Automatic approaching control method based on helicopter

Abstract

The invention discloses an automatic approach control method based on a helicopter. The method comprises the steps that course signals of an instrument landing system and a magnetic course sensor are acquired, high-frequency jitter components in the signals are extracted, smoothing processing is conducted on the signals through a low-pass filter, and high-frequency interference is removed; according to the interaction model of the runway terrain and the rotor wake flow, dynamic characteristics of eddy current disturbance are analyzed, a mathematical expression of an eddy current disturbance propagation rule is constructed, and the change trend of disturbance intensity and direction is determined; an eddy current angle disturbance observer is introduced into the course control law, the influence of eddy current disturbance on course control is estimated in real time, and if the disturbance intensity exceeds a preset range, a periodic variable pitch compensation mechanism is triggered; and adjusting the amplitude and frequency of periodic variable pitch compensation by adopting a self-adaptive control algorithm, and if the crosswind condition changes remarkably, dynamically updating compensation parameters to offset course deflection torque caused by a ground effect.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to an automatic approach control method based on a helicopter. Background Art

[0002] In helicopter automatic flight control systems, the smoothness of the heading channel is affected by multiple factors. First, the heading signal provided by the instrument landing system is subject to electromagnetic interference during reception, resulting in high-frequency jitter. Although coherent demodulation can extract useful information, it cannot completely eliminate the jitter. Furthermore, data from the magnetic heading sensor and fiber optic gyroscope are susceptible to electromagnetic interference at night, causing deviations in the heading angle estimate. To obtain a more accurate heading angle, it is necessary to construct an angle fusion Kalman filter to optimize the data from multiple sensors. However, in this implementation, effectively fusing the data from these different sensors to eliminate the heading angle jitter caused by electromagnetic interference at night remains a technical challenge.

[0003] On the other hand, the interaction between the helicopter's rotor wake and the runway terrain will produce vortex disturbances, which are particularly noticeable under crosswind conditions. The dynamic characteristics and propagation laws of vortex disturbances are complex, which will directly affect the stability of heading control. At present, the research on the vortex interaction model generated by the rotor wake and the runway terrain is not in-depth enough, and it is difficult to accurately predict the intensity and direction of vortex disturbances. Therefore, when introducing a vortex angle disturbance observer into the heading control law, how to estimate the vortex disturbance in real time and perform periodic pitch compensation to offset the heading deflection torque caused by changes in ground effect is also a technical problem that needs to be solved urgently. Summary of the Invention

[0004] This application provides a helicopter-based automatic approach control method, which can achieve more efficient and accurate detection of electricity meters in the power management area and save power resources.

[0005] This application provides a helicopter-based automatic approach control method, comprising:

[0006] S101: Acquire heading signals from the instrument landing system and magnetic heading sensor, extract high-frequency jitter components from the signals, and smooth the signals using a low-pass filter to remove high-frequency interference.

[0007] S102, time-aligning the processed heading signal with the fiber optic gyroscope data, constructing an angle fusion Kalman filter, and optimizing the fusion result of the sensor data by weight distribution to obtain a heading angle estimate;

[0008] S103: To address the heading angle jitter caused by electromagnetic interference at night, a dynamic threshold detection mechanism is introduced. If the heading angle change exceeds the preset threshold, the sensor data is recalibrated to eliminate the impact of outliers on the fusion results.

[0009] S104: Based on the interaction model between runway topography and rotor wake, analyze the dynamic characteristics of vortex disturbances, construct a mathematical expression for the propagation law of vortex disturbances, and determine the changing trends of disturbance intensity and direction;

[0010] S105, introducing an eddy current angle disturbance observer into the heading control law to estimate the impact of the eddy current disturbance on the heading control in real time. If the disturbance intensity exceeds a preset range, a periodic pitch compensation mechanism is triggered;

[0011] S106, using an adaptive control algorithm to adjust the amplitude and frequency of periodic pitch compensation. If the crosswind condition changes significantly, the compensation parameters are dynamically updated to offset the heading deflection moment caused by the ground effect.

[0012] S107, inputting the optimized heading angle estimate and the periodic compensation result into the flight control system to generate a heading control command. If the control command deviates too much from the current flight state, recalibrating the control parameters;

[0013] S108 , by real-time monitoring of the operating status of the heading control system, if abnormal jitter or deviation is detected, the data fusion and disturbance observer re-optimization process is triggered.

[0014] Preferably, the step S101 specifically includes:

[0015] Obtain heading signals based on the landing system and magnetic heading sensor;

[0016] Extract high-frequency jitter components from the heading signal and determine the characteristic frequency range of high-frequency jitter;

[0017] Use the preset low-pass filter parameters to configure the filter to filter the high-frequency jitter components;

[0018] According to the filtered signal, determine whether the high-frequency interference is completely removed. If not, adjust the filtering parameters;

[0019] The filtered signal is smoothed by a smoothing algorithm to obtain a smoothed signal with high-frequency interference removed;

[0020] According to the characteristics of the smoothed signal, determine whether the signal meets the preset smoothness threshold. If not, readjust the smoothing processing parameters;

[0021] The smooth heading signal with high-frequency interference removed is output, completing the signal processing process.

[0022] Preferably, the S102 specifically includes:

[0023] Obtain the sampling timestamps of the heading signal and the fiber optic gyro data, and use the interpolation algorithm to time align the two data sources to obtain synchronized time series data;

[0024] Construct an angle fusion Kalman filter, set the state transfer matrix and observation matrix, and initialize the filter parameters, including the process noise covariance and observation noise covariance;

[0025] Extract the heading angle information from the time-aligned data as the observation input of the Kalman filter, and simultaneously obtain the angular velocity information of the fiber optic gyroscope as the state prediction input;

[0026] The Kalman filter algorithm is used for iterative calculation to obtain the optimal estimate of the heading angle and update the filter state and error covariance matrix;

[0027] According to the sensor accuracy index, the weight distribution coefficient of the heading signal and the fiber optic gyroscope data is calculated to determine the data fusion ratio;

[0028] The optimized heading angle estimate is obtained by combining the Kalman filter output and the sensor raw data through a weighted fusion algorithm.

