Airborne gravity and magnetic vector dynamic measurement method based on multi-source data synchronous fusion

Abstract

The embodiment of the application provides a dynamic measurement method for aviation gravity and magnetic vector based on multi-source data synchronous fusion, relates to the technical field of data acquisition, and outputs the specific force measurement value, the carrier pitch angle, the carrier roll angle and the carrier motion parameter after driving the flight platform to collect the sensing array data; adopts the carrier motion parameter and the specific force measurement value to output the gravity disturbance three components; fuses the total field data of the optical pumping magnetometer to perform the magnetic vector calculation and output of the geomagnetic full element of the flux gate three component data; takes the carrier pitch angle, the carrier roll angle, the carrier position, the carrier speed and the carrier heading angle as the carrier pose data, combines the geomagnetic full element and the carrier pose data, and outputs the dynamic measurement result of the gravity and magnetic vector field. The problems that the gravity and magnetic force measurement are separated in the prior art, the equipment is redundant and inefficient, the acquisition time is asynchronous, and the data precision is reduced are solved. The effect of high-precision, high-dynamic and high-efficiency synchronous acquisition of the gravity and magnetic vector data under a single platform is achieved.

Description

Technical Field

[0001] This invention relates to the field of data acquisition technology, and in particular to a method for dynamic measurement of airborne gravity and magnetic vectors based on synchronous fusion of multi-source data. Background Technology

[0002] Existing airborne gravity and magnetic vector measurement technologies have significant limitations, mainly in that the gravity and magnetic measurement systems are independent of each other, resulting in low detection efficiency, equipment redundancy, and high costs.

[0003] In terms of multi-source data synchronization, the lack of a unified and precise time reference results in significant time deviations between inertial navigation attitude, magnetic three-component, and total field data, which seriously affects the accuracy and reliability of vector data.

[0004] Furthermore, the failure to effectively integrate high-precision attitude references with data from multiple types of sensors resulted in severe error accumulation during vector calculation. In particular, the horizontal component was greatly affected by minute attitude errors, and the lack of an effective dynamic error separation and compensation mechanism resulted in insufficient accuracy of the final gravity and magnetic vector data, making it difficult to meet the high-precision exploration requirements in complex environments.

[0005] In summary, the separation of gravity and magnetic force measurements in existing technologies leads to redundant and inefficient equipment and asynchronous acquisition time, resulting in misaligned data acquisition and insufficient attitude fusion, which in turn causes the accumulation of data errors and reduces data accuracy.

[0006] It should be noted that the information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0007] To address the aforementioned shortcomings or improvement needs of existing technologies, this invention provides a dynamic measurement method for airborne gravity and magnetic vectors based on synchronous fusion of multi-source data. This method solves the problems of existing technologies where gravity and magnetic measurements are separate, leading to equipment redundancy and inefficiency, asynchronous acquisition times, data acquisition misalignment, and insufficient attitude fusion, resulting in accumulated data errors and reduced data accuracy. It achieves high-precision, high-dynamic, and high-efficiency synchronous acquisition of gravity and magnetic vector data on a single platform, meeting the needs of exploration in complex geological environments. The specific technical solution is as follows:

[0008] This invention provides a method for dynamic measurement of airborne gravity and magnetic vectors based on synchronous fusion of multi-source data, the method comprising:

[0009] The multi-source sensor array on the flight platform performs synchronous data acquisition to obtain fluxgate three-component data, optically pumped magnetometer total field data, IMU raw data, and carrier phase raw observations. Based on the IMU raw data, the specific force measurement value, vehicle pitch angle, and vehicle roll angle are output in real time through a strapdown inertial navigation algorithm. The carrier phase raw observations are calculated in parallel along two paths based on the dual-antenna carrier phase difference to output vehicle motion parameters. Gravity vector calculation is performed using the vehicle motion parameters and specific force measurement value to output gravity disturbance three-component data. The magnetic vector calculation of the fluxgate three-component data is performed by fusing the optically pumped magnetometer total field data to output geomagnetic elements. The vehicle pitch angle, vehicle roll angle, vehicle position, vehicle velocity, and vehicle heading angle are used as vehicle attitude data. The gravity disturbance three-component data, geomagnetic elements, and vehicle attitude data are integrated and corrected according to a unified timestamp to output dynamic measurement results of the gravity and magnetic vector fields.

[0010] In one embodiment, the gravity vector is calculated using the carrier motion parameters and the specific force measurement value, and the three components of gravity disturbance are output. The following processing is also performed:

[0011] A dynamic error compensation model is constructed, using the carrier position and carrier velocity as input observations, to estimate the error state parameters of the strapdown inertial navigation system in real time; the error state parameters are used to perform feedback correction on the specific force measurement value to generate compensated specific force data; the compensated specific force data and carrier motion parameters are used to perform gravity vector calculation, and the three components of the gravity disturbance are output, wherein the carrier motion parameters include carrier position, carrier velocity, and carrier heading angle.

[0012] In one implementation, the following processing is also performed:

[0013] A state prediction analysis channel is constructed based on the state prediction equation; an observation generation analysis channel is constructed based on the observation equation; after connecting the state prediction analysis channel and the observation generation analysis channel in parallel, a Kalman filter recursive estimator is configured at the dual-channel output to complete the construction of the dynamic error compensation model.

