Gyroscope vibration error compensation method and system based on hemispherical resonator

Abstract

The application is a gyro vibration error compensation method and system based on a hemispherical resonator, and belongs to the technical field of inertial navigation and precision measurement. Through multi-working condition calibration of a vibration table, working condition grid points are constructed in a dynamic characteristic parameter space with working resonant frequency and damping characteristic quantity as coordinates, forming a compensation coefficient library; during operation, driving signals, output signals and working condition parameters are collected, recursive least square algorithm is used to identify dynamic characteristic parameter estimates on line, vibration error compensation coefficients of target working condition grid points are interpolated in the dynamic characteristic parameter space, and candidate compensation coefficients are obtained; through double threshold value judgment of residual vibration error index and coefficient change after compensation, compensation coefficient write-back update is controlled, and temperature threshold value and time threshold value are combined for whole life cycle management. The application can adaptively adjust vibration error compensation coefficients with working resonant frequency and damping characteristic quantity drift, and reduce the number of factory calibration.

Description

Technical Field

[0001] This invention relates to the field of inertial navigation and precision measurement technology, specifically to a gyroscope vibration error compensation method and system based on a hemispherical harmonic oscillator. Background Technology

[0002] Hemispherical resonator gyroscopes measure angular velocity through the standing wave motion of a hemispherical resonator, offering high accuracy and long-term stability. However, in complex vibration environments, external vibrations can couple to the hemispherical resonator through the mounting structure, causing vibration errors in the output signal. To reduce these errors, current technologies typically employ static calibration on a vibration table or vibration calibration under finite-condition conditions to obtain vibration error compensation coefficients, which are then used to compensate the output signal during operation.

[0003] However, during service, the dynamic characteristic parameters of a hemispherical resonant gyroscope, such as its operating resonant frequency and damping characteristics, drift with changes in ambient temperature, structural aging, and installation stress. Fixed compensation coefficients obtained solely from factory static calibration or finite-condition calibration become insufficient to cover new vibration conditions in the later stages of service, leading to a gradual decline in vibration error compensation. To maintain compensation performance, frequent factory recalibration is required, increasing maintenance costs and downtime. Summary of the Invention

[0004] The purpose of this invention is to provide a solution to one of the aforementioned problems existing in the prior art.

[0005] Specifically, the present invention is achieved through the following technical solution:

[0006] like Figure 1 As shown, the gyroscope vibration error compensation method based on a hemispherical harmonic oscillator includes the following steps:

[0007] Step 1: Calibrate the hemispherical resonant gyroscope on a vibration table under multiple vibration frequencies, vibration amplitudes and ambient temperatures to obtain vibration error compensation coefficients and calibration dynamic characteristic parameters for each working condition. Construct a working condition grid point in the dynamic characteristic parameter space with the working resonant frequency and damping characteristic quantity as coordinates to form a compensation coefficient library.

[0008] Step 2: During operation, collect drive signals, gyroscope output signals, and operating condition parameter sets, and select the target operating condition grid point set from the compensation coefficient library based on the operating condition parameter sets;

[0009] Step 3: Preprocess the drive signal and gyroscope output signal within the sliding time window, and input the preprocessed signal into the recursive least squares algorithm to obtain the estimated values ​​of dynamic characteristic parameters.

[0010] Step 4: Based on the position of the estimated dynamic characteristic parameters in the dynamic characteristic parameter space, select adjacent working condition grid points from the target working condition grid point set that are close to the estimated dynamic characteristic parameters in the direction of the working resonant frequency and the direction of the damping characteristic. Perform interpolation calculation on the vibration error compensation coefficients of the adjacent working condition grid points to obtain the candidate vibration error compensation coefficients.

[0011] Step 5: Use candidate vibration error compensation coefficients to compensate for vibration error in the gyroscope output signal. Calculate the residual vibration error index after compensation and the coefficient change between the candidate vibration error compensation coefficient and the effective vibration error compensation coefficient within the sliding time window. Generate an updated decision result based on the residual vibration error index, residual error threshold, coefficient change, and coefficient change threshold.

[0012] Step 6: When the updated decision result meets the update conditions, write the candidate vibration error compensation coefficient into the compensation coefficient library as the new effective vibration error compensation coefficient, and record the working condition parameter group, the estimated value of dynamic characteristic parameters and the update time. Execute steps 2 to 5 periodically according to the temperature threshold and time threshold.

[0013] Furthermore, the hemispherical resonant gyroscope is calibrated on a vibration table under multiple vibration frequencies, vibration amplitudes, and ambient temperatures to obtain vibration error compensation coefficients and calibration dynamic characteristic parameters for each operating condition. A grid of operating condition points is constructed in the dynamic characteristic parameter space with the operating resonant frequency and damping characteristic quantity as coordinates to form a compensation coefficient library, including:

[0014] Multiple sinusoidal vibration conditions are applied to the vibration table, each with different vibration frequency, vibration amplitude, and ambient temperature. Under each sinusoidal vibration condition, the output signal of the hemispherical resonator gyroscope and the reference angular velocity signal are acquired. The corresponding vibration error compensation coefficient is determined based on the deviation between the output signal and the reference angular velocity signal. Based on the response of the hemispherical resonator gyroscope to the drive signal, the calibration dynamic characteristic parameters for each sinusoidal vibration condition are determined. The calibration dynamic characteristic parameters include at least the calibration working resonant frequency and the calibration damping characteristic quantity. The vibration error compensation coefficient and the calibration dynamic characteristic parameters are mapped to the condition grid points in the dynamic characteristic parameter space, forming the condition grid points in the compensation coefficient library.

[0015] Furthermore, the process of acquiring drive signals, gyroscope output signals, and operating parameter sets during operation, and selecting the target operating condition grid point set from the compensation coefficient library based on the operating parameter sets, includes:

[0016] The drive signal and gyroscope output signal are synchronously acquired at a sampling frequency greater than or equal to twice the working resonant frequency band; the ambient temperature and vibration acceleration at the installation position of the hemispherical resonant gyroscope are acquired through temperature sensors and accelerometers, and the ambient temperature and vibration acceleration are combined with vibration frequency and vibration amplitude to form a set of working condition parameters; the set of working condition parameters is compared with the set of working condition parameters corresponding to each working condition grid point in the compensation coefficient library, and the target set of working condition grid points is obtained from the compensation coefficient library by using the nearest neighbor matching method.

[0017] Furthermore, the preprocessing of the drive signal and gyroscope output signal within the sliding time window, followed by inputting the preprocessed signal into a recursive least squares algorithm to obtain estimated values ​​of dynamic characteristic parameters, includes:

[0018] The drive signal and gyroscope output signal are bandpass filtered to retain the signal within the operating resonant frequency band. The filtered drive signal and the filtered gyroscope output signal are then subjected to amplitude normalization and phase alignment to obtain a preprocessed signal. Within a sliding time window, the preprocessed signal is input into a recursive least squares algorithm. Based on the dynamic characteristic parameter estimates from the previous moment and the preprocessed signal within the current sliding time window, the dynamic characteristic parameter estimates are iteratively updated. The dynamic characteristic parameter estimates include the operating resonant frequency estimate and the damping characteristic quantity estimate.

