Auxiliary installation device and attitude estimation method of geomagnetic observation station vector magnetometer

Abstract

The invention relates to the technical field of geomagnetic measurement, in particular to an auxiliary installation device for a vector magnetometer of a geomagnetic observation station and an attitude estimation method. The device comprises a mounting base, the bottom end of the mounting base is provided with a detachable positioning structure matched with a fluxgate magnetometer shell, and the top end of the mounting base is obliquely provided with an imaging unit containing an image sensor so as to shoot a starry sky map at night; the sensor module continuously collects device position coordinates, UTC time, acceleration and magnetic field data, and collects a star map according to a preset interval; the data processing module performs preprocessing and star point extraction on each frame of starry sky map, and recognizes the identity of a fixed star by using a star map recognition algorithm in combination with sensor data; and the positioning module solves the spatial attitude of the camera coordinate system relative to the inertial coordinate system according to the fixed star pixel coordinates and the corresponding celestial coordinates, and further obtains the stable attitude estimation of the vector magnetometer relative to the inertial coordinate system during observation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geomagnetic measurement, and in particular to an auxiliary installation device and a posture estimation method for a vector magnetometer of a geomagnetic observation station. BACKGROUND

[0002] In modern geomagnetic observation, high-precision vector magnetic field measurement relies on the accurate installation posture calibration of a fluxgate magnetometer. A standard geomagnetic station is usually equipped with a complete observation system, including a scalar magnetometer (such as a proton precession magnetometer, an Overhauser magnetometer, or an optical pumping magnetometer), a DI-Flux (Declination-Inclination Fluxgate) device, and a fluxgate magnetometer for long-term continuous operation (used as a variometer). Among them, the scalar magnetometer is used to provide a measurement reference for the absolute magnetic field total intensity F; the DI-Flux device integrates the fluxgate sensor with a precision optical theodolite, and measures the direction parameters of the magnetic field, namely the declination (D) and the inclination (I), through periodic manual observation, so as to realize accurate measurement of the geomagnetic field vector.

[0003] However, in many temporary, mobile or distributed geomagnetic monitoring projects, there is often no special building and multiple professional equipment required by the standard station. In such scenarios, a fluxgate magnetometer is usually installed on a granite platform with a cement protective cover, or even a simple structure such as a PVC pipe. Although the fluxgate magnetometer itself can output a complete magnetic field vector (including the modulus and direction information), the data measured by it is obtained in the sensor body coordinate system. In order to convert these data to the geographic coordinate system (such as the North-East-Ground coordinate system), it is necessary to accurately know the installation posture of the sensor relative to a certain reference system, especially the pointing direction of its main axis relative to the geographic north and the horizontal plane.

[0004] In actual operation, due to the lack of high-precision posture measurement equipment, the installation process often relies on a handheld compass or a simple compass for rough north alignment, supplemented by a level for leveling. Such methods are highly subjective and easily affected by local magnetic interference, and the installation error often reaches several degrees, which seriously affects the accuracy and usability of subsequent vector data.

[0005] Therefore, there is an urgent need for an auxiliary device that is simple in structure, low in cost, convenient to operate, and has high precision, which can accurately measure the installation posture of the fluxgate magnetometer on site without theodolite and professional survey personnel, and meet the demand for high-precision spatial reference of field, temporary or distributed geomagnetic observation systems. SUMMARY

[0006] The present application aims to overcome the problem of the prior art that it is difficult to determine the high-precision installation posture of a fluxgate magnetometer in a non-standard geomagnetic observation environment, thereby providing an auxiliary installation device and a posture estimation method for a geomagnetic observation station vector magnetometer, which can determine the spatial installation posture of the vector magnetometer during installation, is particularly suitable for a fluxgate magnetometer, and is especially suitable for field or temporary observation scenes that do not have complete geomagnetic station infrastructure, and can realize high-precision installation posture determination without theodolites and professional personnel.