[0014] In one implementation, a dynamic error compensation model is constructed, using the carrier position and carrier velocity as input observations, to estimate the error state parameters of the strapdown inertial navigation system in real time, and the following processing is also performed:

[0015] The IMU raw data is processed using a strapdown inertial navigation algorithm to obtain the inertial navigation position and velocity. In the observation generation and analysis channel, the carrier position and velocity are subtracted from the inertial navigation position and velocity to generate position-velocity deviation observations. In the state prediction and analysis channel, state prediction is performed based on the error state estimate of the previous moment, and the error state prediction value is output. The Kalman filter recursive estimator is used to perform real-time recursive estimation on the position-velocity deviation observations and the error state prediction value, and outputs the error state parameters, wherein the error state parameters include position error, velocity error, attitude error, accelerometer zero bias, and gyroscope zero drift.

[0016] In one implementation, the gravity vector is calculated using the compensated force data and the carrier motion parameters, and the three components of the gravity disturbance are output. The following processing is also performed:

[0017] Using the carrier's position and velocity, combined with the Earth's rotational angular velocity, the Coriolis acceleration and centrifugal acceleration are calculated; based on the carrier's position, the carrier's motion acceleration is calculated; after subtracting the theoretical value of the standard gravity field model, the carrier's motion acceleration, the Coriolis acceleration, and the centrifugal acceleration from the compensated force data, the residual signal is low-pass filtered to extract the three components of the gravitational disturbance.

[0018] In one implementation, the magnetic vector calculation of the fluxgate three-component data is performed by fusing the total field data of the optically pumped magnetometer to output the full geomagnetic elements, and the following processing is also performed:

[0019] The fluxgate three-component data is compensated in real time using a linear compensation model to obtain compensated three-component data. Using the carrier pitch angle, carrier roll angle, and carrier heading angle as rotation parameters, the coordinate system of the compensated three-component data is transformed using a rotation matrix to obtain the magnetic field three-component data. The total field data of the optically pumped magnetometer and the magnetic field three-component data are fused to calculate and generate the geomagnetic full elements, which include magnetic declination, magnetic inclination, and horizontal components.

[0020] In one implementation, a multi-source sensor array on the flight platform performs synchronous data acquisition to obtain fluxgate three-component data, optically pumped magnetometer total field data, IMU raw data, and carrier phase raw observations, and then performs the following processing:

[0021] The multi-source sensor array is pre-deployed on the flight platform. The multi-source sensor array covers a three-axis fluxgate magnetometer, a cesium optically pumped magnetometer, a strapdown inertial navigation system, and a differential GNSS receiver. The multi-source sensor array is connected in parallel to a data synchronization controller via a hardware-triggered link. After receiving the 1PPS pulse output by the differential GNSS receiver, the data synchronization controller broadcasts it to the multi-source sensor array. Through hardware-triggered synchronous data acquisition, the fluxgate three-component data, the optically pumped magnetometer total field data, the IMU raw data, and the carrier phase raw observations are obtained.

[0022] In one implementation, the specific force measurement value is fed back and corrected using the error state parameter to generate compensated specific force data, and the following processing is also performed:

[0023] The attitude error and accelerometer zero bias are extracted from the error state parameters; a correction matrix is ​​constructed using the attitude error, and the coordinate system transformation of the specific force measurement value is performed using the correction matrix to obtain the corrected specific force parameter; the accelerometer zero bias is subtracted from the corrected specific force parameter to output the compensated specific force data.

[0024] In one implementation, the position error, velocity error, and attitude error are fed back to the strapdown inertial navigation system for real-time iterative correction of navigation calculation parameters.

[0025] Beneficial effects of the embodiments of the present invention:

[0026] The solution provided in this invention synchronously acquires fluxgate three-component data, optical pump total field data, IMU raw data, and carrier phase observations via a flight platform. Combined with a strapdown inertial navigation algorithm, it outputs specific force measurements and attitude angles in real time. Dual-antenna carrier phase dual-path calculation is used to obtain the carrier's position, velocity, and heading angle. An inertial navigation system error is corrected based on a dynamic error compensation model, and compensated specific force data is output. By subtracting Coriolis acceleration, centrifugal acceleration, carrier motion acceleration, and normal gravity values, and using low-pass filtering, the gravity disturbance three-component is extracted. Simultaneously, optical pump total field data is fused to perform attitude coordinate system transformation and vector calculation on the real-time compensated fluxgate three-component data to generate full geomagnetic elements. Finally, a unified timestamp is used to integrate and correct the gravity disturbance three-component, full geomagnetic elements, and carrier attitude data, achieving high-precision, high-dynamic dynamic measurement output of the gravity and magnetic vector fields. This significantly improves detection efficiency and reduces system complexity while eliminating attitude error accumulation and data spatiotemporal misalignment, meeting the needs of exploration in complex geological environments. Of course, implementing any product or method of this invention does not necessarily require achieving all of the above advantages simultaneously. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 A schematic diagram of the process of the dynamic measurement method for airborne gravity and magnetic vectors based on synchronous fusion of multi-source data provided by the present invention is shown.

[0029] Figure 2 The diagram illustrates the process of calculating the three components of gravity disturbance in the airborne gravity and magnetic vector dynamic measurement method based on multi-source data synchronous fusion provided by the present invention. Detailed Implementation

[0030] To facilitate understanding of the present invention, a more complete description of the invention will be given below with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein; rather, these embodiments are provided to enable a more thorough and complete understanding of the disclosure of the invention.