[0019] Furthermore, based on the position of the estimated dynamic characteristic parameters in the dynamic characteristic parameter space, adjacent working condition grid points that are close to the estimated dynamic characteristic parameters in the direction of the working resonant frequency and the direction of damping characteristics are selected from the target working condition grid point set. Interpolation calculations are then performed on the vibration error compensation coefficients of the adjacent working condition grid points to obtain candidate vibration error compensation coefficients, including:

[0020] In the dynamic characteristic parameter space, with the estimated dynamic characteristic parameter as the center, select the working condition grid points located on both sides of the estimated working resonant frequency in the direction of the working resonant frequency and on both sides of the estimated damping characteristic in the direction of the damping characteristic to form the adjacent working condition grid point set. Based on the difference between the estimated dynamic characteristic parameter and the corresponding calibrated dynamic characteristic parameter of the adjacent working condition grid points, perform piecewise linear interpolation on the vibration error compensation coefficient of the adjacent working condition grid points to obtain the candidate vibration error compensation coefficient.

[0021] Furthermore, the method of using candidate vibration error compensation coefficients to compensate for vibration errors in the gyroscope output signal, calculating the residual vibration error index after compensation and the coefficient change between the candidate vibration error compensation coefficients and the effective vibration error compensation coefficients within a sliding time window, and generating an updated decision result based on the residual vibration error index, residual error threshold, coefficient change, and coefficient change threshold, includes:

[0022] Within a sliding time window, candidate vibration error compensation coefficients are used to compensate the gyroscope output signal for vibration error, resulting in a compensated output signal. The compensated output signal is compared with a reference angular velocity signal, and the mean square value of the compensated residual vibration error is calculated as the compensated residual vibration error index. The difference between the candidate vibration error compensation coefficient and the effective vibration error compensation coefficient is calculated, and the maximum absolute value of each component of the difference is taken as the coefficient change. An updated decision result is generated based on the comparison results between the compensated residual vibration error index and the residual error threshold, as well as the comparison results between the coefficient change and the coefficient change threshold.

[0023] Furthermore, the update conditions include that the residual vibration error index after compensation is less than the residual error threshold and the coefficient change is less than the coefficient change threshold; when the update decision result does not meet the update conditions, the effective vibration error compensation coefficient remains unchanged, the current update decision is recorded as invalid update, and the number of invalid updates is accumulated. When the number of invalid updates reaches the invalid update count threshold, a fault prompt message is output.

[0024] Furthermore, the recorded operating condition parameter set, dynamic characteristic parameter estimate, and update time include:

[0025] The working condition parameter set corresponding to the effective vibration error compensation coefficient, the estimated value of dynamic characteristic parameters, and the time of writing into the compensation coefficient library are combined to form the compensation coefficient evolution record. The compensation coefficient evolution record is stored in association with the compensation coefficient library to analyze the evolution trend of vibration error compensation performance with working time and changes in the working condition parameter set.

[0026] Furthermore, the temperature threshold and time threshold are set based on the design life and allowable vibration error variation range of the hemispherical resonator gyroscope. When the change in ambient temperature exceeds the temperature threshold, it indicates that the ambient temperature has changed across intervals. When the increase in cumulative working time exceeds the time threshold, it indicates that the aging degree of the structure has increased to a new stage. Steps two to five are executed periodically based on the comparison results between the change in ambient temperature and the increase in cumulative working time and the temperature threshold and time threshold, in order to maintain the consistency between the effective vibration error compensation coefficient and the estimated value of the dynamic characteristic parameter.

[0027] A gyroscope vibration error compensation system based on a hemispherical harmonic oscillator, employing the aforementioned gyroscope vibration error compensation method based on a hemispherical harmonic oscillator, includes: a vibration calibration module, an online data acquisition module, a dynamic characteristic parameter identification module, a parameter space interpolation module, a compensation and evaluation module, an update and management module, and a data processing module; the vibration calibration module, online data acquisition module, dynamic characteristic parameter identification module, parameter space interpolation module, compensation and evaluation module, and update and management module are respectively connected to the data processing module;

[0028] The vibration calibration module is used to calibrate the hemispherical resonant gyroscope on a vibration table under multiple vibration frequencies, vibration amplitudes and ambient temperatures, to obtain vibration error compensation coefficients and calibration dynamic characteristic parameters for each working condition, and to construct working condition grid points in the dynamic characteristic parameter space with the working resonant frequency and damping characteristic quantity as coordinates to form a compensation coefficient library.

[0029] The online data acquisition module is used to acquire drive signals, gyroscope output signals, and operating condition parameter sets during operation; the operating condition association module is used to select the target operating condition grid point set from the compensation coefficient library based on the operating condition parameter set.

[0030] The dynamic characteristic parameter identification module is used to preprocess the drive signal and gyroscope output signal within a sliding time window, and obtain the estimated value of the dynamic characteristic parameter through a recursive least squares algorithm.

[0031] The parameter space interpolation module is used to select adjacent working condition grid points from the target working condition grid point set based on the position of the estimated dynamic characteristic parameter value in the dynamic characteristic parameter space, and to perform interpolation calculation on the vibration error compensation coefficient of the adjacent working condition grid points to obtain candidate vibration error compensation coefficients.

[0032] The compensation and evaluation module is used to compensate the gyroscope output signal for vibration error using candidate vibration error compensation coefficients, calculate the residual vibration error index and coefficient change after compensation, and generate an updated decision result based on the residual vibration error index, residual error threshold, coefficient change and coefficient change threshold after compensation.

[0033] The update and management module is used to write candidate vibration error compensation coefficients into the compensation coefficient library as new effective vibration error compensation coefficients when the update judgment result meets the update conditions. It records the working condition parameter group, the estimated value of dynamic characteristic parameters and the update time, manages the evolution record of compensation coefficients, and controls the operation of the online data acquisition module, working condition association module, dynamic characteristic parameter identification module, parameter space interpolation module and compensation and evaluation module according to the temperature threshold and time threshold period.

[0034] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0035] A compensation coefficient library is constructed in the dynamic characteristic parameter space, and the vibration error compensation coefficient is mapped to the working resonant frequency and damping characteristic quantity, so that the compensation coefficient can be adjusted as the dynamic characteristic parameters change.

[0036] The estimated values ​​of dynamic characteristic parameters are identified by the recursive least squares algorithm within the sliding time window, so that the compensation coefficients are updated to reflect the current actual structural state.

[0037] The dual threshold update condition is constructed by the coefficient change between the candidate vibration error compensation coefficient and the effective vibration error compensation coefficient, as well as the residual vibration error index after compensation. This ensures the compensation effect while controlling the update amplitude of the compensation coefficient and preventing compensation divergence.

[0038] By recording the evolution of compensation coefficients and triggering periodic updates using temperature and time thresholds, vibration error compensation management can be achieved throughout the entire life cycle, reducing the frequency of return-to-factory calibration. Attached Figure Description

[0039] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0040] Figure 1 This is a flowchart illustrating the vibration error compensation method for gyroscopes based on hemispherical harmonic oscillators. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0042] Example 1

[0043] like Figure 1 As shown, the gyroscope vibration error compensation method based on a hemispherical harmonic oscillator includes the following steps:

[0044] Step 1: Calibrate the hemispherical resonant gyroscope on a vibration table under multiple vibration frequencies, vibration amplitudes and ambient temperatures to obtain vibration error compensation coefficients and calibration dynamic characteristic parameters for each working condition. Construct a working condition grid point in the dynamic characteristic parameter space with the working resonant frequency and damping characteristic quantity as coordinates to form a compensation coefficient library.