[0007] To solve the above technical problems, the technical solution of the present application provides an auxiliary installation device for a geomagnetic observation station vector magnetometer, which comprises: An installation seat, which is provided at the bottom end with a positioning structure matched with the shell of the fluxgate magnetometer, the positioning structure being used for detachable mechanical coupling with the shell of the fluxgate magnetometer; and is provided at the top end with an obliquely installed imaging unit, the imaging unit containing an image sensor, which is used for shooting starry sky images at night; A sensor module, which is used for continuously collecting position coordinate data and UTC time data of the auxiliary installation device, is used for continuously collecting acceleration data and magnetic field data of the auxiliary installation device, and is also used for collecting starry sky images at a preset interval; A data processing module, which is used for preprocessing and star point extraction of each starry sky image, is used for identifying the identity of stars in each starry sky image by using a star map recognition algorithm in combination with the position coordinates, UTC time data, acceleration and magnetic field data of the auxiliary installation device; and A positioning module, which is used for solving the spatial posture of the camera coordinate system relative to the inertial coordinate system according to the pixel coordinates of the stars and the corresponding star celestial coordinates, thereby obtaining the stable posture estimation of the geomagnetic observation station vector magnetometer relative to the inertial coordinate system during observation.

[0008] As an improvement of the above-mentioned device, the installation seat is a cylindrical structure; the positioning structure is arranged at the axial bottom end of the cylindrical structure, and the positioning structure is a clamping groove or a clamp; the imaging unit is arranged at the axial top end of the cylindrical structure; and the optical axis of the image sensor is inclined upward by 45°±5° relative to the axis of the cylinder.

[0009] As an improvement of the above device, the sensor module comprises a GNSS receiving unit, an inertial measurement unit, a data storage unit, a power supply unit, and a control and interface unit; wherein the GNSS receiving unit is used to continuously collect position coordinate data and UTC time data of the auxiliary installation device; the inertial measurement unit is used to continuously collect acceleration and magnetic field data of the auxiliary installation device; the data storage unit uses a local non-volatile storage medium to store data and star image; the power supply unit comprises a rechargeable battery pack; the control and interface unit is used to control the running state of each unit of the sensor module.

[0010] As an improvement of the above device, the data processing module comprises: a preprocessing unit, for each frame of star image, used to extract the corresponding UTC time data, position coordinate data, acceleration data and magnetic field data of the star image, and used to perform dark field correction and flat field correction on the star image; an extraction unit, for each frame of star image, used to extract bright star targets in the preprocessed star image, and used to calculate pixel coordinates and brightness information of the bright star targets; and an identification unit, for each frame of star image, used to calculate the visible star area of local stars according to the position coordinate data and UTC time data corresponding to the frame of star image; used to predict the pointing range of the camera of the imaging unit in the celestial coordinate system according to the acceleration data and magnetic field data corresponding to the frame of star image; used to search for bright stars in the star table within the pointing range area, and used to perform star identification to determine the identity of at least four stars in the frame of star image.

[0011] As an improvement of the above device, the positioning module uses three-vector method, quaternion optimal estimation algorithm or singular value decomposition to solve the spatial attitude of the camera coordinate system relative to the inertial coordinate system.

[0012] As an improvement of the above device, it further comprises a fusion module, which is used to time align the spatial attitude of the camera coordinate system relative to the inertial coordinate system solved for each frame of star image, minimize the relative rotation deviation caused by the earth rotation between adjacent frames; use nonlinear optimization to perform smooth correction on the overall attitude sequence, and finally determine the optimal estimated attitude of the vector magnetometer relative to the inertial system.

[0013] As an improvement of the above device, it further comprises a calibration module; the calibration module comprises a camera calibration unit and a transfer matrix calibration unit, wherein the camera calibration unit uses a standard checkerboard or a star point simulator to calibrate the focal length, principal point, radial and tangential distortion parameters of the camera; to establish an accurate mapping relationship between the pixel coordinates in the image coordinate system and the line-of-sight direction; wherein the line-of-sight direction corresponding to any pixel point on the image is the direction of a ray that starts from the projection center, passes through the physical point corresponding to the pixel on the imaging plane, and is directed to the external three-dimensional world; the transfer matrix calibration unit is used to collect the spatial pose of the camera coordinate system relative to the inertial coordinate system and the magnetic field vector direction output by the vector magnetometer, and fit the fixed rotation matrix between the two by the least square method .

