A Joint Control Method for Vibration Isolation and Image Stabilization of Airborne Optoelectronic Platforms Based on Spectral Peak Identification

Abstract

This invention relates to the field of photoelectric detection equipment technology, and particularly to a joint control method for vibration isolation and image stabilization of airborne photoelectric platforms based on spectral peak identification. The method involves acquiring vibration acceleration signals and attitude error signals, identifying vibration spectral peaks and constructing a time-series evolution curve, and extracting a set of time-series evolution feature parameters. During imaging idle periods, micro-perturbations are injected for online calibration. The calibrated feature parameter set is analyzed to construct a system performance evolution trend curve and determine the comprehensive performance degradation parameters, thus judging the system's operating status. Based on the operating status, corresponding strategies are selected, and the parameters of the active vibration isolator and image stabilization loop are adjusted for different anomaly types. Signals are reacquired and the adjustment effect is verified, and the comprehensive performance degradation parameters are used to determine if the system returns to a healthy baseline. This invention achieves early prediction, graded self-healing, and closed-loop protection of the vibration isolation and image stabilization system performance degradation, improving the imaging stability and mission reliability of airborne photoelectric platforms during long-term flight missions.

Description

Technical Field

[0001] This invention relates to the field of photoelectric detection equipment technology, and in particular to a method for joint control of vibration isolation and image stabilization of an airborne photoelectric platform based on spectral peak identification. Background Technology

[0002] Airborne optoelectronic platforms are subjected to multi-source broadband vibrations during flight from the engine, rotor, aerodynamic turbulence, and maneuvering overloads, directly affecting the imaging clarity and target tracking accuracy of the optoelectronic payload. To suppress such vibrations, existing solutions typically employ a two-stage joint control architecture consisting of an active vibration isolator and an optoelectronic image stabilization platform. The active vibration isolator is responsible for attenuating the transmission of mid-to-high frequency airframe vibrations to the optoelectronic payload, while the image stabilization loop provides closed-loop compensation for residual line-of-sight sway. The parameter matching of the vibration isolator and the image stabilization loop is highly dependent on the real-time identification of the main vibration spectral peaks. Traditional methods involve extracting the peak frequencies and amplitudes from the acceleration signal through spectral analysis and adjusting the notch filter parameters of the vibration isolator or the bandwidth of the image stabilization loop accordingly.

[0003] However, this static identification method based on instantaneous peak detection has significant shortcomings. Firstly, the vibration environment of an airborne platform is not a time-invariant system. Over long-term use, the equivalent stiffness and damping parameters of the vibration isolators slowly drift unidirectionally due to factors such as rubber component aging, preload relaxation, and magnetorheological fluid sedimentation. This causes the frequency center of the vibration peak to gradually deviate from the design value. Traditional instantaneous peak identification lacks the ability to capture this temporal evolution trend and cannot predict performance degradation in advance. Secondly, most existing control strategies perform passive feedback adjustment based on the current peak information, lacking feedforward prediction methods for impending transient impacts or resonant jumps, making it difficult to avoid short-term imaging blur caused by control lag. This makes it difficult for existing peak-based vibration isolation and image stabilization joint control methods to continuously guarantee photoelectric imaging quality during long-term flight missions. Summary of the Invention

[0004] To address this, the present invention provides a joint control method for vibration isolation and image stabilization of an airborne optoelectronic platform based on spectral peak identification. This method overcomes the problem in existing technologies that fail to perceive and calibrate the temporal evolution trend of vibration spectral peaks online, thus failing to distinguish the performance degradation stage of the vibration isolation system and execute hierarchical closed-loop control, resulting in difficulty in ensuring imaging stability during long-term flight.

[0005] To achieve the above objectives, this invention provides a method for joint control of vibration isolation and image stabilization of an airborne optoelectronic platform based on spectral peak identification, comprising: The vibration acceleration signal of the active vibration isolator of the airborne optoelectronic platform and the attitude error signal of the optoelectronic image stabilization platform are collected, and the vibration spectrum peaks are identified by spectrum analysis of the vibration acceleration signal. Construct the time-series evolution curve of the spectral peak and extract the time-series evolution feature parameter set of the spectral peak, wherein the time-series evolution feature parameter set includes frequency drift rate and amplitude growth rate; During the imaging idle period, a known micro-perturbation is injected into the active vibration isolator, the perturbation response spectrum is obtained, and the currently identified spectral peak characteristics are calibrated online. The calibrated time-series evolution characteristic parameter set was analyzed to construct the system performance evolution trend curve, determine the comprehensive performance degradation parameters, and preliminarily determine the working status of the airborne optoelectronic platform vibration isolation and image stabilization system. Based on the operating status, a joint control and analysis strategy for vibration isolation and image stabilization is determined to identify the type of system anomaly or to implement protection, including: conventional monitoring mode, self-healing adjustment mode, and emergency protection mode. Based on the system anomaly type and the characteristic information of the corresponding spectral peaks, the operating parameters of the active vibration isolator and the image stabilization circuit are adjusted accordingly. The vibration acceleration signal is reacquired and the comprehensive performance degradation parameter is calculated. If the comprehensive performance degradation parameter returns to the preset healthy benchmark range for several consecutive times, the control adjustment is exited; otherwise, the control intensity is gradually increased.

[0006] As a preferred technical solution for the joint control method of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification, the time-series evolution feature parameter set also includes spectral peak drift rate and spectral peak drift rate. The peak drift rate is calculated based on the instantaneous kurtosis frequency drift curve, and the kurtosis variability is the statistical dispersion of the kurtosis of the peaks in each sliding window within a set monitoring period.