[0029] If there is a large deviation between the heading signal and the fiber optic gyroscope data, the weight distribution coefficient is adjusted and the fusion result is recalculated until the deviation is less than the preset threshold.

[0030] Preferably, the S103 specifically includes:

[0031] Obtain the heading angle data stream, calculate the heading angle jitter value, and obtain the change;

[0032] According to the dynamic threshold detection mechanism, determine whether the change exceeds the preset threshold;

[0033] If the change exceeds the preset threshold, the sensor data stream is obtained and the calibration value is recalculated; if it does not exceed the preset threshold, no subsequent operation is required;

[0034] Eliminate outliers and retain the calibrated sensor data stream;

[0035] Perform data fusion on the calibrated sensor data stream and the fusion result;

[0036] Analyze the impact of interference sources on heading angle data stream and extract interference features;

[0037] According to the interference characteristics, the dynamic threshold detection mechanism is updated and the preset threshold is updated.

[0038] Preferably, the S104 specifically includes:

[0039] For the eddy current model, the frequency characteristics of the disturbance wave are extracted through Fourier transform to determine the main propagation direction of the eddy current disturbance;

[0040] According to the propagation characteristics of the disturbance wave, the mathematical formula of eddy current disturbance propagation is constructed using partial differential equations to obtain the spatial distribution of disturbance intensity values.

[0041] Based on the mathematical calculation results, the changing trend of the intensity value is analyzed through gradient operation to determine the increase and decrease rules of the disturbance intensity at different locations;

[0042] According to the distribution characteristics of intensity values, the vector analysis method is used to calculate the variation characteristics of the direction angle and determine the variation law of the main direction of the eddy current disturbance.

[0043] Based on the characteristics of directional angle changes, a dynamic model of eddy current disturbance is constructed through statistical analysis to obtain the spatiotemporal evolution law of the disturbance field.

[0044] For the dynamic model, a machine learning algorithm is used to fit the relationship between the force and the change law to predict the evolution trend of the eddy disturbance.

[0045] Preferably, the S105 specifically includes:

[0046] Obtain real-time heading data from the heading control system and determine the heading deviation value based on the preset heading control model;

[0047] The eddy current angle disturbance observer is used to extract the eddy current disturbance angle from the heading deviation value to obtain the eddy current disturbance intensity value;

[0048] According to the preset disturbance intensity range, determine whether the eddy current disturbance intensity value exceeds the preset range;

[0049] If the eddy current disturbance intensity value exceeds the preset range, the execution parameters of the periodic variable pitch compensation mechanism are obtained; if it does not exceed the range, no subsequent operations are required;

[0050] Through the periodic pitch compensation mechanism, a pitch compensation signal is generated to adjust the dynamic response of the heading control system;

[0051] According to the pitch compensation signal, the output of the heading control model is updated to obtain the corrected heading data;

[0052] The corrected heading data is used to recalculate the heading deviation value to complete the closed-loop adjustment of the heading control system.

[0053] Preferably, the step S106 specifically includes:

[0054] Obtain real-time crosswind data from the aircraft to determine whether the crosswind condition exceeds a preset threshold;

[0055] If the crosswind condition exceeds the threshold, the adaptive control algorithm is used to calculate the amplitude and frequency of the periodic pitch compensation; if the crosswind condition does not exceed the preset threshold, no subsequent operation is required;

[0056] Update compensation parameters according to calculation results and generate new control instructions;

[0057] Obtain the aircraft's heading deflection data to determine whether there is a moment caused by ground effect;

[0058] If there is a ground effect torque, the adaptive control algorithm is used to adjust the compensation parameters; if there is no ground effect torque, no subsequent operation is required;

[0059] Applying the adjusted compensation parameters to the periodic variable pitch control system;

[0060] Compensation action is performed through the periodic pitch control system to offset the heading deflection moment.

[0061] Preferably, the step S107 is specifically:

[0062] Obtain the optimized heading angle estimate and periodic compensation results, and input them into the flight control system to generate heading control instructions;

[0063] Use a preset threshold to compare the deviation between the control command and the current flight status to determine whether the deviation exceeds the preset threshold;

[0064] If the deviation value exceeds the preset threshold, the control parameters are recalculated according to the deviation value and the control parameter table is updated;

[0065] The updated control parameters are used to regenerate heading control instructions and input them into the flight control system;

[0066] Obtain the deviation between the updated heading control command and the current flight status, and determine whether the deviation is within a preset threshold;

[0067] If the deviation value is within the preset threshold, the current heading control instruction is determined to be the final control instruction;

[0068] The final control command is used to execute flight control and complete the heading control process.

[0069] Preferably, the step S108 specifically includes:

[0070] Use sensors to obtain real-time operating data of the heading control system and analyze the heading control parameters in the data;

[0071] Determine whether the heading control parameters have abnormal jitter or deviation values ​​based on the preset threshold. If an abnormality is detected, mark the trigger bar;

[0072] For the marked trigger bar, the multi-source sensor data is extracted from the data fusion module and data fusion processing is performed;

[0073] The disturbance observation module analyzes the fused data to identify the source of the disturbance and its impact on heading control;

[0074] According to the disturbance analysis results, the optimization algorithm is used to re-optimize the parameters of the disturbance observation module;

[0075] Integrate the optimized disturbance observation module with the heading control system to generate new control instructions and adjust the channel shape;

[0076] The adjusted heading control system is continuously tracked through the real-time monitoring module.

[0077] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0078] By fusing and filtering data from the instrument landing system, magnetic heading sensor, and fiber optic gyroscope, a more accurate heading angle estimation is achieved. A dynamic threshold detection mechanism and eddy current disturbance observer are introduced to effectively identify and compensate for abnormal heading changes. An adaptive control algorithm is used to adjust periodic pitch compensation to offset heading deflection caused by ground effect. This invention significantly improves the accuracy and stability of heading control through technologies such as multi-sensor data fusion, dynamic threshold detection, eddy current disturbance modeling, and adaptive control. It is particularly suitable for flight control at night and in complex terrain conditions, providing a strong guarantee for improving flight safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 The figure is a flow chart of an automatic approach control method based on a helicopter according to an embodiment of the present invention. DETAILED DESCRIPTION

[0080] To facilitate understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0081] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains; the terms used herein in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more of the associated listed items.