[0045] Step 2: During operation, collect drive signals, gyroscope output signals, and operating condition parameter sets, and select the target operating condition grid point set from the compensation coefficient library based on the operating condition parameter sets;

[0046] Step 3: Preprocess the drive signal and gyroscope output signal within the sliding time window, and input the preprocessed signal into the recursive least squares algorithm to obtain the estimated values ​​of dynamic characteristic parameters.

[0047] Step 4: Based on the position of the estimated dynamic characteristic parameters in the dynamic characteristic parameter space, select adjacent working condition grid points from the target working condition grid point set that are close to the estimated dynamic characteristic parameters in the direction of the working resonant frequency and the direction of the damping characteristic. Perform interpolation calculation on the vibration error compensation coefficients of the adjacent working condition grid points to obtain the candidate vibration error compensation coefficients.

[0048] Step 5: Use candidate vibration error compensation coefficients to compensate for vibration error in the gyroscope output signal. Calculate the residual vibration error index after compensation and the coefficient change between the candidate vibration error compensation coefficient and the effective vibration error compensation coefficient within the sliding time window. Generate an updated decision result based on the residual vibration error index, residual error threshold, coefficient change, and coefficient change threshold.

[0049] Step 6: When the updated decision result meets the update conditions, write the candidate vibration error compensation coefficient into the compensation coefficient library as the new effective vibration error compensation coefficient, and record the working condition parameter group, the estimated value of dynamic characteristic parameters and the update time. Execute steps 2 to 5 periodically according to the temperature threshold and time threshold.

[0050] Specifically, a vibration table is used to calibrate the hemispherical resonator gyroscope under multiple operating conditions. Several vibration frequency values ​​and vibration amplitudes, as well as several ambient temperature points, are selected. The vibration frequency, vibration amplitude, and ambient temperature are combined to form a set of discrete vibration calibration conditions. Each vibration calibration condition corresponds to a unique set of vibration frequency, vibration amplitude, and ambient temperature.

[0051] Under each vibration calibration condition: a sinusoidal excitation is applied to the vibration table via the vibration table controller, causing the vibration table to output vibrations with the corresponding vibration frequency and amplitude; the ambient temperature is kept stable near the target temperature point; the output signal of the hemispherical resonator gyroscope is acquired; and the reference angular velocity signal is acquired synchronously. The reference angular velocity signal can be obtained from the output angular velocity of a high-precision rotary table or from a high-precision reference gyroscope.

[0052] For each vibration calibration condition, the determination of the vibration error compensation coefficient includes the following process:

[0053] Within the recorded time sequence, for each sampling moment, the output signal of the hemispherical resonator gyroscope and the corresponding reference angular velocity signal are read.

[0054] The error signal at that sampling moment is obtained by subtracting the reference angular velocity signal from the output signal of the hemispherical resonator gyroscope. The error signals from all sampling moments are then arranged in chronological order to form an error signal sequence.

[0055] Before calibration, a vibration error compensation structure is pre-set. This structure can be constructed using a combination of zero-bias compensation, proportional compensation, and higher-order term compensation. Zero-bias compensation is used to offset a fixed offset in the output signal under the corresponding vibration calibration conditions, making the average output close to the reference angular velocity signal when there is no vibration input. Proportional compensation adjusts the compensation amount based on the magnitude of vibration-related observations, which can be vibration acceleration characteristic values, drive signal amplitudes, or other vibration-related signal quantities. By making the compensation amount proportional to these observations, the output error caused by changes in vibration amplitude is reduced. Higher-order term compensation describes the nonlinear components present in the vibration effect. It constructs the compensation amount by combining vibration-related observations in a higher-order form, where the higher-order form can be the square of the observation, the cube of the observation, or the product of multiple observations, thereby compensating for errors that vary nonlinearly with vibration intensity. By combining zero-bias compensation, proportional compensation, and higher-order term compensation to form a vibration error compensation structure, fixed offsets, linear vibration errors, and nonlinear vibration errors can be covered, achieving precise compensation for vibration errors under the target vibration calibration conditions.

[0056] Based on the above structure, the number of vibration error compensation coefficients to be solved and the physical meaning of each compensation coefficient are determined. For example, a certain compensation coefficient corresponds to a fixed offset, a certain compensation coefficient corresponds to a compensation amount proportional to the vibration acceleration, and a certain compensation coefficient corresponds to a compensation amount proportional to the square of the vibration acceleration.

[0057] Based on the pre-selected vibration error compensation structure, the various compensation quantities involved in zero-bias compensation, proportional compensation, and higher-order term compensation are uniformly expressed as a linear combination relationship between several compensation coefficients and the corresponding signal quantities, specifically including:

[0058] For each sampling moment, collect the vibration-related signal quantities at that moment, such as vibration acceleration characteristic value, the square of vibration acceleration characteristic value, driving signal amplitude, the square of driving signal amplitude, vibration frequency related quantities, etc.; arrange the above signal quantities according to the pre-selected compensation structure to form a set of signal quantity combinations used to constitute the compensation quantity; assign a vibration error compensation coefficient to be solved to each signal quantity combination; form a set of vibration error compensation coefficients to be solved to all the vibration error compensation coefficients, and each compensation coefficient in the set corresponds to a compensation component in the compensation structure.

[0059] Through the above process, a correspondence between the error signal sequence and a set of vibration error compensation coefficients to be solved was established under each vibration calibration condition.

[0060] To determine a suitable set of vibration error compensation coefficients so that the compensated error under the vibration calibration condition is as small as possible, the following specific steps are taken:

[0061] In the initial stage, an initial value is set for each vibration error compensation coefficient. The initial value can be zero or an approximate value set based on experience. Under the current set of vibration error compensation coefficients, for each sampling moment in the error signal sequence, the compensation amount for that sampling moment is calculated according to the pre-selected vibration error compensation structure. The compensation amount is subtracted from the output signal of the hemispherical resonator gyroscope to obtain the compensated output signal. Since the compensation amount in this invention is essentially the estimated vibration error itself, this part of the error needs to be removed from the output, hence the subtraction. The compensated output signal is subtracted from the reference angular velocity signal to obtain the compensated error signal. The compensated error signals for all sampling moments in the calibration time sequence are statistically analyzed, and the statistical index of the compensated error signal is calculated. The statistical index can be the mean square value or the absolute average value.

[0062] Compare the statistical indicators with the statistical indicators corresponding to the compensation coefficients in the previous round. If the statistical indicators decrease, it means that the current combination of vibration error compensation coefficients reduces the error, and the compensation coefficients can be fine-tuned along the current adjustment direction. If the statistical indicators increase, it means that the current adjustment direction is not conducive to reducing the error, and the adjustment direction or adjustment step size of the compensation coefficients needs to be changed.

[0063] In each iteration, for each vibration error compensation coefficient, the compensation coefficient is increased or decreased according to the goal of reducing the statistical index, until the statistical index no longer decreases significantly or the number of iterations reaches the preset upper limit.

