Optical fiber gyroscope earth rotation north-seeking method and device

Abstract

The invention provides an optical fiber gyroscope earth rotation north-seeking method and device, and the method comprises the steps: driving an optical fiber gyroscope through a mechanical rotation control device according to the local latitude to carry out multi-position sampling, and obtaining an original angular velocity sequence; performing spectrum feature extraction and real-time speed correction on the original signal to obtain net rotation speed data; a segmented adaptive window function processing technology is adopted to obtain an earth rotation signal segment set; and finally, carrying out segmented cross-correlation calculation by combining an embedded computer with a geographic position constraint condition to obtain a true north azimuth angle. According to the method, the random influence of rotary table jitter is dispersed into each independent segment through a segment processing strategy, so that the jitter in each segment is relatively small and can be predicted, processing parameters are dynamically adjusted according to signal quality by adopting an adaptive window function, and noise interference is further suppressed; results of all sections are optimized and fused through a geographical constraint weighting algorithm, and the influence of mechanical jitter on north-seeking precision under the condition of a high-speed rotary table is effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a fiber optic gyroscope earth rotation north finding method and device. Background Art

[0002] Fiber-optic gyro (FOG) north-finding technology, an important inertial navigation technique, is widely used in precision measurement, navigation positioning, and establishing azimuth references. Traditional FOG dynamic north-finding methods typically use a turntable to continuously rotate the FOG, analyzing the Earth's rotation component in the FOG's output signal to determine true north. This method offers advantages such as high autonomy and immunity to external interference, making it valuable for both military and civilian applications.

[0003] However, a major problem with existing technologies is that when the turntable's rotational speed is increased to shorten the north-finding time, the speed jitter of the turntable's mechanical system significantly affects the frequency stability of the fiber optic gyroscope's output signal. This results in the inability of traditional cross-correlation algorithms to accurately extract the Earth's rotation signal at high speeds, severely impacting north-finding accuracy. Especially at high rotational speeds, signal contamination caused by turntable jitter dramatically increases north-finding errors, hindering the development of fast, high-precision north-finding technologies. Summary of the Invention

[0004] The main purpose of the present invention is to solve the technical problem that the north-seeking accuracy is reduced due to mechanical jitter when the turntable rotates at high speed in the existing fiber optic gyroscope dynamic north-seeking technology; A first aspect of the present invention provides a fiber optic gyroscope (FOG) method for finding north by earth rotation. The FOG north finder includes a fiber optic gyroscope, an accelerometer, a mechanical rotation control device, and an embedded computer. The FOG method includes: The mechanical rotation control device drives the fiber optic gyroscope to perform multi-position sampling of the earth's rotation characteristics according to the local latitude of the local geographical location, thereby obtaining an original angular velocity sequence containing projection information of the horizontal component of the earth's rotation; Performing spectrum feature extraction and real-time speed correction processing on the original angular velocity sequence to obtain corrected net rotational velocity data; performing piecewise adaptive window function processing on the net rotation speed data to obtain a set of Earth rotation signal segments; The embedded computer performs segmented cross-correlation north-finding processing on each signal segment of the earth rotation signal segment set according to the constraints of the local geographical location to obtain the true north azimuth.

[0005] Optionally, in a first implementation of the first aspect of the present invention, the mechanical rotation control device drives the fiber optic gyroscope to perform multi-position sampling of the Earth's rotation characteristics according to the local latitude of the local geographic location, and obtaining an original angular velocity sequence containing projection information of the horizontal component of the Earth's rotation includes: Calculate and process the horizontal component of the earth's rotation for the local latitude of the local geographical location to obtain the theoretical horizontal component of the earth's rotation; Adaptively adjusting sampling parameters according to the signal strength of the theoretical earth rotation horizontal component to obtain sampling parameters that match the signal strength, wherein the sampling parameters include sampling time and sampling density parameters; The mechanical rotation control device drives the fiber optic gyroscope to perform variable speed rotation sampling processing according to the sampling parameters, thereby obtaining an original angular velocity sequence containing projection information of the horizontal component of the earth's rotation.

[0006] Optionally, in a second implementation of the first aspect of the present invention, performing spectrum feature extraction and real-time speed correction processing on the original angular velocity sequence to obtain corrected net rotational velocity data includes: Performing fast Fourier transform processing on the original angular velocity sequence to obtain a power spectral density distribution of a preset frequency band; Identifying and processing the natural frequency characteristics of the mechanical rotation control device according to the power spectrum density distribution to obtain a mechanical characteristic spectrum template including a servo control frequency, a resonance frequency, and an external interference frequency; Performing real-time comparative analysis on the power spectrum density distribution and the mechanical characteristic spectrum template to obtain a jitter signal of the mechanical rotation control device; Performing intensity determination processing on the jitter signal according to a preset jitter threshold, and starting least squares modeling processing when the jitter intensity exceeds the preset jitter threshold to obtain a real-time speed compensation amount; The actual rotation speed of the mechanical rotation control device and the real-time speed compensation amount are subjected to superposition correction processing to obtain corrected net rotation speed data.

[0007] Optionally, in a third implementation of the first aspect of the present invention, performing intensity determination processing on the jitter signal according to a preset jitter threshold, initiating least squares modeling processing when the jitter intensity exceeds the preset jitter threshold, and obtaining the real-time speed compensation amount includes: Performing root mean square (RMS) calculation on the jitter signal to obtain a strength quantification index of the jitter signal; Performing a numerical comparison process based on the intensity quantification index and a preset jitter threshold, and triggering a compensation algorithm start condition when the intensity quantification index exceeds the preset jitter threshold; Perform least squares mathematical modeling on the jitter signal that meets the startup conditions and establish a linear regression equation between the jitter signal and the time variable; Predicting and calculating the jitter trend at a future moment according to the linear regression equation to obtain a predicted jitter value; The predicted jitter value is subjected to sign inversion and amplitude adjustment processing to obtain a real-time speed compensation amount for offsetting the influence of the jitter.

[0008] Optionally, in a fourth implementation of the first aspect of the present invention, performing piecewise adaptive window function processing on the net rotation speed data to obtain a set of Earth rotation signal segments includes: performing three-dimensional evaluation processing of the net rotation velocity data in terms of signal-to-noise ratio, spectrum purity, and phase stability to obtain an Earth rotation signal quality evaluation index; Dynamically adjusting the window function processing parameters according to the signal quality evaluation index to obtain an adaptive window function parameter combination including a window length and a window function type; Performing window function filtering on the net rotation speed data according to the adaptive window function parameter combination to obtain an Earth rotation signal processed by the window function; Performing temperature compensation algorithm correction processing on the earth rotation signal processed by the window function according to the current operating temperature of the fiber optic gyroscope to obtain a temperature-corrected earth rotation signal segment; The temperature-corrected earth rotation signal segments are segmented and organized to obtain an earth rotation signal segment set.