[0082] Example 1: Figure 1The figure is a flowchart of an automatic approach control method based on a helicopter according to an embodiment of the present invention.

[0083] like Figure 1 As shown, a helicopter-based automatic approach control method includes the following steps:

[0084] S101, obtain the heading signals of the instrument landing system and the magnetic heading sensor, extract the high-frequency jitter components in the signals, and use a low-pass filter to smooth the signals to remove high-frequency interference.

[0085] High-frequency jitter refers to the higher-frequency, more rapidly varying portion of the heading signal. This is typically caused by factors such as magnetic heading sensor noise, electromagnetic interference, or mechanical vibration. These high-frequency components can obscure the true heading information, affecting signal accuracy and stability. Spectral analysis is required to determine the characteristic frequency range of high-frequency jitter. This can typically be done by observing the signal's spectrum to identify the frequency range of the high-frequency components. For example, a frequency threshold can be set to deem frequency components above this threshold as high-frequency jitter. The threshold for determining high-frequency jitter components can be set based on application requirements and signal characteristics.

[0086] Specifically, connect the data acquisition module to the landing system and magnetic heading sensor to obtain the heading signal from the landing system and magnetic heading sensor, ensuring stable signal transmission. Set the sampling frequency of the data acquisition device to continuously capture the heading signal and store it in a raw data format (.txt or .csv file).

[0087] Extract high-frequency jitter components from the heading signal and determine its characteristic frequency range. Use a fast Fourier transform algorithm to perform spectrum analysis on the continuously captured heading signal, converting the time-domain signal into a frequency-domain signal. Observe the spectrum to determine the characteristic frequency range of high-frequency jitter.

[0088] Using preset low-pass filter parameters, configure a filter to remove high-frequency jitter. Based on the characteristic frequency range of high-frequency jitter, configure the low-pass filter parameters (such as cutoff frequency, filter type, and order) to filter the original heading signal and remove high-frequency jitter. Perform another spectrum analysis on the filtered signal to determine whether the high-frequency interference has been completely removed. If not, adjust the filter parameters and re-filter.

[0089] The filtered signal is smoothed using a smoothing algorithm to produce a smoothed signal that removes high-frequency interference. Based on the characteristics of the smoothed signal, a determination is made as to whether the signal meets a preset smoothness threshold. If not, the smoothing parameters are readjusted. The final smoothed heading signal, free of high-frequency interference, is output, completing the signal processing process.

[0090] For example, when acquiring heading signals from the instrument landing system and magnetic heading sensor, the data acquisition module captures the raw signals at a sampling rate of 100 times per second. These signals often contain high-frequency jitter components. To extract these high-frequency components, a fast Fourier transform (FFT) algorithm is used to perform spectral analysis on the signals, identifying jitter components with frequencies above 20 Hz. A Butterworth low-pass filter is used to smooth these high-frequency jitters, with the filter cutoff frequency set to 15 Hz to ensure effective removal of high-frequency interference. During the filtering process, a fourth-order Butterworth filter is used, with design parameters including a passband ripple of 5 dB and a stopband attenuation of 40 dB to ensure a smooth and distortion-free signal. After filtering, the high-frequency components in the signal are effectively suppressed, retaining stable heading information. Finally, by comparing the signal spectra before and after filtering, the filtering effect is verified, ensuring that the processed signal meets the accuracy requirements of the subsequent navigation system.

[0091] Step S102 , time-aligning the processed heading signal with the fiber optic gyroscope data, constructing an angle fusion Kalman filter, and optimizing the fusion result of the heading signal sensor and the fiber optic gyroscope sensor data through weight distribution to obtain a heading angle estimation value.

[0092] Specifically, the sampling timestamps of the heading signal and the fiber optic gyro data are obtained, and the interpolation algorithm is used to time-align the two data sources to obtain synchronized time series data. The respective timestamp information is extracted from the heading signal data source and the fiber optic gyro data source. Since the two data sources may have inconsistent sampling frequencies or asynchronous sampling times, an interpolation algorithm is required to perform time alignment. Select a reference time series (such as the timestamp of the heading signal) and interpolate the fiber optic gyro data to align it with the timestamp of the heading signal. The interpolation method can be linear interpolation, spline interpolation, etc., and the interpolation method is selected according to the data characteristics and requirements. After interpolation processing, synchronized time series data is obtained, that is, the heading signal and the fiber optic gyro data are synchronized in time. Assume that the timestamp of the heading signal is t h and the timestamp of the fiber optic gyroscope is t g , it is necessary to interpolate the fiber optic gyro data to the timestamp of the heading signal:

[0093]

[0094] x1 and x2 are adjacent sampling time points of the FOG, y1 and y2 are corresponding sampling values, x is the sampling time point of the heading signal, and y is the interpolated FOG data.

[0095] Based on the relationship between heading angle and angular velocity, an angle fusion Kalman filter is constructed. The Kalman filter's state transition matrix (which describes how the state of the vehicle's heading angle and angular velocity changes over time) and observation matrix (which describes how to obtain observations from the state of the vehicle's heading angle and angular velocity over time) are set. The Kalman filter parameters (process noise covariance and observation noise covariance) are initialized. The heading angle information is extracted from the time-aligned data as the observation input of the Kalman filter. The angular velocity information of the fiber optic gyroscope is obtained from the time-aligned data as the state prediction input.