[0064] When the statistical indicators do not change much in multiple consecutive iterations, or when the pre-set stopping conditions are met, the set of vibration error compensation coefficients at this time is taken as the target set of vibration error compensation coefficients corresponding to the vibration calibration condition and stored in the compensation coefficient library.

[0065] Under each vibration calibration condition, the calibration dynamic characteristic parameters of the hemispherical harmonic oscillator are obtained: based on the driving signal and gyroscope output signal during the vibration calibration period, frequency analysis and recursive estimation methods are used to obtain the operating resonant frequency and damping characteristic parameters of the hemispherical harmonic oscillator under that condition, specifically including:

[0066] The determination of the operating resonant frequency includes: on the one hand, performing frequency domain analysis on the gyroscope output signal within the calibration time period, finding the position of the main peak with the largest amplitude in the spectrum, and taking the frequency corresponding to the main peak as the operating resonant frequency; on the other hand, establishing an autoregressive structure for the output signal in the time domain, adjusting the frequency-related parameters in the autoregressive structure in a recursive manner, so that the error between the predicted signal generated according to the autoregressive structure and the actual output signal gradually decreases. When the prediction error meets the preset criteria, the frequency-related parameters at this time are read, and the frequency corresponding to the parameter is taken as the operating resonant frequency.

[0067] The damping characteristic can be determined by analyzing the time-dependent decay characteristics of the output signal: During vibration calibration, several time periods are selected, and the envelope of the output signal is extracted for each period. The decreasing rate of the envelope amplitude at different times is compared, and the rate of decrease of the envelope amplitude over time characterizes the energy dissipation characteristic, which is then converted into a damping characteristic. Alternatively, during recursive estimation, a parameter related to the decay rate can be introduced into the parameter set. This parameter reflects the decreasing trend of the output signal amplitude over time, and can be used as a damping characteristic or for calculating damping characteristics. Through the above frequency analysis and recursive estimation process, the corresponding operating resonant frequency and damping characteristic can be determined for each vibration calibration condition and stored in the compensation coefficient library as calibration dynamic characteristic parameters.

[0068] A dynamic characteristic parameter space is defined, with the operating resonant frequency as the abscissa and the damping characteristic quantity as the ordinate. For each vibration calibration condition, the vibration error compensation coefficient is mapped to the calibration operating resonant frequency and the calibration damping characteristic quantity, forming a condition grid point. All condition grid points are stored in the compensation coefficient library. Each condition grid point simultaneously records the vibration frequency, vibration amplitude, ambient temperature, calibration operating resonant frequency, calibration damping characteristic quantity, and vibration error compensation coefficient. This forms a discrete dynamic characteristic parameter space grid, with each grid point bound to a complete set of condition parameters and a set of vibration error compensation coefficients.

[0069] Online data acquisition is performed during the operation of the hemispherical resonator gyroscope, and a set of target operating condition grid points is selected from the compensation coefficient library. A sampling module is used to synchronously acquire the drive signal and the gyroscope output signal at a sampling frequency greater than or equal to twice the operating resonant frequency band; a temperature sensor is used to acquire the ambient temperature near the hemispherical resonator gyroscope; and an accelerometer is used to acquire the vibration acceleration at the installation location of the hemispherical resonator gyroscope.

[0070] The operating parameter group consists of ambient temperature, vibration acceleration characteristic value at the installation location, current drive vibration frequency, and current drive vibration amplitude. The vibration acceleration characteristic value at the installation location can be obtained by calculating the effective value or peak value of the vibration acceleration within a fixed time window. The current drive vibration frequency and current drive vibration amplitude can be directly read from the drive signal control command.

[0071] During the working condition association process, for each working condition grid point in the compensation coefficient library, the working condition parameter group recorded for that grid point is read. The ambient temperature, vibration acceleration characteristic value, driving vibration frequency, and driving vibration amplitude in the current working condition parameter group are compared with the ambient temperature, vibration acceleration characteristic value, driving vibration frequency, and driving vibration amplitude in the working condition parameter group of the current working condition grid point, respectively. The differences in ambient temperature, vibration acceleration characteristic value, driving vibration frequency, and driving vibration amplitude are calculated, and the absolute values ​​of these differences are taken. The differences in ambient temperature, vibration acceleration characteristic value, driving vibration frequency, and driving vibration amplitude are weighted and summed according to preset weights to obtain the comprehensive working condition difference for that grid point. The preset weights are set according to application requirements. The comprehensive working condition differences of all working condition grid points in the compensation coefficient library are sorted in ascending order. Several working condition grid points with the smallest comprehensive working condition differences are selected. The number of these selected grid points is determined by a preset integer, and these selected grid points form the target working condition grid point set.

[0072] Through the above process, the target working condition grid point set contains a set of working condition grid points that are similar to the current working condition parameter set in terms of external vibration frequency, vibration amplitude, and temperature.

[0073] Define a sliding time window with a fixed length on the time axis, for example, containing a fixed number of sampling points; whenever a new segment of data is sampled, move the time window forward by a certain length so that the time window always contains the latest continuous segment of data.

[0074] Bandpass filtering is performed on the drive signal and gyroscope output signal within the sliding time window to retain the signal components within the working resonant frequency band and suppress low-frequency drift interference and high-frequency noise. Amplitude normalization is performed on the filtered drive signal and the filtered gyroscope output signal respectively, that is, the amplitude of the signal is scaled to a uniform range by a fixed ratio. Phase alignment is performed on the filtered drive signal and the filtered gyroscope output signal to reduce the phase difference between the two on the time axis to a predetermined range, thus obtaining the preprocessed signal.

[0075] In the first sliding time window, the estimated values ​​of the dynamic characteristic parameters are initialized to the average or predicted values ​​of the calibrated dynamic characteristic parameters. Within each sliding time window, the estimated values ​​of the dynamic characteristic parameters obtained in the previous sliding time window are used as the initial parameters for the current sliding time window. Within the current sliding time window, parameter estimation equations are constructed based on the preprocessed signal, and the estimated values ​​of the operating resonant frequency and damping characteristics are adjusted recursively. The recursive method is as follows:

[0076] At each sampling moment, a predicted output is calculated based on the current parameter estimate. The predicted output is compared with the actual output to obtain the error. The parameter estimate is adjusted according to the magnitude and direction of the error to reduce the difference between the next predicted output and the actual output. After the entire sliding time window ends, the updated dynamic characteristic parameter estimate is obtained, which includes the working resonant frequency estimate and the damping characteristic quantity estimate.

[0077] By using a sliding time window and iterative recursive least squares algorithm, this method can track the changes in the estimated operating resonant frequency and damping characteristic values ​​over time.

[0078] Based on the position of the estimated dynamic characteristic parameters in the dynamic characteristic parameter space, adjacent working condition grid points are selected from the target working condition grid point set, and candidate vibration error compensation coefficients are generated.

[0079] In the target operating condition grid point set, read the calibration operating resonant frequency and calibration damping characteristic of each operating condition grid point; in the direction of the operating resonant frequency, find the operating condition grid point with the largest calibration operating resonant frequency among the operating condition grid points with a calibration operating resonant frequency less than the estimated value of the operating resonant frequency, and record it as the operating condition grid point below the frequency; find the operating condition grid point with the smallest calibration operating resonant frequency among the operating condition grid points with a calibration operating resonant frequency greater than the estimated value of the operating resonant frequency, and record it as the operating condition grid point above the frequency.

