Satellite rotation maneuver accelerometer calibration method based on data optimization

Abstract

The application discloses a satellite rotating maneuvering accelerometer calibration method based on data optimization, belongs to the technical field of accelerometer calibration, obtains star sensor data of on-orbit measurement of a gravity satellite, adopts a first processing method to perform first preprocessing on the star sensor data, and calculates satellite angular acceleration data according to the star sensor data; obtains accelerometer data of on-orbit measurement of the gravity satellite, and adopts a second processing method to perform second preprocessing; obtains a first correlation coefficient; obtains a first star sensor data section and a first accelerometer data section; obtains a first weight factor from the first correlation coefficient; adopts an optimized data selection method to calculate an accelerometer scale factor according to the first star sensor data section and the first accelerometer data section; and obtains on-orbit calibration results of the satellite accelerometer scale factor according to the accelerometer scale factor and the first weight factor, so that the technical effect of improving the data quality of satellite maneuvering calibration is achieved.

Description

Technical Field

[0001] This invention relates to the field of accelerometer calibration technology, and in particular to a data-optimized method for calibrating a satellite rotational maneuvering accelerometer. Background Technology

[0002] Currently, three types of gravity field measurement satellites have been developed internationally. To fully utilize their performance during in-orbit operation, while developing satellite technology, particular emphasis is placed on ground-based application research of the data acquired by the satellites. For example, the CHAMP and GRACE satellites of the United States and Germany, and the GOCE satellite of the European Space Agency, have all developed corresponding in-orbit calibration methods for their payloads to ensure the production of high-precision gravity field models.

[0003] The accelerometer onboard the gravity satellite has two high-sensitivity axes and one low-sensitivity axis; the test mass is 40×40×10mm. 3 The accelerometer measures a cube with a mass of 70g. Its function is to detect non-conservative forces acting on the satellite. If the accelerometer is not located at the satellite's center of mass, the detected non-conservative forces will include angular velocity, angular acceleration, and acceleration introduced by the gravitational gradient. The accelerometer can be divided into two parts: a displacement detection module, used to detect the position of the test mass within the electrode frame, and a feedback control module, which controls the position of the test mass through feedback electrostatic force, ensuring it remains at the center of the electrode frame. The accelerometer's scaling factor is calibrated using the rotating satellite method, providing initial acceleration data parameters for the production of gravity field models. Summary of the Invention

[0004] This invention provides a data-optimized calibration method for satellite rotational maneuvering accelerometers, which addresses the technical problem in existing technologies where data mismatch between accelerometers and star sensors during satellite maneuvers leads to low signal-to-noise ratios and increased accelerometer scaling factor estimation errors. This method improves the quality of satellite maneuvering calibration data, enabling accelerometer data to more accurately reflect the satellite's real-time angular acceleration, and provides precise data support for enhancing the calibration of gravity satellite accelerometer scaling factors.

[0005] This invention provides a data-optimized calibration method for satellite rotational maneuvering accelerometers, comprising: Step 1: obtaining star sensor data measured by a gravity satellite in orbit, and performing a first preprocessing method on the star sensor data, and calculating satellite angular acceleration data based on the star sensor data; Step 2: obtaining accelerometer data measured by the gravity satellite in orbit, and performing a second preprocessing method on the accelerometer data; Step 3: performing correlation analysis to calculate the correlation between the star sensor data after the first preprocessing and the accelerometer data after the second preprocessing, thereby obtaining the correlation between the star sensor data after the first preprocessing and the accelerometer data after the second preprocessing. Step 4: Based on the correlation principle, after segmenting the first preprocessed star sensor data and the second preprocessed accelerometer data, obtain the first star sensor data segment and the first accelerometer data segment; Step 5: From the first correlation coefficient, obtain the first weight factor corresponding to the first star sensor data segment and the first accelerometer data segment; Step 6: Based on the first star sensor data segment and the first accelerometer data segment, calculate the accelerometer scaling factor using an optimized data selection method; Step 7: Based on the accelerometer scaling factor and the first weight factor, obtain the on-orbit calibration result of the satellite accelerometer scaling factor.

[0006] Preferably, in step 1, the first processing method includes: performing a deduplication operation on the star sensor data by data measurement time matching and time monotonicity; and performing a data gross error removal operation on the star sensor data by the median method.