[0096] The Kalman filter algorithm is used for iterative calculations to obtain the optimal estimate of the heading angle and update the filter state and error covariance matrix. The state transition matrix and the state estimate from the previous time series are used to predict the current state. The error covariance matrix is ​​updated based on the process noise covariance matrix and the state transition matrix. The Kalman gain is calculated based on the observation matrix, the error covariance matrix, and the observation noise covariance matrix. The Kalman gain determines the weight of the observation in the state update. The current state estimate is updated using the observation, the Kalman gain, and the predicted state. The error covariance matrix is ​​updated based on the Kalman gain and the observation noise covariance matrix to prepare for filtering at the next time. For the detailed prediction calculation process, please refer to the Kalman filter steps and will not be detailed in this article.

[0097] Based on the accuracy of the heading signal sensor and the fiber optic gyro sensor, the weight distribution coefficient of the heading signal and fiber optic gyro data is calculated to determine the data fusion ratio. The weight distribution coefficient reflects the importance of each sensor data in the fusion process.

[0098] The optimized heading angle estimate is obtained by combining the Kalman filter output and the raw sensor data through a weighted fusion algorithm. The fusion algorithm can consider factors such as the accuracy and stability of each sensor and reasonably assign weights.

[0099] If there is a large deviation between the heading signal and the fiber optic gyroscope data (the deviation threshold can be set according to the specific application and sensor accuracy. For example, it can be set to ±5° or ±10°, depending on the accuracy requirements of the system and the performance of the sensor), it indicates that the current weight distribution may be unreasonable. At this time, it is necessary to adjust the weight distribution coefficient and recalculate the fusion result. The weight distribution coefficient is adjusted by continuous iteration (the iteration number threshold can be set to a reasonable upper limit, such as 10 times or 20 times, to avoid infinite iteration. If the number of iterations exceeds this threshold and the deviation is still large, it may be necessary to check the sensor data or adjust the weight distribution strategy) until the deviation is less than the preset threshold and a satisfactory fusion result is obtained.

[0100] For example, when processing heading signals and fiber optic gyroscope data, the two must first be time-aligned. Assuming the sampling frequency of the heading signal is 100 Hz and the sampling frequency of the fiber optic gyroscope is 50 Hz, the interpolation algorithm can be used to increase the data frequency of the fiber optic gyroscope to 100 Hz to ensure that the two are completely aligned on the time axis. Next, an angle fusion Kalman filter is constructed, assuming that the state vector is the heading angle θ, the state transfer matrix is ​​F = [1], the observation matrix is ​​H = [1], the process noise covariance matrix is ​​Q =

[01] , and the observation noise covariance matrix is ​​R =

[02] . Through the prediction and update steps of the Kalman filter, the predicted state θ_k|k-1=F*θ_k-1|k-1 and the predicted covariance P_k|k-1=F*P_k-1|k-1*F^T+Q are calculated respectively, and then the Kalman gain K_k=P_k|k-1*H^T*(H*P_k|k-1*H^T+R)^(-1) is calculated, and finally the state θ_k|k=θ_k|k-1+K_k*(z_k-H*θ_k|k-1) and covariance P_k|k=(I-K_k*H)*P_k|k-1 are updated. In order to optimize the fusion result of sensor data, a weight distribution strategy is adopted. The weight of the heading signal sensor is set to 6, and the weight of the fiber optic gyroscope sensor is set to 4. The fused heading angle estimate θ_fused = 6*θ_gps + 4*θ_fog is calculated by the weighted average method, where θ_gps is the angle value of the heading signal and θ_fog is the angle value of the fiber optic gyroscope.

[0101] S103: To address the heading angle jitter caused by electromagnetic interference at night, a dynamic threshold detection mechanism is introduced. If the heading angle change exceeds the preset threshold, the sensor data is recalibrated to eliminate the impact of outliers on the fusion results.

[0102] Specifically, the heading angle data stream is acquired, the heading angle jitter value is calculated, and the variation is obtained. Heading angle data (each data point contains a timestamp and corresponding heading angle value) is acquired from the heading sensor in real time. The jitter value is quantified as the variation by calculating the heading angle difference between adjacent data points. The jitter value refers to the degree of change in the heading angle within a time series.

[0103] Using the dynamic threshold detection mechanism, the change is determined to see if it exceeds a preset threshold. A preset threshold is set to determine whether the heading angle change is normal. The threshold can be dynamically adjusted based on historical data, sensor characteristics, or application requirements. The calculated change is compared with the preset threshold. If the change exceeds the threshold, the heading angle data is considered to be abnormal or interfered with.

[0104] If the change exceeds a preset threshold, the sensor data stream is acquired and the calibration value is recalculated. When the change exceeds the preset threshold, the complete data stream from the relevant sensor (heading signal sensor or fiber optic gyro sensor) is acquired. Using this acquired data, the sensor calibration value is recalculated. If the change does not exceed the preset threshold, the current heading angle data can be assumed to be normal, without significant anomalies or interference. There is no need to recalculate the calibration value because the current data is already considered reliable.

[0105] Eliminate outliers and retain the calibrated sensor data stream. Based on the calibration results, identify and eliminate outliers in the sensor data stream. Outliers may be caused by sensor failure, environmental interference, or data transmission errors. Save the calibrated sensor data stream for subsequent data fusion and analysis.

[0106] The calibrated sensor data stream is then fused with the fusion result. The calibrated sensor data stream is then fused with the previous fusion result (Kalman filter output). The fusion process can use weighted averaging, Kalman filtering, or other data fusion algorithms. Through data fusion, a more accurate and stable heading angle estimate is obtained.

[0107] Analyze the impact of interference sources on the heading angle data stream to extract interference features. Based on these features, update and optimize the dynamic threshold detection mechanism and update the preset threshold. Analyze interference sources that cause abnormal heading angle data, such as magnetic field interference and mechanical vibration, and extract features related to the interference source from the sensor data stream, such as the frequency, amplitude, and duration of the interference. Based on the extracted interference features, adjust the parameters or strategy of the dynamic threshold detection mechanism. For example, if the interference primarily manifests as high-frequency noise, increase filtering of the high-frequency component or reduce the sensitivity of the threshold. Update the preset threshold based on the optimized dynamic threshold detection mechanism. Ensure that the new threshold is adapted to the current environmental conditions and sensor characteristics, improving the accuracy and stability of heading angle estimation.