[0080] In the direction of damping characteristics, the working condition grid points on the upper and lower sides of the estimated value of damping characteristic quantity are found in the same way and are denoted as the lower damping working condition grid point and the upper damping working condition grid point, respectively.

[0081] As needed, frequency upper and lower side operating condition grid points and damping upper and lower side operating condition grid points can be combined to form multiple adjacent operating condition grid points within a rectangular neighborhood, thus forming a set of adjacent operating condition grid points.

[0082] For each component of the vibration error compensation coefficient, perform the following steps:

[0083] Calculate the difference between the estimated operating resonant frequency and the calibrated operating resonant frequency at the lower frequency operating condition grid point, and calculate the difference between the calibrated operating resonant frequency at the upper frequency operating condition grid point and the estimated operating resonant frequency. Take the absolute value of these differences. Based on the ratio of the two differences, determine the interpolation weight in the frequency direction. The interpolation weight is used to represent the position of the current estimated operating resonant frequency between the two calibrated operating resonant frequencies. Using the same method, calculate the difference between the estimated damping characteristic quantity and the calibrated damping characteristic quantity at the lower damping operating condition grid point, and the difference between the calibrated damping characteristic quantity at the upper damping operating condition grid point and the estimated damping characteristic quantity, to obtain the interpolation weight in the damping direction.

[0084] First, linear interpolation is performed between the vibration error compensation coefficient components corresponding to the grid points of the lower frequency operating condition and the grid points of the upper frequency operating condition in the frequency direction to obtain the frequency direction interpolation result.

[0085] Then, in the damping direction, linear interpolation is performed between the interpolation results on the lower and upper damping sides to obtain the final interpolation result; the final interpolation result is used as the candidate vibration error compensation coefficient component for this component.

[0086] By repeating the above process, all components of the candidate vibration error compensation coefficient can be obtained, thus forming a complete candidate vibration error compensation coefficient.

[0087] Within the sliding time window, for each sampling moment, the gyroscope output signal is read; the candidate vibration error compensation coefficients are used to compensate the gyroscope output signal for vibration error, resulting in a compensated output signal. The compensation logic can be as follows: based on the current vibration error compensation coefficient components, the corresponding compensation amount is calculated, and the compensation amount is subtracted from the output signal to obtain the compensated output signal.

[0088] For each sampling moment within the sliding time window, the compensated output signal is subtracted from the reference angular velocity signal to obtain the residual error signal at that moment. For each residual error signal, it is multiplied by itself to obtain the squared residual error value. The squared residual error values ​​of all sampling moments within the sliding time window are summed to obtain the sum of the squared residual error values. The sum of the squared residual error values ​​is divided by the total number of sampling points within the sliding time window to obtain the mean square value of the compensated residual vibration error. This mean square value is used as the index of the compensated residual vibration error.

[0089] Read the currently active vibration error compensation coefficient from the compensation coefficient library; for each component of the vibration error compensation coefficient, calculate the difference between the candidate vibration error compensation coefficient component and the active vibration error compensation coefficient component; for each component's difference, take the absolute value of the difference to obtain the absolute value of the difference for each component; among all the absolute values ​​of the difference for all components, find the one with the largest value and use that value as the coefficient change.

[0090] The compensated residual vibration error index is compared with a pre-set residual error threshold: if the compensated residual vibration error index is less than the residual error threshold, a residual error judgment is generated as passed; if the compensated residual vibration error index is greater than or equal to the residual error threshold, a residual error judgment is generated as failed.

[0091] The coefficient change is compared with a pre-set coefficient change threshold: if the coefficient change is less than the coefficient change threshold, the coefficient change is judged as passed; if the coefficient change is greater than or equal to the coefficient change threshold, the coefficient change is judged as failed; when both the residual error and coefficient change are judged as passed, an updated judgment result is generated as passed; when either the residual error or coefficient change is judged as failed, an updated judgment result is generated as failed.

[0092] The update conditions are defined as follows: the residual vibration error index after compensation is less than the residual error threshold, and the coefficient change is less than the coefficient change threshold. When the update decision result meets the update conditions: the candidate vibration error compensation coefficient is written into the corresponding working condition grid point position in the compensation coefficient library, replacing the original effective vibration error compensation coefficient; the current working condition parameter group, the current dynamic characteristic parameter estimate, and the current time are recorded to form a compensation coefficient evolution record; the compensation coefficient evolution record is associated with and stored in the compensation coefficient library.

[0093] When the update decision does not meet the update conditions: do not update the effective vibration error compensation coefficients in the compensation coefficient library; record this update decision as an invalid update; increase the invalid update count; when the invalid update count reaches the invalid update count threshold, output a fault prompt message to indicate that there is an abnormal state of the hemispherical resonator gyroscope or the installation environment.

[0094] The temperature and time thresholds are set based on the design life of the hemispherical resonator gyroscope and the allowable range of vibration error variation. The current ambient temperature and current cumulative working time are recorded in each update cycle. The ambient temperature change is calculated by subtracting the ambient temperature recorded during the last compensation coefficient update from the current ambient temperature, and taking the absolute value of the difference. The cumulative working time increase is calculated by subtracting the cumulative working time recorded during the last compensation coefficient update from the current cumulative working time. When the ambient temperature change exceeds the temperature threshold, it indicates that the ambient temperature has changed across ranges. When the cumulative working time increase exceeds the time threshold, it indicates that the structural aging has entered a new stage. When the ambient temperature change exceeds the temperature threshold or the cumulative working time increase exceeds the time threshold, a new update cycle is triggered. In the new update cycle, steps two through five are executed, thereby periodically updating the effective vibration error compensation coefficients.

[0095] Example 2

[0096] A gyroscope vibration error compensation system based on a hemispherical harmonic oscillator, employing the aforementioned gyroscope vibration error compensation method based on a hemispherical harmonic oscillator, includes: a vibration calibration module, an online data acquisition module, a dynamic characteristic parameter identification module, a parameter space interpolation module, a compensation and evaluation module, an update and management module, and a data processing module; the vibration calibration module, online data acquisition module, dynamic characteristic parameter identification module, parameter space interpolation module, compensation and evaluation module, and update and management module are respectively connected to the data processing module;

[0097] The vibration calibration module is used to calibrate the hemispherical resonant gyroscope on a vibration table under multiple vibration frequencies, vibration amplitudes and ambient temperatures, to obtain vibration error compensation coefficients and calibration dynamic characteristic parameters for each working condition, and to construct working condition grid points in the dynamic characteristic parameter space with the working resonant frequency and damping characteristic quantity as coordinates to form a compensation coefficient library.

[0098] The online data acquisition module is used to acquire drive signals, gyroscope output signals, and operating condition parameter sets during operation; the operating condition association module is used to select the target operating condition grid point set from the compensation coefficient library based on the operating condition parameter set.

[0099] The dynamic characteristic parameter identification module is used to preprocess the drive signal and gyroscope output signal within a sliding time window, and obtain the estimated value of the dynamic characteristic parameter through a recursive least squares algorithm.

[0100] The parameter space interpolation module is used to select adjacent working condition grid points from the target working condition grid point set based on the position of the estimated dynamic characteristic parameter value in the dynamic characteristic parameter space, and to perform interpolation calculation on the vibration error compensation coefficient of the adjacent working condition grid points to obtain candidate vibration error compensation coefficients.