[0014] As an improvement of the above device, it further comprises a coordinate conversion module for converting the magnetic field vector output by the vector magnetometer to the geocentric coordinate system, so as to obtain the coordinates of the geomagnetic vector in the geocentric coordinate system , and the conversion process is as follows: ; wherein: is the fixed rotation matrix obtained by calibration, is the spatial pose obtained by solving the camera coordinate system relative to the inertial coordinate system; is the rotation matrix from the inertial coordinate system to the geocentric coordinate system.

[0015] To achieve another object of the present application, the present application also provides a geomagnetic observation station vector magnetometer attitude estimation method, which is realized based on the above-mentioned auxiliary installation device of the geomagnetic observation station vector magnetometer, comprising: The sensor module continuously collects the position coordinate data and UTC time data of the auxiliary installation device, continuously collects the acceleration data and magnetic field data of the auxiliary installation device, and collects the star image according to the preset interval; The data processing module pre-processes and extracts star points from each frame of star image, combines the position coordinates, UTC time data, acceleration and magnetic field data of the auxiliary installation device, and uses star identification algorithm to identify the identity of each star in the star image; The positioning module calculates the spatial pose of the camera coordinate system relative to the inertial coordinate system according to the pixel coordinates of the stars and the corresponding star celestial coordinates, so as to obtain the stable attitude estimation of the geomagnetic observation station vector magnetometer relative to the inertial coordinate system during observation.

[0016] ​The auxiliary installation device and attitude estimation method for the vector magnetometer of the geomagnetic observation station provided by this invention proposes several highly innovative technical improvements based on the existing technology, mainly reflected in the following aspects: 1. A static degree data acquisition device is proposed for the installation attitude calibration of fluxgate magnetometers: Unlike the traditional manual alignment of theodolites, this invention introduces non-real-time, long-term continuous star observation into the installation and calibration scenario of geomagnetic sensors for the first time. It uses the Earth's rotation to provide multi-angle star map observation data, and improves the attitude calculation accuracy through post-processing fusion, thus solving the problem of high-precision attitude determination under low-cost conditions.

[0017] 2. Use IMU (Inertial Measurement Unit) for coarse attitude determination + use star map for fine attitude determination: Utilize the precise spatiotemporal information provided by GNSS (Global Navigation Satellite System) and the coarse attitude output by IMU to pre-estimate the camera pointing range, significantly reduce the search space of the star map database, improve pattern matching efficiency, and reduce the demand for computing resources. 3. Achieving a traceable coordinate transformation chain from the inertial frame to the Earth-fixed frame: This invention not only obtains the camera's attitude in the J2000 inertial frame, but also transforms it to the ECEF coordinate system through a standard astronomical model (such as IAU2006 / 2010), and completes the spatial benchmark unification of the magnetometer measurement data by combining the transfer matrix, providing a high-precision geometric basis for subsequent cross-station comparison, modeling and inversion of geomagnetic vector data.

[0018] 4. The device has a highly integrated structure, is easy to operate, and is suitable for unattended field scenarios: The entire system is integrated into a portable cylindrical structure with local storage and independent power supply capabilities. It supports automatic start-up, timed data collection, and all-night operation without manual intervention, making it particularly suitable for use at field geomagnetic observation points with inconvenient transportation and power shortages. Attached Figure Description

[0019] Figure 1 A schematic diagram of the auxiliary installation device for the vector magnetometer of the geomagnetic observation station provided by the present invention; Figure 2 A flowchart of the attitude estimation method for the vector magnetometer of the geomagnetic observation station provided by the present invention. Detailed Implementation

[0020] The technical solutions provided by the present invention will be further illustrated below with reference to the embodiments.