[0007] Preferably, in step 1, the star sensor data is subjected to gross error removal using the median method, including:

[0008] |y i |>(m+n·MAD),

[0009] Wherein, if the formula conditions are met, the sampled point data are gross errors, where m = Median(yi), MAD = Median{|yi-m| / 0.6745}, y i Let y be the data sequence to be removed for outliers, n be the number of data points, and Median be the middle number in the sorted data sequence. n / 2 .

[0010] Preferably, in step 1, calculating the satellite angular acceleration data based on the star sensor data includes:

[0011] After removing outliers from the star sensor data using the median method, the satellite angular acceleration data is obtained by performing a second differential calculation.

[0012] Preferably, in step 2, the second processing method includes:

[0013] The accelerometer data is debiased and de-drifted using a low-pass filter.

[0014] Preferably, in step 2, the process of using a low-pass filter to debias and de-drift the accelerometer data specifically includes:

[0015] y(i)=Med[x(iN),…,x(i),…,x(i+N)],

[0016] Where N = (L-1) / 2, L is the length of the accelerometer data sequence, x(i) is the i-th element of the accelerometer data sequence, and Med is the low-pass median filter calculation.

[0017] Preferably, in step 3, the correlation calculation of the star sensor data after the first preprocessing and the accelerometer data after the second preprocessing includes:

[0018] Cov(A,S)=E{[AE(A)][SE(S)]},

[0019]

[0020] Where A is the accelerometer data sequence, S is the star sensor data sequence, E is the expected value of the calculated data sequence, D is the variance of the calculated data sequence, Cov is the calculated covariance, and cof is the correlation coefficient of the calculated data.

[0021] Preferably, in step 6, the step of calculating the accelerometer scaling factor using an optimized data selection method based on the first star sensor data segment and the first accelerometer data segment includes:

[0022] Let the satellite's center of mass be the origin of the coordinate system. Then the acceleration introduced by the satellite's rotation is:

[0023] A in = -ω×(ω×r)+r×β+2υ×ω,

[0024] Among them, A in The acceleration introduced by the satellite rotation is r, the coordinate of the center of mass of the accelerometer test mass is r, ω is the angular velocity of the satellite rotation is ω, β=dω / dt is the angular acceleration of the satellite rotation is β=dω / dt is υ is the velocity of the test mass relative to the satellite. For the accelerometer in the control state, the test mass remains stationary relative to the capacitor plate, then υ≈0.

[0025] When the rotating satellite rotates around the X-axis, that is The satellite's rotation angle is:

[0026] θ = θ0cosω0t,

[0027] Where θ0 is the initial rotation angle, ω0 is the initial rotation angular velocity, and the acceleration introduced by the rotating satellite is:

[0028]

[0029] Among them, A in The triaxial angular accelerations introduced for rotation are a x,in a y,in a z,in Let y and z be the centroid deviations of the satellite in the y and z directions, respectively. Ignoring the minor influence of coordinate axis installation deviations, the triaxial accelerometer measurement value Δa of the rotating satellite's angular acceleration is... x,out , Δa y,out , Δa z,out

[0030] for:

[0031]

[0032] The scaling factors s in the y and z directions were calculated. y s z for:

[0033]

[0034] Among them, s y s is the y-axis scaling factor. z is the scaling factor in the z-direction.

[0035] Preferably, in step 7, obtaining the on-orbit calibration result of the satellite accelerometer scaling factor based on the accelerometer scaling factor and the first weighting factor includes:

[0036] After weighted calculation based on the accelerometer scaling factor and the first weight factor, the on-orbit calibration result of the satellite accelerometer scaling factor is obtained.

[0037] The above-described one or more technical solutions in the embodiments of the present invention have at least one or more of the following technical effects:

[0038] This invention provides a data-optimized satellite rotational maneuvering accelerometer calibration method, comprising the following steps: Step 1: Obtaining star sensor data measured by a gravity satellite in orbit, and performing a first preprocessing method on the star sensor data, and calculating satellite angular acceleration data based on the star sensor data; Step 2: Obtaining accelerometer data measured by the gravity satellite in orbit, and performing a second preprocessing method on the accelerometer data; Step 3: Using correlation analysis, calculating the correlation between the first preprocessed star sensor data and the second preprocessed accelerometer data, and obtaining a first correlation coefficient between the first preprocessed star sensor data and the second preprocessed accelerometer data during the satellite maneuvering time period; Step 4: Based on the correlation principle, performing data segment truncation on the first preprocessed star sensor data and the second preprocessed accelerometer data. Next, the first star sensor data segment and the first accelerometer data segment are obtained; Step 5: From the first correlation coefficient, the first weighting factor corresponding to the first star sensor data segment and the first accelerometer data segment is obtained; Step 6: Based on the first star sensor data segment and the first accelerometer data segment, the accelerometer scaling factor is calculated using an optimized data selection method; Step 7: Based on the accelerometer scaling factor and the first weighting factor, the on-orbit calibration result of the satellite accelerometer scaling factor is obtained. This solves the technical problem in the prior art where the mismatch between accelerometer and star sensor data during satellite maneuvers leads to a low data signal-to-noise ratio, which in turn increases the calibration estimation error of the accelerometer scaling factor. This achieves the technical effect of improving the quality of satellite maneuver calibration data, making the accelerometer data more accurately reflect the real-time angular acceleration of the satellite, and providing accurate data support for improving the calibration of the gravity satellite accelerometer scaling factor.

[0039] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0040] Figure 1 This is a flowchart illustrating a data-optimized satellite rotational maneuvering accelerometer calibration method according to an embodiment of the present invention;

[0041] Figure 2 This is another flowchart illustrating a data-optimized satellite rotational maneuvering accelerometer calibration method according to an embodiment of the present invention;

[0042] Figure 3 This is a schematic diagram comparing the translational acceleration along the Y-axis before and after the center of mass moves when rotating around the X-axis in an embodiment of the present invention. Detailed Implementation

[0043] This invention provides a data-optimized satellite rotational maneuvering accelerometer calibration method to solve the technical problem in the prior art where the mismatch between accelerometer and star sensor data during satellite maneuvering leads to a low data signal-to-noise ratio, which in turn increases the calibration estimation error of the accelerometer scaling factor.

[0044] The overall concept of the technical solutions in the embodiments of the present invention is as follows:

[0045] This invention provides a data-optimized satellite rotational maneuvering accelerometer calibration method, comprising the following steps: Step 1: Obtaining star sensor data measured by a gravity satellite in orbit, and performing a first preprocessing method on the star sensor data, and calculating satellite angular acceleration data based on the star sensor data; Step 2: Obtaining accelerometer data measured by the gravity satellite in orbit, and performing a second preprocessing method on the accelerometer data; Step 3: Using correlation analysis, calculating the correlation between the first preprocessed star sensor data and the second preprocessed accelerometer data, and obtaining a first correlation coefficient between the first preprocessed star sensor data and the second preprocessed accelerometer data during the satellite maneuvering time period; Step 4: According to the correlation principle, adjusting the first preprocessed data... After segmenting the star sensor data and the accelerometer data after the second preprocessing, the first star sensor data segment and the first accelerometer data segment are obtained; Step 5: Obtain the first weight factor corresponding to the first star sensor data segment and the first accelerometer data segment from the first correlation coefficient; Step 6: Calculate the accelerometer scaling factor using an optimized data selection method based on the first star sensor data segment and the first accelerometer data segment; Step 7: Obtain the on-orbit calibration result of the satellite accelerometer scaling factor based on the accelerometer scaling factor and the first weight factor. This achieves the technical effect of improving the quality of satellite maneuver calibration data, making the accelerometer data more accurately reflect the real-time angular acceleration of the satellite, and providing accurate data support for improving the calibration of the gravity satellite accelerometer scaling factor.

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] Example

[0048] Figure 1 This invention provides a data-optimized calibration method for satellite rotational maneuvering accelerometers, as shown in the following embodiment. Figure 1 , 2 As shown, the method includes:

[0049] Step 1: Obtain star sensor data measured by the gravity satellite in orbit, and perform a first preprocessing on the star sensor data using a first processing method, and calculate the satellite angular acceleration data based on the star sensor data.