[0108] For example, to address the issue of heading angle jitter caused by electromagnetic interference at night, a magnetic heading sensor first collects heading angle data at a sampling frequency of 100 Hz to ensure real-time data. A dynamic threshold detection mechanism is then designed with an initial threshold of 5 degrees. Based on historical data statistics, the standard deviation of the heading angle is calculated to be 2 degrees, and the dynamic threshold is set to 5 times the standard deviation, or 5 degrees. When the heading angle changes by more than 5 degrees, the system automatically triggers a recalibration process. During the calibration process, a Kalman filter algorithm is used to optimize the magnetic heading sensor data through two steps: state prediction and measurement update. In the state prediction phase, the current state is predicted using the state estimate from the previous time series and the system model. In the measurement update phase, the current measured value and the predicted value are combined and weighted using the Kalman gain to obtain the optimal estimate.

[0109] S104: Analyze the dynamic characteristics of the vortex disturbance based on the interaction model between the runway terrain and the rotor wake, obtain the propagation law of the vortex disturbance, and determine the changing trend of the disturbance intensity and direction.

[0110] Specifically, based on the terrain model and wake field data, the dynamic model of the vortex state is constructed using the fluid mechanics equation to obtain the initial distribution characteristics of the vortex field. . Collect terrain model data, including terrain height h(x,y), slope θ, roughness R q etc., as well as wake field data (such as velocity v(x,y,z,t), flow direction φ, vorticity ω(x,y,z,t)), etc.), pre-process the data, including denoising, interpolation, normalization, etc., to ensure the accuracy and consistency of the data. According to the characteristics of the eddy field, the fluid mechanics equation is selected, and the influence of the terrain on the flow is considered (it may be necessary to add terrain-related source terms or boundary conditions to the equation). The fluid mechanics equation is discretized using numerical methods (such as finite difference method, finite element method, spectral method, etc.), and the initial and boundary conditions are set by combining the terrain model and wake field data. Through numerical solution, the initial distribution characteristics of the eddy field are obtained, including the position, size, intensity, etc. of the vortex.

[0111] For the eddy state model, the frequency characteristics of the disturbance wave are extracted through Fourier transform to determine the main propagation direction of the eddy disturbance. According to the propagation direction characteristics of the disturbance wave, the partial differential equation is used to construct the mathematical formula for the propagation of the eddy disturbance, and the spatial distribution of the disturbance intensity value is obtained. Based on the calculation results of the mathematical formula, the changing trend of the intensity value is analyzed through gradient operation to determine the increase and decrease law of the disturbance intensity at different locations. According to the distribution characteristics of the intensity value, the vector analysis method is used to calculate the changing characteristics of the direction angle to determine the main direction change law of the eddy disturbance. Based on the changing characteristics of the direction angle, the dynamic model of the eddy disturbance is constructed through statistical analysis to obtain the spatiotemporal evolution law of the disturbance field. For the dynamic model, the machine learning algorithm is used to fit the relationship between the force and the change law to predict the long-term evolution trend of the eddy disturbance.

[0112] For example, in the interaction model between runway terrain and rotor wake, the dynamic characteristics of vortex disturbances are first analyzed by numerical simulation methods. Computational fluid dynamics (CFD) technology is used, the rotor speed is set to 1200 rpm, the wake velocity is set to 15 m / s, and the Navier-Stokes equation is used for solution. Through discretization processing, the calculation domain is divided into 1 million grid cells to ensure calculation accuracy. In the simulation process, turbulence models (such as the k-ε model) are introduced to capture the turbulent characteristics in the wake, and the boundary conditions are set to no-slip wall conditions to simulate the actual influence of runway terrain. Through iterative calculation, the velocity field and pressure field distribution of the rotor wake are obtained, and the generation and evolution process of the vortex is further analyzed. When constructing the mathematical expression of the propagation law of vortex disturbances, the vortex transport equation is used, combined with the continuity equation and momentum equation, to derive the relationship between vortex intensity and propagation distance. Assuming an initial vortex intensity of 5 and a propagation distance of 10 meters, the equation was solved using numerical integration methods to determine the attenuation pattern of vortex intensity with distance. It was found that the vortex intensity decays exponentially during propagation, with an attenuation coefficient of 1. To determine the changing trends in the intensity and direction of the disturbance, vector field analysis was used to calculate the gradient of the vortex velocity field and determine the spatial distribution of the disturbance intensity. Statistical analysis revealed that the disturbance intensity is greatest below the rotor and gradually decreases with increasing distance. The disturbance direction is primarily along the rotor's rotation direction, with a deflection angle of 5 degrees. These results provide a scientific basis for optimizing rotor design and runway layout.

[0113] S105, introducing an eddy current angle disturbance observer into the heading control law to estimate the impact of the eddy current disturbance on the heading control in real time. If the disturbance intensity exceeds a preset range, the periodic pitch compensation mechanism is triggered.

[0114] Specifically, real-time navigation data from the heading control system is acquired. A heading sensor is used to collect heading data from the heading control system, ensuring that the frequency and accuracy of data acquisition meet the requirements of the heading control system. The collected heading data is preprocessed through noise removal and filtering to improve its accuracy and reliability. The preprocessed heading data is then converted into a format usable by the heading control system.

[0115] Determine the heading deviation value by combining it with a preset heading control model. Based on the characteristics and requirements of the heading control system, a heading control model is preset. The heading control model should be able to describe the desired heading trajectory or target heading of the heading control system. Real-time heading data is compared with the heading control model to calculate the heading deviation value (the difference between the actual heading and the desired heading).

[0116] A vortex angle disturbance observer is used to extract the vortex disturbance angle from the heading deviation value. A vortex angle disturbance observer is designed to extract the vortex disturbance angle from the heading deviation value. The vortex angle disturbance observer should be able to accurately identify and isolate the impact of vortex disturbances on the heading. The heading deviation value is input into the vortex angle disturbance observer to extract the vortex disturbance angle. The vortex disturbance angle indicates the direction of the vortex disturbance on the heading control system.