[0009] Optionally, in a fifth implementation of the first aspect of the present invention, performing segmented cross-correlation north-finding processing on each signal segment of the Earth rotation signal segment set according to the constraints of the local geographical location by the embedded computer to obtain the true north azimuth includes: Calculating a geographic constraint weight coefficient by the embedded computer according to a local magnetic declination, a gravity anomaly, and an earth ellipsoid parameter in the local geographic location; performing cross-correlation coefficient calculation on each signal segment in the set of Earth rotation signal segments to obtain a sine cross-correlation coefficient and a cosine cross-correlation coefficient of each signal segment; performing arc tangent function calculation on each signal segment according to the sine cross-correlation coefficient and the cosine cross-correlation coefficient to obtain a preliminary azimuth angle of each signal segment; Analyzing the phase differences between adjacent signal segments in the Earth rotation signal segment set to obtain a zero offset compensation amount, and performing zero correction processing on the preliminary azimuth to obtain corrected azimuths of each segment; The corrected azimuths are processed using a weighted average algorithm according to the geographic constraint weight coefficient to obtain the true north azimuth.

[0010] Optionally, in a sixth implementation of the first aspect of the present invention, performing cross-correlation coefficient calculation processing on each signal segment in the set of Earth rotation signal segments to obtain a sine cross-correlation coefficient and a cosine cross-correlation coefficient of each signal segment includes: Analyzing and processing the frequency characteristics of each signal segment in the set of earth rotation signal segments to obtain an actual rotation frequency corresponding to each signal segment; generating a sine reference signal and a cosine reference signal that match the frequency of each signal segment according to the actual rotation frequency; Zero-delay cross-correlation operation is performed on each signal segment data and the corresponding sine reference signal and the corresponding cosine reference signal to obtain the sine cross-correlation coefficient and cosine cross-correlation coefficient of each signal segment.

[0011] A second aspect of the present invention provides a fiber optic gyroscope earth rotation north-finding device, the fiber optic gyroscope north-finding device comprising a fiber optic gyroscope, an accelerometer, a mechanical rotation control device, and an embedded computer, the fiber optic gyroscope earth rotation north-finding device comprising: a sampling module, configured to drive the fiber optic gyroscope to perform multi-position sampling of the Earth's rotation characteristics according to the local latitude of the local geographical location through the mechanical rotation control device, and obtain an original angular velocity sequence containing projection information of the horizontal component of the Earth's rotation; a correction module, configured to perform spectrum feature extraction and real-time speed correction processing on the original angular velocity sequence to obtain corrected net rotational velocity data; a processing module, configured to perform piecewise adaptive window function processing on the net rotation speed data to obtain a set of Earth rotation signal segments; The solution module is used to perform segmented cross-correlation north-seeking solution processing on each signal segment of the earth rotation signal segment set according to the constraints of the local geographical location through the embedded computer to obtain the true north azimuth.

[0012] The above-mentioned fiber optic gyroscope Earth rotation north-finding method and device uses a mechanical rotation control device to drive the fiber optic gyroscope to perform multi-position sampling according to the local latitude to obtain a raw angular velocity sequence; spectral feature extraction and real-time velocity correction are performed on the raw signal to obtain net rotational velocity data; a segmented adaptive window function processing technique is used to obtain a set of Earth rotation signal segments; and finally, an embedded computer performs segmented cross-correlation calculations in combination with geographic location constraints to obtain the true north azimuth. The present invention uses a segmented processing strategy to disperse the random effects of turntable jitter into each independent segment, making the jitter within each segment relatively small and predictable. An adaptive window function is used to dynamically adjust processing parameters based on signal quality to further suppress noise interference. The results of each segment are optimized and integrated using a geographic constraint weighted algorithm, effectively addressing the impact of mechanical jitter on north-finding accuracy under high-speed turntable conditions.

[0013] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or understood by practicing the present invention. The purposes and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0014] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 Schematic diagram of a first embodiment of a method for finding north using an earth rotation optical fiber gyroscope according to an embodiment of the present invention; Figure 2 FIG. 1 is a schematic diagram of an embodiment of a fiber optic gyroscope earth rotation north-finding device in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0017] The terms "including," "having," and any variations thereof, as used in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to the process, method, product, or device.

[0018] To facilitate understanding of this embodiment, a fiber optic gyroscope earth rotation north finding method disclosed in an embodiment of the present invention is first introduced in detail. Figure 1 As shown, this method includes the following steps: S101, driving the fiber optic gyroscope through a mechanical rotation control device to perform multi-position sampling of the Earth's rotation characteristics according to the local latitude of the local geographic location, thereby obtaining an original angular velocity sequence containing projection information of the horizontal component of the Earth's rotation; In one embodiment of the present invention, the mechanical rotation control device drives the fiber optic gyroscope to perform multi-position sampling of the earth's rotation characteristics according to the local latitude of the local geographical location to obtain an original angular velocity sequence containing projection information of the horizontal component of the earth's rotation, including: calculating the horizontal component of the earth's rotation at the local latitude of the local geographical location to obtain a theoretical horizontal component of the earth's rotation; adaptively adjusting sampling parameters according to the signal strength of the theoretical horizontal component of the earth's rotation to obtain sampling parameters that match the signal strength, wherein the sampling parameters include sampling time and sampling density parameters; and driving the fiber optic gyroscope to perform variable speed rotation sampling according to the sampling parameters by the mechanical rotation control device to obtain an original angular velocity sequence containing projection information of the horizontal component of the earth's rotation.

[0019] Specifically, when a fiber optic gyro (FOG) north finder begins operation, it first needs to obtain the device's current geographic latitude. Upon receiving the local latitude value, L, the embedded computer immediately initiates a theoretical calculation program for the horizontal component of Earth's rotation. The Earth's rotational angular velocity, ωe, has different horizontally projected components at different latitudes. This horizontal component, ωeh, follows the mathematical relationship ωeh = ωe × cosL. When the device is near the equator, the latitude, L, approaches zero degrees and the cosine value approaches 1, at which point the horizontal component reaches its maximum value. When the device is near the poles, the latitude, L, approaches 90 degrees and the cosine value approaches zero, minimizing the horizontal component. The embedded computer then performs a cosine function calculation based on the input latitude value, multiplying the calculated cosine value by the Earth's standard rotational angular velocity to obtain the theoretical horizontal component value for the current location. This theoretical value serves as a reference for all subsequent processing steps and also determines the theoretical upper limit of the Earth's rotation signal that the FOG can sense.