[0031] Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0032] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0033] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0034] The present invention provides a dynamic measurement method for airborne gravity and magnetic vectors based on synchronous fusion of multi-source data, which solves the problem that in the prior art, gravity and magnetic force measurements are separated, resulting in redundant and inefficient equipment and asynchronous acquisition time, causing data acquisition misalignment and insufficient attitude fusion, leading to the accumulation of data errors and reduced data accuracy.

[0035] Example: See Figure 1 The flowchart of the airborne gravity and magnetic vector dynamic measurement method based on multi-source data synchronous fusion provided in this embodiment of the invention includes:

[0036] Y100: Drives the multi-source sensor array on the flight platform to perform synchronous data acquisition, obtaining fluxgate three-component data, optically pumped magnetometer total field data, IMU raw data, and carrier phase raw observations.

[0037] In one implementation, a multi-source sensor array on a flight platform performs synchronous data acquisition to obtain fluxgate three-component data, optically pumped magnetometer total field data, IMU raw data, and carrier phase raw observations. Step Y100 may further include:

[0038] Y110: The multi-source sensor array is pre-deployed on the flight platform, wherein the multi-source sensor array covers a three-axis fluxgate magnetometer, a cesium optically pumped magnetometer, a strapdown inertial navigation system, and a differential GNSS receiver.

[0039] Y120: The multi-source sensor array is connected to the data synchronization controller in parallel via a hardware-triggered link.

[0040] Y130: After receiving the 1PPS pulse output by the differential GNSS receiver, the data synchronization controller broadcasts it to the multi-source sensor array and triggers synchronous data acquisition through hardware to obtain the fluxgate three-component data, the optical pump magnetometer total field data, the IMU raw data, and the carrier phase raw observation.

[0041] This embodiment physically integrates gravity and magnetic sensors on the same carrier platform. Specifically, a multi-source sensor array consisting of a three-axis fluxgate magnetometer, a cesium optical pump magnetometer, a strapdown inertial navigation system, and a differential GNSS receiver is pre-deployed on the flight platform to achieve spatial co-location of gravity and magnetic sensing hardware and eliminate the position offset error of traditional separate installation.

[0042] Based on this, a dedicated hardware trigger link, such as LVDS or optical fiber, is used to connect all sensors directly to the data synchronization controller in parallel, forming a star topology network. The hardware trigger link avoids the delay of the operating system protocol stack and software interruption, realizing point-to-point physical layer connection from the sensor to the controller, ensuring that the signal transmission path is of equal length and without buffering.

[0043] The data synchronization controller uses the 1PPS pulse output by the differential GNSS as the global time reference and broadcasts it to all sensors through a hardware trigger link. Each sensor synchronously starts data acquisition on the rising edge of the pulse, so that the fluxgate sampling, IMU data frame, and GNSS carrier phase observation are strictly aligned with UTC time. Finally, the fluxgate three-component data, optical pump magnetometer total field data, IMU raw data, and carrier phase raw observations corresponding to the sensing devices are obtained.

[0044] Y200: Based on the raw data from the IMU, the specific force measurement value, the vehicle pitch angle, and the vehicle roll angle are output in real time through the strapdown inertial navigation algorithm.

[0045] The strapdown inertial navigation system is the main body for IMU data acquisition. It includes gyroscope and accelerometer hardware modules. The raw data of the IMU acquired by the strapdown inertial navigation system specifically includes the angular velocity vector output by the three-axis gyroscope and the specific force vector output by the accelerometer.

[0046] The specific process of using the strapdown inertial navigation algorithm to calculate the specific force measurement value, the vehicle pitch angle, and the roll angle in real time is as follows:

[0047] First, the raw IMU data is preprocessed. Specifically, the gyroscope data is corrected for temperature drift error using a preset temperature drift compensation model. The compensation coefficient is determined by calibration experiments. The accelerometer data is filtered using a moving average to suppress high-frequency noise.

[0048] Next, attitude calculation is performed. The three-axis angular velocity values ​​output in real time by the gyroscope are used to update the carrier attitude through quaternion differential equations. The second-order Runge-Kutta numerical integration method is used for discretization calculation to ensure the calculation stability under high dynamic conditions.

[0049] The updated quaternion is converted into a direction cosine matrix from the carrier coordinate system to the navigation coordinate system. The carrier pitch angle, representing the rotation angle about the carrier Y-axis, is analytically calculated from the element in the third row and first column of the matrix. The carrier roll angle, representing the rotation angle about the carrier X-axis, is analytically calculated from the ratio of the element in the third row and second column to the element in the third column.

[0050] In this process, the projection component of the Earth's rotation angular velocity in the navigation coordinate system is calculated based on the real-time latitude of the carrier, and the direction cosine matrix is ​​compensated and corrected. Then, the original specific force value measured by the accelerometer is rotated from the carrier coordinate system to the navigation coordinate system through the direction cosine matrix, and the local theoretical gravity value calculated by the WGS84 Earth model is subtracted to obtain the corrected specific force measurement value.

[0051] The entire solution process in this embodiment is executed in real time in a loop at the sampling frequency of the IMU. Parallel pipelined processing of quaternion updates, matrix operations and coordinate transformations is implemented through FPGA hardware to ensure that the processing latency is controlled at the microsecond level.