[0007] As a preferred technical solution for the combined vibration isolation and image stabilization control method of airborne optoelectronic platforms based on spectral peak identification, the online calibration of the currently identified spectral peak characteristics includes: During the imaging idle period, a swept-frequency micro-perturbation signal is injected into the active vibration isolator, and the perturbation response spectrum is acquired. The position of the response spectrum peak is compared with the position of the currently identified spectrum peak, the frequency deviation and amplitude deviation are calculated, and the frequency center and amplitude of the spectrum peak are corrected.

[0008] As a preferred technical solution for the joint control method of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification, the steps for constructing the system performance evolution trend curve include: The Weibull distribution function was used to fit each time series evolution characteristic parameter to obtain the single-parameter evolution trend curve; The evolution trend curves of each single parameter are weighted and fused according to the weight coefficients of each parameter's influence on system performance to construct the overall system performance evolution trend curve. The overall performance degradation parameter is determined based on the shape and scale parameters of the overall performance evolution trend curve.

[0009] As a preferred technical solution for the joint control method of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification, the working state of the airborne optoelectronic platform vibration isolation and image stabilization system is initially determined, including: If the overall performance degradation parameter is less than the preset stability threshold, the system is determined to be in a steady state. If the overall performance degradation parameter is between a preset stable threshold and a preset critical threshold, then the system is determined to be in the early stage of degradation. If the overall performance degradation parameter is greater than a preset critical threshold, the system is determined to be in a critical fluctuation period.

[0010] As a preferred technical solution for the joint control method of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification, the joint control analysis strategy for vibration isolation and image stabilization is determined according to the operating state, including: If the system is in a steady state, the conventional monitoring mode is adopted, keeping the parameters of the active vibration isolator and the image stabilization circuit at their default values, and only periodically monitoring the overall performance degradation parameters. If the system is in the early stage of degradation, the self-healing adjustment mode is adopted to shorten the sliding time window interval, determine the abnormality type by combining the spectral peak evolution characteristics, and perform online parameter compensation. If the system is in a critical fluctuation period, the emergency protection mode is adopted to further shorten the sliding time window interval, immediately execute the limit protection control and generate an early warning signal. If the overall performance degradation parameter does not return to the healthy benchmark after control, the control intensity is upgraded.

[0011] As a preferred technical solution for the joint control method of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification, in response to the self-healing adjustment mode, the abnormality type is determined and online parameter compensation is performed, including: If the frequency drift rate shows a continuous unidirectional drift and the kurtosis variability exceeds the preset value, it is determined to be an abnormality such as aging of the vibration isolator performance or loosening of the structural connection. The stiffness compensation and damping compensation are calculated according to the drift direction and kurtosis variability, and the parameters of the active vibration isolator are adjusted online and the low cutoff frequency of the image stabilization circuit is matched synchronously. If the amplitude growth rate continues to rise and the frequency drift rate is within the preset range, it is determined to be a steady-state resonance anomaly. The parameters of the active vibration isolator notch filter are adjusted according to the center of the spectrum peak frequency. If the overall performance degradation parameter triggers the threshold and both the frequency and amplitude are within the preset range, it is determined to be sensor drift or abnormal electrical noise interference. The data of the interfered channel is downweighted and a digital notch filter is added to the image stabilization circuit.

[0012] As a preferred technical solution for the joint control method of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification, in response to the emergency protection mode, the ultimate protection control is executed, including: Switch the image stabilization system to high-bandwidth image lock mode, switch the active vibration isolator to maximum energy consumption mode, generate system performance warning signals, and record the spectral peak evolution curve and control adjustment process throughout the entire process; If the overall performance degradation parameters are continuously calculated after execution and still do not return to the preset healthy benchmark range, the active vibration isolator drive power supply will be cut off and the system will switch to passive vibration isolation mode.

[0013] As a preferred technical solution for the joint control method of vibration isolation and image stabilization of airborne optoelectronic platform based on spectral peak identification, in response to the conventional monitoring mode, if the spectral peak amplitude growth rate suddenly increases within a preset short-time window and the duration is less than a preset short-time threshold, it is determined to be a transient impact anomaly, and the bandwidth of the image stabilization loop is increased to a preset high bandwidth value. If the amplitude growth rate of the spectral peak continues to rise and the frequency is stable, it is determined to be a steady-state resonance anomaly. The center frequency and bandwidth of the notch filter of the active vibration isolator are adjusted according to the center of the spectral peak frequency.

[0014] As a preferred technical solution for the combined vibration isolation and image stabilization control method of airborne optoelectronic platform based on spectral peak identification, the active vibration isolator is a magnetorheological fluid damping isolator or a piezoelectric ceramic active vibration isolator; the vibration acceleration signal is collected by accelerometers arranged at the upper and lower ends of the active vibration isolator; and the attitude error signal is collected by gyroscopes inside the image stabilization platform.

