Dynamic drift error detection method based on resonant inertial navigation

Abstract

The invention relates to a dynamic drift error detection method based on resonant inertial navigation. The method comprises the following steps: step 1, constructing a hemispherical resonator gyroscope error model; 2, analyzing attitude error characteristics of the hemispherical resonant inertial navigation system; and step 3, carrying out dynamic drift detection based on resonant inertial navigation based on the hemispherical resonant gyro drift error model constructed in the step 1 and the attitude error characteristic analysis result of the hemispherical resonant inertial navigation system in the step 2. The gyroscope drift can be detected under the dynamic condition, a detection test is designed according to the characteristic that the gyroscope drift periodically changes along with the vibration mode angle, and finally the gyroscope precision can be improved through compensation.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of quantum gyro drift detection, and relates to a dynamic drift error detection method, in particular to a dynamic drift error detection method based on resonant inertial navigation. BACKGROUND

[0002] The hemispherical resonator inertial navigation system has advantages of few components, long service life, short start-up time and stable physical characteristics, and the hemispherical resonator gyro, a core component of the hemispherical resonator inertial navigation system, is a high-precision solid vibration gyro based on Coriolis vibration principle and having inertial navigation level performance. With the hemispherical resonator inertial navigation system receiving more and more attention and application in the field of inertial navigation, it has become one of the research hotspots in the field of inertial technology at home and abroad.

[0003] The drift of the hemispherical resonator gyro is related to the angle of the resonator standing wave mode and has periodicity. When measuring the circumferential drift of the hemispherical resonator gyro, currently, an external reference device or a force feedback working mode is mainly used to change the angle of the gyro standing wave mode, and the average value of the gyro angular velocity output in a period of time corresponding to the angle of the standing wave mode is recorded to obtain a gyro angle-mode angular velocity output sequence, and finally a circumferential drift error compensation model of the hemispherical resonator gyro is identified.

[0004] In a dynamic environment, the gyro angular velocity output is a dynamic change quantity, and the circumferential drift and the gyro output angular velocity cannot be separated, so the traditional static calibration method is not applicable.

[0005] To solve the above problems, the present application provides a dynamic drift error detection method based on resonant inertial navigation.

[0006] After searching, no existing technical patent literature similar to or similar to the present application has been found. SUMMARY

[0007] The present application proposes a dynamic drift error detection method based on resonant inertial navigation to solve the problems of the prior art, which can detect gyro drift under dynamic conditions and design a detection test according to the periodic change of the gyro drift with the angle of the standing wave mode, and finally improve the gyro precision through compensation.

[0008] The above-mentioned object of the present application is realized by the following technical scheme: A dynamic drift error detection method based on resonant inertial navigation comprises the following steps: Step 1, constructing a hemispherical resonator gyro error model; Step 2, analyzing the attitude error characteristics of the hemispherical resonator inertial navigation system; Step 3, based on the hemispherical resonator gyro drift error model constructed in step 1 and the analysis results of the attitude error characteristics of the hemispherical resonator inertial navigation system in step 2, performing dynamic drift detection based on resonant inertial navigation; Furthermore, the specific steps of step 1 include: (1) Based on Lynch's non-ideal harmonic oscillator error model, the equation of motion can be obtained as follows: (1) In the formula, The Brian coefficient is approximately 0.27; External input angular velocity; The modal mass of the harmonic oscillator; , The detection axes are mutually orthogonal , Displacement at; , This refers to the control force applied along the corresponding axis of the harmonic oscillator; This represents the average frequency of the gyroscope; This represents the frequency splitting value of the gyroscope; Indicates the angle between the frequency axis and the electrode axis; This indicates the circumferential average damping of the harmonic oscillator. This indicates that the circumferential damping of the harmonic oscillator is uneven. Indicates the time constant of the maximum damping axis; Indicates the time constant of the minimum damping axis; This indicates the angle between the damping axis and the electrode axis.