[0050] Specifically, a star sensor is a high-precision spatial attitude measurement device that uses stars as a reference frame and the starry sky as its working object. By detecting stars at different positions on the celestial sphere and performing calculations, it provides accurate spatial orientation and reference for aerospace vehicles such as satellites and spacecraft. Like inertial gyroscopes, it also possesses autonomous navigation capabilities and has significant application value. In this embodiment, star sensor data measured by a gravity satellite in orbit is first acquired. Then, the acquired star sensor data undergoes preprocessing operations such as deduplication and gross error removal. Specifically, data measurement time matching and time monotonicity are used to remove duplicates from the star sensor data; the median method is used to remove gross errors from the star sensor data. Furthermore, satellite angular acceleration data can be estimated using star sensor attitude data, i.e., by selecting two star sensors and using star sensor quaternion data to estimate the satellite angular acceleration data.

[0051] During long-term operation in orbit, star sensors are subject to various changes in the space environment and preprocessing anomalies. Therefore, the acquired satellite star sensor data often contains gross errors and other anomalies. This embodiment employs the median method (MAD), which has good robustness, to handle these gross errors. The formula is as follows:

[0052] |y i |>(m+n·MAD),

[0053] If the above formula conditions are met, the data at this sampling point is considered gross error, where m = Median(yi), MAD = Median{|yi-m| / 0.6745}. i This is the data sequence to be removed for gross errors, where n is the number of data points. Median is the middle number in the sorted data sequence, i.e., y. n / 2 This method performs outlier detection after differentiating the star sensor data once. Typically, the outlier data is set to 0. In this embodiment, after detecting outlier data, the corresponding star sensor data is set to empty, meaning the star sensor data for that epoch is missing. Median indicates that the data is removed in pairs, and the last remaining data point (the average of the two remaining data points) is the median of this data set.

[0054] The satellite angular acceleration data can be estimated by performing a second differential calculation on the star sensor data after removing outliers from the median. In other words, the satellite angular acceleration data can be estimated by performing a second differential calculation on the processed star sensor data with a high signal-to-noise ratio under the condition of continuous time.

[0055] Step 2: Obtain the accelerometer data measured by the gravity satellite in orbit, and perform a second preprocessing on the accelerometer data using the second processing method.

[0056] Specifically, after obtaining the accelerometer data from the gravity satellite's on-orbit measurement, the accelerometer data needs to be preprocessed accordingly. During the on-orbit measurement process of the gravity satellite accelerometer, there will be long-period drift. In order to avoid the jump introduced by solar radiation pressure, the calibration maneuver period should be selected during the period when no satellite enters or exits the shadow area. A low-pass filter is used to complete the debiasing and de-drift processing of the accelerometer data. Then, the accelerometer data and the star sensor data are time-aligned.

[0057] After the satellite enters steady-state operation, the output value of the electrostatic accelerometer can be expressed as: Where s is the scaling factor of the accelerometer. A is the non-gravitational acceleration within the satellite's own system, b is the zero bias value of the accelerometer, and in this embodiment, the zero bias value measured on the ground is used for calculation. n Measure noise.

[0058] During on-orbit measurements, gravity satellite accelerometers experience long-period drift. Therefore, this embodiment utilizes median filtering to remove drift and bias from the accelerometer data. The specific filtering method is as follows.

[0059] y(i)=Med[x(iN),…,x(i),…,x(i+N)],

[0060] Where N = (L-1) / 2, L is the length of the accelerometer data sequence, x(i) is the i-th element of the accelerometer data sequence, and Med indicates that low-pass median filtering is performed.

[0061] Step 3: Using correlation analysis, perform correlation calculation on the star sensor data after the first preprocessing and the accelerometer data after the second preprocessing to obtain the first correlation coefficient between the star sensor data after the first preprocessing and the accelerometer data after the second preprocessing during the satellite maneuvering time period.

[0062] Specifically, after the star sensor data and accelerometer data are processed, the star sensor data after the first preprocessing and the accelerometer data after the second preprocessing can be obtained. Then, the matching factor needs to be calculated for the star sensor data after the first preprocessing and the accelerometer data after the second preprocessing.

[0063] Correlation analysis was performed on the data from the satellite accelerometer calibration maneuver phase, and the cross-correlation coefficient was calculated. The correlation coefficient was calculated using the following formula:

[0064] Cov(A,S)=E{[AE(A)][SE(S)]},

[0065]

[0066] Where A represents the accelerometer data sequence, S represents the star sensor data sequence, E represents the expected value of the calculated data sequence, D represents the variance of the calculated data sequence, Cov represents the covariance, and cof represents the correlation coefficient of the calculated data.