[0020] Specifically, based on the calculated theoretical horizontal component of Earth's rotation, the embedded computer enters the adaptive adjustment phase of sampling parameters. Signal strength differences directly affect the FOG's measurement accuracy and required data acquisition time. When the theoretical horizontal component is large, the FOG can obtain a relatively strong effective signal, enabling accurate measurements with a shorter sampling time and lower sampling density. Conversely, when the theoretical horizontal component is small, the effective signal is weak, requiring a longer sampling time and increased sampling density to accumulate sufficient valid data. The embedded computer uses a pre-set table of signal strength and sampling parameter correspondences to search for sampling time parameters that match the current theoretical horizontal component. The sampling time parameter determines the duration of the entire data acquisition process, while the sampling density parameter determines the frequency of the FOG's data output per unit time. The optimized combination of these two parameters ensures consistent angular velocity measurement data at all latitudes.

[0021] Specifically, after receiving the sampling parameter instruction, the mechanical rotation control device begins executing the variable speed rotation sampling driver. The device first calculates the total number of rotation cycles based on the sampling time parameter, and then determines the number of data collection points within each cycle based on the sampling density parameter. Variable speed rotation sampling adopts a staged speed control strategy. In the initial stage, a lower turntable speed is used for a rough azimuth scan. The main purpose of this stage is to quickly determine the possible north-facing area range. After the rough scan is completed, the mechanical rotation control device increases the turntable speed to a preset high-speed gear and limits the rotation range to the candidate area determined by the rough scan for precise sampling. The time the turntable stays at each sampling position is dynamically adjusted according to the sampling density parameter to ensure that the fiber optic gyroscope has sufficient time to complete stable angular velocity measurements. Throughout the rotation process, the mechanical rotation control device continuously monitors the actual speed and position of the turntable, transmitting accurate azimuth information to the fiber optic gyroscope in real time.

[0022] Specifically, driven by a mechanical rotation control device, the fiber optic gyroscope (FOG) rotates its sensitive axis along a preset trajectory in the horizontal plane. As the direction of the sensitive axis continuously changes, the projected component of the Earth's rotational angular velocity sensed by the FOG exhibits a periodic variation. When the sensitive axis points toward true north, the projected component reaches its maximum value, equal to the theoretical horizontal component of Earth's rotation. When the sensitive axis points toward true south, the projected component reaches its minimum value, a negative of the theoretical value. When the sensitive axis points east-west, the projected component approaches zero. The FOG outputs these continuously changing angular velocity measurements to an embedded computer at a preset sampling frequency. Each output data point contains the angular velocity value at the corresponding sampling moment and the precise azimuth angle of the turntable at that moment. The embedded computer organizes the received data into a complete raw angular velocity sequence in chronological order. This sequence not only contains the valid signal of Earth's rotation but also incorporates various error components such as turntable mechanical jitter, internal noise within the FOG, and external environmental interference.

[0023] S102, performing spectrum feature extraction and real-time velocity correction processing on the original angular velocity sequence to obtain corrected net rotation velocity data; In one embodiment of the present invention, the spectral feature extraction and real-time speed correction processing of the original angular velocity sequence to obtain the corrected net rotational speed data includes: performing fast Fourier transform processing on the original angular velocity sequence to obtain a power spectrum density distribution of a preset frequency band; identifying the natural frequency characteristics of the mechanical rotation control device according to the power spectrum density distribution to obtain a mechanical characteristic spectrum template including a servo control frequency, a resonance frequency and an external interference frequency; performing real-time comparative analysis processing on the power spectrum density distribution and the mechanical characteristic spectrum template to obtain a jitter signal of the mechanical rotation control device; performing intensity judgment processing on the jitter signal according to a preset jitter threshold, and initiating least squares modeling processing when the jitter intensity exceeds the preset jitter threshold to obtain a real-time speed compensation amount; and superimposing and correcting the actual rotational speed of the mechanical rotation control device and the real-time speed compensation amount to obtain the corrected net rotational speed data.

[0024] Specifically, upon receiving the raw angular velocity sequence, the embedded computer immediately initiates the Fast Fourier Transform (FFT) processing routine. The FFT is a mathematical algorithm that converts time-domain signals into frequency-domain signals, decomposing continuously varying angular velocity data into combinations of distinct frequency components. The embedded computer segments the raw angular velocity sequence into segments according to a preset window length, with each segment containing a sufficient number of sampling points to ensure the required frequency resolution for analysis. The transformation routine performs complex operations on each segment, converting the time-domain angular velocity amplitude information into frequency-domain amplitude and phase information. After the transformation is complete, the embedded computer calculates the power spectral density (PSD) at each frequency point. This density reflects the energy distribution intensity of the original signal at a specific frequency. Because the turntable's mechanical jitter and the Earth's rotation signal have different frequency characteristics, the PSD distribution clearly displays the frequency locations and intensity comparisons of these different signal components. The embedded computer limits the frequency band of interest to the 0.1 Hz to 50 Hz range, which covers the primary distribution areas of the turntable's servo control frequencies, mechanical resonance frequencies, and ambient vibration frequencies.

[0025] Based on the obtained power spectral density distribution, the embedded computer initiates the mechanical signature spectrum identification algorithm. The algorithm first searches the power spectrum for frequency peaks whose amplitudes are significantly higher than the background noise. These peaks correspond to the characteristic frequency components of the turntable mechanism. The servo control frequency originates from the turntable motor's closed-loop control circuit. Its frequency value is directly related to the controller's response speed and gain parameters, and appears as a relatively stable narrowband peak in the power spectrum. Resonant frequencies arise from the natural vibration modes of the turntable's mechanical structure, including bearing vibration, gear meshing frequency, and bending and torsional resonances of structural components. These frequencies appear as multiple discrete, sharp peaks in the power spectrum. External interference frequencies primarily include power grid frequency, vibration frequencies of surrounding equipment, and ambient noise. Their appearance in the power spectrum is relatively irregular and their amplitudes vary widely. The identification algorithm classifies and labels different types of frequency components by analyzing characteristic parameters such as frequency position, amplitude, bandwidth, and temporal stability of each peak. The algorithm organizes the identification results into a mechanical signature spectrum template, which records the center frequency, frequency band, typical amplitude, and frequency type of each characteristic frequency, serving as a reference for subsequent jitter signal identification.

[0026] The real-time comparative analysis program continuously monitors the degree of match between the current power spectral density distribution and the mechanical characteristic spectrum template. The program examines the amplitude changes of each characteristic frequency recorded in the template within the current power spectrum. When the actual amplitude of a characteristic frequency deviates significantly from the typical value recorded in the template, it indicates that the frequency component has abnormal jitter. The comparative analysis uses a sliding window technique, comparing multiple consecutive power spectra to distinguish short-term random fluctuations from sustained jitter trends. The program converts the identified abnormal frequency components back into time domain signals through an inverse Fourier transform, obtaining the jitter signal of the mechanical rotation control device. This jitter signal reflects the deviation between the actual rotation speed of the turntable and the ideal uniform rotation speed, and its amplitude is directly related to the degree of impact on the measurement accuracy of the fiber optic gyroscope.