[0021] Example 1 The auxiliary installation device for the vector magnetometer at the geomagnetic observatory provided in this embodiment integrates mechanical positioning, multi-source sensing, and star map recognition. This device is rigidly connected to the outer shell of the fluxgate magnetometer. Utilizing nighttime star imaging combined with GNSS / IMU prior information, it accurately calculates the spatial attitude of the device body through offline data processing. Furthermore, through pre-calibrated attitude transfer relationships, it precisely derives the installation attitude of the fluxgate magnetometer's sensor coordinate system relative to the geographic coordinate system.

[0022] I. Device Structure The device is a rigid, non-magnetic cylindrical structure with good dimensional stability and consistent thermal expansion. It is preferably made of carbon fiber and is approximately 0.5 meters long to balance structural stability and portability. Figure 1 As shown, functional modules are provided at both ends of the cylindrical structure along the axial direction: 1. The bottom is equipped with a precision slot or clamp structure for rapid, repeatable, and high-precision mechanical coupling with the housing of the target fluxgate magnetometer. This slot is customized according to the external dimensions of the target fluxgate sensor to ensure that the relative position and orientation height between the device and the magnetometer are consistent each time it is installed, with a mechanical repeatability positioning accuracy better than 0.1°.

[0023] 2. A tilted imaging unit is located at the top, including an image sensor whose optical axis is tilted upwards at 45°±5° relative to the cylinder axis, facing the sky, for capturing images of the night sky. This angle design ensures sufficient elevation to avoid obstruction from ground objects while maintaining coverage of the main navigation stars.

[0024] 3. Internal sensor module: GNSS receiver unit: used to acquire the precise geographic coordinates and UTC time of the device's location, and supports PPS (Pulse Per Second) output; Inertial Measurement Unit (IMU): Contains a triaxial accelerometer and a triaxial magnetometer, used to continuously record the device's coarse attitude in the Earth's gravitational field and geomagnetic field; Data storage and power supply unit: Equipped with local non-volatile storage media (such as microSD card) and rechargeable battery pack, supporting the device to work continuously overnight (≥8 hours) in an unattended state, and automatically triggering image acquisition once at a preset interval (such as every 30 seconds).

[0025] Control and Interface Module: The microcontroller is responsible for coordinating the operation of each module, managing data packaging and storage, and enabling remote status monitoring and configuration via Wi-Fi or Bluetooth when necessary.

[0026] II. Attitude Measurement Method Flow After the fluxgate magnetometer is physically installed and leveled, the device of this invention is securely mounted to its housing via the front-end interface. The system is then started, and the following data acquisition process is executed: The GNSS receiving unit continuously records the location of the acquisition device and the precise UTC time; The inertial measurement unit continuously acquires acceleration and magnetic field data at a high frequency (≥10Hz); Under clear nighttime conditions, the camera automatically takes a picture of the starry sky at fixed time intervals (such as 30s or 60s), with the exposure time of each picture controlled within 1 second to ensure that the stars are clear and without trailing. The entire device remains stationary, but due to the Earth's rotation, the distribution of stars in the night sky images captured by the camera at different times will show regular changes, which is equivalent to observing inertial space from multiple perspectives.

[0027] This process can last all night, yielding dozens to hundreds of sets of synchronized data (time, location, IMU, images), providing rich information for subsequent high-precision attitude calculations.

[0028] III. Attitude Calculation Methods (Post-Processing) All collected data will be processed offline after the observation is completed. The specific steps are as follows: 1. Establishment of time-space reference: Using the precise location provided by the GNSS receiver unit and the UTC timestamp of each frame of image, calculate the zenith direction, true north direction, and visible range of stars in the local celestial coordinate system at that moment; 2. Image preprocessing and star point extraction: Perform dark field correction and flat field correction on each image (based on pre-calibrated camera response characteristics); Extract bright star targets from the image and calculate their pixel coordinates and brightness information.