[0009] Based on the Lynch averaging method analysis, we can obtain: (2) In the formula, SF is the conversion scale from the applied voltage signal to the precession signal, i.e., the force scale; The amplitude of the vibration wave at the antinode on the harmonic oscillator; The amplitude of the vibration wave node; The reference frequency set for the frequency control loop; It is the mode angle; Input angular velocity from the outside; , , , These represent the amplitude control, quadrature control, frequency control, and mode angle control signals output by the resonant gyroscope control circuit, respectively. Specifically, the angle control loop can be simplified to: (3) In the formula, it is called The term represents resonant gyroscope drift and exhibits mode-dependent angle. It exhibits a periodic change in the form of a sine function.

[0010] (2) Let the drift of the resonant gyroscope be denoted as , which has the following form: (4) In the formula, , The periodic drift sine and cosine coefficients of the gyro, c is a constant drift term, and the gyro drift value changes by 90° with the mode angle as a period.

[0011] Moreover, the specific steps of step 2 include: (1) Analyze the attitude error characteristics of the inertial navigation system in the undamped navigation solution state; (2) Analyze the attitude error characteristics of the inertial navigation system in the Kalman filter fine alignment.

[0012] Moreover, the specific method of step 2 step (1) is: Select the "East-North-Sky (E-N-U)" geographic coordinate system as the navigation reference coordinate system of the inertial navigation system, denoted as n system, and the attitude differential equation of the inertial navigation system with the n system as the reference system is: (5) In the formula, is the attitude matrix of the carrier system (b system) relative to the navigation coordinate system (n system); is the component of the angular velocity of the b system relative to the inertial system (i system) in the b system; is the component of the angular velocity of the n system relative to the i system in the n system.

[0013] In actual calculation, there is a certain error in each quantity in formula (5), and the form of the attitude differential equation of the inertial navigation system is calculated as follows: (6) In the formula, is the calculated attitude matrix, and the corresponding navigation coordinate system is called the calculated navigation coordinate system, denoted as system; , is the gyro measurement error; , is the navigation system calculation error.

[0014] Taking the n system as the reference coordinate system, the equivalent rotation vector from the n system to the system is denoted as , which is simply called misalignment angle error, and the relationship between the calculated attitude matrix and the ideal attitude matrix is as follows: (7) Solving (5)~(7) together, the inertial navigation attitude error differential equation is obtained as follows, which reflects the calculated navigation system The misalignment angle error of the relative navigation system n with respect to the ideal navigation system n The change rule (8) In general, when the inertial navigation system enters the undamped navigation stage after alignment, at this time and The misalignment angle error of the system in a short time The change rule is as follows: (9) For a resonant inertial navigation system, the gyro measurement error is mainly composed of gyro drift and gyro noise Further, we have: (10) Moreover, the specific method of step 2 step (2) is: Based on the characteristics of the inertial navigation system, when the system alignment reaches the limit accuracy, the misalignment angle error is: (11) In the formula, , is the eastward and northward accelerometer bias of the system in the n system, L is the local latitude of the system, is the northward component of the earth's rotation angular velocity in the n system, is the eastward gyro drift of the system in the n system, .

[0015] Considering that the accelerometer bias characteristics in the hemispherical resonant inertial navigation system are stable and easy to compensate, its influence on the misalignment angle error is much smaller than the gyro drift, so formula (11) can be further simplified as: (12) In the fine alignment of the inertial navigation system, the Kalman filter calculates the estimated value of the system misalignment angle error in real time, denoted as .

[0016] When the Kalman filter converges sufficiently and the system fine alignment reaches the limit accuracy, the following relationship exists: (13) Let the system fine alignment at time t be has completely converged, denoted as , after a short time , denoted as , by combining formulas (5), (10), and (14), and ignoring the noise, the difference between the two is approximately: (14) wherein, , , , , , .

