library or through a custom function. Changes in the geomagnetic field will interfere with the output of the gyroscope. By calculating the vector product, a vector describing this interference can be obtained, so as to correct the output of the gyroscope. In a dynamic environment, the attitude of the carrier will continue to change, thereby affecting the relative direction of the geomagnetic field and the magnetic field strength. The operation of the vector product can adapt to this change and ensure the accuracy of the correction process.

[0034] It should be noted that the fusion process in step 2 satisfies: ;in, Represent quaternions respectively.

[0035] It should also be noted that the quaternion is initialized according to the initial measurement values ​​of the gyroscope and the magnetic field sensor. During the operation of the system, the gyroscope and the magnetic field sensor will continue to provide new measurement data. These data need to be converted into quaternions in real time and used to update the current state of the system; the fusion process involves combining the angular velocity measured by the gyroscope and the magnetic field strength measured by the magnetic field sensor through quaternions. After the fusion process is completed, the quaternion is updated according to the fusion result. Through the quaternion fusion process, the gyroscope output can be accurately corrected. This process can effectively reduce the errors caused by gyroscope drift or magnetic field interference, and improve the navigation and positioning accuracy of the system.

[0036] It should be noted that in step 2, the quaternion correction process also satisfies: .

[0037] It should also be noted that the sources of error in the gyroscope output are analyzed, including zero bias, scale factor error, nonlinear error, etc. These errors will affect the accuracy of the quaternion. Based on the error analysis, an error compensation model is established, and an algorithm for updating the quaternion is developed to incorporate error compensation. The corrected quaternion is fused with other sensor data (such as accelerometers and magnetometers) to further improve the robustness and accuracy of the system. Through the quaternion correction process, the measurement accuracy of the HRG can be significantly improved, and the drift in long-term operation can be reduced, thereby improving the reliability of the navigation system.

[0038] It should be noted that the observed quantity satisfy: It should also be noted that in a dynamic environment, the output of the gyroscope may be affected by a variety of interferences. By using high-precision sensors and updating the observed quantities in real time, the system can more accurately capture key parameters such as angular velocity, acceleration, and geomagnetic components, thereby improving the overall measurement accuracy.

[0039] It should be noted that the complementary filtering algorithm in step 3 includes adjusting the frequency characteristics through a proportional link. Adjusting the frequency characteristics through a proportional link provides an effective technical solution to improve the output drift optimization capability of a hemispherical resonant gyroscope in a fluctuating magnetic field. By analyzing the frequency characteristics of the gyroscope and accelerometer and designing a suitable proportional link, the system can quickly respond to external changes in a dynamic environment, enhance the accuracy of data fusion, reduce output drift, and improve the robustness of the system.

[0040] It should be noted that the secondary optimization process satisfies: ;in: represents the measured gravity component; Represents the gravity component after the operation transformation; Indicates gravity correction error; Indicates the measured geomagnetic component; Represents the geomagnetic component after computational transformation; Indicates the geomagnetic correction error; represents the quadratic optimization error; represents the proportionality coefficient; represents the integral coefficient; represents the secondary optimization amount.

[0041] It should also be noted that the measurement accuracy and stability of the system are not only improved, but also the long-term drift is reduced and the accuracy of multi-sensor data fusion is improved. Through the refined system dynamic model, enhanced observation model, in-depth error compensation and optimized data fusion strategy, the secondary optimization process ensures the high performance of the system in complex environments. At the same time, in the face of technical challenges such as algorithm complexity, computing resource limitations and real-time requirements, through solutions such as algorithm optimization, hardware acceleration and parallel computing, the secondary optimization process can achieve efficient operation in resource-constrained systems and meet the requirements of modern navigation and positioning systems for high accuracy and high stability.