[0027] The embedded computer quantifies the extracted jitter signal intensity, calculating the root mean square (RMS) value to obtain a quantitative jitter strength index. The RMS value effectively reflects the overall energy level of the jitter signal, avoiding the problem of positive and negative jitter canceling out in simple amplitude statistics. The program compares the calculated strength index with a preset jitter threshold, which is determined based on the FOG's accuracy requirements and the turntable's mechanical properties. When the strength index exceeds the threshold, indicating that the current mechanical jitter has reached a level that affects north-finding accuracy, the embedded computer immediately initiates a least-squares modeling program. The least-squares method is a mathematical fitting algorithm that seeks the optimal fitting parameters to minimize the sum of squared errors between the model's predicted values ​​and the actual observed values. The modeling program uses the jitter signal as the dependent variable and time as the independent variable to construct a linear or polynomial regression equation to describe the jitter signal's changing pattern. This equation predicts jitter trends over a short period of time, and the program uses this equation to calculate the appropriate velocity compensation. The sign of the speed compensation is opposite to the predicted jitter value, and the amplitude is appropriately scaled to ensure that the compensation effectively offsets the effects of jitter without introducing overcorrection. Finally, the embedded computer algebraically superimposes the actual rotational speed of the mechanical rotation control device with the calculated real-time speed compensation to obtain corrected net rotational speed data, significantly suppressing the mechanical jitter component in this data.

[0028] Furthermore, the jitter signal is subjected to intensity judgment processing according to a preset jitter threshold, and when the jitter intensity exceeds the preset jitter threshold, least squares modeling processing is initiated to obtain a real-time speed compensation amount, including: performing root mean square value calculation processing on the jitter signal to obtain an intensity quantification index of the jitter signal; performing numerical comparison processing based on the intensity quantification index and the preset jitter threshold, and triggering a compensation algorithm start condition when the intensity quantification index exceeds the preset jitter threshold; performing least squares mathematical modeling processing on the jitter signal that meets the start condition to establish a linear regression equation between the jitter signal and the time variable; predicting and calculating the jitter trend at a future moment according to the linear regression equation to obtain a predicted jitter value; and performing sign inversion and amplitude adjustment processing on the predicted jitter value to obtain a real-time speed compensation amount for offsetting the influence of jitter.

[0029] Specifically, upon receiving the jitter signal from the mechanical rotation control device, the embedded computer immediately initiates a root mean square (RMS) calculation program to quantitatively assess the jitter signal strength. The RMS value is a statistical metric that effectively reflects the overall energy level of a changing signal. Its calculation process involves squaring the value at each sampling point in the signal, averaging it, and then performing a square root calculation. This calculation method has the advantage of eliminating the effect of positive and negative values ​​canceling out each other in the jitter signal, accurately reflecting the true jitter intensity. The calculation program segments the jitter signal into pre-set time windows, each containing a sufficient number of sampling points to ensure the reliability of the statistical results. The program squares each jitter data point within each time window, then accumulates and sums all the squared values, dividing the sum by the total number of sampling points to obtain the RMS value. Finally, the RMS value for that time window is obtained by performing a square root calculation on the RMS value. This segmented calculation method tracks the time-varying characteristics of jitter intensity and promptly detects sudden changes in jitter intensity. After the calculation is completed, the program uses the root mean square value of each time window as a quantitative indicator of the jitter signal strength in the corresponding period. This indicator is expressed in angular velocity units and intuitively reflects the potential impact of turntable jitter on the measurement accuracy of the fiber optic gyroscope.

[0030] Upon receiving the intensity quantification index, the numerical comparison processing program immediately compares and analyzes it against a pre-set jitter threshold. This pre-set jitter threshold is a critical value determined based on the FOG's measurement accuracy requirements, the turntable's mechanical performance parameters, and the north-seeking accuracy target. This threshold represents the maximum tolerable jitter intensity level. The comparison program uses real-time monitoring to continuously check whether the newly calculated intensity quantification index exceeds the pre-set threshold. When the intensity quantification index value is less than or equal to the pre-set threshold, the current mechanical jitter is within an acceptable range and will not significantly affect north-seeking accuracy. The program continues monitoring without taking any compensation measures. When the intensity quantification index exceeds the pre-set threshold, the program immediately triggers the compensation algorithm startup condition and sends a start command to the least-squares modeling module. This judgment mechanism utilizes a hysteresis comparison strategy, setting rising and falling thresholds to avoid frequent switching near the thresholds and ensure the stable operation of the compensation algorithm.

[0031] Upon receiving the start command, the least squares mathematical modeling program begins in-depth analysis and mathematical modeling of jitter signals that meet the start-up conditions. Least squares is a parameter estimation method that determines optimal model parameters by minimizing the sum of squared errors between observed values ​​and model predictions. The modeling program first collects historical jitter signal data and builds a dataset with time as the independent variable and jitter amplitude as the dependent variable. The program analyzes the jitter signal's time-domain characteristics, including its trend, periodic components, and random fluctuations, and selects an appropriate regression model structure based on these characteristics. For jitter signals exhibiting a linear trend, the program constructs a linear regression equation with two parameters: slope and intercept. For jitter signals with significant nonlinear characteristics, the program uses polynomial regression or piecewise linear regression. The modeling program uses matrix operations to determine the optimal parameters for the regression equation, minimizing the sum of squared distances between all historical data points and the regression line. During the solution process, the program also calculates the goodness-of-fit index of the regression equation to assess the model's accuracy in describing the jitter signal's variation patterns. The established linear regression equation can not only describe the historical variation pattern of the jitter signal, but more importantly, it has the ability to predict future jitter trends.

[0032] The prediction calculation process quantitatively predicts jitter trends over a short period of time in the future based on an established linear regression equation. Starting from the current time, the program advances forward according to a preset prediction time step, gradually calculating the predicted jitter values ​​for each future time. The prediction time range is determined based on the turntable's dynamic response characteristics and the fiber optic gyroscope's sampling period, ensuring both timely and accurate predictions. The program directly calculates the corresponding predicted jitter values ​​by substituting the future time values ​​into the regression equation. To enhance prediction reliability, the program also calculates confidence intervals for the predicted values ​​to assess the uncertainty level of the prediction results. Once the predicted jitter values ​​are obtained, the sign inversion and amplitude adjustment processes immediately convert them into compensation values. The sign inversion operation reverses the sign of the predicted jitter value, ensuring that the compensation value is in the opposite direction of the jitter, thereby offsetting the jitter. The amplitude adjustment process optimizes the compensation value based on the turntable's dynamic response characteristics to avoid over-compensation or under-compensation. The adjusted value becomes the real-time velocity compensation value, which is directly added to the turntable's control instructions to actively suppress mechanical jitter. The entire compensation process adopts a closed-loop control strategy. The program continuously monitors the compensation effect and dynamically adjusts the compensation parameters according to changes in actual jitter intensity to ensure the continued effectiveness of the jitter suppression effect.