[0052] Y300: Based on the dual-antenna carrier phase difference dual-path parallel solution, the original carrier phase observation is calculated, and the carrier motion parameters are output.

[0053] The original carrier phase observations are obtained from a dual-antenna differential GNSS receiver, and the original carrier phase observations include L1 / L2 dual-frequency carrier phase and Doppler frequency shift.

[0054] The dual-antenna channel data is aligned to a time synchronization accuracy of <1μs by hardware timestamps to eliminate signal transmission delay between antennas; cycle slip detection and repair are performed on the carrier phase data, and the ionospheric delay error is suppressed by using a dual-frequency ionospheric combination model, followed by dual-path parallel solution.

[0055] In the positioning solution path, based on carrier phase differential technology, with the forward antenna as the reference station and the backward antenna as the rover station, a double-difference observation equation is constructed. The least squares adjustment is used to solve the three-dimensional position and velocity vector of the carrier in real time. Among them, the ambiguity is fixed and the LAMBDA algorithm is used to search for the optimal integer ambiguity combination.

[0056] In the heading solution path, based on the dual-antenna baseline vector constraint, the original carrier phase difference is corrected by the post-processed antenna phase center deviation calibration value, the dual-antenna carrier phase single difference observation value is calculated in real time, and the projection of the baseline vector on the horizontal plane is calculated by combining the carrier pitch angle and carrier roll angle provided by the Y200 step, and then the carrier heading angle is inverted.

[0057] The final dual-path output results include the three-dimensional position of the vehicle, the velocity vector of the vehicle, and the heading angle of the vehicle. The dual-path output results are fused by Kalman filtering, and the final output is the vehicle position representing the three-dimensional position in the WGS84 coordinate system, the vehicle velocity representing the three-axis components of the navigation coordinate system, and the vehicle heading angle representing the direction angle from 0° to 360° of the true north reference.

[0058] In this implementation, the calculation process is executed in real time at the GNSS update rate, and the parallel pipelined processing of carrier phase difference calculation and heading angle calculation is implemented through FPGA hardware.

[0059] Y400: The gravity vector is calculated using the carrier motion parameters and specific force measurement values, and the three components of gravity disturbance are output.

[0060] In one implementation, Figure 2 The diagram illustrates the calculation of the gravity vector using the carrier motion parameters and specific force measurements, outputting three components of gravity disturbance. Step Y400 may further include:

[0061] Y410: Construct a dynamic error compensation model, using the carrier position and carrier velocity as input observations, to estimate the error state parameters of the strapdown inertial navigation system in real time.

[0062] Y420: Use the error state parameters to perform feedback correction on the specific force measurement value to generate compensated specific force data.

[0063] Y430: The gravity vector is calculated using the compensation force data and the carrier motion parameters, and the three components of the gravity disturbance are output. The carrier motion parameters include the carrier position, the carrier velocity, and the carrier heading angle.

[0064] In one implementation, it may further include:

[0065] Y410-1: Constructing a state prediction analysis channel based on state prediction equations.

[0066] Y410-2: Constructing observation generation and analysis channels based on observation equations.

[0067] Y410-3: After connecting the state prediction analysis channel and the observation generation analysis channel in parallel, a Kalman filter recursive estimator is configured at the dual-channel output to complete the construction of the dynamic error compensation model.

[0068] In one implementation, a dynamic error compensation model is constructed, using the carrier position and carrier velocity as input observations, to estimate the error state parameters of the strapdown inertial navigation system in real time. Step Y410 may further include:

[0069] Y411: The raw IMU data is solved using the strapdown inertial navigation algorithm to obtain the inertial navigation position and velocity.

[0070] Y412: In the observation generation and analysis channel, the carrier position and carrier velocity are subtracted from the inertial navigation position and inertial navigation velocity, respectively, to generate a position-velocity deviation observation.

[0071] Y413: In the state prediction analysis channel, state prediction is performed based on the error state estimation of the previous moment, and the error state prediction value is output.

[0072] Y414: The Kalman filter recursive estimator is used to perform real-time recursive estimation of the position and velocity deviation observations and the error state predictions, and outputs the error state parameters, wherein the error state parameters include position error, velocity error, attitude error, accelerometer zero bias and gyroscope zero drift.

[0073] In one implementation, the error state parameter is used to perform feedback correction on the specific force measurement value to generate compensated specific force data. Step Y420 may further include:

[0074] Y421: Extract the attitude error and accelerometer zero bias from the error state parameters.

[0075] Y422: A correction matrix is ​​constructed using the attitude error, and the coordinate system transformation of the specific force measurement value is performed using the correction matrix to obtain the corrected specific force parameter.

[0076] Y423: Subtract the accelerometer zero bias from the corrected force parameter and output the compensated force data.

[0077] In one implementation, the gravity vector is calculated using the compensation force data and the carrier motion parameters, and the three components of the gravity disturbance are output. Step Y430 may further include:

[0078] Y431: Using the carrier's position and velocity, combined with the Earth's rotational angular velocity, calculate the Coriolis acceleration and centrifugal acceleration.

[0079] Y432: Calculate the acceleration of the carrier motion based on the carrier's position.

[0080] Y433: After subtracting the theoretical value of the standard gravity field model, the carrier's motion acceleration, Coriolis acceleration, and centrifugal acceleration from the compensated force data, low-pass filtering is performed on the residual signal to extract the three components of the gravity disturbance.