[0117] The eddy current disturbance intensity value is obtained. According to the changes in the eddy current disturbance angle and the heading deviation value, the eddy current disturbance intensity value is calculated. The eddy current disturbance intensity value represents the magnitude of the disturbance caused by the eddy current to the heading control system.

[0118] Based on the preset disturbance intensity range, determine whether the eddy current disturbance intensity value exceeds the preset range. Based on the stability and performance requirements of the heading control system, a preset eddy current disturbance intensity range is set to ensure that the heading control system remains stable despite eddy current disturbances. The calculated eddy current disturbance intensity value is compared with the preset range to determine whether it exceeds the preset range. If the eddy current disturbance intensity value exceeds the preset range, compensatory adjustments are required; if it does not exceed the preset range, no subsequent operations are required.

[0119] Obtain the execution parameters of the periodic pitch compensation mechanism. Design a periodic pitch compensation mechanism to adjust the dynamic response of the heading control system. The pitch compensation mechanism should be able to dynamically adjust the control parameters of the heading control system based on changes in the eddy current disturbance intensity. Based on the eddy current disturbance intensity and the characteristics of the heading control system, obtain the execution parameters of the periodic pitch compensation mechanism. The execution parameters include the pitch compensation amplitude, frequency, and phase.

[0120] A pitch compensation signal is generated through a periodic pitch compensation mechanism. Based on the execution parameters, a pitch compensation signal is generated through the periodic pitch compensation mechanism. The pitch compensation signal should be able to offset the impact of eddy current disturbances on the heading control system, so that the heading control system remains stable.

[0121] Adjust the dynamic response of the heading control system. Input the pitch compensation signal into the heading control system to adjust its dynamic response. By adjusting the pitch compensation signal, the heading control system can quickly respond to eddy current disturbances and maintain heading stability.

[0122] Update the output of the heading control model based on the pitch compensation signal. Based on the pitch compensation signal and the actual response of the heading control system, update the output of the heading control model. The updated heading control model output should be able to more accurately reflect the actual heading of the heading control system.

[0123] The corrected heading data is used to recalculate the heading deviation value, completing the closed-loop adjustment of the heading control system. The corrected heading data is used to recalculate the heading deviation value, which should more accurately reflect the difference between the heading control system and the desired heading. This recalculated heading deviation value is input into the heading control system for the next round of adjustment and control. Through continuous closed-loop adjustment and control, the heading control system can always maintain a stable heading.

[0124] For example, the heading control system obtains real-time heading data through sensors such as gyroscopes and magnetic compasses. The system sets a standard heading of 30 degrees east of north. When the measured heading is 35 degrees east of north, the heading deviation is 5 degrees. The vortex angle disturbance observer, designed based on the Kalman filter algorithm, can separate the angular changes caused by vortex disturbances from the heading deviation. The observer compares the normal heading change rate with the actual heading change rate to extract the disturbance angle caused by vortexes. For example, when the instantaneous heading change rate exceeds 2 degrees per second, it is determined that vortex disturbances are present. The vortex disturbance intensity can be expressed as a dimensionless parameter, typically ranging from 0 to 10, with values ​​below 5 representing the normal disturbance range. When the disturbance intensity calculated by the observer is 7.5, it exceeds the preset safety range and requires the activation of a compensation mechanism. The core of the periodic pitch compensation mechanism is to offset the influence of vortexes by adjusting the rotor blade angle. The execution parameters include pitch amplitude and phase angle. When the vortex disturbance intensity reaches 7.5, the system automatically sets the pitch amplitude to 4 degrees and the phase angle to 90 degrees, generating a corresponding compensation signal. This pitch compensation signal acts on the heading control system's actuator, causing the rotor blades to produce periodic pitch changes, thereby offsetting the vortex disturbance. The compensated heading is gradually corrected from the original 35 degrees east of north to the desired 30 degrees east of north. The heading control model is based on the proportional-integral-derivative control principle, optimizing system response by adjusting control parameters. In practical applications, a proportional coefficient of 0.8, an integral time constant of 2 seconds, and a derivative time constant of 0.5 seconds achieve good control results. The system continuously acquires heading data at a 20 Hz sampling rate, recalculating the heading deviation after each sampling cycle to achieve closed-loop control. This method enables the heading control system to respond to vortex disturbances in real time and maintain a stable heading. As the vortex disturbance weakens, the system reduces the compensation amount to avoid overcompensation. Practice has demonstrated that this control strategy can keep heading deviation within a range of plus or minus 2 degrees, effectively improving flight quality. By accumulating and analyzing heading data, a characteristic model of eddy current disturbances can be established, providing a basis for further optimizing control strategies. An eddy current angle disturbance observer is introduced into the heading control law. First, a Kalman filter algorithm is used to estimate the impact of eddy current disturbances on heading control in real time. For example, at a flight speed of 150 meters per second, the observer can update the eddy current angle disturbance value at a sampling frequency of 0.1 seconds. In the specific algorithm, the state equation and observation equation are based on the aircraft's dynamic model and sensor data, respectively, and the state estimate is updated by minimizing the prediction error. When the disturbance intensity exceeds a preset range (e.g., ±5 degrees), the system automatically triggers a periodic pitch compensation mechanism. This compensation mechanism uses a PID control algorithm with a proportional coefficient of 8, an integral coefficient of 2, and a differential coefficient of 1, adjusting the rudder deflection angle with a period of 0.5 seconds. During the analysis process, the system compares the real-time disturbance value with historical data to determine whether the disturbance is a normal fluctuation. If the disturbance persists beyond the preset range, the compensation parameters are further adjusted.Throughout the entire process, the system ensures the stability and accuracy of heading control through real-time data stream processing and adaptive algorithm optimization. For example, when the disturbance intensity is 8 degrees, the compensation mechanism can control the heading deviation within ±5 degrees within 1 second.

[0125] S106, using an adaptive control algorithm to adjust the amplitude and frequency of the periodic pitch compensation. If the crosswind condition changes significantly, the compensation parameters are dynamically updated to offset the heading deflection torque caused by the ground effect.