[0033] S103, performing segmented adaptive window function processing on the net rotation speed data to obtain a set of Earth rotation signal segments; In one embodiment of the present invention, the segmented adaptive window function processing of the net rotation speed data to obtain a set of earth rotation signal segments includes: performing three-dimensional evaluation processing of the net rotation speed data on the signal-to-noise ratio, spectral purity and phase stability to obtain an earth rotation signal quality evaluation index; dynamically adjusting the window function processing parameters according to the signal quality evaluation index to obtain an adaptive window function parameter combination including a window length and a window function type; performing window function filtering processing on the net rotation speed data according to the adaptive window function parameter combination to obtain an earth rotation signal processed by a window function; performing temperature compensation algorithm correction processing on the earth rotation signal processed by the window function according to the current operating temperature of the fiber optic gyroscope to obtain a temperature-corrected earth rotation signal segment; and segmented organization processing on the temperature-corrected earth rotation signal segment to obtain a set of earth rotation signal segments.

[0034] Specifically, after receiving the corrected net rotation velocity data, the embedded computer immediately launches a three-dimensional signal quality assessment program to conduct a comprehensive data analysis. The signal-to-noise ratio assessment program first converts the net rotation velocity data into the frequency domain using a fast Fourier transform, identifies the primary frequency band corresponding to the Earth's rotation frequency, and calculates the signal power density within this band. The program also analyzes noise bands away from the primary frequency band, calculates the average level of noise power density, and derives a quantitative signal-to-noise ratio metric from the ratio of signal power to noise power. The spectral purity assessment algorithm analyzes the spectral distribution characteristics around the primary frequency band and quantifies the degree of spectral concentration by calculating the ratio of the primary peak energy to the total energy. A high-purity signal appears as a sharp single peak in the spectrum, while an interfered signal exhibits a diffuse or multi-peaked spectrum. The phase stability assessment program monitors the phase variation of the Earth's rotation signal over consecutive measurement cycles, assessing the signal's temporal consistency by calculating the standard deviation of the phase difference. A phase-stable signal indicates that no significant systematic phase drift occurred during the measurement. The results of the three evaluation dimensions are combined into a comprehensive Earth rotation signal quality assessment index through a weighted fusion algorithm. This index reflects the overall reliability level of the current data with a numerical range of 0 to 1.

[0035] Based on the signal quality assessment index, the dynamic adjustment processing program initiates an intelligent window parameter selection algorithm. This algorithm incorporates a built-in database of mappings between quality indexes and window parameters, automatically determining the optimal processing strategy based on the current assessment results. When the signal quality index exceeds 0.8, indicating a clear Earth rotation signal with weak interference, the program selects a shorter window length of 5 to 6 rotation periods, combined with a Hanning window. The Hanning window has a frequency response characteristic of narrow main lobes and low side lobes, enabling accurate extraction of the frequency components of the Earth rotation signal. When the quality index is between 0.5 and 0.8, the program uses a medium window length of 6 to 8 rotation periods, employing the Blackman window. The Blackman window attenuates side lobes more quickly, effectively suppressing leakage interference from adjacent frequency components. When the quality index is below 0.5, indicating strong signal noise, the program selects a longer window length of 8 to 10 rotation periods, employing a parameter-adjustable Kaiser window. The Kaiser window optimizes the trade-off between frequency resolution and sidelobe suppression by adjusting the shape parameter β. The program dynamically calculates the β parameter based on the specific quality indicator value to ensure that the window function characteristics perfectly match the current signal characteristics.

[0036] After receiving the optimized parameter combination, the window filter processing program begins performing sophisticated signal processing. The program first segments the net rotation velocity data into overlapping segments based on the specified window length. The overlap between adjacent segments is set to 50% to prevent information loss due to window edge effects. The program then generates a coefficient sequence based on the selected window function type. The coefficients of the Hanning window follow a distribution of 0.5×(1-cos(2πn / N)), the Blackman window consists of a linear combination of three cosine terms, and the Kaiser window coefficients are calculated based on a zero-order modified Bessel function. During the filtering process, the program multiplies each data segment by the corresponding window coefficient point by point, performing a time-domain windowing operation. The windowed data segments exhibit better spectral characteristics in the frequency domain, reducing spectral leakage. Furthermore, the Earth rotation signal component is enhanced based on the selectivity of the window function. The program then recombines the processed data segments using an overlap-and-add method to recover the complete time-domain signal, resulting in the windowed Earth rotation signal.

[0037] The temperature compensation algorithm applies corrections based on the fiber optic gyro's current operating temperature reading. The program consults a pre-calibrated temperature characteristic table, calculates the bias compensation value and scale factor correction coefficient at the current temperature, and then applies a linear correction to the Earth rotation signal. The segmented organization process divides the temperature-corrected signal into pre-set segment lengths, assigning a time tag and azimuth angle to each segment. This ultimately creates a structured collection of Earth rotation signal segments, each containing complete rotation period information and corresponding quality assessment results.

[0038] S104 , performing segmented cross-correlation north-finding processing on each signal segment of the earth rotation signal segment set according to the constraints of the local geographical location by an embedded computer to obtain a true north azimuth.

[0039] In one embodiment of the present invention, the embedded computer performs segmented cross-correlation north-finding and solving processing on each signal segment of the earth rotation signal segment set according to the constraints of the local geographical location to obtain the true north azimuth, including: calculating the geographical constraint weight coefficient according to the local magnetic declination, gravity anomaly and earth ellipsoid parameters in the local geographical location by the embedded computer; performing cross-correlation coefficient calculation processing on each signal segment in the earth rotation signal segment set to obtain the sine cross-correlation coefficient and cosine cross-correlation coefficient of each signal segment; performing inverse tangent function solution processing on each signal segment according to the sine cross-correlation coefficient and cosine cross-correlation coefficient to obtain the preliminary azimuth of each signal segment; analyzing the phase difference between adjacent signal segments in the earth rotation signal segment set to obtain the zero-position offset compensation amount, and performing zero-position correction processing on the preliminary azimuth to obtain the corrected azimuth of each segment; performing weighted average algorithm processing on the corrected azimuth of each segment according to the geographical constraint weight coefficient to obtain the true north azimuth.