[0029] 3. Star map identification and pointing calculation: Based on the initial IMU value and GNSS position / time, predict the approximate direction of the camera's optical axis on the celestial sphere (right ascension α, declination δ). The system retrieves reference stars for the region from the star catalog database and uses a pattern matching algorithm to identify the stars in the image. Using algorithms such as TRIAD (Tri-Axis Attitude Determination), QUEST (Quaternion ESTimator), and SVD (Singular Value Decomposition), the precise attitude (rotation matrix or quaternion) of the camera coordinate system relative to the inertial coordinate system (such as J2000) is calculated.

[0030] 4. Multi-frame fusion optimization: In theory, attitude estimation can be performed in a single frame, but multi-view observations brought about by the Earth's rotation can be used to further improve the accuracy and robustness of attitude estimation. The final result is a stable attitude estimate of the device relative to the inertial frame during the entire observation period, and its attitude relative to the geocentric-fixed coordinate system (ECEF coordinate system) should theoretically remain unchanged.

[0031] 5. Coordinate system transformation: The calculated inertial frame attitude is transformed to the Earth-centered Earth-fixed (ECEF) coordinate system, which serves as a common reference frame for subsequent data processing.

[0032] IV. Key Calibration Process The accuracy of this invention depends on two key calibration steps: 1. In-camera distortion calibration: In a laboratory environment, using a standard checkerboard or star point simulator, the camera's focal length, principal point, radial and tangential distortion parameters were calibrated. Establish a precise mapping relationship between pixel coordinates and viewing direction in the image coordinate system; where the viewing direction corresponding to any pixel in the image is the direction of a ray that starts from the projection center, passes through the physical point corresponding to that pixel on the imaging plane, and is directed towards the external three-dimensional world; the projection center is the optical center of the camera.

[0033] The calibration results are stored in the system or saved along with the data for later processing.

[0034] 2. Transfer matrix calibration (device coordinate system) fluxgate coordinate system): In a known attitude reference environment (such as a three-dimensional turntable), the device of the present invention is installed on a fluxgate magnetometer; The attitude output by the synchronous acquisition device (obtained by star map calculation) and the direction of the magnetic field vector output by the fluxgate (known real geographical direction). The fixed rotation matrix between the two is fitted using the least squares method. That is, the "transfer matrix"; V. Application of Magnetometer Data Coordinate Transformation Once the transfer matrix is ​​obtained, the original magnetic field vector measured by the fluxgate magnetometer under any installation scenario can be used to obtain the magnetic field vector. Transform to the ECEF coordinate system to obtain the coordinates of the geomagnetic vector in the ECEF (Earth Center Earth Fixed) coordinate system and its corresponding coordinates in the ECEF coordinate system. :

[0035] in: The transfer matrix obtained from calibration; The pose derived from the star map solution; is the rotation matrix from the inertial frame to the geocentric Earth-fixed coordinate system.

[0036] This enables a complete coordinate chain transformation from the sensor's body coordinate system to the Earth-fixed coordinate system, providing a high-precision spatial reference for subsequent geomagnetic modeling, data assimilation, and cross-station comparison.

[0037] The following describes the specific implementation process of the present invention in detail with reference to a typical embodiment.

[0038] Example: Attitude calibration of a temporary geomagnetic observation point in the field 1. Device Configuration Parameters Cylindrical material: carbon fiber square tube, 50cm in length, 15cm in outer diameter; Backend interface: Custom card slot, matching the diameter of a certain type of fluxgate sensor housing; Camera: Raspberry Pi HQ Camera + f=25mm fixed focus lens, 1.55μm pixel size; GNSS receiver unit: Model U-blox NEO-M8N, supports NMEA navigation data output format standard, positioning accuracy <2m, time synchronization accuracy ±1ms; IMU: Model BMI160 (three-axis accelerometer + three-axis gyroscope + three-axis magnetometer), sampling rate 100Hz, communicates with the main controller via I²C (Inter-Integrated Circuit). Main control unit: Raspberry Pi 5, running Linux system; Storage: 32GB microSD card; Power supply: 5000mAh lithium battery, supporting 12 hours of continuous operation; Software: Python scripts control the acquisition process, triggering the packaging and storage of image and sensor data every 60 seconds.