[0017] Moreover, the specific steps of step 3 include: (1) As shown in formula (4), the system gyro drift adopts a polynomial model, therefore the least square method is used to fit the samples, and the specific calculation method is as follows: (15) wherein, is the system drift coefficient to be estimated, is the change amount of misalignment angle estimation value under different alignment times-drift coefficient sequence, is the change amount of misalignment angle estimation value under different alignment times.

[0018] (2) The test experiment is performed on the hemispherical resonator inertial navigation system, one test period T=300s, a total of 20 periods, and the system recording frequency is 100hz.

[0019] Moreover, the specific steps of step 3, step (2) include: 1) The inertial navigation system is initialized, the gyro adopts force feedback mode, the gyro mode angle is pulled to mode angle 0°, the system starts to align, and the system enters the Kalman filter fine alignment stage is waited.

[0020] 2) After the system enters the fine alignment stage, 200s is waited, the system misalignment angle estimation value reaches the steady state, the current time misalignment angle estimation value is recorded as ; 100s data is started to be recorded, the current time misalignment angle estimation value change amount-drift coefficient and misalignment angle estimation value are calculated and recorded at an interval of 0.01s, a group of misalignment angle estimation value change amount-drift coefficient sequence and misalignment angle estimation value change amount sequence , at the current mode angle are obtained, at the end of 100s, the current misalignment angle error estimation value is used to feedback and correct the system attitude, and then the misalignment angle error estimation value is set to zero, so as to avoid that the system attitude error diverges too much.

[0021] 3) 2) is recorded as a test period, and the current mode angle is recorded as ​, the gyro mode angle is pulled to Then, the new test cycle is repeated in 2) and the corresponding sequence of the change of the misalignment angle estimation value , .

[0022] 4) After completing the test of the whole cycle, the sequence of the change of the misalignment angle estimation value and the sequence of the change of the misalignment angle estimation value are combined as the input of formula (15), and the system drift coefficient is calculated.

[0023] The advantages and beneficial effects of the present application are as follows: 1) Based on the error model of the hemispherical resonator gyro, the present application designs a working mode based on force feedback to detect the drift error of the gyro in the hemispherical resonator inertial navigation system under dynamic conditions. The gyro drift error is calculated by the change of the alignment angle estimation value of the inertial navigation system under different mode angles through the force feedback mode, and the test is designed according to the characteristics of the change of the drift error with the mode angle. The final result can be applied to the drift compensation of the resonator gyro to improve the alignment and navigation accuracy and stability of the hemispherical resonator inertial navigation system, and has engineering application value.

[0024] 2) The present application adopts the force feedback mode to control the precession of the gyro mode angle to a fixed mode angle to keep the gyro drift constant in the same test cycle, thereby improving the test accuracy.

[0025] 3) The present application uses the change of the system misalignment angle estimation value as the output sequence to avoid the influence of the accumulation of other error items in the alignment process. After completing a test cycle, the current misalignment angle estimation value is used to feedback and calibrate the system posture to avoid the situation that the change of the misalignment angle estimation value and the system drift value are approximately linearly related due to the too large attitude error of the system.

[0026] 4) The present application establishes the correlation model between the change of the alignment angle estimation value when the system fine alignment tends to be stable and the drift error of the inertial navigation system, and designs the corresponding test method, so that the gyro drift error model of the hemispherical resonator inertial navigation system can be measured without static conditions. DETAILED DESCRIPTION

[0027] Figure 1 is the gyro mode angle-drift curve diagram of the present application; Figure 2 is the time-misalignment angle error estimation value curve diagram of the present application. DETAILED DESCRIPTION

[0028] The structure of the present application will be further described below in combination with the drawings and through embodiments. It should be noted that the present embodiment is descriptive rather than limiting.