[0015] Compared with existing technologies, the advantages of this invention are as follows: This invention achieves real-time calibration of spectral peak features through online micro-perturbation injection and response comparison, eliminating identification errors caused by sensor drift and environmental changes; on this basis, it utilizes the Weibull distribution function to fuse multiple time-series evolution characteristic parameters to construct the overall system performance evolution trend curve, and extracts normalized comprehensive performance degradation parameters to confirm the working state of the vibration isolation and image stabilization system; then, it selects corresponding strategies according to the working state, and with the gradual shortening of the sliding time window interval, it maintains the lowest parameter perturbation and computational load during the steady state period, performs differentiated online parameter compensation for different anomaly types to achieve self-healing repair, and provides extreme protection during the critical fluctuation period through high-bandwidth image locking, maximum energy consumption vibration isolation, and full-process data recording and early warning; and when self-healing is ineffective, it automatically cuts off the active drive power supply and switches to passive vibration isolation to form a fallback isolation, effectively avoiding image blurring caused by vibration isolator performance degradation or sudden changes in operating conditions, and significantly improving the long-term stability and mission reliability of photoelectric imaging quality in complex flight environments. Attached Figure Description

[0016] Figure 1This is a flowchart of the joint control method for vibration isolation and image stabilization of an airborne optoelectronic platform based on spectral peak identification, according to an embodiment of the present invention. Figure 2 A flowchart for constructing the system performance evolution trend curve according to an embodiment of the present invention; Figure 3 A logic diagram for determining the working state of the airborne optoelectronic platform vibration isolation and image stabilization system in this embodiment of the invention. Detailed Implementation

[0017] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0018] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0020] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] Please see Figure 1 As shown, this is a method for joint control of vibration isolation and image stabilization of an airborne optoelectronic platform based on spectral peak identification, comprising: Step S1: Collect the vibration acceleration signal of the active vibration isolator of the airborne optoelectronic platform and the attitude error signal of the optoelectronic image stabilization platform, and perform spectral analysis on the vibration acceleration signal to identify the vibration spectrum peaks; Step S2: Construct the time-series evolution curve of the spectral peak and extract the time-series evolution feature parameter set of the spectral peak, wherein the time-series evolution feature parameter set includes the frequency drift rate and the amplitude growth rate. Step S3: During the imaging idle period, inject a known micro-perturbation into the active vibration isolator, obtain the perturbation response spectrum, and perform online calibration on the currently identified spectral peak characteristics; Step S4: Analyze the calibrated time-series evolution characteristic parameter set, construct the system performance evolution trend curve, determine the comprehensive performance degradation parameters, and preliminarily determine the working status of the airborne optoelectronic platform vibration isolation and image stabilization system. Step S5: Determine the joint control and analysis strategy of vibration isolation and image stabilization based on the working state to determine the type of system anomaly or to carry out protection, including: conventional monitoring mode, self-healing adjustment mode, and emergency protection mode. Step S6: Based on the system anomaly type and the characteristic information of the corresponding spectral peaks, adjust the operating parameters of the active vibration isolator and the image stabilization circuit accordingly. Step S7: Reacquire vibration acceleration signal and calculate comprehensive performance degradation parameter. If the comprehensive performance degradation parameter returns to the preset healthy benchmark range for several consecutive times, exit control adjustment; otherwise, upgrade control intensity step by step.

[0022] Specifically, the peak shift rate is calculated based on the instantaneous kurtosis and frequency shift curve, reflecting the rate of change in the correlation between the frequency center and sharpness of the peak. In peak identification over several consecutive sliding time windows, the correspondence between the peak frequency center and the kurtosis within each window is recorded simultaneously, forming an instantaneous kurtosis-frequency shift curve with frequency on the horizontal axis and kurtosis on the vertical axis. The peak shift rate is obtained by calculating the rate of change of kurtosis with respect to frequency over adjacent time periods on this curve. For example, if the peak frequency slowly shifts from approximately 45 Hz to approximately 47 Hz over ten window intervals, while the kurtosis decreases from approximately 3.0 to approximately 2.2, a shift rate of 0.4 can be calculated.

[0023] Kurtosis variability is the statistical dispersion of the kurtosis values ​​of the spectral peaks identified in each sliding window within a set monitoring period, typically expressed as standard deviation or interquartile range. Its purpose is to characterize the stability of vibration energy distribution. When the vibration isolator is operating normally, the spectral peak shape is stable, and the kurtosis dispersion is small; when the vibration isolator performance deteriorates or intermittent abnormal excitation occurs, the spectral peaks become wider and narrower, kurtosis fluctuations intensify, and kurtosis variability increases significantly.

[0024] Through a limited number of accelerated aging tests and flight measurements, the distribution ranges of peak shift rate and kurtosis variability in healthy, deteriorated, and failed states were established, and the upper quantile of the healthy state data was used as the safety threshold. In this invention, the peak shift rate captures the rate of change in the correlation between the peak frequency center and kurtosis, while kurtosis variability characterizes the stable change in vibration energy concentration. Together, they extend traditional peak identification from the static frequency amplitude level to the dynamic characteristic level. When the vibration isolator experiences stiffness degradation or structural loosening, the unidirectional frequency drift and increased kurtosis fluctuation will produce significant responses on these two parameters. This allows the system to obtain clear early warning signals in the early stages of performance degradation, providing a highly discriminative basis for subsequent state determination and anomaly classification.

[0025] In practice, the system determines whether the optoelectronic platform is currently in a non-imaging task period, such as when the infrared thermal imager detector is in a cooling break or the visible light camera is in an inter-frame interval. The determination can be based on the detector's integral synchronization signal and the mission status indication from the flight management system, ensuring that the injection of micro-disturbance signals will not affect the high-resolution images being acquired. When the idle condition is met, a small-amplitude sinusoidal sweep signal with a linearly changing frequency over time is injected through the actuator of the active vibration isolator. Its amplitude is controlled within a few percentage points of the isolator's rated driving force, for example, not exceeding 5% of the rated output, and its frequency range covers the adjacent frequency band of the current main vibration spectrum peak, for example, within a few hertz plus or minus the current peak frequency.