[0126] Specifically, real-time crosswind data from the aircraft is acquired. Using an aircraft-mounted anemometer and wind direction sensor, wind speed and direction data at the aircraft's current location are collected in real time. This ensures frequent sensor data updates to accurately capture rapid changes in crosswinds. The raw sensor data is smoothed through a filter to remove noise and outliers. The crosswind component, defined as the wind speed and direction perpendicular to the aircraft's flight direction, is then calculated.

[0127] Determine whether crosswind conditions exceed preset thresholds. Based on the aircraft's design specifications and performance limitations, set thresholds for crosswind speed and direction. Real-time crosswind data is compared with these thresholds. If the crosswind speed or direction exceeds the threshold, the adaptive control algorithm is triggered. If the crosswind condition does not exceed the threshold, the impact of the current crosswind on the aircraft is considered acceptable, and no further action is required.

[0128] An adaptive control algorithm is used to calculate the amplitude and frequency of periodic pitch compensation. An adaptive control algorithm suitable for the aircraft's dynamic characteristics, such as Model Reference Adaptive Control (MRAC) or Self-Tune Control (STC), is selected. Based on the magnitude and direction of deviations in the crosswind data, the algorithm dynamically adjusts the amplitude (i.e., the amount of change in blade angle) and frequency (i.e., the period of change). The algorithm also considers the aircraft's current speed, altitude, attitude, and other state information to ensure the accuracy and effectiveness of the compensation action.

[0129] Based on the calculation results, the compensation parameters are updated and new control instructions are generated. The pitch change amplitude and frequency calculated by the algorithm are used as new compensation parameters and written into the aircraft's control system, replacing the original compensation parameters. Based on the updated compensation parameters, new control instructions are generated to adjust the aircraft's blade angle or engine thrust.

[0130] Acquire the aircraft's heading deflection data. Using sensors such as gyroscopes and magnetic compasses, the aircraft's heading angle and heading deflection rate are collected in real time. The raw data is filtered and calibrated to ensure accuracy and reliability.

[0131] Determine whether there is a moment caused by ground effect. Analyze heading deflection data to identify abnormal heading deflections caused by ground effect (such as ground reflected airflow, ground friction, etc.). Set the identification threshold of the ground effect moment to determine whether there is a significant ground effect.

[0132] An adaptive control algorithm is used to adjust compensation parameters. If ground effect torque is detected, the adaptive control algorithm is reactivated. The algorithm dynamically adjusts the compensation parameters for periodic pitch variation based on the magnitude and direction of the ground effect torque. If no ground effect torque is detected, no further action is required.

[0133] Apply the adjusted compensation parameters to the periodic pitch control system. Write the adjusted compensation parameters into the aircraft's periodic pitch control system to ensure the control system can respond to changes in these parameters in real time and accurately perform compensation actions.

[0134] Compensation is performed through the cyclical pitch control system to offset the heading torque. Based on updated compensation parameters, the cyclical pitch control system adjusts the aircraft's blade angle or engine thrust. This periodic pitch action generates a compensating torque opposite to the ground effect torque, thereby offsetting heading deviation. The aircraft's heading deviation is monitored in real time, and compensation parameters are further adjusted as needed to ensure the aircraft maintains a stable flight attitude and heading in complex environmental conditions.

[0135] S107: Input the optimized heading angle estimation value and the periodic compensation result into the flight control system to generate a heading control instruction. If the control instruction deviates too much from the current flight state, recalibrate the control parameters.

[0136] Specifically, the optimized heading angle estimate and periodic compensation results are obtained and input into the flight control system to generate a heading control instruction. The deviation value between the control instruction and the current flight state is compared using a preset threshold to determine whether the deviation value exceeds the preset threshold. If the deviation value exceeds the preset threshold, the control parameters are recalculated based on the deviation value, and the control parameter table is updated. The heading control instruction is regenerated using the updated control parameters and input into the flight control system. The deviation value between the updated heading control instruction and the current flight state is obtained to determine whether the deviation value is within the preset threshold. If the deviation value is within the preset threshold, the current heading control instruction is determined to be the final control instruction. The final control instruction is used to execute flight control, completing the heading control process.

[0137] For example, the optimized heading angle estimate is further processed using a Kalman filter. Assuming the current heading angle estimate is 125 degrees, the filtered result is 123 degrees, the filter gain is 95, and the state covariance matrix is ​​[1, 0; 0, 1]. Subsequently, the periodic compensation results are analyzed using a Fourier transform algorithm to extract a compensation signal with a period of 10 seconds, an amplitude of 2 degrees, and a phase of 30 degrees. The processed heading angle estimate and compensation signal are input into the flight control system. The system generates a heading control command based on a preset PID control algorithm with a proportional coefficient of 2, an integral time of 5 seconds, and a derivative time of 5 seconds. The resulting heading control command is 120 degrees. The system monitors the current flight state in real time. Assuming the actual heading angle is 128 degrees, the deviation between the calculated control command and the current flight state is 2 degrees. If the deviation exceeds a preset threshold of 5 degrees, a control parameter recalibration process is triggered. The PID parameters are optimized using the least squares method, resulting in a proportional coefficient of 15, an integral time of 2 seconds, and a differential time of 48 seconds. This ensures that the deviation between control commands and flight status remains within the permitted range. The entire process is automated by the embedded system, ensuring accurate and stable flight control.

[0138] S108, by real-time monitoring of the operating status of the heading control system, if abnormal jitter or deviation is detected, the data fusion and disturbance observer re-optimization process is triggered to ensure the smoothness and stability of the heading channel.

[0139] Specifically, sensors are used to obtain real-time operating data of the heading control system, and the heading control parameters in the data are analyzed. The preset threshold is used to determine whether the heading control parameters have abnormal jitter or deviation values. If an abnormality is detected, the trigger bar is marked. For the marked trigger bar, multi-source sensor data is extracted from the data fusion module and data fusion processing is performed. The fused data is analyzed by the disturbance observation module to identify the source of the disturbance and its impact on the heading control. Based on the disturbance analysis results, the optimization algorithm is used to re-optimize the parameters of the disturbance observation module. The optimized disturbance observation module is integrated with the heading control system to generate new control instructions and adjust the channel state. The adjusted heading control system is continuously tracked through the real-time monitoring module to confirm whether the smoothness and stability meet expectations.