[0067] Step 4: Based on the correlation principle, after segmenting the first preprocessed star sensor data and the second preprocessed accelerometer data, the first star sensor data segment and the first accelerometer data segment are obtained.

[0068] Specifically, based on different correlations, the data segments involved in the calculation need to be truncated to obtain the first star sensor data segment and the first accelerometer data segment. For example, a data sequence may contain 300 seconds of maneuver data. Correlation calculations need to be performed on these 300 seconds of data every 100 seconds, and the 100 seconds of data with the highest correlation are selected for calculation. In other words, the 300 seconds of data will have three 100-second segments, resulting in three correlation coefficients. When truncating, the 100-second segment with the highest correlation is selected.

[0069] Step 5: Obtain the first weighting factor corresponding to the first star sensor data segment and the first accelerometer data segment from the first correlation coefficient.

[0070] Specifically, by using correlation analysis, the first correlation coefficient between star sensor data and accelerometer data during the satellite maneuvering period is calculated. Then, the first correlation coefficient can be used as a weighting factor for data selection. The first weighting factor corresponding to the first star sensor data segment and the first accelerometer data segment can be obtained from the first correlation coefficient.

[0071] Step 6: Calculate the accelerometer scaling factor using an optimized data selection method based on the first star sensor data segment and the first accelerometer data segment.

[0072] Step 7: Obtain the on-orbit calibration result of the satellite accelerometer scaling factor based on the accelerometer scaling factor and the first weighting factor.

[0073] Specifically, based on the satellite maneuvering time period measurement data obtained through an optimized selection strategy—namely, the first star sensor data segment and the first accelerometer data segment—an optimized data selection method is used to estimate the accelerometer scaling factor. Multiple sets of calibration results are weighted and calculated based on the correlation weight factor of the accelerometer and star sensor data, thereby achieving on-orbit calibration of the satellite accelerometer scaling factor. In other words, based on the satellite maneuvering time period measurement data obtained through the optimized selection strategy, a rotating satellite maneuvering method is used, combined with the accelerometer and star sensor satellite maneuvering time period data, to achieve on-orbit calibration of the satellite accelerometer scaling factor. By adding the data correlation coefficient as a weight factor, the data selection strategy for the satellite rotation maneuvering time period is optimized, and then the on-orbit calibration of the accelerometer scaling factor parameters is achieved. This effectively improves the quality of the satellite maneuvering calibration data, making the accelerometer data more accurately reflect the real-time angular acceleration of the satellite, and providing precise data support for improving the calibration of the gravity satellite accelerometer scaling factor.

[0074] Furthermore, by optimizing the data selection method, the accelerometer scaling factor is estimated. This method combines the accelerometer and star sensor measurement data processing during the gravity satellite maneuvering period, effectively overcoming the problems of low accuracy and poor stability of traditional optimal estimation methods.

[0075] Furthermore, the scaling factor estimation for the rotating satellite accelerometer is as follows:

[0076] When an accelerometer is in orbit, it can be calibrated using the rotating satellite method. Assuming the satellite's center of mass is the origin, the acceleration introduced by the satellite's rotation is:

[0077] A in = -ω×(ω×r)+r×β+2υ×ω,

[0078] Where r represents the centroid coordinates of the accelerometer test mass, ω and β = dω / dt represent the angular velocity and angular acceleration of the satellite rotation, respectively, and υ represents the velocity of the test mass relative to the satellite. For the accelerometer in control mode, the test mass remains stationary relative to the capacitor plates, i.e., υ ≈ 0.

[0079] Assuming rotation about the X-axis, i.e. Let the satellite rotate by the following angle:

[0080] θ = θ0cosω0t, where θ0 is the initial rotation angle and ω0 is the initial rotation angular velocity.

[0081] The acceleration introduced by the rotating satellite is:

[0082]

[0083] Among them, A in a represents the triaxial angular acceleration introduced by rotation. x,in a y,in a z,in Let y and z be the centroid deviations of the satellite in the y and z directions, respectively. Ignoring the minor effects of coordinate axis installation deviations, the triaxial accelerometer measurement value Δa for the angular acceleration introduced by the rotating satellite is... x,out , Δa y,out , Δa z,out It can then be written as:

[0084]

[0085] Then the scaling factors in the y and z directions can be obtained as follows:

[0086]

[0087] Among them, s y s is the y-axis scaling factor. z is the scaling factor in the z-direction.