[0081] Specifically, this implementation first constructs a dynamic error compensation model to estimate the error state of the strapdown inertial navigation system in real time. The structure of the dynamic error compensation model includes a dual-channel parallel state prediction analysis channel and an observation generation analysis channel, and a Kalman filter recursive estimator is configured at the dual-channel output end.

[0082] The construction process of the dynamic error compensation model is as follows:

[0083] By establishing an error propagation mathematical model that describes the error propagation law of the strapdown inertial navigation system, the error propagation mathematical model is essentially a set of differential equations that describe the evolution of position error, velocity error, attitude error, etc. over time.

[0084] An independent state prediction analysis channel is constructed based on the equations of the aforementioned error propagation mathematical model. This channel uses the estimated error state parameters from the previous time step as initial conditions and advances one solution cycle using numerical integration methods such as the Euler method or the Runge-Kutta method. It outputs the predicted error state value for the current time step, which depends only on the internal error propagation law of the inertial navigation system and does not involve external observation data.

[0085] The error state parameters include, but are not limited to, position error, velocity error, and attitude error angle. The predicted error state values ​​include predicted position error, velocity error, attitude error angle, accelerometer zero bias, and gyroscope zero drift.

[0086] The state prediction analysis channel predicts the evolution trend of the error state at the current moment based on the error dynamics characteristics, thereby realizing open-loop error propagation simulation.

[0087] Based on the mathematical relationship between external high-precision observation data and inertial navigation output, an observation equation is constructed, and an independent observation generation and analysis channel is built based on the observation equation. The observation generation and analysis channel receives the true values ​​of the carrier position and velocity provided by the Y300 differential GNSS in step Y300, and subtracts them in real time from the inertial navigation position and velocity output by the strapdown inertial navigation algorithm to generate position deviation and velocity deviation observations, directly quantifying the real-time error of the inertial navigation system.

[0088] The state prediction analysis channel and the observation generation analysis channel are connected in parallel. A Kalman filter recursive estimator is deployed at the dual-channel output. The state prediction channel outputs the error state prediction value, and the observation generation channel outputs the error observation residual. The Kalman filter performs weighted fusion on the two signals. The predicted value provides prior information on error evolution, and the observation value provides a basis for real-time correction. Finally, the optimal error state estimate is output, and the closed-loop construction of the dynamic error compensation model is completed.

[0089] Based on the real-time output of the three-axis gyroscope angular velocity and accelerometer force data from the IMU, a strapdown inertial navigation algorithm is used to perform pure inertial navigation calculations, outputting the inertial navigation position and velocity.

[0090] First, the carrier attitude is updated by integrating the angular velocity of the gyroscope to determine the rotational relationship between the carrier coordinate system and the navigation coordinate system. Then, the specific force value measured by the accelerometer is rotated from the carrier coordinate system to the navigation coordinate system, and the theoretical gravity value under the navigation system is subtracted to obtain the carrier's motion acceleration.

[0091] Then, the trapezoidal integral method is used to integrate the vehicle's motion acceleration over time to calculate the three-axis velocity components of the vehicle in the navigation system. The three-axis velocity components include eastward, northward, and upward velocities, thus obtaining the inertial navigation velocity.

[0092] Finally, by combining the latitude, longitude, and elevation conversion model, the inertial navigation velocity is integrated over time to update the latitude, longitude, and elevation coordinates of the carrier, thus obtaining the inertial navigation position.

[0093] It should be understood that, since there is no external feedback in the solution process, the inertial navigation position and velocity include attitude errors caused by gyroscope zero drift, acceleration deviations caused by accelerometer zero bias, and error accumulation effects during the integration process.

[0094] In the independent observation generation and analysis channel, the high-precision carrier position and carrier velocity obtained from differential GNSS are used as the reference true values ​​and subtracted from the inertial navigation position and inertial navigation velocity output by Y411 in real time.

[0095] The difference in position coordinates generates a three-dimensional position deviation observation, which includes latitude deviation, longitude deviation, and elevation deviation. The difference in velocity vectors generates a three-axis velocity deviation observation, which includes eastward, northward, and celestial velocity differences.

[0096] The position and velocity deviation observations output by the analysis channel through the observations directly quantify the real-time error of the strapdown inertial navigation system.

[0097] In the independent state prediction analysis channel, the error state parameters estimated at the previous moment are used as initial conditions. The numerical integration method is used to advance one solution cycle and output the error state prediction value at the current moment, including the predicted position error, velocity error, attitude error angle, accelerometer zero bias and gyroscope zero drift.

[0098] The Kalman filter recursive estimator receives two parallel input signals: one is the position and velocity deviation observation from the observation generation analysis channel, which serves as the direct observation input for the error; the other is the error state prediction value from the state prediction analysis channel, which serves as the prior input for the state evolution.

[0099] The estimator performs recursive estimation using the Kalman filter algorithm. First, in the time update stage, it extrapolates the state estimate at the current time using the error state prediction value and its covariance matrix. Then, in the measurement update stage, it compares the position and velocity deviation observations with the prediction values ​​to generate observation residuals, dynamically calculates the Kalman gain matrix, and fuses the residuals to correct the prediction bias.

[0100] Finally, the optimally estimated error state parameters are output in real time, including position error, velocity error, attitude error, accelerometer zero bias, and gyroscope zero drift.