[0039] 2. On-site operation procedures Install the fluxgate magnetometer on the granite platform and connect the cables and data acquisition equipment; Accurately fit the front slot of the device of the present invention into the designated position on the magnetic fluxgate housing, ensuring that there is no looseness; When the power is turned on, the system automatically acquires GNSS positioning information and begins recording IMU data. On a clear night, starting at 20:00, the camera automatically takes a picture of the starry sky every 60 seconds, with the exposure time set to 1.0s and the gain automatically adjusted. The entire device remained stationary and continuously collected data until 4:00 AM the following morning, acquiring approximately 480 images and 8 hours of IMU and GNSS data. The device was disassembled and taken back to the laboratory for data processing.

[0040] 3. Data Processing Flow Preprocessing stage: Extract the timestamp, GNSS location, and initial IMU attitude value for each frame of image; Dark field and flat field corrections were performed on the image (based on a laboratory-calibrated response function). The centroid algorithm is used to extract the pixel coordinates of star points.

[0041] Star map recognition and pose calculation: For each frame of image, calculate the local sidereal time and visible star region based on its UTC time and location; Predict the camera's pointing direction using initial IMU values ​​(error approximately ±5°); Retrieve bright stars (V < 6 magnitude) in the region from the star catalog and use the triangle matching method for star map identification; After successfully identifying at least four stars, the QUEST algorithm is invoked to calculate the camera's attitude quaternions in the J2000 coordinate system.

[0042] Multi-frame fusion optimization: Align all successfully calculated pose results in time. Construct an error function: minimize the relative rotational deviation between adjacent frames caused by the Earth's rotation; Nonlinear optimization is used to smooth the overall attitude sequence, and the optimal estimated attitude of the device relative to the inertial frame is finally determined.

[0043] Coordinate transformation and transfer: The optimal pose is transformed to the ECEF coordinate system using the IAU2006 model; Load the pre-calibrated transfer matrix (Obtained through turntable experiments); Calculate the installation attitude matrix .

[0044] 4. Results Output Finally, an attitude report is generated, which includes: Install attitude matrix: All raw data and processing logs are packaged and archived for subsequent auditing.

[0045] The result was written into the station's metadata file for later coordinate system correction of the magnetometer vector data.

[0046] Example 2 This embodiment provides an attitude estimation method for a vector magnetometer at a geomagnetic observatory, implemented based on the auxiliary installation device for the vector magnetometer at a geomagnetic observatory provided in Embodiment 1, such as... Figure 2 As shown, it includes the following steps: The sensor module continuously collects the position coordinate data and UTC time data of the auxiliary installation device, continuously collects the acceleration data and magnetic field data of the auxiliary installation device, and collects starry sky images at preset intervals. The data processing module preprocesses and extracts star points from each frame of the starry sky image. Combined with the position coordinates of the auxiliary installation device, UTC time data, acceleration and magnetic field data of the auxiliary installation device, the star map recognition algorithm is used to identify the stars in each frame of the starry sky image. The positioning module calculates the spatial attitude of the camera coordinate system relative to the inertial coordinate system based on the pixel coordinates of the star and the corresponding celestial sphere coordinates of the star, thereby obtaining a stable attitude estimate of the vector magnetometer of the geomagnetic observatory relative to the inertial coordinate system during the observation period.

[0047] Compared with the prior art, the present invention can bring the following significant technical advantages and practical benefits: 1. Significantly improved installation attitude measurement accuracy: Through star map recognition and multi-frame fusion algorithm, the device attitude calculation accuracy (which depends on weather conditions and the number of stars) is much higher than that of traditional compasses for north.