[0042] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for optimizing the drift of a fluctuating magnetic field output of a hemispherical resonant gyroscope, characterized in that: The following steps are involved: Step 1, primary optimization, using the measured magnetic field strength and acceleration to solve the quaternion using the Gauss-Newton iteration method as the state value of the Kalman filter, and then obtaining a primary optimization estimate of the quaternion based on the state value; Step 2, conversion correction, after completing step 1, the gravity acceleration and geomagnetic component in the geographic coordinate system are converted to the carrier coordinate system through the conversion matrix between the carrier coordinate system and the geographic coordinate system, and then the magnetic field strength and acceleration measured in the carrier coordinate system are vector-producted, and the sum of the vector products is subjected to proportional integral low-pass filtering and the result is fused with the angular velocity measured by the gyroscope to obtain the angular velocity after compensation of the gyroscope, and the correction quaternion is obtained by using the complementary filtering algorithm and the fourth-order Runge-Kutta method; Step 3, secondary optimization. After step 2 is completed, the corrected quaternion obtained by the complementary filtering algorithm and the fourth-order Runge-Kutta method is used as the state quantity, and the first optimized estimate of the quaternion obtained by the first Kalman filtering in step 1 is used as the observation quantity for the second Kalman filtering to complete the secondary optimization of the quaternion, and then the attitude matrix is ​​established through the quaternion to obtain the optimized output.

2. The method for optimizing the output drift of a fluctuating magnetic field of a hemispherical resonator gyroscope according to claim 1, characterized in that: The magnetic field strength and acceleration measured in step 1 are the update quantities with time as reference, and the state equation of the Kalman filter satisfies: ;in, Indicates the moment; Indicates time The state vector of Represents the state transfer matrix, describing the transition from time To time state transfer; express The state vector of represents the gain matrix of the observation vector; Indicates time The observation vector of represents the observation noise; represents the control matrix; represents process noise.

3. The method for optimizing the output drift of a hemispherical resonator gyroscope fluctuating magnetic field according to claim 2, characterized in that: The state transfer matrix of the state value in step 1 satisfies: ； The measurement process in step 1 is is the sampling period and satisfies: , represents the angle error of the gyroscope on the X axis, represents the angular velocity around the X axis; , represents the angular error of the gyroscope on the Y axis, represents the angular velocity around the Y axis; , represents the angular error of the gyroscope on the Z axis, Represents the angular velocity around the Z axis; , represents the initial output angle error of the gyroscope.

4. The method for optimizing the output drift of a hemispherical resonant gyroscope fluctuating magnetic field according to claim 1, characterized in that: In step 2, the conversion process of the gravitational acceleration and geomagnetic component in the geographic coordinate system to the carrier coordinate system satisfies: ； ； Represents the measured value of gravitational acceleration along the x direction; Represents the measured value of gravity acceleration along the y direction; Represents the measured value of gravity acceleration along the z direction; represents the transformation matrix; represents the acceleration due to gravity; Represents the pose matrix.

5. The method for optimizing the fluctuating magnetic field output drift of a hemispherical resonant gyroscope according to claim 4, characterized in that: The calculation process of the vector product in step 2 satisfies: ； ； in, Respectively represent the geomagnetic component in the x direction, the geomagnetic component in the y direction, and the geomagnetic component in the z direction; They represent the measured geomagnetic component in the x direction, the geomagnetic component in the y direction, and the geomagnetic component in the z direction respectively.

6. The method for optimizing the output drift of a fluctuating magnetic field of a hemispherical resonator gyroscope according to claim 5, characterized in that: The fusion process in step 2 satisfies: ； in, Represent quaternions respectively.

7. The method for optimizing the output drift of a fluctuating magnetic field of a hemispherical resonator gyroscope according to claim 6, characterized in that: In step 2, the quaternion correction process also satisfies: 。 8. A method for optimizing the drift of a fluctuating magnetic field output of a hemispherical resonator gyroscope according to claim 6 or 7, characterized in that: Observable satisfy: .

9. The method for optimizing the drift of the fluctuating magnetic field output of a hemispherical resonator gyroscope according to claim 1, characterized in that: The complementary filtering algorithm in step 3 includes adjusting the frequency characteristics through a proportional link.

10. The method for optimizing the output drift of a fluctuating magnetic field of a hemispherical resonator gyroscope according to claim 9, characterized in that: The secondary optimization process satisfies: ； in: represents the measured gravity component; Represents the gravity component after the operation transformation; Indicates gravity correction error; Indicates the measured geomagnetic component; Represents the geomagnetic component after computational transformation; Indicates the geomagnetic correction error; represents the quadratic optimization error; represents the proportionality coefficient; represents the integral coefficient; represents the secondary optimization amount.

Images