[0026] Based on this, the acceleration response at the output of the vibration isolator is simultaneously acquired, and spectral analysis is performed on the response signal to obtain the actual disturbance response spectrum. The positions of the spectral peaks identified in the measured response spectrum are compared with the positions of the main vibration spectral peaks obtained by the conventional spectral peak identification module around the same time to establish a mapping relationship from excitation to response, and the frequency deviation and amplitude deviation are calculated. For example, if the injected sweep frequency signal should theoretically produce a response at 50Hz, but the measured response spectral peak is located at approximately 50.3Hz, there is a frequency deviation of approximately 0.3Hz; similarly, if there is a systematic deviation of approximately 3% between the response amplitude and the theoretical expectation, this deviation is synchronously incorporated into the correction model. Finally, these deviations are used as calibration offsets to correct the spectral peak frequency center and amplitude in real time for the current period and for a subsequent period until the next calibration event occurs.

[0027] In this invention, online calibration of spectral peak identification results is achieved by injecting known micro-perturbations during imaging idle periods and comparing the responses. This method utilizes the system's own actuators as the excitation source, eliminating frequency and amplitude measurement deviations caused by factors such as temperature changes, sensor zero drift, or device aging without additional hardware, ensuring that the data basis for time-series evolution analysis and performance degradation determination remains accurate and reliable. Furthermore, calibration is performed during non-imaging intervals, without affecting the image acquisition quality of normal tasks.

[0028] Understandably, selecting frequency drift rate, kurtosis variability, and spectral peak drift rate as the main parameters for curve construction from the set of characteristic parameters is reasonable. All three parameters exhibit a monotonic trend in system degradation and have a clear physical direction. Frequency drift rate reflects the long-term unidirectional shift of the equivalent stiffness or mass of the vibration isolation system; kurtosis variability reflects the stability degradation of vibration energy concentration; and spectral peak drift rate reflects the rate of degradation of the correlation between the frequency center and kurtosis. The amplitude growth rate and the rate of change of half-width at half-maximum (HWHM) are significantly affected by the instantaneous changes in flight conditions and are therefore not included as main parameters for fusion but are retained for subsequent anomaly type determination. Data is collected for each parameter according to a time series, accumulating data points over a sufficiently long time span, such as several hours of data corresponding to multiple consecutive flights. Subsequently, with time as the independent variable and the cumulative change of each characteristic parameter as the dependent variable, the Weibull distribution function is used to perform least-squares fitting on the evolution trend of each parameter, yielding their respective single-parameter evolution trend curves.

[0029] The shape parameter of the Weibull distribution determines whether the degradation trend is early-accelerated, linear, or late-accelerated, while the scale parameter determines the overall time span of degradation, thus enabling the capture of the inflection point characteristics of system performance from slow to rapid degradation. By fitting each feature parameter separately, the corresponding shape and scale parameters are obtained.

[0030] Based on the relative importance of each parameter to the performance of the vibration isolation and image stabilization system, weighting coefficients are set for fusion. The weighting coefficients are determined by establishing the correlation coefficient between each parameter and the final image sway using historical data, and then using linear regression or grey relational analysis to obtain the contribution of each parameter change to image quality degradation, which is then normalized into weight values. The shape and scale parameters of each single-parameter curve are weighted and fused according to their respective weights to obtain a weighted composite curve, which represents the overall performance evolution trend curve of the system.

[0031] The overall performance degradation parameter is calculated based on the shape and scale parameters of the overall performance evolution trend curve.

[0032] The original comprehensive degradation index R = β / η; Where β is the shape parameter of the fusion curve and η is the scale parameter. An increase in β indicates that the degradation curve rises more rapidly, while a decrease in η indicates that the degradation completion time is shortened. Under the combined effect of both, the R value increases monotonically, and the more severe the system degradation, the larger the R value. The magnitude of this R value depends on the specific parameters of the vibration isolation system and the unit system of the data.

[0033] To facilitate engineering interpretation and subsequent threshold comparison, R is linearly normalized and mapped to obtain the comprehensive performance degradation parameter D=(R-Rmin) / (R_max-R_min); Where Rmin is the upper quantile of the R value under healthy system conditions, and Rmax is the R value corresponding to when the system reaches the technical indicator of complete failure.

[0034] The calibration method for Rmin and Rmax is as follows: In accelerated aging tests, the entire process of the vibration isolator operating from a brand-new state to the point where its transmissibility deteriorates to the limit of its technical specifications is recorded, and the R value at each moment throughout the process is calculated. The top tenths of all R values ​​during the healthy and stable phase are taken as Rmin, and the R value at the moment when the transmissibility deteriorates to the specified limit is taken as Rmax. After this mapping, the D value falls between 0 and 1, with D during the healthy period approximately between 0 and 0.3, and the failure point being 1.

[0035] In this invention, the Weibull distribution function is used to fit and fuse multiple characteristic parameters, which can capture the inflection point characteristics of the system from slow degradation to accelerated decline, and integrate parameters with different dimensions and directions of change into a single overall system performance evolution trend curve. Based on this, the extracted comprehensive performance degradation parameters, in the form of normalized scalars, uniformly reflect the overall health level of the vibration isolation and image stabilization system, making the degradation degree comparable under different flight operations and operating conditions, and providing a simple and reliable quantitative basis for the subsequent three-level determination of the working status.

[0036] In practice, the stability threshold and critical threshold were determined through a combination of accelerated life testing on the ground and statistical analysis of historical flight data. Multiple sets of the same type of vibration isolators were subjected to continuous vibration stress exceeding rated operating conditions in the laboratory to accelerate their aging. The evolution characteristics of their spectral peaks were recorded throughout the process, and the overall performance degradation parameters were calculated. Simultaneously, the image stabilization accuracy output by the optoelectronic platform was recorded.