[0101] The compensation and evaluation module is used to compensate the gyroscope output signal for vibration error using candidate vibration error compensation coefficients, calculate the residual vibration error index and coefficient change after compensation, and generate an updated decision result based on the residual vibration error index, residual error threshold, coefficient change and coefficient change threshold after compensation.

[0102] The update and management module is used to write candidate vibration error compensation coefficients into the compensation coefficient library as new effective vibration error compensation coefficients when the update judgment result meets the update conditions. It records the working condition parameter group, the estimated value of dynamic characteristic parameters and the update time, manages the evolution record of compensation coefficients, and controls the operation of the online data acquisition module, working condition association module, dynamic characteristic parameter identification module, parameter space interpolation module and compensation and evaluation module according to the temperature threshold and time threshold period.

[0103] Example 3

[0104] This embodiment is based on Embodiment 1, with a simplified configuration of the vibration error compensation structure. It adopts a combination of zero-bias compensation and proportional compensation, without introducing higher-order terms for compensation.

[0105] In this embodiment, the vibration error compensation structure consists only of a zero-bias compensation component and a proportional compensation component:

[0106] The zero-bias compensation component corresponds to a vibration error compensation coefficient, which is used to offset the fixed offset under the current vibration calibration conditions; the proportional compensation component corresponds to several vibration error compensation coefficients, each of which is multiplied by a vibration-related observation to form a compensation component.

[0107] The selection of vibration-related observations includes: vibration acceleration characteristic values, as observations reflecting the vibration intensity at the installation location; and drive signal amplitude, as observations reflecting the drive vibration intensity. Each observation corresponds to a proportional compensation coefficient, and the compensation amount changes proportionally to the observation during calculation.

[0108] Under this structure, the vibration error compensation coefficients that need to be solved in each vibration calibration condition include a zero-bias compensation coefficient and several proportional compensation coefficients, which constitute a set of vibration error compensation coefficients.

[0109] During the vibration calibration phase, for each vibration calibration condition, the compensation coefficient is calculated according to the steps in "Determination of Vibration Error Compensation Coefficient" in Example 1. This example provides a detailed explanation of the compensation amount construction process:

[0110] Within the recorded time sequence, an error signal sequence is constructed. For each sampling time, the vibration acceleration characteristic value and the driving signal amplitude are read. The zero-bias compensation component, the compensation component proportional to the vibration acceleration characteristic value, and the compensation component proportional to the driving signal amplitude are added to obtain the total compensation amount for that sampling time. The total compensation amount is subtracted from the output signal of the hemispherical resonator gyroscope to obtain the compensated output signal. Within the recorded time sequence, a compensated error signal sequence is constructed by comparing the compensated output signal with the reference angular velocity signal. The statistical index of the compensated error signal sequence is calculated, and the mean square value is selected as the statistical index. During the search for compensation coefficients, the zero-bias compensation coefficients are first scanned with a coarse step size to select a range of coefficients that results in a smaller mean square value. Then, the scan is refined within this range with a smaller step size. After determining the zero-bias compensation coefficients, for each proportional compensation coefficient, an iterative search method is used to adjust the value of the proportional compensation coefficient. After each adjustment, the mean square value is recalculated and compared with the mean square value before adjustment. Only adjustments that reduce the mean square value are accepted.

[0111] Repeat the above process until the improvement of the mean square value by the adjustment of all proportional compensation coefficients is lower than the preset improvement threshold, or the number of iterations reaches the preset upper limit.

[0112] When the mean square value does not change much in several rounds of adjustment, the current combination of zero bias compensation coefficient and proportional compensation coefficient is considered to be the set of target vibration error compensation coefficients for the vibration calibration condition, and is stored in the compensation coefficient library.

[0113] Steps two through six of the online compensation phase are the same as in Example 1. The difference in this example lies in the calculation method of the compensation amount:

[0114] Within the sliding time window, for each sampling moment, the zero-bias compensation coefficient, the proportional compensation coefficient corresponding to the vibration acceleration characteristic value, and the proportional compensation coefficient corresponding to the driving signal amplitude are read from the candidate vibration error compensation coefficients. The zero-bias compensation coefficient and the proportional compensation coefficient are multiplied and summed with the vibration-related observations at the current sampling moment to obtain the compensation amount at that sampling moment. The compensation amount is subtracted from the gyroscope output signal at that sampling moment to obtain the compensated output signal.

[0115] The calculation of residual vibration error index, coefficient change, update decision, and compensation coefficient write-back logic after subsequent compensation are all the same as in Example 1.

[0116] This embodiment simplifies the compensation structure, reduces the number of vibration error compensation coefficients, and simplifies the solution process. It is suitable for application scenarios where vibration error mainly consists of fixed offset and components that are approximately linearly related to vibration intensity, while maintaining the spatial interpolation of dynamic characteristic parameters and the dual threshold update mechanism.

[0117] Example 4:

[0118] This embodiment illustrates the application of the method of the present invention on an aircraft platform and explains how to use the compensation coefficient evolution record for performance monitoring and maintenance decision-making.

[0119] This embodiment considers a scenario where a hemispherical resonant gyroscope is mounted on an aircraft platform:

[0120] During takeoff, climb, cruise, maneuvering, and landing, aircraft are subjected to structural vibrations of different frequencies and amplitudes; the ambient temperature may change from normal ground temperature to low temperature at high altitudes, and then to a higher temperature range near the engine; the hemispherical resonant gyroscope is required to maintain stable vibration error compensation performance throughout the entire flight mission.

[0121] The online data acquisition module shares a clock signal with the flight control computer to ensure that the sampling time is synchronized with the flight status record; the temperature sensor is placed near the hemispherical resonator gyroscope to collect the shell temperature at the location of the gyroscope; the accelerometer is placed on the mounting structure of the hemispherical resonator gyroscope to collect the vibration acceleration at the mounting location; the update and management module establishes a data interface with the maintenance management system to export the compensation coefficient evolution record.

[0122] In this embodiment, the temperature threshold and time threshold are set in combination with the aircraft mission characteristics and the design life of the hemispherical resonant gyroscope:

[0123] Temperature threshold setting: Divide the range of ambient temperatures that the aircraft may experience into several temperature intervals, such as low temperature interval, medium temperature interval and high temperature interval; statistically analyze the temperature difference between the boundaries of adjacent temperature intervals, and select a temperature threshold based on the sensitivity of the hemispherical resonant gyroscope to temperature changes, so that the temperature threshold is approximately equal to the temperature difference between the boundaries of adjacent temperature intervals; when the change in ambient temperature exceeds the temperature threshold, it is considered that the aircraft has entered from one temperature interval to another temperature interval, and steps two to five need to be repeated to complete one update cycle.

[0124] Time threshold setting: Based on the design life of the hemispherical resonator gyroscope, the total working time corresponding to the life is divided into several stages, such as the initial stage, the middle stage, and the late stage. A time range is assigned to each stage, and a time threshold is selected based on the characteristics of material fatigue and structural aging. When the increase in cumulative working time exceeds the time threshold, it indicates that the device has entered a new aging stage. When the increase in cumulative working time exceeds the time threshold, even if the change in ambient temperature does not exceed the temperature threshold, steps two to five are triggered to perform an update cycle.