[0048] 2. Eliminates dependence on specialized equipment and personnel: No theodolites, manual observation, or scalar magnetometers are required. Ordinary technicians can complete high-precision installation and calibration, reducing the construction threshold and operation and maintenance costs of geomagnetic observation systems.

[0049] 3. Highly adaptable and can be deployed in various environments: It supports use on simple platforms without dedicated observation rooms or permanent power supply, and is suitable for distributed or temporary geomagnetic projects such as earthquake precursor monitoring, space weather field observation, polar scientific expeditions, and mineral exploration.

[0050] 4. Data is traceable and verifiable: All raw data (images, GNSS, IMU) are fully recorded, supporting subsequent reprocessing and error analysis, which enhances the credibility and scientific value of geomagnetic observation data.

[0051] 5. Supports long-term attitude stability monitoring: The device can be periodically and repeatedly installed and data collected to detect whether the magnetometer has attitude drifted due to foundation settlement, thermal expansion and contraction, etc., so as to achieve long-term quality control of the sensor status.

[0052] 6. Low cost and easy to promote: The core components use consumer-grade hardware (such as Raspberry Pi camera, MEMS IMU, GNSS module), the overall cost is controllable, and it has the potential for large-scale promotion and application.

[0053] 7. Promote the standardization of geomagnetic observation: Provide a unified attitude calibration tool for non-standard stations, which will help improve the data consistency and interoperability of distributed geomagnetic networks across the country and even the world.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art should understand that modifications or equivalent substitutions to the technical solutions of the present invention do not depart from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An auxiliary installation device for a vector magnetometer at a geomagnetic observation station, characterized in that, include: The mounting base has a positioning structure at its bottom that matches the housing of the fluxgate magnetometer. The positioning structure is used for detachable mechanical coupling with the housing of the fluxgate magnetometer. The top of the mounting base has an inclined imaging unit, which includes an image sensor for capturing images of the night sky. The sensor module is used to continuously collect position coordinate data and UTC time data of the auxiliary installation device, to continuously collect acceleration data and magnetic field data of the auxiliary installation device, and to collect starry sky images at preset intervals. The data processing module is used to preprocess and extract stars from each frame of the starry sky image. It combines the position coordinates of the auxiliary installation device, UTC time data, acceleration and magnetic field data of the auxiliary installation device, and uses a star map recognition algorithm to identify the stars in each frame of the starry sky image. and, The positioning module is used to calculate the spatial attitude of the camera coordinate system relative to the inertial coordinate system based on the pixel coordinates of the star and the corresponding celestial sphere coordinates of the star, thereby obtaining a stable attitude estimate of the vector magnetometer of the geomagnetic observatory relative to the inertial coordinate system during the observation period.

2. The auxiliary installation device for the vector magnetometer of the geomagnetic observation station according to claim 1, characterized in that, The mounting base is a cylindrical structure; the positioning structure is located at the axial bottom end of the cylindrical structure, and the positioning structure is a slot or a clamp; the imaging unit is located at the axial top end of the cylindrical structure; the optical axis of the image sensor is tilted upward at 45°±5° relative to the axis of the cylindrical body.

3. The auxiliary installation device for the vector magnetometer of the geomagnetic observation station according to claim 1, characterized in that, The sensor module includes a GNSS receiving unit, an inertial measurement unit, a data storage unit, a power supply unit, and a control and interface unit. The GNSS receiving unit continuously acquires the position coordinates and UTC time data of the auxiliary installation device. The inertial measurement unit continuously acquires the acceleration and magnetic field data of the auxiliary installation device. The data storage unit uses a local non-volatile storage medium to store data and starscape images. The power supply unit includes a rechargeable battery pack. The control and interface unit controls the operating status of each unit of the sensor module.