[0029] A dynamic drift error detection method based on resonant inertial navigation, comprising the following steps: Step 1, constructing a drift error model of a hemispherical resonator gyro; The specific method of step 1 is: In this embodiment, the hemispherical resonator gyro is a Coriolis gyro, and based on the non-ideal resonator error model of Lynch, the motion equation of the hemispherical resonator gyro is: Based on the non-ideal resonator error model of Lynch, the motion equation is: (1) In the formula, is the Bryan coefficient, about 0.27; is the external input angular velocity; is the modal mass of the resonator; 、 is the displacement at the position 、 is the control force applied to the resonator in the corresponding axis direction; 、 is the average frequency of the gyro; is the frequency split value of the gyro; is the angle between the frequency axis and the electrode axis; is the circumferential average damping of the resonator, is the circumferential damping unevenness of the resonator is the time constant of the maximum damping axis; is the time constant of the minimum damping axis; is the angle between the damping axis and the electrode axis.

[0030] Based on the Lynch average method analysis, we can get: (2) In the formula, SF is the conversion scale of the applied voltage signal to the precession signal, that is, the force application scale; is the amplitude of the vibration antinode point on the resonator; is the amplitude of the vibration node; is the reference frequency set by the frequency control loop; is the mode angle; is the external input angular velocity; 、 、 、 respectively represent the amplitude control, orthogonal control, frequency control and mode angle control signals output by the control loop of the resonator gyro; In particular, the angle control loop can be simplified as:​ (3) In the formula, the term is the resonant gyro drift, and has the characteristic of periodic change in the form of a sine function with the mode angle .

[0031] Further, the resonant gyro drift is denoted as , which has the form as follows: (4) In the formula, , are the periodic drift sine and cosine coefficients of the gyro, c is a constant drift term, and the gyro drift value changes by 90° with the mode angle as a period.

[0032] Under static conditions, the external input angular velocity is constant, the mode angle can be driven to precess through force feedback mode, the gyro angular velocity output under different mode angles is obtained, and the mode angle-drift corresponding relationship is calculated based on methods such as least squares after deducting the ground speed; but under dynamic conditions, the gyro angular velocity output changes at any time, which makes such methods unsuitable, and it is difficult to directly separate the drift related term from the angular velocity.

[0033] Step 2, analyze the attitude error characteristics of the spherical resonant inertial navigation system; In the embodiment, the specific steps of step 2 include: (1) analyze the attitude error characteristics of the inertial navigation system under the undamped navigation solution state; The specific method of step 2 step (1) is: Select the “East-North-Sky (E-N-U)” geographic coordinate system as the navigation reference coordinate system of the inertial navigation system, denoted as the n system, and the attitude differential equation of the inertial navigation system with the n system as the reference system is: (5) In the formula, is the attitude matrix of the carrier system (b system) relative to the navigation coordinate system (n system); is the component of the b system relative to the inertial system (i system) in the b system; is the component of the n system relative to the i system in the n system.

[0034] In actual calculation, there is a certain error in each quantity in formula (5), and the calculation of the attitude differential equation of the inertial navigation system is as follows: (6) In the formula, is the calculated attitude matrix, and the corresponding navigation coordinate system is called the calculated navigation coordinate system, denoted as system; , is the gyro measurement error; , is the navigation system calculation error.

[0035] Taking n system as the reference coordinate system, the equivalent rotation vector from n system to system is denoted as , which is called misalignment angle error, and the relationship between the calculated attitude matrix and the ideal attitude matrix is as follows: (7) By combining (5)~(7) and simplifying, the inertial navigation attitude error differential equation is obtained as follows, which reflects the change rule of the misalignment angle error of the calculated navigation system relative to the ideal navigation system n system: (8) Generally, when the inertial navigation system enters the undamped navigation stage after alignment, the and terms are small quantities, and the change rule of the misalignment angle error of the system in a short time is as follows: (9) For a resonant inertial navigation system, the gyro measurement error is mainly composed of gyro drift and gyro noise , and further: (10) (2) analyze the attitude error characteristics of the inertial navigation system in the Kalman filter fine alignment, the specific method of step 2 of the (2) step is: Based on the characteristics of the inertial navigation system, when the system alignment reaches the limit accuracy, the misalignment angle error is: (11) In the formula, , are the eastward and northward accelerometer biases of the system in the n system, L is the local latitude of the system, is the northward component of the earth rotation angular velocity in the n system, is the eastward gyro drift of the system in the n system.