[0040] Specifically, upon receiving the local geographic location information, the embedded computer immediately initiates a program to accurately calculate the geographic constraint weight coefficient. This program first reads the local magnetic declination data. Magnetic declination is the angle between magnetic north and true north, and its value varies depending on geographic location and time. The program then queries a global magnetic declination database to obtain the declination value for the current location. It then analyzes the magnetic field gradient changes around that location. If the device is located in an area with a large declination, indicating a complex magnetic environment that could interfere with the fiber optic gyro (FOG) measurement, the program accordingly reduces the weight coefficient for that direction. Gravity anomaly parameter processing is more complex. The program accesses a database of Earth's gravity field models to calculate the deviation between the local gravitational acceleration and the standard gravity value. Gravity anomalies can affect the horizontal mounting accuracy of the FOG, and thus the accuracy of the measurement of the horizontal component of Earth's rotation. Based on the magnitude and direction of the gravity anomaly, the program calculates the gravity correction coefficient for each azimuth. The calculation of the Earth ellipsoid parameters involves a local geodetic coordinate system transformation. Based on the device's latitude and longitude, the program calculates geometric parameters such as the radius of curvature and the normal tilt angle at that location. These parameters affect the projection of the Earth's rotational angular velocity in the local coordinate system, directly impacting the degree of match between theoretically calculated and measured values. The program integrates these three geographic parameters using a composite function to calculate a geographic constraint weight coefficient for each azimuth interval. This coefficient reflects the reliability differences in measurement data in different directions.

[0041] The cross-correlation coefficient calculation program performs sophisticated mathematical operations on each segment in the Earth rotation signal segment set. Cross-correlation is a signal processing technique used to analyze the similarity and phase relationship between two signals. For each segment, the program generates a corresponding sine and cosine reference signal, whose frequencies are aligned with the theoretical rotation frequency of the turntable. The sine reference signal is set to zero phase, while the cosine reference signal leads the sine signal by 90 degrees, forming an orthogonal reference signal pair. During the calculation, the program multiplies each segment by the sine reference signal point by point. The product is then integrated and summed over the entire length of the segment to obtain the sine cross-correlation coefficient for that segment. The magnitude of this coefficient reflects the degree of correlation between the Earth rotation signal and the sine reference signal, while its sign indicates the phase relationship between the two. Similarly, the program calculates the cross-correlation between each segment and the cosine reference signal to obtain the cosine cross-correlation coefficient. The sine and cosine cross-correlation coefficients form a complex number pair whose magnitude represents the strength of the Earth's rotation signal, while its phase angle directly relates to the angle of true north. The program repeats the same calculation process for all signal segments, building a complete database of cross-correlation coefficients.

[0042] The inverse tangent function calculation process uses the obtained cross-correlation coefficients to calculate the preliminary azimuth angle for each signal segment. The inverse tangent function is a trigonometric operation that calculates the corresponding angle value based on the sine and cosine values. The program uses a four-quadrant inverse tangent function, which can accurately determine the quadrant in which the azimuth angle is located based on the sign combination of the sine and cosine cross-correlation coefficients, avoiding the quadrant ambiguity problem of the traditional inverse tangent function. During the calculation process, the program uses the sine cross-correlation coefficient as the numerator and the cosine cross-correlation coefficient as the denominator, and uses the inverse tangent operation to obtain the angle value in radians, which is then converted to degrees. Each signal segment corresponds to a preliminary azimuth angle, which represents the angle between the FOG sensitive axis and true north when the turntable orientation corresponding to the signal segment is correct. Due to the possibility of systematic offset of the turntable mechanical zero position, these preliminary azimuth angles require further correction processing.

[0043] The zero offset analysis program identifies and compensates for the turntable's mechanical zero position error by examining the phase differences between adjacent signal segments. The program calculates preliminary azimuth angle differences between adjacent signal segments. Theoretically, this difference should equal the corresponding turntable rotation angle increment. If the actual difference systematically deviates from the theoretical value, this indicates that the turntable's mechanical zero position has shifted. Using a least-squares fitting method, the program performs a linear regression analysis on the angle differences between all adjacent segments and the theoretical turntable increment. The intercept of the regression line represents the zero offset compensation. The correction program applies this compensation to all preliminary azimuth angles to obtain the corrected azimuth angles for each segment. Finally, the weighted average algorithm combines these corrected azimuth angles based on geographic constraint weights. The program multiplies each azimuth angle by its corresponding weight, then sums all weighted results and divides them by the sum of the weights to obtain the final true north azimuth angle. This angle, optimized for geographic factors and compensated for mechanical errors, represents the precise true north direction of the device's current location.

[0044] Furthermore, the cross-correlation coefficient calculation processing of each signal segment in the Earth rotation signal segment set to obtain the sine cross-correlation coefficient and cosine cross-correlation coefficient of each signal segment includes: analyzing and processing the frequency characteristics of each signal segment in the Earth rotation signal segment set to obtain the actual rotation frequency corresponding to each signal segment; generating a sine reference signal and a cosine reference signal that match the frequency of each signal segment according to the actual rotation frequency; and performing zero-delay cross-correlation operation processing on each signal segment data and the corresponding sine reference signal and the corresponding cosine reference signal to obtain the sine cross-correlation coefficient and cosine cross-correlation coefficient of each signal segment.

[0045] Specifically, after receiving a collection of Earth rotation signal segments, the embedded computer immediately initiates a frequency signature analysis program for each segment. This program uses a high-precision spectrum analysis algorithm to convert the time-domain signal into a frequency-domain representation by performing a fast Fourier transform on the signal segment data. Due to speed fluctuations during actual turntable operation, the actual rotation frequency of each signal segment may not exactly equal the theoretically set value, necessitating precise frequency identification for each segment. The analysis program searches the frequency domain for the frequency peak with the largest amplitude, which corresponds to the primary rotation frequency component of the turntable within that signal segment. To improve the accuracy of frequency identification, the program employs interpolation refinement technology, performing high-density frequency interpolation calculations near the primary peak, increasing the frequency resolution to more than ten times the original resolution. The program also analyzes the spectral characteristics of the primary peak, including peak sharpness, bandwidth, and deviation from the theoretical frequency. These characteristic parameters are used to assess the frequency stability of the signal segment. Using a spectrum peak tracking algorithm, the program accurately extracts the actual rotation frequency corresponding to each signal segment. This frequency value reflects the actual motion state of the turntable during that time period. The frequency identification results are stored in Hertz, providing an accurate frequency reference for subsequent reference signal generation.