[0101] Among them, position error represents the coordinate deviation of the carrier in the latitude, longitude and elevation directions; velocity error represents the velocity deviation in the east, north and sky directions in the navigation coordinate system; attitude error represents the three-axis deviation angle between the carrier coordinate system and the navigation coordinate system around the X / Y / Z axes; accelerometer zero bias represents the inherent zero position deviation vector of the three-axis accelerometer; and gyroscope zero drift represents the systematic drift vector of the angular velocity measurement of the three-axis gyroscope.

[0102] The dynamic error compensation model constructed in this embodiment breaks through the GNSS update rate limitation by integrating the high-frequency prediction in step Y413 and the low-frequency high-precision observation in step Y412, achieving error estimation at the same frequency as the inertial navigation output, real-time output of attitude error and accelerometer zero bias, directly blocking the distortion of the horizontal component of gravity, and suppressing attitude calculation divergence by gyroscope zero drift estimation, providing a clean specific force input for gravity vector calculation in the subsequent step Y430.

[0103] From the error state parameters output by the dynamic error compensation model, the attitude error component and the accelerometer zero bias component are separated.

[0104] A rotation correction matrix is ​​generated by the attitude error. The rotation correction matrix is ​​the error inverse of the direction cosine matrix. This matrix is ​​multiplied by the original specific force measurement value to realize the coordinate system error compensation transformation and output the intermediate corrected specific force parameter that has eliminated the projection deviation caused by the attitude error.

[0105] By subtracting the accelerometer zero-bias component from the corrected force parameters, the influence of the inherent zero-position offset of the sensor hardware on the measured value is eliminated. The final output compensated force data is a three-axis force vector in the navigation coordinate system, and its systematic deviation has been dynamically suppressed.

[0106] Based on the carrier's position and velocity, and combined with the Earth's rotational angular velocity, two types of inertial acceleration effects are calculated. Coriolis acceleration is generated by the coupling of the carrier's velocity relative to the Earth and the Earth's rotation, while centrifugal acceleration is formed by the carrier's centripetal motion due to the Earth's rotation. It should be understood that neither of these accelerations is a gravitational effect and must be separated from the gravity calculation.

[0107] The position numerical differential method is used to perform a second time difference on the high-frequency carrier position sequence provided by differential GNSS, and the motion acceleration of the carrier in three-dimensional space is directly calculated. This motion acceleration reflects the dynamic effect generated by the carrier's maneuver.

[0108] Four non-gravity components are deducted sequentially from the compensation force data. First, the theoretical value of the standard gravity field calculated by the WGS84 Earth gravity model based on the carrier's position is deducted. Second, the carrier motion acceleration, which characterizes the carrier's own maneuvering acceleration, is deducted in step Y432. Then, the Coriolis acceleration and centrifugal acceleration calculated in step Y431 are deducted.

[0109] The residual signal after subtraction represents the deviation between the actual gravity field and the theoretical gravity field. This signal is filtered by a low-pass filter with a cutoff frequency of 0.01 to 0.1 Hz to remove non-gravity effects such as carrier vibration noise and high-frequency interference from sensors. The final output of the three components of gravity disturbance in the navigation coordinate system specifically includes the eastward disturbance component, the northward disturbance component, and the vertical disturbance component.

[0110] This embodiment achieves the direct output of the three components of gravity disturbance in the navigation coordinate system without relying on external vertical deviation separation. Its accuracy and dynamic performance meet the requirements of airborne gravity vector measurement and provide reliable input for subsequent gravity anomaly and vertical deviation conversion. At the same time, the closed-loop architecture significantly reduces system complexity and computational resource consumption.

[0111] Y500: Integrates the total field data of the optical pump magnetometer to perform magnetic vector calculation of the fluxgate three-component data and outputs all geomagnetic elements.

[0112] In one implementation, the magnetic vector calculation of the fluxgate three-component data is performed by fusing the total field data of the optically pumped magnetometer to output the full geomagnetic elements. Step Y500 may further include:

[0113] Y510: The fluxgate three-component data is compensated in real time using a linear compensation model to obtain compensated three-component data.

[0114] Y520: Using the vehicle pitch angle, vehicle roll angle and vehicle heading angle as rotation parameters, the coordinate system transformation of the compensated three-component data is performed through a rotation matrix to obtain the three components of the magnetic field.

[0115] Y530: By integrating the total field data and the three components of the magnetic field from the optically pumped magnetometer, the geomagnetic elements are calculated and generated, wherein the geomagnetic elements include magnetic declination, magnetic inclination, and horizontal components.

[0116] Specifically, in this embodiment, the linear compensation model is based on the sensor error parameters calibrated on the ground. It performs a linear transformation on the original measurement value through the coefficient matrix and offset vector, directly eliminating the inherent hardware error and outputting the compensated three-component data, ensuring that the three-component data is not affected by the sensor's own defects.

[0117] A three-dimensional rotation matrix is ​​constructed using the vehicle's pitch angle, roll angle, and yaw angle. This matrix rotates the compensated three-component data from the vehicle coordinate system to the geographic coordinate system. Specifically, the rotation matrix is ​​multiplied by the compensated three-component vectors to achieve coordinate system transformation, outputting the magnetic field three components in the geographic coordinate system. These magnetic field three components include an eastward component, a northward component, and a vertical component. This unifies the magnetic field data to the Earth's reference frame, eliminating the influence of vehicle attitude changes on the measurement direction.