[0037] When the image sway begins to exceed the requirements specified in the system's technical specifications, the value of the overall performance degradation parameter at this point is recorded as a reference for the critical threshold. The stability threshold is usually taken as the upper quantile of the overall performance degradation parameter of multiple healthy vibration isolators operating under normal conditions for a long period of time, typically 10%.

[0038] If the calculated comprehensive performance degradation parameter value at a certain moment is less than the stability threshold (for example, if the stability threshold is assumed to be a normalized value of 0.3, and the current comprehensive parameter is 0.2), the system is considered to be in a steady state, meaning the vibration isolation and image stabilization system is operating well. If the comprehensive performance degradation parameter is between the stability threshold and the critical threshold (for example, between 0.3 and 0.7), the system is considered to be in the early stage of degradation, indicating that the performance has shown a measurable degradation trend but still has a certain margin, and online compensation measures need to be initiated. If the comprehensive performance degradation parameter is greater than the critical threshold (for example, exceeding 0.7), the system is considered to have entered the critical fluctuation period, with a rapid risk of a sharp performance drop or loss of control, and extreme protection measures must be taken immediately.

[0039] In this invention, by presetting stable and critical thresholds, the continuously changing comprehensive performance degradation parameters are discretized into three operating state intervals with clear physical meanings: steady state, pre-deterioration stage, and critical fluctuation stage. This three-level division enables the system to automatically identify its own health stage and switch control strategies in a timely manner during the gradual deterioration process, avoiding missed or false judgments caused by a single threshold, and laying an accurate decision-making foundation for subsequent hierarchical closed-loop control.

[0040] During implementation, when the system is determined to be in a steady-state period, it adopts a conventional monitoring mode. In this mode, the system maintains the stiffness and damping parameters of the active vibration isolator, and the gain and bandwidth parameters of the image stabilization circuit at their factory default values ​​or optimal values ​​set before the initial flight, without applying any additional human interference. Vibration signal peak identification continues, but the update interval of the sliding time window remains at a relatively loose level; for example, the step size of each window movement is maintained at the default value in seconds, and the value of the overall performance degradation parameter is only periodically updated with a low computational load. The purpose of this mode is to minimize the potential negative impact of unnecessary parameter perturbations on normal imaging tasks.

[0041] When the system is identified as being in the early stages of degradation, it enters a self-healing adjustment mode. In this mode, the system automatically shortens the update interval of the sliding time window, for example, reducing it from the default normal step size to half of its original value, to improve the timeliness of capturing spectral peak evolution characteristics. In this mode, the system, based on the specific manifestations of the spectral peak evolution characteristics, further determines whether the degradation anomaly is due to vibration isolator performance aging, loose structural connections, steady-state resonance, or sensor interference, and performs targeted online parameter compensation, striving to bring the system back to a healthy range without interrupting the imaging task. After compensation, it continuously monitors whether the overall performance degradation parameters have fallen below the stable threshold.

[0042] When a critical fluctuation period is identified, the system enters emergency protection mode. At this time, the update interval of the sliding time window is further shortened, for example, to a quarter or even less of the initial default value, to achieve near real-time status awareness and response. In this mode, fine-grained anomaly classification is no longer performed; instead, extreme protection control is immediately executed, switching the vibration isolator to maximum energy dissipation mode, locking the stabilization system to high bandwidth, and simultaneously generating an early warning. If, after several consecutive cycles, the overall performance degradation parameters still fail to return to the healthy baseline range—for example, remaining above the critical threshold or showing a continuing upward trend—the control intensity is escalated. The escalation process progresses step by step until the power supply to the active vibration isolator is finally cut off, switching to passive vibration isolation. This hierarchical closed-loop control mechanism ensures that even under the most extreme operating conditions, the system will not enter a state of complete loss of control.

[0043] In this invention, three modes—routine monitoring, self-healing adjustment, and emergency protection—are automatically selected based on the operating status. Combined with progressively shortening sliding time window intervals, this achieves a tiered progression from lenient monitoring to refined diagnosis and rapid response. During the steady-state period, low computational load and zero parameter disturbances are maintained; precise self-healing is initiated in the early stages of degradation; and extreme protection is immediately implemented and its effectiveness verified in a closed-loop manner during critical fluctuations. This forms a complete adaptive management chain, enabling the system to maintain optimal performance when healthy, autonomously repair itself during degradation, and provide full protection during dangerous situations.

[0044] In practice, when analysis reveals that the concentrated frequency drift rate of the time-series evolution parameters continuously drifts in a single direction, and the kurtosis variability exceeds a preset value, it is determined to be an anomaly such as aging of the vibration isolator performance or loosening of structural connections. Unidirectional frequency drift signifies an irreversible change in the equivalent stiffness or equivalent mass of the vibration isolation system, such as hardening of rubber components or a decrease in bolt preload; while increased kurtosis variability indicates that the system's response to random excitation has become unstable. In this case, the system selects the polarity of compensation based on the drift direction and determines the stiffness compensation and damping compensation amounts as follows.