[0046] The reference signal generation program creates specifically matched sine and cosine reference signals for each signal segment based on the identified actual rotation frequency. The sine reference signals are generated using digital signal processing techniques. The program calculates the corresponding digital frequency parameters based on the actual rotation frequency and the sampling rate of the signal segment. A sine function generator then generates a continuous sine waveform. The generated sine reference signal has frequency characteristics identical to those of the corresponding signal segment, with its initial phase set to zero degrees and its amplitude normalized to unity. The cosine reference signal generation process is similar to the sine signal, but its initial phase leads the sine signal by 90 degrees, forming an orthogonal reference signal pair. This orthogonal design allows for simultaneous extraction of both amplitude and phase information from the Earth's rotation signal, providing a complete mathematical foundation for accurate azimuth determination. The program ensures that the length of each reference signal exactly matches the corresponding Earth's rotation signal segment, maintaining a consistent number of sampling points, thus avoiding computational errors caused by length mismatches. The generated reference signals exhibit ideal spectral purity, free of harmonic components or noise interference, providing high-quality reference signals for subsequent cross-correlation operations.

[0047] The zero-delay cross-correlation processing routine performs the core mathematical calculations, calculating the correlation between each Earth rotation signal segment and its corresponding sine and cosine reference signals. Zero-delay cross-correlation refers to a cross-correlation operation performed with zero time delay. This operation directly reflects the similarity and phase relationship between two signals at the same moment. During the calculation, the program multiplies the Earth rotation signal segment with the sine reference signal point by point. This calculation multiplies the value of each sampling point in the signal segment by the value of the sine reference signal at the corresponding moment, and then accumulates all the products. The accumulated result is divided by the signal segment length to obtain the normalized sine cross-correlation coefficient, which ranges from -1 to +1. The sign of the sine cross-correlation coefficient reflects the phase relationship between the Earth rotation signal and the sine reference signal, with a positive value indicating in-phase and a negative value indicating out-of-phase. The absolute value reflects the strength of the correlation. Similarly, the program calculates the cross-correlation between the Earth rotation signal segment and the cosine reference signal to obtain the cosine cross-correlation coefficient. The cosine cross-correlation coefficient provides phase information that is orthogonal to the sine coefficients. The two coefficients combine to form a complete complex number representation, with the real part corresponding to the cosine coefficients and the imaginary part corresponding to the sine coefficients.

[0048] The program repeatedly performs the same cross-correlation operation on all signal segments in the Earth rotation signal segment set, establishing a complete database of cross-correlation coefficients. Each signal segment corresponds to a pair of sine and cosine cross-correlation coefficients. These coefficient pairs contain complete characteristic information about the Earth rotation signal within that signal segment. The program organizes the calculation results according to the chronological order of the signal segments, assigning each coefficient pair a corresponding time label and turntable azimuth angle identifier. After the calculation is complete, the program also performs a quality check process to analyze the numerical rationality and consistency of each cross-correlation coefficient and identify data segments that may contain anomalies. This quality check includes amplitude range checks, continuity analysis between adjacent segments, and comparison with theoretical expectations. Through these checks, the program ensures that the obtained cross-correlation coefficients are sufficiently reliable and accurate, providing a reliable data foundation for the final azimuth angle solution.

[0049] In this embodiment, a mechanical rotation control device drives a fiber optic gyroscope to perform multi-position sampling according to local latitude, obtaining a raw angular velocity sequence. Spectral feature extraction and real-time velocity correction are performed on the raw signal to obtain net rotational velocity data. A segmented adaptive window function processing technique is used to obtain a set of Earth rotation signal segments. Finally, an embedded computer performs segmented cross-correlation analysis in conjunction with geographic location constraints to obtain the true north azimuth. This invention utilizes a segmented processing strategy to disperse the random effects of turntable jitter into individual segments, making the jitter within each segment relatively small and predictable. An adaptive window function is then used to dynamically adjust processing parameters based on signal quality to further suppress noise interference. A geographic constraint weighted algorithm is then used to optimize and fuse the results from each segment, effectively addressing the impact of mechanical jitter on north-seeking accuracy under high-speed turntable conditions.

[0050] The above describes the fiber optic gyroscope earth rotation north finding method in the embodiment of the present invention. The fiber optic gyroscope earth rotation north finding device in the embodiment of the present invention is described below. The fiber optic gyroscope north finding device includes a fiber optic gyroscope, an accelerometer, a mechanical rotation control device and an embedded computer. Figure 2 In one embodiment of the present invention, a fiber optic gyroscope earth rotation north-finding device includes: Sampling module 201, configured to drive the fiber optic gyroscope to perform multi-position sampling of the Earth's rotation characteristics according to the local latitude of the local geographic location through the mechanical rotation control device, and obtain an original angular velocity sequence containing projection information of the horizontal component of the Earth's rotation; The correction module 202 is used to perform spectrum feature extraction and real-time speed correction processing on the original angular velocity sequence to obtain corrected net rotational velocity data; A processing module 203 is configured to perform piecewise adaptive window function processing on the net rotation speed data to obtain a set of Earth rotation signal segments; The solving module 204 is configured to perform a segmented cross-correlation north-finding solution on each signal segment of the earth rotation signal segment set according to the constraints of the local geographical location through the embedded computer to obtain a true north azimuth.

[0051] In an embodiment of the present invention, the fiber optic gyroscope earth rotation north-finding device operates the fiber optic gyroscope earth rotation north-finding method described above. The fiber optic gyroscope earth rotation north-finding device drives the fiber optic gyroscope to perform multi-position sampling according to the local latitude through a mechanical rotation control device to obtain an original angular velocity sequence; performs spectrum feature extraction and real-time velocity correction on the original signal to obtain net rotation velocity data; adopts a segmented adaptive window function processing technology to obtain a set of earth rotation signal segments; finally, an embedded computer is used to perform segmented cross-correlation solution in combination with geographic location constraints to obtain the true north azimuth. The present invention disperses the random effects of turntable jitter into each independent segment through a segmented processing strategy, so that the jitter within each segment is relatively small and predictable, and adopts an adaptive window function to dynamically adjust the processing parameters according to the signal quality to further suppress noise interference; optimizes and fuses the results of each segment through a geographic constraint weighted algorithm, effectively solving the impact of mechanical jitter on north-finding accuracy under high-speed turntable conditions.

[0052] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0053] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0054] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fiber optic gyroscope earth rotation north finding method, characterized in that: The fiber optic gyroscope north finder includes a fiber optic gyroscope, an accelerometer, a mechanical rotation control device and an embedded computer. The fiber optic gyroscope earth rotation north finder method includes: The mechanical rotation control device drives the fiber optic gyroscope to perform multi-position sampling of the earth's rotation characteristics according to the local latitude of the local geographical location, thereby obtaining an original angular velocity sequence containing projection information of the horizontal component of the earth's rotation; Performing spectrum feature extraction and real-time speed correction processing on the original angular velocity sequence to obtain corrected net rotational velocity data; performing piecewise adaptive window function processing on the net rotation speed data to obtain a set of Earth rotation signal segments; The embedded computer performs segmented cross-correlation north-finding processing on each signal segment of the earth rotation signal segment set according to the constraints of the local geographical location to obtain the true north azimuth.