[0118] After synthesizing the horizontal component of the magnetic field from the eastward and northward components, the magnetic declination is calculated based on the vector relationship of the horizontal component of the magnetic field, reflecting the horizontal offset characteristics of the geomagnetic field; the magnetic inclination is calculated by combining the vertical component with the total field scalar value extracted from the total field data of the optically pumped magnetometer, characterizing the degree of tilt of the magnetic field in the vertical direction.

[0119] Simultaneously, the vector sum of the horizontal component synthesized from the eastward and northward components is used as the horizontal intensity of the geomagnetic field. Finally, the geomagnetic full elements, including magnetic declination, magnetic inclination, and horizontal component, are output, which fully describe the geomagnetic field vector distribution characteristics of the current spatial location and provide directly analyzable magnetic field vector field information for geological exploration.

[0120] Y600: The carrier pitch angle, carrier roll angle, carrier position, carrier velocity and carrier heading angle are used as carrier attitude data. The three components of gravity disturbance, all elements of geomagnetism and carrier attitude data are integrated and corrected according to a unified timestamp, and the dynamic measurement results of gravity and magnetic vector fields are output.

[0121] The vehicle pitch angle, roll angle, heading angle, position, and velocity are uniformly classified into vehicle attitude data. Based on the 1PPS pulse trigger generated by the data synchronization controller in step Y100, the three components of gravity disturbance, all elements of geomagnetism, and vehicle attitude data are spatiotemporally aligned and integrated. Position lag correction is performed on the three components of gravity disturbance using the position and velocity information in the vehicle attitude data. At the same time, the attitude consistency of the horizontal component in the all elements of geomagnetism is verified by using the vehicle attitude angle. Finally, the dynamic measurement results of gravity and magnetic vector fields that are time-synchronized, spatially consistent, and physically coupled are output, forming a comprehensive vector field dataset that can be directly used for geological inversion.

[0122] This embodiment eliminates the spatial offset and temporal asynchronous errors of traditional split-type installations by deploying multi-source sensor hardware in a co-located manner and using 1PPS global synchronization. It uses a dynamic error compensation model to correct inertial navigation attitude errors and accelerometer zero bias in real time, and directly outputs high-precision gravity disturbance three components by combining four-fold non-gravity stripping, eliminating the post-processing step of vertical deviation separation. Based on linear real-time compensation and attitude conversion, it fuses the optical pump total field and fluxgate three components to generate geomagnetic full elements that can be directly used for geological interpretation. Finally, it outputs dynamic measurement results of carrier pose and strong coupling of gravity and magnetic vector fields through spatiotemporal integration correction, realizing high-precision, high-dynamic, and high-efficiency synchronous acquisition of gravity and magnetic vector data on a single platform, meeting the exploration needs of complex geological environments.

[0123] In one implementation, the position error, velocity error, and attitude error are fed back to the strapdown inertial navigation system for real-time iterative correction of navigation calculation parameters.

[0124] In this embodiment, the rotation matrix from the carrier coordinate system to the navigation coordinate system is corrected by attitude error to eliminate the influence of attitude calculation deviation on coordinate system transformation; the cumulative deviation in the inertial navigation velocity integration process is compensated by velocity error to block the linear growth of velocity error; and the initial value of position integral is adjusted in real time by position error to suppress the secondary accumulation of position error.

[0125] This closed-loop feedback mechanism is dynamically iterated within each cycle of inertial navigation calculation, forming a real-time loop of error estimation, parameter correction, and calculation update. This suppresses the transmission and amplification of errors such as sensor zero bias and temperature drift at the source, ensuring the long-term stability of navigation calculation parameters.

[0126] This embodiment achieves the following technical effects:

[0127] 1. By physically co-locating a quartz flexible accelerometer, vector fluxgate, optically pumped magnetometer, strapdown inertial navigation system, and differential GNSS receiver on a single flight platform, and using a hardware link to achieve microsecond-level synchronous acquisition with 1PPS pulse triggering, the redundancy and spatiotemporal misalignment of equipment in traditional split systems are completely eliminated, significantly improving detection efficiency and applicability to complex terrain engineering.

[0128] 2. In gravity vector measurement, the attitude error and accelerometer zero bias are corrected in real time through a dynamic error compensation model. Combined with quadruple non-gravity stripping, high-precision gravity disturbance three components are directly output. In magnetic vector measurement, the geomagnetic elements with extremely low magnetic declination accuracy are generated by fusion of linear real-time compensation and attitude conversion with optical pumping total field. This eliminates the need for vertical deviation separation and repetition line correction, ensuring high-quality data output.

[0129] 3. FPGA hardware acceleration enables end-to-end microsecond-level processing to support high-mobility scenarios of the carrier. Kalman filter feedback correction blocks the error accumulation chain. Combined with spatiotemporal integration, it outputs carrier pose and gravity and magnetic vector field data in a unified manner, which has the comprehensive advantages of high adaptability, low cost and stable reliability.

[0130] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0131] The foregoing description of specific exemplary embodiments of the invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. The scope of the invention is intended to be defined by the claims and their equivalents.