4. The auxiliary installation device for the vector magnetometer of the geomagnetic observation station according to claim 1, characterized in that, The data processing module includes: The preprocessing unit extracts the corresponding UTC time data, position coordinate data, acceleration data, and magnetic field data for each frame of starry sky image, and performs dark field correction and flat field correction on the starry sky image. The extraction unit, for each frame of the starry sky image, is used to extract bright star targets from the preprocessed starry sky image and calculate the pixel coordinates and brightness information of the bright star targets; and, The identification unit, for each frame of star image, calculates the visible star region of local stars based on the position coordinate data and UTC time data corresponding to that frame of star image; predicts the pointing range of the imaging unit's camera in the celestial coordinate system using the acceleration data and magnetic field data corresponding to that frame of star image; and uses this to search for bright stars within the pointing range region in the star catalog and perform star map identification to determine the identities of at least four stars in that frame of star image.

5. The auxiliary installation device for the vector magnetometer of the geomagnetic observation station according to claim 1, characterized in that, The positioning module uses the three-vector method, quaternion optimal estimation algorithm or singular value decomposition to calculate the spatial attitude of the camera coordinate system relative to the inertial coordinate system.

6. The auxiliary installation device for the vector magnetometer of the geomagnetic observation station according to claim 1, characterized in that, It also includes a fusion module, which is used to time-align the spatial attitude of the camera coordinate system relative to the inertial coordinate system calculated for each frame of starry sky image, minimizing the relative rotation deviation caused by the Earth's rotation between adjacent frames; and to use nonlinear optimization to smooth the overall attitude sequence, and finally determine the optimal estimated attitude of the vector magnetometer relative to the inertial frame.

7. The auxiliary installation device for the vector magnetometer of the geomagnetic observation station according to claim 1, characterized in that, Also includes: Calibration module; The calibration module includes: a camera calibration unit and a transfer matrix calibration unit, wherein... The camera calibration unit uses a standard checkerboard or star point simulator to calibrate the camera's focal length, principal point, radial and tangential distortion parameters; it is used to establish a precise mapping relationship between pixel coordinates and viewing direction in the image coordinate system; wherein, the viewing direction corresponding to any pixel in the image is the direction of a ray that starts from the projection center, passes through the physical point corresponding to that pixel on the imaging plane, and shoots out into the external three-dimensional world. The transfer matrix calibration unit is used to acquire the spatial attitude of the camera coordinate system relative to the inertial coordinate system and the direction of the magnetic field vector output by the vector magnetometer, and to fit a fixed rotation matrix between the two using the least squares method. .

8. The auxiliary installation device for the vector magnetometer of the geomagnetic observation station according to claim 7, characterized in that, It also includes a coordinate transformation module, used to transform the magnetic field vector output by the vector magnetometer. Transform to the Earth-centered Earth-fixed coordinate system to obtain the coordinates of the geomagnetic vector in the Earth-centered Earth-fixed coordinate system. The conversion process is as follows: ； in: For the fixed rotation matrix obtained from calibration, To calculate the spatial attitude of the camera coordinate system relative to the inertial coordinate system; This is the rotation matrix from the inertial coordinate system to the geocentric coordinate system.

9. An attitude estimation method for a vector magnetometer at a geomagnetic observatory, implemented based on the auxiliary installation device for the vector magnetometer at a geomagnetic observatory as described in any one of claims 1-8, characterized in that, include: The sensor module continuously collects the position coordinate data and UTC time data of the auxiliary installation device, continuously collects the acceleration data and magnetic field data of the auxiliary installation device, and collects starry sky images at preset intervals. The data processing module preprocesses and extracts star points from each frame of the starry sky image. Combined with the position coordinates of the auxiliary installation device, UTC time data, acceleration and magnetic field data of the auxiliary installation device, the star map recognition algorithm is used to identify the stars in each frame of the starry sky image. The positioning module calculates the spatial attitude of the camera coordinate system relative to the inertial coordinate system based on the pixel coordinates of the star and the corresponding celestial sphere coordinates of the star, thereby obtaining a stable attitude estimate of the vector magnetometer of the geomagnetic observatory relative to the inertial coordinate system during the observation period.

Images