[0125] Through the above settings, the update and management module can reasonably control the update frequency of the compensation coefficient based on the temperature change and the increase in cumulative working time during flight missions.

[0126] During flight missions, the system operates as follows: Before the mission begins, the vibration calibration module loads the compensation coefficient library, which is derived from ground-based multi-condition calibration. After the mission begins, the online data acquisition module acquires drive signals, gyroscope output signals, ambient temperature, and vibration acceleration at a set sampling frequency. The condition association module selects a set of target condition grid points from the compensation coefficient library based on the current condition parameter set. The dynamic characteristic parameter identification module updates the estimated working resonant frequency and damping characteristic quantity within a sliding time window. The parameter space interpolation module performs interpolation calculations on the vibration error compensation coefficients corresponding to the target condition grid points within the dynamic characteristic parameter space, generating candidate vibration error compensation coefficients. The compensation and evaluation module performs vibration error compensation on the gyroscope output signal within a sliding time window, calculates the residual vibration error index and coefficient change after compensation, and generates an update decision result based on dual thresholds. The update and management module decides whether to write the result into the compensation coefficient library based on the update decision result and records the evolution of the compensation coefficients.

[0127] When the aircraft experiences temperature changes across different ranges or its cumulative operating time exceeds a time threshold, the update and management module triggers a new update cycle to adapt to the new environment and aging state.

[0128] In this embodiment, the compensation coefficient evolution record is used not only for online compensation coefficient management, but also for the formulation of aircraft maintenance strategies:

[0129] Each time a compensation coefficient is written to the compensation coefficient library, the following information is recorded: the corresponding operating condition parameter group, including ambient temperature, vibration acceleration characteristic value, driving vibration frequency, and driving vibration amplitude; the corresponding dynamic characteristic parameter estimate, including the working resonant frequency estimate and damping characteristic quantity estimate; the effective vibration error compensation coefficient; the update time and the current cumulative working time.

[0130] During the maintenance cycle, the evolution record of the compensation coefficient is exported and arranged in chronological order; the time series of the estimated working resonant frequency is analyzed to determine whether the estimated working resonant frequency continuously shifts in a certain direction, such as continuously increasing or continuously decreasing; the time series of the estimated damping characteristic is analyzed to determine whether the damping characteristic continuously changes; the changes in the effective vibration error compensation coefficient are analyzed, and the magnitude and direction of the change in the compensation coefficient are statistically analyzed for each update; the above trends are compared with the allowable frequency drift range, damping change range, and compensation coefficient change range to determine whether the hemispherical resonator gyroscope is in an acceptable aging state.

[0131] When the analysis results show that the estimated working resonant frequency and damping characteristic values ​​fluctuate within the predetermined range, and the change in the compensation coefficient is within the acceptable range, it indicates that the vibration error compensation can effectively adapt to the aging process under the update mechanism, and can extend the maintenance cycle. When the analysis results show that the estimated working resonant frequency or damping characteristic value is close to the preset limit, or the change in the compensation coefficient is continuously close to the coefficient change threshold in multiple updates, it indicates that the device performance is close to the acceptable boundary, and the maintenance task can be marked as priority. When the number of invalid updates repeatedly reaches the invalid update number threshold in multiple flight missions, it indicates that the residual vibration error index after compensation is difficult to meet the requirements under multiple operating conditions, and it can be determined that there is a structural abnormality or installation abnormality, and it is recommended to arrange disassembly and inspection.

[0132] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method of compensating for vibration errors in a gyro based on a hemispherical resonator, characterized in that, The method comprises the following steps: Step one, calibrating the hemispherical resonator gyroscope under multiple vibration frequency, vibration amplitude and environmental temperature combinations on a vibration table, obtaining vibration error compensation coefficients and calibration dynamic characteristic parameters under each working condition, constructing working condition grid points in a dynamic characteristic parameter space with working resonant frequency and damping characteristic quantity as coordinates, and forming a compensation coefficient library; Step two, collecting driving signals, gyroscope output signals and working condition parameter groups during operation, and selecting a target working condition grid point set from the compensation coefficient library according to the working condition parameter groups; Step three, pre-processing the driving signals and gyroscope output signals in a sliding time window, inputting the pre-processed signals into a recursive least squares algorithm, and obtaining dynamic characteristic parameter estimates; Step four, selecting adjacent working condition grid points adjacent to the dynamic characteristic parameter estimates in the working resonant frequency direction and the damping characteristic direction from the target working condition grid point set according to the position of the dynamic characteristic parameter estimates in the dynamic characteristic parameter space, and performing interpolation operation on the vibration error compensation coefficients of the adjacent working condition grid points to obtain candidate vibration error compensation coefficients; Step five, compensating the gyroscope output signals for vibration errors by using the candidate vibration error compensation coefficients, calculating residual vibration error indicators after compensation and coefficient change amounts of the candidate vibration error compensation coefficients and effective vibration error compensation coefficients in the sliding time window, and generating an update decision result according to the residual vibration error indicators, a residual error threshold, the coefficient change amounts and a coefficient change threshold; Step six, when the update decision result meets the update condition, writing the candidate vibration error compensation coefficients into the compensation coefficient library as new effective vibration error compensation coefficients, recording the working condition parameter groups, the dynamic characteristic parameter estimates and the update time, and performing steps two to five periodically according to a temperature threshold and a time threshold.

2. The method of claim 1, wherein the vibration error compensation of the hemispherical resonator based gyroscope is performed by a controller. The method of calibrating the hemispherical resonator gyroscope under multiple vibration frequency, vibration amplitude and environmental temperature combinations on a vibration table, obtaining vibration error compensation coefficients and calibration dynamic characteristic parameters under each working condition, constructing working condition grid points in a dynamic characteristic parameter space with working resonant frequency and damping characteristic quantity as coordinates, and forming a compensation coefficient library comprises: Applying multiple groups of sinusoidal vibration working conditions on the vibration table respectively, each group of sinusoidal vibration working conditions having different vibration frequency, vibration amplitude and environmental temperature; collecting the hemispherical resonator gyroscope output signals and reference angular velocity signals under each sinusoidal vibration working condition, determining the corresponding vibration error compensation coefficients according to the deviation between the output signals and the reference angular velocity signals; determining the calibration dynamic characteristic parameters of each sinusoidal vibration working condition according to the response of the hemispherical resonator gyroscope to the driving signals, the calibration dynamic characteristic parameters at least including calibration working resonant frequency and calibration damping characteristic quantity, and corresponding the vibration error compensation coefficients and the calibration dynamic characteristic parameters to the working condition grid points in the dynamic characteristic parameter space to form the working condition grid points in the compensation coefficient library.

3. The method of claim 1, wherein the vibration error compensation of the hemispherical resonator based gyroscope is performed by a controller. The method of collecting driving signals, gyroscope output signals and working condition parameter groups during operation, and selecting a target working condition grid point set from the compensation coefficient library according to the working condition parameter groups comprises: The driving signal and the gyroscope output signal are synchronously collected at a sampling frequency greater than or equal to twice the working resonance frequency band; the environmental temperature and the vibration acceleration of the installation position of the hemispherical resonator gyroscope are collected by a temperature sensor and an accelerometer, and the environmental temperature and the vibration acceleration are combined with the vibration frequency and the vibration amplitude to form a working condition parameter group; the working condition parameter group is compared with corresponding working condition parameter groups of each working condition grid point in the compensation coefficient library, and a nearest neighbor matching method is used to obtain a target working condition grid point set from the compensation coefficient library.