Claims

1. An aviation gravity-magnetic vector dynamic measurement method based on multi-source data synchronous fusion, characterized in that, include: The multi-source sensor array on the flight platform performs synchronous data acquisition to obtain fluxgate three-component data, optically pumped magnetometer total field data, IMU raw data, and carrier phase raw observations. Based on the raw IMU data, the specific force measurement value, the vehicle pitch angle and the vehicle roll angle are output in real time through the strapdown inertial navigation algorithm. Based on the dual-antenna carrier phase difference dual-path parallel solution of the original carrier phase observation, the carrier motion parameters are output. The gravity vector is calculated using the carrier motion parameters and specific force measurement values, and the three components of gravity disturbance are output. By integrating the total field data from the optically pumped magnetometer, the magnetic vector of the fluxgate three-component data is calculated, and the complete geomagnetic elements are output. The carrier pitch angle, carrier roll angle, carrier position, carrier velocity and carrier heading angle are used as carrier attitude data. The three components of gravity disturbance, all elements of geomagnetism and carrier attitude data are integrated and corrected according to a unified timestamp, and the dynamic measurement results of gravity and magnetic vector fields are output. Specifically, the gravity vector is calculated using the carrier motion parameters and specific force measurements, outputting three components of gravity disturbance, including: A dynamic error compensation model is constructed, using the carrier position and carrier velocity as input observations, to estimate the error state parameters of the strapdown inertial navigation system in real time. The specific force measurement value is corrected by feedback using the error state parameters to generate compensated specific force data; The gravity vector is calculated using the compensated force data and the carrier motion parameters, and the three components of the gravity disturbance are output. The carrier motion parameters include the carrier position, the carrier velocity, and the carrier heading angle. Specifically, the gravity vector is calculated using the compensated force data and the carrier motion parameters, and the three components of the gravity disturbance are output, including: Using the carrier's position and velocity, combined with the Earth's rotational angular velocity, calculate the Coriolis acceleration and centrifugal acceleration; Calculate the acceleration of the carrier based on its position; After subtracting the theoretical value of the standard gravity field model, the carrier's motion acceleration, Coriolis acceleration, and centrifugal acceleration from the compensated force data, the residual signal is low-pass filtered to extract the three components of the gravity disturbance. Specifically, the magnetic vector calculation of the fluxgate three-component data is performed by integrating the total field data of the optically pumped magnetometer, and the complete geomagnetic elements are output, including: The fluxgate three-component data is compensated in real time using a linear compensation model to obtain compensated three-component data. Using the vehicle pitch angle, vehicle roll angle and vehicle heading angle as rotation parameters, coordinate system transformation of the compensated three-component data is performed through a rotation matrix to obtain the three components of the magnetic field. By integrating the total field data and the three components of the magnetic field from the optically pumped magnetometer, the geomagnetic elements are calculated and generated, including magnetic declination, magnetic inclination, and horizontal components.

2. The method of claim 1, wherein the method is characterized by, Also includes: A state prediction analysis channel is constructed based on the state prediction equation; An observation generation and analysis channel is constructed based on the observation equation; After connecting the state prediction analysis channel and the observation generation analysis channel in parallel, a Kalman filter recursive estimator is configured at the dual-channel output to complete the construction of the dynamic error compensation model.

3. The method of claim 2, wherein the method further comprises: A dynamic error compensation model is constructed, using the carrier position and carrier velocity as input observations, to estimate the error state parameters of the strapdown inertial navigation system in real time, including: The IMU's raw data is solved using a strapdown inertial navigation algorithm to obtain the inertial navigation position and velocity. In the observation generation and analysis channel, the carrier position and carrier velocity are subtracted from the inertial navigation position and inertial navigation velocity, respectively, to generate position and velocity deviation observations; In the state prediction analysis channel, state prediction is performed based on the error state estimate of the previous time step, and the error state prediction value is output. The Kalman filter recursive estimator is used to perform real-time recursive estimation of the position and velocity deviation observations and the error state predictions, and outputs the error state parameters, wherein the error state parameters include position error, velocity error, attitude error, accelerometer zero bias, and gyroscope zero drift.

4. The method of claim 1, wherein the method further comprises: The multi-source sensor array on the flight platform performs synchronous data acquisition, obtaining fluxgate three-component data, optically pumped magnetometer total field data, IMU raw data, and carrier phase raw observations, including: The multi-source sensor array is pre-deployed on the flight platform, wherein the multi-source sensor array covers a three-axis fluxgate magnetometer, a cesium optically pumped magnetometer, a strapdown inertial navigation system, and a differential GNSS receiver; The multi-source sensor array is connected to the data synchronization controller in parallel via a hardware-triggered link. After receiving the 1PPS pulse output by the differential GNSS receiver, the data synchronization controller broadcasts it to the multi-source sensor array and triggers synchronous data acquisition through hardware to obtain the fluxgate three-component data, the optical pump magnetometer total field data, the IMU raw data, and the carrier phase raw observations.

5. The method of claim 1, wherein the method further comprises: The specific force measurement value is fed back and corrected using the error state parameters to generate compensated specific force data, including: The attitude error and accelerometer zero bias are extracted from the error state parameters; A correction matrix is ​​constructed using the attitude error, and the coordinate system transformation of the specific force measurement value is performed using the correction matrix to obtain the corrected specific force parameter; The compensated force data is output by subtracting the accelerometer zero bias from the corrected force parameter.

6. The method of claim 3, wherein the method further comprises: The position error, velocity error, and attitude error are fed back to the strapdown inertial navigation system for real-time iterative correction of navigation calculation parameters.

Images