[0045] For every unit shift in frequency drift towards lower frequencies, the corresponding stiffness is compensated upwards by a fixed percentage of the current value. This percentage is determined by the slope of the linear regression between the frequency drift and the optimal stiffness compensation in ground calibration experiments. For every unit increase in kurtosis variability beyond a preset safety limit, the corresponding damping is increased upwards by a fixed step size. This step size is determined by the slope of the linear regression between kurtosis variability and the optimal damping increment. The stiffness compensation is proportional to both the frequency drift and kurtosis variability, and the damping compensation is also proportional to both. All proportionality coefficients are obtained through accelerated aging experiments using multiple linear regression calibration with the goal of minimizing image sway. A frequency drift towards lower frequencies represents a decrease in stiffness, and the system increases the virtual stiffness parameter of the vibration isolator according to the above proportional relationships; conversely, it decreases. The compensation amount is converted into drive current or voltage by the motion controller and applied to the actuator. At the same time, the low cutoff frequency of the image stabilization circuit is matched to the center point of the new residual vibration frequency to avoid low-frequency image blurring caused by changes in the spectral characteristics of residual vibration after vibration isolation.

[0046] The preset safety limit is the statistical baseline value of kurtosis variability under healthy system conditions. The calibration method involves collecting vibration data from multiple healthy vibration isolators operating under normal conditions for an extended period during the ground phase. The mean and standard deviation of kurtosis variability for each device within a set monitoring period are calculated, and the upper decimal of the kurtosis variability for all healthy samples is taken as the safety limit. When kurtosis variability exceeds this limit, it indicates that the fluctuation in vibration energy concentration has exceeded the normal distribution range, requiring the activation of damping compensation. This limit is dimensionless and depends on the system sampling rate and kurtosis calculation window. It is typically calculated to be approximately three times the healthy mean or the healthy mean plus three times the standard deviation. The specific multiple is determined by the inflection point where kurtosis variability begins to monotonically increase during accelerated aging experiments.

[0047] When the amplitude growth rate continues to rise but the frequency drift rate remains within the preset range, excluding stiffness drift factors, it is determined to be a steady-state resonance anomaly. This situation is common when the excitation force of the vibration source itself increases or a certain mode of the machine body is excited, but the vibration isolator itself does not degrade. In this case, the system should adjust the notch filter to precisely lock its center frequency to the currently identified spectral peak frequency, and adjust the depth and quality factor of the notch filter according to the amplitude.

[0048] When the overall performance degradation parameter has triggered a threshold, but both the frequency drift rate and amplitude growth rate are within the preset normal range, it is determined to be sensor drift or abnormal electrical noise interference. In this case, neither the vibration isolator itself nor the vibration environment has substantially deteriorated; rather, there is a deviation in the measurement link. Adjusting the mechanical parameters based on this false appearance would introduce real disturbances. Therefore, the system downweights or temporarily isolates the sensor data from the interfered channel and simultaneously adds a digital notch filter at the corresponding frequency point in the image stabilization control loop to cut off the influence path of false frequency components on control decisions.

[0049] In this invention, under the self-healing adjustment mode, anomalies are categorized into three types based on different combinations of parameters such as frequency drift rate, amplitude growth rate, and overall performance degradation: vibration isolator performance aging or structural loosening, steady-state resonance, and sensor drift or electrical noise interference. Differentiated online compensation strategies are implemented for each root cause, achieving targeted parameter repair and avoiding control mismatch or spurious disturbances that may result from general adjustments, thus improving the accuracy and success rate of self-healing intervention.

[0050] When the emergency protection mode is activated, the image stabilization system immediately switches from the normal tracking mode to the high-bandwidth image lock mode. This significantly increases the closed-loop bandwidth of the servo control loop, enabling it to suppress rapid and large-amplitude line-of-sight sway. The trade-off is that it may introduce higher noise and power consumption, but this is within acceptable limits in this extreme situation. Simultaneously, the active vibration isolator switches to its maximum energy dissipation mode. When using a magnetorheological fluid damper, this mode means adjusting the excitation current to near its saturation value, pushing the damping coefficient to its physical limit, and maximizing the conversion of vibration energy into heat for dissipation.

[0051] Throughout this process, the system records the spectral peak evolution curve and every operational command and response result of control adjustments at a high frequency, forming a continuous black-box-style data log. Simultaneously, it triggers the airborne early warning system, transmitting status information via data link downlink or post-flight analysis interface for timely investigation by ground crew. This data link includes timestamps, current and past peak drift rates, kurtosis variability, overall performance degradation parameters, and the specific actions taken.

[0052] If, after the emergency protection is implemented, the comprehensive performance degradation parameters obtained from re-collection and calculation for several consecutive cycles remain above the critical threshold, or even continue to increase, it indicates that the isolator itself or its structure has suffered a serious failure. Continuing to maintain active control is not only useless, but may also exacerbate the damage due to the injection of driving energy. At this time, the system activates an emergency isolation strategy, cutting off the active isolator drive power supply by controlling relays or solid-state switches, and releasing the lock on the passive damping element, causing it to automatically switch to a pure passive vibration isolation mode. Although this passive mode cannot actively compensate for high-frequency disturbances, it avoids the negative damping effect caused by active drive failure, ensuring that the platform does not completely lose its vibration isolation capability.

[0053] In this invention, the emergency protection mode utilizes the coordinated execution of high-bandwidth image locking and maximum energy consumption mode to temporarily sacrifice some noise performance in exchange for structural safety and imaging continuity. Simultaneously, the spectral peak evolution curve and control actions are recorded throughout the process to form a traceable data log. In the event of self-healing failure, the power supply to the active vibration isolator is automatically cut off, switching to passive vibration isolation mode, forming a destructive fallback isolation to prevent the negative damping effect caused by active drive failure, ensuring that the platform does not completely lose its vibration isolation capability under extreme conditions.