[0036] Considering that the accelerometer bias characteristics in the hemispherical resonant inertial navigation system are stable and easy to compensate, the influence of the misalignment angle error caused by the accelerometer bias is much smaller than that of the gyro drift, so formula (11) can be further simplified as:​​​ (12) In the precise alignment of an inertial navigation system, the Kalman filter calculates the misalignment angle error of the system in real time. The estimated value is denoted as .

[0037] When the Kalman filter converges fully and the system alignment reaches its limit accuracy, the following relationship holds: (13) Suppose the system is precisely aligned at time t. It has fully converged, recorded as After a short period of time Afterwards, recorded as Combining equations (5), (10), and (14), and neglecting noise, we can approximate the difference between the two as follows: (14) In the formula, , , , , , .

[0038] Step 3: Based on the hemispherical resonant gyroscope drift error model constructed in Step 1 and the attitude error characteristic analysis results of the hemispherical resonant inertial navigation system in Step 2, perform dynamic drift detection based on resonant inertial navigation. The specific steps of step 3 include: (1) As shown in equation (4), the system gyroscope drift adopts a polynomial model, therefore the least squares method is used to calculate the drift. The samples are fitted using the following specific calculation method: (15) In the formula, The system drift coefficient to be estimated is... This represents the sequence of changes in the misalignment angle estimate versus the drift coefficient at different alignment times. This is a sequence of changes in the estimated misalignment angle at different alignment times.

[0039] (2) Test the hemispherical resonant inertial navigation system. One test period is T=300s, and there are 20 cycles in total. The system recording frequency is 100hz. The specific steps of step 3, step (2) include: 5) Initialize the inertial navigation system, the gyro adopts force feedback mode, the gyro mode angle is pulled to 0°, the system starts to align, and waits for the system to enter the Kalman filter fine alignment phase.

[0040] 6) After the system enters the fine alignment phase, wait for 200s, wait for the system misalignment angle estimate value to reach steady state, record the current time misalignment angle estimate value as ; start recording 100s data, calculate and record the current time misalignment angle estimate value change amount-drift coefficient and misalignment angle estimate value , get a set of misalignment angle estimate value-drift coefficient sequence and misalignment angle estimate value change amount sequence , At the end of 100s, use the current misalignment angle error estimate value to feedback correct the system attitude, and then set the misalignment angle error estimate value to zero to avoid the system attitude error divergence too large.

[0041] 7) Record 2) as a test period, record the current mode angle as , before starting the next period, pull the gyro mode angle to , then repeat the operation in 2) to complete a new test period and get the corresponding sequence , , the test misalignment angle estimate value change curve is shown in Figure 2 .

[0042] 8) After completing the whole period test, based on the obtained misalignment angle estimate value change amount-drift coefficient sequence and misalignment angle estimate value change amount sequence , as the input of formula (15), the system drift coefficient can be calculated.

[0043] Record a set of mode angle-drift curves obtained by long-time static detection of a set of harmonic inertial navigation as a reference, and the mode angle-drift curve obtained by fitting the results of the detection method designed by the application as a comparison, the comparison results are shown in Figure 1 , it can be seen that the designed method is effective.

[0044] Although the embodiments and drawings of the application are disclosed for the purpose of illustration, those skilled in the art can understand that various substitutions, changes and modifications are possible without departing from the spirit of the application and the appended claims, therefore, the scope of the application is not limited to the disclosed content of the embodiments and drawings.