4. The method of claim 1, wherein the vibration error compensation of the hemispherical resonator based gyroscope is performed by a microprocessor. The pre-processing of the driving signal and the gyroscope output signal in the sliding time window, the input of the pre-processed signal into the recursive least squares algorithm, and the estimation of the dynamic characteristic parameter are included. The driving signal and the gyroscope output signal are band-pass filtered to retain signals in the working resonance frequency band, and the filtered driving signal and the filtered gyroscope output signal are subjected to amplitude normalization processing and phase alignment processing to obtain a pre-processed signal; the pre-processed signal is input into the recursive least squares algorithm in the sliding time window, and the dynamic characteristic parameter estimation value is iteratively updated based on the previous dynamic characteristic parameter estimation value and the pre-processed signal in the current sliding time window; the dynamic characteristic parameter estimation value includes the working resonance frequency estimation value and the damping characteristic quantity estimation value.

5. The method of claim 1, wherein The candidate vibration error compensation coefficient is obtained by interpolating the vibration error compensation coefficients of the adjacent working condition grid points in the working resonance frequency direction and the damping characteristic direction, including: The adjacent working condition grid points are selected from the target working condition grid point set in the dynamic characteristic parameter space with the dynamic characteristic parameter estimation value as the center, and the working condition grid points located on both sides of the working resonance frequency estimation value in the working resonance frequency direction and the working condition grid points located on both sides of the damping characteristic quantity estimation value in the damping characteristic direction are selected to form an adjacent working condition grid point set; the vibration error compensation coefficients of the adjacent working condition grid points are subjected to piecewise linear interpolation to obtain the candidate vibration error compensation coefficient.

6. The method of claim 1, wherein The candidate vibration error compensation coefficient is obtained by interpolating the vibration error compensation coefficients of the adjacent working condition grid points in the working resonance frequency direction and the damping characteristic direction, including: The candidate vibration error compensation coefficient is obtained by interpolating the vibration error compensation coefficients of the adjacent working condition grid points in the working resonance frequency direction and the damping characteristic direction, including: The candidate vibration error compensation coefficient is obtained by interpolating the vibration error compensation coefficients of the adjacent working condition grid points in the working resonance frequency direction and the damping characteristic direction, including: The candidate vibration error compensation coefficient is used to compensate the output signal of the gyroscope in a sliding time window to obtain a compensated output signal; the compensated output signal is compared with a reference angular velocity signal to calculate a mean square value of the residual vibration error after compensation as a residual vibration error index after compensation; a difference between the candidate vibration error compensation coefficient and the effective vibration error compensation coefficient is calculated, and a maximum value of absolute values of components of the difference is taken as a coefficient variation, and an update decision result is generated according to a comparison result of the residual vibration error index after compensation and a residual error threshold value and a comparison result of the coefficient variation and a coefficient variation threshold value.

7. The method of claim 1, wherein the vibration error compensation of the hemispherical resonator based gyroscope is performed by a microprocessor. The update condition includes that the residual vibration error index after compensation is less than the residual error threshold value and the coefficient variation is less than the coefficient variation threshold value; when the update decision result does not satisfy the update condition, the effective vibration error compensation coefficient is kept unchanged, the current update decision is recorded as invalid update, and the number of invalid updates is accumulated; and a fault prompt information is output when the number of invalid updates reaches an invalid update number threshold value.

8. The method of claim 1, wherein the vibration error compensation of the hemispherical resonator based gyroscope is performed by a microprocessor. The recorded working condition parameter group, the dynamic characteristic parameter estimation value and the update time include: The working condition parameter group corresponding to the effective vibration error compensation coefficient, the dynamic characteristic parameter estimation value and the time when the compensation coefficient is written into the compensation coefficient library are combined to form a compensation coefficient evolution record, and the compensation coefficient evolution record is stored in association with the compensation coefficient library, and is used to analyze the evolution trend of the vibration error compensation performance with the working time and the working condition parameter group.

9. The method of claim 1, wherein, The temperature threshold value and the time threshold value are set according to the design life of the hemispherical resonator gyroscope and the allowed vibration error variation range; when the environmental temperature variation exceeds the temperature threshold value, it indicates that the environmental temperature changes across the interval, and when the accumulated working time increases by more than the time threshold value, it indicates that the structural aging degree increases to a new stage; steps 2 to 5 are periodically executed according to the comparison results of the environmental temperature variation and the accumulated working time increase and the temperature threshold value and the time threshold value, so as to maintain the consistency of the effective vibration error compensation coefficient and the dynamic characteristic parameter estimation value.

10. A vibrational error compensation system for a gyro based on a hemispherical resonator, characterized by The vibration error compensation method for the hemispherical resonator gyroscope according to any one of claims 1-9 comprises a vibration calibration module, an online data acquisition module, a dynamic characteristic parameter identification module, a parameter space interpolation module, a compensation and evaluation module, an update and management module and a data processing module; the vibration calibration module, the online data acquisition module, the dynamic characteristic parameter identification module, the parameter space interpolation module, the compensation and evaluation module and the update and management module are connected with the data processing module respectively; The vibration calibration module is used for calibrating the hemispherical resonator gyroscope under a combination of multiple vibration frequencies, vibration amplitudes and environmental temperatures on a vibration table to obtain vibration error compensation coefficients and calibration dynamic characteristic parameters under each working condition, and constructing working condition grid points in a dynamic characteristic parameter space with the working resonant frequency and the damping characteristic quantity as coordinates to form a compensation coefficient library; The online data acquisition module is used for acquiring driving signals, gyroscope output signals and working condition parameter groups during operation; and a working condition association module is used for selecting a target working condition grid point set from the compensation coefficient library according to the working condition parameter group. The dynamic characteristic parameter identification module is configured to preprocess the driving signal and the gyroscope output signal in a sliding time window and obtain an estimated value of the dynamic characteristic parameter by using a recursive least square algorithm. The parameter space interpolation module is configured to select adjacent working condition grid points from the target working condition grid point set according to the position of the estimated value of the dynamic characteristic parameter in the dynamic characteristic parameter space, and perform interpolation operation on the vibration error compensation coefficients of the adjacent working condition grid points to obtain candidate vibration error compensation coefficients. The compensation and evaluation module is configured to perform vibration error compensation on the gyroscope output signal by using the candidate vibration error compensation coefficients, calculate a residual vibration error index after compensation and a coefficient variation amount, and generate an update decision result according to the residual vibration error index after compensation, a residual error threshold, the coefficient variation amount and a coefficient variation threshold. The update and management module is configured to write the candidate vibration error compensation coefficients into the compensation coefficient library as new effective vibration error compensation coefficients when the update decision result meets an update condition, record a working condition parameter group, the estimated value of the dynamic characteristic parameter and an update time, manage compensation coefficient evolution records, and control the operation of the online data acquisition module, the working condition association module, the dynamic characteristic parameter identification module, the parameter space interpolation module and the compensation and evaluation module according to a temperature threshold and a time threshold period.

Images