[0054] In implementation, while monitoring spectral peaks in the conventional monitoring mode, special attention is paid to sudden changes in the amplitude growth rate within a short period. When a sudden increase in the spectral peak amplitude growth rate is detected within a preset short-term window, and the duration of this sudden increase is extremely short, for example, less than a few milliseconds shorter than the preset timing, it is determined to be a transient shock anomaly. This threshold is set by statistically analyzing the response time of instantaneous acceleration spikes caused by gusts or weapon drops at the spectral peak level in measured flight data. For example, recording the duration distribution of a large number of transient events from a sudden increase in amplitude to a return to steady state, and taking its lower quantile or typical statistical value. Since the duration of such shocks is much shorter than the transient response time constant of the image stabilization servo system, it may cause short-term blurring of the next frame image. Therefore, the system's corresponding measure is to significantly increase the bandwidth of the image stabilization loop within a short period of time, equal to or close to the maximum allowable servo bandwidth of the system, thereby improving its ability to follow rapidly changing disturbances.

[0055] If the amplitude growth rate continues to rise while the frequency center of the spectral peak remains relatively stable, transient impacts are ruled out, and the problem is determined to be a steady-state resonance anomaly. This situation is common when the gap between a mechanical component gradually widens or the imbalance of a rotating component increases. In response, the system adjusts the center frequency of the notch filter in the active vibration isolator to the current spectral peak frequency and dynamically adjusts the notch depth and quality factor based on the amplitude.

[0056] The specific number of consecutive observations can be determined by experimentally measuring the dynamic response characteristics of the system after self-healing or protection actions. For example, in ground tests, different degrees of degradation are simulated and the parameter stabilization time after self-healing compensation is recorded. The number of observations required for the overall performance degradation parameter to converge and eventually fall below the stabilization threshold within several cycles after the action is counted. The upper limit of this number is used as the number of consecutive regressions required for judgment. If the overall parameter continuously decreases and stabilizes below the stabilization threshold within the set number of observations after compensation, the system is considered to have recovered and the control mode is gradually downgraded back to the conventional monitoring mode. If the overall parameter oscillates back and forth or increases further within the number of observations, it is determined that this round of control is ineffective, and the control intensity is increased according to the preset steps. The upgrade gradient of control intensity can include: the first upgrade increases the damping coefficient to the preset high value level, the second upgrade simultaneously increases the vibration isolator stiffness and notch filter depth, and the third upgrade directly calls the emergency protection mode until emergency isolation is performed.

[0057] In this invention, under conventional monitoring mode, the abrupt changes and duration of the spectral peak amplitude growth rate are used to distinguish between transient impact anomalies and steady-state resonance anomalies. For transient impacts, the image stabilization bandwidth is rapidly increased to suppress rapidly changing disturbances; for steady-state resonances, the notch filter parameters are precisely adjusted to suppress continuous vibrations. This achieves the ability to identify and quickly handle occasional disturbances even when the system is generally healthy, preventing small disturbances from escalating into major problems.

[0058] Active vibration isolators employ either magnetorheological fluid damping isolators or piezoelectric ceramic active vibration isolators. The former, due to its damping being continuously and rapidly adjustable with the magnetic field, exhibits excellent energy efficiency in low-frequency, large-amplitude vibration isolation applications. The latter, with its high stiffness and high bandwidth, is suitable for precise compensation of medium- and high-frequency micro-amplitude vibrations. Two vibration isolators can be orthogonally or obliquely arranged in the installation space to simultaneously handle vibration components in multiple directions.

[0059] The acquisition of vibration acceleration signals is accomplished by accelerometers installed at the upper end of the active vibration isolator (connected to the photoelectric load) and the lower end (connected to the aircraft body), respectively. Differential analysis or transmissibility calculation of the two signals effectively isolates interference from the rigid motion of the aircraft body, extracts relative vibration information caused by the vibration isolator itself, and provides pure vibration source data for the evolution analysis of spectral peaks. The accelerometers selected are low-noise, high-resolution microelectromechanical systems (MEMS) or piezoelectric accelerometers, with a range covering several times the typical vibration intensity of the airborne platform. The attitude error signal of the photoelectric stabilization platform is acquired by a high-precision gyroscope built into the stabilization platform. The type can be a fiber optic gyroscope or a ring laser gyroscope, and its zero-bias stability and angular random walk indicators meet the requirements of imaging sharpness for residual sway angular velocity. The technical solution of the present invention has now been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of this invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of this invention.

[0060] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for joint control of vibration isolation and image stabilization of an airborne optoelectronic platform based on spectral peak identification, characterized in that, include: The vibration acceleration signal of the active vibration isolator of the airborne optoelectronic platform and the attitude error signal of the optoelectronic image stabilization platform are collected, and the vibration spectrum peaks are identified by spectrum analysis of the vibration acceleration signal. Construct the time-series evolution curve of the spectral peak and extract the time-series evolution feature parameter set of the spectral peak, wherein the time-series evolution feature parameter set includes frequency drift rate and amplitude growth rate; During the imaging idle period, a known micro-perturbation is injected into the active vibration isolator, the perturbation response spectrum is obtained, and the currently identified spectral peak characteristics are calibrated online. The calibrated time-series evolution characteristic parameter set was analyzed to construct the system performance evolution trend curve, determine the comprehensive performance degradation parameters, and preliminarily determine the working status of the airborne optoelectronic platform vibration isolation and image stabilization system. Based on the operating status, a joint control and analysis strategy for vibration isolation and image stabilization is determined to identify the type of system anomaly or to implement protection, including: conventional monitoring mode, self-healing adjustment mode, and emergency protection mode. Based on the system anomaly type and the characteristic information of the corresponding spectral peaks, the operating parameters of the active vibration isolator and the image stabilization circuit are adjusted accordingly. The vibration acceleration signal is reacquired and the comprehensive performance degradation parameter is calculated. If the comprehensive performance degradation parameter returns to the preset healthy benchmark range for several consecutive times, the control adjustment is exited; otherwise, the control intensity is gradually increased.