2. The fiber optic gyroscope earth rotation north finding method according to claim 1, characterized in that: The mechanical rotation control device drives the fiber optic gyroscope to perform multi-position sampling of the earth's rotation characteristics according to the local latitude of the local geographical location to obtain the original angular velocity sequence containing the projection information of the horizontal component of the earth's rotation, which includes: Calculate and process the horizontal component of the earth's rotation for the local latitude of the local geographical location to obtain the theoretical horizontal component of the earth's rotation; Adaptively adjusting sampling parameters according to the signal strength of the theoretical earth rotation horizontal component to obtain sampling parameters that match the signal strength, wherein the sampling parameters include sampling time and sampling density parameters; The mechanical rotation control device drives the fiber optic gyroscope to perform variable speed rotation sampling processing according to the sampling parameters, thereby obtaining an original angular velocity sequence containing projection information of the horizontal component of the earth's rotation.

3. The fiber optic gyroscope earth rotation north finding method according to claim 1, characterized in that: The extracting spectrum features and performing real-time speed correction processing on the original angular velocity sequence to obtain the corrected net rotational velocity data includes: Performing fast Fourier transform processing on the original angular velocity sequence to obtain a power spectral density distribution of a preset frequency band; Identifying and processing the natural frequency characteristics of the mechanical rotation control device according to the power spectrum density distribution to obtain a mechanical characteristic spectrum template including a servo control frequency, a resonance frequency, and an external interference frequency; Performing real-time comparative analysis on the power spectrum density distribution and the mechanical characteristic spectrum template to obtain a jitter signal of the mechanical rotation control device; Performing intensity determination processing on the jitter signal according to a preset jitter threshold, and starting least squares modeling processing when the jitter intensity exceeds the preset jitter threshold to obtain a real-time speed compensation amount; The actual rotation speed of the mechanical rotation control device and the real-time speed compensation amount are subjected to superposition correction processing to obtain corrected net rotation speed data.

4. The fiber optic gyroscope earth rotation north finding method according to claim 3, characterized in that: The performing intensity determination processing on the jitter signal according to a preset jitter threshold, and starting least squares modeling processing when the jitter intensity exceeds the preset jitter threshold to obtain a real-time speed compensation amount includes: Performing root mean square (RMS) calculation on the jitter signal to obtain a strength quantification index of the jitter signal; Performing a numerical comparison process based on the intensity quantification index and a preset jitter threshold, and triggering a compensation algorithm start condition when the intensity quantification index exceeds the preset jitter threshold; Perform least squares mathematical modeling on the jitter signal that meets the startup conditions and establish a linear regression equation between the jitter signal and the time variable; Predicting and calculating the jitter trend at a future moment according to the linear regression equation to obtain a predicted jitter value; The predicted jitter value is subjected to sign inversion and amplitude adjustment processing to obtain a real-time speed compensation amount for offsetting the influence of the jitter.

5. The fiber optic gyroscope earth rotation north finding method according to claim 1, characterized in that: The performing of segmented adaptive window function processing on the net rotation speed data to obtain a set of Earth rotation signal segments comprises: performing three-dimensional evaluation processing of the net rotation velocity data in terms of signal-to-noise ratio, spectrum purity, and phase stability to obtain an Earth rotation signal quality evaluation index; Dynamically adjusting the window function processing parameters according to the signal quality evaluation index to obtain an adaptive window function parameter combination including a window length and a window function type; Performing window function filtering on the net rotation speed data according to the adaptive window function parameter combination to obtain an Earth rotation signal processed by the window function; Performing temperature compensation algorithm correction processing on the earth rotation signal processed by the window function according to the current operating temperature of the fiber optic gyroscope to obtain a temperature-corrected earth rotation signal segment; The temperature-corrected earth rotation signal segments are segmented and organized to obtain an earth rotation signal segment set.

6. The fiber optic gyroscope earth rotation north finding method according to claim 1, characterized in that: The embedded computer performs segmented cross-correlation north-finding processing on each signal segment of the earth rotation signal segment set according to the constraints of the local geographical location to obtain the true north azimuth, which includes: Calculating a geographic constraint weight coefficient by the embedded computer according to a local magnetic declination, a gravity anomaly, and an earth ellipsoid parameter in the local geographic location; performing cross-correlation coefficient calculation on each signal segment in the set of Earth rotation signal segments to obtain a sine cross-correlation coefficient and a cosine cross-correlation coefficient of each signal segment; performing arc tangent function calculation on each signal segment according to the sine cross-correlation coefficient and the cosine cross-correlation coefficient to obtain a preliminary azimuth angle of each signal segment; Analyzing the phase differences between adjacent signal segments in the Earth rotation signal segment set to obtain a zero offset compensation amount, and performing zero correction processing on the preliminary azimuth to obtain corrected azimuths of each segment; The corrected azimuths are processed using a weighted average algorithm according to the geographic constraint weight coefficient to obtain the true north azimuth.

7. The fiber optic gyroscope earth rotation north finding method according to claim 6, characterized in that: The performing cross-correlation coefficient calculation on each signal segment in the set of earth rotation signal segments to obtain the sine cross-correlation coefficient and the cosine cross-correlation coefficient of each signal segment includes: Analyzing and processing the frequency characteristics of each signal segment in the set of earth rotation signal segments to obtain an actual rotation frequency corresponding to each signal segment; generating a sine reference signal and a cosine reference signal that match the frequency of each signal segment according to the actual rotation frequency; Zero-delay cross-correlation operation is performed on each signal segment data and the corresponding sine reference signal and the corresponding cosine reference signal to obtain the sine cross-correlation coefficient and cosine cross-correlation coefficient of each signal segment.

8. A fiber optic gyroscope earth rotation north-finding device, characterized in that: The fiber optic gyro north finder includes a fiber optic gyroscope, an accelerometer, a mechanical rotation control device and an embedded computer. The fiber optic gyroscope earth rotation north finder device includes: a sampling module, configured to drive the fiber optic gyroscope to perform multi-position sampling of the Earth's rotation characteristics according to the local latitude of the local geographical location through the mechanical rotation control device, and obtain an original angular velocity sequence containing projection information of the horizontal component of the Earth's rotation; a correction module, configured to perform spectrum feature extraction and real-time speed correction processing on the original angular velocity sequence to obtain corrected net rotational velocity data; a processing module, configured to perform piecewise adaptive window function processing on the net rotation speed data to obtain a set of Earth rotation signal segments; The solution module is used to perform segmented cross-correlation north-seeking solution processing on each signal segment of the earth rotation signal segment set according to the constraints of the local geographical location through the embedded computer to obtain the true north azimuth.

Images