Claims

1. A dynamic drift error detection method based on resonant inertial navigation, characterized in that: Includes the following steps: Step 1: Construct the error model of the hemispherical resonant gyroscope; Step 2: Analyze the attitude error characteristics of the hemispherical resonant inertial navigation system; Step 3: Based on the hemispherical resonant gyroscope drift error model constructed in Step 1 and the attitude error characteristic analysis results of the hemispherical resonant inertial navigation system in Step 2, dynamic drift detection based on resonant inertial navigation is carried out.

2. The dynamic drift error detection method based on resonant inertial navigation according to claim 1, characterized in that: The specific steps of step 1 include: (1) Based on Lynch's non-ideal harmonic oscillator error model, the equation of motion can be obtained as follows: (1) In the formula, The Brian coefficient is approximately 0.27; External input angular velocity; The modal mass of the harmonic oscillator; , The detection axes are mutually orthogonal , Displacement at; , This refers to the control force applied along the corresponding axis of the harmonic oscillator; This represents the average frequency of the gyroscope; This represents the frequency splitting value of the gyroscope; Indicates the angle between the frequency axis and the electrode axis; This indicates the circumferential average damping of the harmonic oscillator. This indicates that the circumferential damping of the harmonic oscillator is uneven. Indicates the time constant of the maximum damping axis; Indicates the time constant of the minimum damping axis; Indicates the angle between the damping axis and the electrode axis; Based on the Lynch averaging method analysis, we can obtain: (2) In the formula, SF is the conversion scale from the applied voltage signal to the precession signal, i.e., the force scale; The amplitude of the vibration wave at the antinode on the harmonic oscillator; The amplitude of the vibration wave node; The reference frequency set for the frequency control loop; It is the mode angle; Input angular velocity from the outside; , , , These represent the amplitude control, quadrature control, frequency control, and mode angle control signals output by the resonant gyroscope control circuit, respectively. Specifically, the angle control loop can be simplified to: (3) In the formula, it is called The term represents resonant gyroscope drift and exhibits mode-dependent angle. Its characteristic of changing periodically in the form of a sine function; (2) Let the drift of the resonant gyroscope be denoted as It has the following forms: (4) In the formula, , represents the sine and cosine coefficients of the gyroscope's periodic drift, and c is a constant drift term. The gyroscope drift value changes with the mode angle by 90° for one period.

3. The dynamic drift error detection method based on resonant inertial navigation according to claim 1, characterized in that: The specific steps of step 2 include: (1) Analyze the attitude error characteristics of the inertial navigation system under the undamped navigation solution state; (2) Analyze the attitude error characteristics of the Kalman filter-aligned inertial navigation system.