[0088] Based on GRACE-Fo satellite data, accelerometer data and star sensor data during the satellite maneuvering period were simulated. When the satellite rotates along the X-axis, the y-axis acceleration before and after the change of the center of mass, according to the demonstrated parameters, is shown in Figure 3. Moving the center of mass adjustment block along the Y-direction, for the satellite, the change in the spacecraft's center of mass Δy is mainly ensured by the accuracy of the center of mass adjustment device; θ0 is ensured by the measurement accuracy of the star sensor. According to the gravity satellite payload specifications, the uncertainties of Δy, θ0, and ω0 are δ(Δy) = 10 μm, δ(θ0) = 240 μrad, and δ(ω0) = 10 μrad, respectively. -3 mrad / s. The angular acceleration β obtained by still considering the method of using the torque generated by the thrusters to achieve satellite rotation is... x Let θ0 = 1 rad and ω0 = 7.7 × 10⁻⁶. -3 If the change in the center of mass is Δy = 2 mm (corresponding to the maximum center of mass offset of the gravity satellite), then this method can achieve an accuracy of δ(sy) / sy ~ 0.5% for the on-orbit calibration of the scaling factor.

[0089] In this embodiment, the accelerometer on the gravity satellite is mainly used to detect non-conservative forces acting on the satellite. Utilizing translational acceleration data measured by the accelerometer after the satellite's rotational maneuver and angular acceleration data obtained from star sensor measurements, the correlation coefficient between the star sensor data and the accelerometer data is used as a weighting factor. This embodiment employs an optimized data selection method to estimate the accelerometer scaling factor, resulting in a more accurate accelerometer scaling factor and providing a technical foundation for subsequent calibration of domestic gravity satellites. Specifically, starting with the satellite maneuvering method and maneuvering calibration data processing, an on-orbit calibration of the accelerometer scaling factor, along with related data processing methods and procedures, is constructed, providing a fundamental analytical tool for on-orbit calibration of gravity satellites.

[0090] The above-described one or more technical solutions in the embodiments of the present invention have at least one or more of the following technical effects:

[0091] This invention provides a data-optimized satellite rotational maneuvering accelerometer calibration method, comprising the following steps: Step 1: Obtaining star sensor data measured by a gravity satellite in orbit, and performing a first preprocessing method on the star sensor data, and calculating satellite angular acceleration data based on the star sensor data; Step 2: Obtaining accelerometer data measured by the gravity satellite in orbit, and performing a second preprocessing method on the accelerometer data; Step 3: Using correlation analysis, calculating the correlation between the first preprocessed star sensor data and the second preprocessed accelerometer data, and obtaining a first correlation coefficient between the first preprocessed star sensor data and the second preprocessed accelerometer data during the satellite maneuvering time period; Step 4: Based on the correlation principle, performing data segment truncation on the first preprocessed star sensor data and the second preprocessed accelerometer data. Next, the first star sensor data segment and the first accelerometer data segment are obtained; Step 5: From the first correlation coefficient, the first weighting factor corresponding to the first star sensor data segment and the first accelerometer data segment is obtained; Step 6: Based on the first star sensor data segment and the first accelerometer data segment, the accelerometer scaling factor is calculated using an optimized data selection method; Step 7: Based on the accelerometer scaling factor and the first weighting factor, the on-orbit calibration result of the satellite accelerometer scaling factor is obtained. This solves the technical problem in the prior art where the mismatch between accelerometer and star sensor data during satellite maneuvers leads to a low data signal-to-noise ratio, which in turn increases the calibration estimation error of the accelerometer scaling factor. This achieves the technical effect of improving the quality of satellite maneuver calibration data, making the accelerometer data more accurately reflect the real-time angular acceleration of the satellite, and providing accurate data support for improving the calibration of the gravity satellite accelerometer scaling factor.

[0092] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the invention.

[0093] Obviously, those skilled in the art can make various modifications and variations to the embodiments of the present invention without departing from the spirit and scope of the embodiments of the present invention. Thus, if these modifications and variations to the embodiments of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention also intends to include these modifications and variations.