2. The method for joint control of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification according to claim 1, characterized in that, The set of parameters for the temporal evolution of the spectral peak includes the spectral peak drift rate calculated based on the instantaneous kurtosis frequency drift curve, and the kurtosis variability, which is the statistical dispersion of the kurtosis of the spectral peak in each sliding window within a set monitoring period.

3. The method for joint control of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification according to claim 1, characterized in that, The online calibration of the currently identified spectral peak features includes: During the imaging idle period, a swept-frequency micro-perturbation signal is injected into the active vibration isolator, and the perturbation response spectrum is acquired. The position of the response spectrum peak is compared with the position of the currently identified spectrum peak, the frequency deviation and amplitude deviation are calculated, and the frequency center and amplitude of the spectrum peak are corrected.

4. The method for joint control of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification according to claim 1, characterized in that, The steps for constructing the system performance evolution trend curve include: The Weibull distribution function was used to fit each time series evolution characteristic parameter to obtain the single-parameter evolution trend curve; The evolution trend curves of each single parameter are weighted and fused according to the weight coefficients of each parameter's influence on system performance to construct the overall system performance evolution trend curve. The overall performance degradation parameter is determined based on the shape and scale parameters of the overall performance evolution trend curve.

5. The method for joint control of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification according to claim 4, characterized in that, A preliminary assessment of the operational status of the airborne optoelectronic platform vibration isolation and image stabilization system includes: If the overall performance degradation parameter is less than the preset stability threshold, the system is determined to be in a steady state. If the overall performance degradation parameter is between a preset stable threshold and a preset critical threshold, then the system is determined to be in the early stage of degradation. If the overall performance degradation parameter is greater than a preset critical threshold, the system is determined to be in a critical fluctuation period.

6. The method for joint control of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification according to claim 5, characterized in that, Based on the described operating state, the joint control analysis strategy for vibration isolation and image stabilization is determined, including: If the system is in a steady state, the conventional monitoring mode is adopted, keeping the parameters of the active vibration isolator and the image stabilization circuit at their default values, and only periodically monitoring the overall performance degradation parameters. If the system is in the early stage of degradation, the self-healing adjustment mode is adopted to shorten the sliding time window interval, determine the abnormality type by combining the spectral peak evolution characteristics, and perform online parameter compensation. If the system is in a critical fluctuation period, the emergency protection mode is adopted to further shorten the sliding time window interval, immediately execute the limit protection control and generate an early warning signal. If the overall performance degradation parameter does not return to the healthy benchmark after control, the control intensity is upgraded.

7. The method for joint control of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification according to claim 6, characterized in that, In response to the self-healing adjustment mode, the abnormality type is determined and online parameter compensation is performed, including: If the frequency drift rate shows a continuous unidirectional drift and the kurtosis variability exceeds the preset value, it is determined to be an abnormality such as aging of the vibration isolator performance or loosening of the structural connection. The stiffness compensation and damping compensation are calculated according to the drift direction and kurtosis variability, and the parameters of the active vibration isolator are adjusted online and the low cutoff frequency of the image stabilization circuit is matched synchronously. If the amplitude growth rate continues to rise and the frequency drift rate is within the preset range, it is determined to be a steady-state resonance anomaly. The parameters of the active vibration isolator notch filter are adjusted according to the center of the spectral peak frequency. If the overall performance degradation parameter triggers the threshold and both the frequency and amplitude are within the preset range, it is determined to be sensor drift or abnormal electrical noise interference. The data of the interfered channel is downweighted and a digital notch filter is added to the image stabilization circuit.

8. The method for joint control of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification according to claim 6, characterized in that, In response to the emergency protection mode, the limit protection control is executed, including: Switch the image stabilization system to high-bandwidth image lock mode, switch the active vibration isolator to maximum energy consumption mode, generate system performance warning signals, and record the spectral peak evolution curve and control adjustment process throughout the entire process; If the overall performance degradation parameters are continuously calculated after execution and still do not return to the preset healthy benchmark range, the active vibration isolator drive power supply will be cut off and the system will switch to passive vibration isolation mode.

9. The method for joint control of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification according to claim 6, characterized in that, In response to the conventional monitoring mode, if the growth rate of the spectral peak amplitude suddenly increases within a preset short-term window and the duration is less than a preset short-term threshold, it is determined to be a transient impact anomaly, and the bandwidth of the image stabilization loop is increased to a preset high bandwidth value. If the amplitude growth rate of the spectral peak continues to rise and the frequency is stable, it is determined to be a steady-state resonance anomaly. The center frequency and bandwidth of the notch filter of the active vibration isolator are adjusted according to the center of the spectral peak frequency.

10. The method for joint control of vibration isolation and image stabilization of airborne optoelectronic platforms based on spectral peak identification according to claim 9, characterized in that, The active vibration isolator is a magnetorheological fluid damping vibration isolator or a piezoelectric ceramic active vibration isolator; the vibration acceleration signal is collected by accelerometers arranged at the upper and lower ends of the active vibration isolator; the attitude error signal is collected by gyroscopes inside the image stabilization platform.

Images