4. The dynamic drift error detection method based on resonant inertial navigation according to claim 3, characterized in that: The specific method for step 2, step (1) is as follows: The "East-North-Sky (ENU)" geographic coordinate system is selected as the navigation reference coordinate system of the inertial navigation system, denoted as the n-system. The attitude differential equation of the inertial navigation system with the n-system as the reference frame is: (5) In the formula, The attitude matrix of the carrier system (b-frame) relative to the navigation coordinate system (n-frame); The component of the angular velocity of the b-frame relative to the inertial frame (i-frame) output by the resonant gyroscope in the b-frame; Let n be the component of the rotational angular velocity of the n system relative to the i system in the n system; In actual calculations, there are certain errors in each quantity in equation (5). The form of the attitude differential equation of the inertial navigation system is as follows: (6) In the formula, To calculate the attitude matrix, the corresponding navigation coordinate system is called the computational navigation coordinate system, abbreviated as . Tie; , This refers to the measurement error of the gyroscope. , Calculate the error for the navigation system; Using the n-frame as the reference coordinate system, let the coordinate system from the n-frame to the coordinate system be denoted as . The equivalent rotation vector of the system is This is simply referred to as the misalignment angle error, which can be obtained by calculating the attitude matrix. With ideal posture array The relationship between them is as follows: (7) Combining (5) to (7), the simplified differential equation of the inertial navigation attitude error is as follows, reflecting the computational navigation system Misalignment angle error relative to the ideal navigation system n Pattern of change; (8) Normally, after the inertial navigation system completes alignment and enters the undamped navigation phase, at this time... and All items are small quantities, representing the system's misalignment angle error over a short period of time. The pattern of change is as follows: (9) For a resonant inertial navigation system, the gyroscope measurement error is mainly caused by gyroscope drift. and gyroscope noise From this, we can further conclude: (10)。 5. The dynamic drift error detection method based on resonant inertial navigation according to claim 3, characterized in that: The specific method for step (2) of step 2 is as follows: Based on the characteristics of inertial navigation systems, the misalignment angle error when the system alignment reaches its limit accuracy for: (11) In the formula, , The system's eastward and northward accelerometer biases are zero in the n-frame, and L is the local latitude of the system. Let be the northward component of the Earth's rotational angular velocity in the n-frame. This represents the eastward gyroscope drift of the system in the n-system. ; Considering that the zero-bias characteristic of the added gauge in the hemispherical resonant inertial navigation system is stable and easy to compensate, its impact on the misalignment angle error is much smaller than that of the gyroscope drift. Therefore, equation (11) can be further simplified to: (12) In the precise alignment of an inertial navigation system, the Kalman filter calculates the misalignment angle error of the system in real time. The estimated value is denoted as ; When the Kalman filter converges fully and the system alignment reaches its limit accuracy, the following relationship holds: (13) Suppose the system is precisely aligned at time t. It has fully converged, recorded as After a short period of time Afterwards, recorded as Combining equations (5), (10), and (14), and neglecting noise, we can approximate the difference between the two as follows: (14) In the formula, , , ， ， ， 。 6. The dynamic drift error detection method based on resonant inertial navigation according to claim 1, characterized in that: The specific steps of step 3 include: (1) As shown in equation (4), the system gyroscope drift adopts a polynomial model, therefore the least squares method is used to calculate the drift. The samples are fitted using the following specific calculation method: (15) In the formula, The system drift coefficient to be estimated is... This represents the sequence of changes in the misalignment angle estimate versus the drift coefficient at different alignment times. This is a sequence of changes in the estimated misalignment angle at different alignment times; (2) The hemispherical resonant inertial navigation system was tested. One test period was T=300s, and a total of 20 periods were conducted. The system recording frequency was 100hz.

7. The dynamic drift error detection method based on resonant inertial navigation according to claim 6, characterized in that: The specific steps of step 3, step (2) include: 1) Initialize the inertial navigation system. The gyroscopes are in force feedback mode. All gyroscope mode angles are pulled to 0°. The system begins alignment and waits for the system to enter the Kalman filter fine alignment stage. 2) After the system enters the fine alignment stage, wait 200 seconds for the estimated misalignment angle to reach a steady state. Record the estimated misalignment angle at the current moment as . Start recording 100 seconds of data, and calculate and record the change in the estimated misalignment angle - drift coefficient at 0.01-second intervals. And the estimated value of the misalignment angle This yields a sequence of changes in the estimated misalignment angle and the drift coefficient under the current mode shape angle. and the sequence of changes in the inaccuracy angle estimates , At the end of 100 seconds, use the current estimated value of the misalignment angle error. Feedback correction system attitude, then estimate misalignment angle error value Set to zero to avoid excessive divergence in system attitude error; 3) Denote 2) as one test cycle, and denote the current mode angle as... Before starting the next cycle, the gyroscope's mode angle is uniformly pulled to... Then repeat step 2) to complete the new test cycle and obtain the corresponding sequence. , ; 4) After completing the full cycle of testing, the sequence of changes in the estimated misalignment angle and the drift coefficient is combined. Combined sequence of changes in the estimated inaccuracy angle As input to equation (15), the system drift coefficient can be calculated. .

Images