Claims

1. A data-optimization-based satellite rotational maneuver accelerometer calibration method, characterized in that, include: Step 1: Obtain star sensor data measured by the gravity satellite in orbit, and use the first processing method to perform a first preprocessing on the star sensor data, and calculate the satellite angular acceleration data based on the star sensor data; Step 2: Obtain the accelerometer data measured by the gravity satellite in orbit, and perform a second preprocessing on the accelerometer data using the second processing method; Step 3: Using correlation analysis, perform correlation calculation on the first preprocessed star sensor data and the second preprocessed accelerometer data to obtain the first correlation coefficient between the first preprocessed star sensor data and the second preprocessed accelerometer data during the satellite maneuvering time period. Step 4: Based on the correlation principle, after segmenting the first preprocessed star sensor data and the second preprocessed accelerometer data, the first star sensor data segment and the first accelerometer data segment are obtained. Step 5: Obtain the first weighting factor corresponding to the first star sensor data segment and the first accelerometer data segment from the first correlation coefficient, wherein the first correlation coefficient is used as the first weighting factor; Step 6: Calculate the accelerometer scaling factor using an optimized data selection method based on the first star sensor data segment and the first accelerometer data segment; Step 7: Obtain the on-orbit calibration result of the satellite accelerometer scaling factor based on the accelerometer scaling factor and the first weighting factor, including: performing a weighted calculation based on the accelerometer scaling factor and the first weighting factor to obtain the on-orbit calibration result of the satellite accelerometer scaling factor; In step 1, the first processing method includes: The star sensor data is deduplicated by measuring time matching and time monotonicity. The star sensor data was subjected to gross error removal using the median method. In step 2, the second processing method includes: The accelerometer data is debiased and de-drifted using a low-pass filter.

2. The method of claim 1, wherein the satellite rotational maneuver accelerometer calibration is performed by: In step 1, the star sensor data is subjected to gross error removal using the median method, including: ， Wherein, if the formula condition is met, the sampling point data is gross error, wherein , , y i is a sequence of gross error data to be removed, n is the number of data, and Median is the middle number in the sorted data sequence, that is .

3. The method of claim 1, wherein the satellite rotational maneuver accelerometer calibration is performed by: In step 1, the satellite angular acceleration data is calculated based on the star sensor data, including: After removing outliers from the star sensor data using the median method, the satellite angular acceleration data is obtained by performing a second differential calculation.

4. The method of claim 1, wherein the satellite rotational maneuver accelerometer calibration is performed by: In step 2, the process of using a low-pass filter to debias and de-drift the accelerometer data specifically includes: ， in, L is the length of the accelerometer data sequence. x ( i ) represents the i-th element of the accelerometer data sequence, and Med is the low-pass median filter calculation.

5. The method of claim 1, wherein the satellite rotational maneuver accelerometer calibration is performed by: In step 3, the correlation calculation of the star sensor data after the first preprocessing and the accelerometer data after the second preprocessing includes: ， ， wherein, A is a sequence of accelerometer data, S is a sequence of star sensor data, E is a sequence of computed data, D is a sequence of computed data, Cov is a computed covariance, cof is a computed correlation coefficient of data.

6. The satellite rotational accelerometer calibration method as described in claim 1, characterized in that, In step 6, the step of calculating the accelerometer scaling factor using an optimized data selection method based on the first star sensor data segment and the first accelerometer data segment includes: Let the satellite's center of mass be the origin of the coordinate system. Then the acceleration introduced by the satellite's rotation is: ， in, A in Acceleration introduced for satellite rotation, To determine the center-of-mass coordinates of the accelerometer's test mass, The angular velocity of the satellite's rotation. The angular acceleration of the satellite's rotation. To verify the velocity of the mass relative to the satellite, where, for the accelerometer in controlled mode, the mass remains stationary relative to the capacitor plates, then... ; When the rotating satellite rotates around the X-axis, that is Then the satellite's rotation angle is: ， in, The initial rotation angle, Given the initial rotational angular velocity, the acceleration introduced by the rotating satellite is: ， in, The triaxial angular accelerations introduced for rotation are respectively , , Let y and z be the centroid deviations of the satellite in the y and z directions, respectively. Ignoring the minor influence of coordinate axis installation deviations, the triaxial accelerometer measurements of the angular acceleration introduced by the rotating satellite are... , , for: ， The scaling factors in the y and z directions were calculated. , for: ， in, s y The y-axis scaling factor. s z is the scaling factor in the z-direction.

Images