[0140] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0141] By fusing and filtering data from the instrument landing system, magnetic heading sensor, and fiber optic gyroscope, a more accurate heading angle estimation is achieved. A dynamic threshold detection mechanism and eddy current disturbance observer are introduced to effectively identify and compensate for abnormal heading changes. An adaptive control algorithm is used to adjust periodic pitch compensation to offset heading deflection caused by ground effect. This invention significantly improves the accuracy and stability of heading control through technologies such as multi-sensor data fusion, dynamic threshold detection, eddy current disturbance modeling, and adaptive control. It is particularly suitable for flight control at night and in complex terrain conditions, providing a strong guarantee for improving flight safety.

[0142] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Various modifications and variations are readily apparent to those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A helicopter-based automatic approach control method, characterized in that: include: S101: Acquire heading signals from the instrument landing system and magnetic heading sensor, extract high-frequency jitter components from the signals, and smooth the signals using a low-pass filter to remove high-frequency interference. S102, time-aligning the processed heading signal with the fiber optic gyroscope data, constructing an angle fusion Kalman filter, and optimizing the fusion result of the sensor data by weight distribution to obtain a heading angle estimate; S103: To address the heading angle jitter caused by electromagnetic interference at night, a dynamic threshold detection mechanism is introduced. If the heading angle change exceeds the preset threshold, the sensor data is recalibrated to eliminate the impact of outliers on the fusion results. S104: Based on the interaction model between runway topography and rotor wake, analyze the dynamic characteristics of vortex disturbances, construct a mathematical expression for the propagation law of vortex disturbances, and determine the changing trends of disturbance intensity and direction; S105, introducing an eddy current angle disturbance observer into the heading control law to estimate the impact of the eddy current disturbance on the heading control in real time. If the disturbance intensity exceeds a preset range, a periodic pitch compensation mechanism is triggered; S106, using an adaptive control algorithm to adjust the amplitude and frequency of periodic pitch compensation. If the crosswind condition changes significantly, the compensation parameters are dynamically updated to offset the heading deflection moment caused by the ground effect. S107, inputting the optimized heading angle estimate and the periodic compensation result into the flight control system to generate a heading control command. If the control command deviates too much from the current flight state, recalibrating the control parameters; S108 , by real-time monitoring of the operating status of the heading control system, if abnormal jitter or deviation is detected, the data fusion and disturbance observer re-optimization process is triggered.

2. The automatic approach control method based on a helicopter according to claim 1, characterized in that: The S101 specifically includes: Obtain heading signals based on the landing system and magnetic heading sensor; Extract high-frequency jitter components from the heading signal and determine the characteristic frequency range of high-frequency jitter; Use the preset low-pass filter parameters to configure the filter to filter the high-frequency jitter components; According to the filtered signal, determine whether the high-frequency interference is completely removed. If not, adjust the filtering parameters; The filtered signal is smoothed by a smoothing algorithm to obtain a smoothed signal with high-frequency interference removed; According to the characteristics of the smoothed signal, determine whether the signal meets the preset smoothness threshold. If not, readjust the smoothing processing parameters; The smooth heading signal with high-frequency interference removed is output, completing the signal processing process.

3. The helicopter-based automatic approach control method according to claim 1, wherein: The S102 specifically includes: Obtain the sampling timestamps of the heading signal and the fiber optic gyro data, and use the interpolation algorithm to time align the two data sources to obtain synchronized time series data; Construct an angle fusion Kalman filter, set the state transfer matrix and observation matrix, and initialize the filter parameters, including the process noise covariance and observation noise covariance; Extract the heading angle information from the time-aligned data as the observation input of the Kalman filter, and simultaneously obtain the angular velocity information of the fiber optic gyroscope as the state prediction input; The Kalman filter algorithm is used for iterative calculation to obtain the optimal estimate of the heading angle and update the filter state and error covariance matrix; Based on the sensor accuracy index, the weight distribution coefficient of the heading signal and the fiber optic gyroscope data is calculated to determine the data fusion ratio; through the weighted fusion algorithm, the Kalman filter output and the sensor raw data are combined to obtain the optimized heading angle estimate; If there is a large deviation between the heading signal and the fiber optic gyroscope data, the weight distribution coefficient is adjusted and the fusion result is recalculated until the deviation is less than the preset threshold.

4. The helicopter-based automatic approach control method according to claim 1, wherein: The S103 specifically includes: Obtain the heading angle data stream, calculate the heading angle jitter value, and obtain the change; According to the dynamic threshold detection mechanism, determine whether the change exceeds the preset threshold; If the change exceeds the preset threshold, the sensor data stream is obtained and the calibration value is recalculated; if it does not exceed the preset threshold, no subsequent operation is required; Eliminate outliers and retain the calibrated sensor data stream; Perform data fusion on the calibrated sensor data stream and the fusion result; Analyze the impact of interference sources on heading angle data stream and extract interference features; According to the interference characteristics, the dynamic threshold detection mechanism is updated and the preset threshold is updated.

5. The helicopter-based automatic approach control method according to claim 1, wherein: The S104 specifically includes: For the eddy state model, the frequency characteristics of the disturbance wave are extracted through Fourier transform to determine the main propagation direction of the eddy disturbance. Based on the propagation direction characteristics of the disturbance wave, the mathematical formula of the eddy disturbance propagation is constructed using partial differential equations to obtain the spatial distribution of the disturbance intensity value. Based on the mathematical calculation results, the changing trend of the intensity value is analyzed through gradient operation to determine the increase and decrease rules of the disturbance intensity at different locations; According to the distribution characteristics of intensity values, the vector analysis method is used to calculate the variation characteristics of the direction angle and determine the variation law of the main direction of the eddy current disturbance. Based on the characteristics of directional angle changes, a dynamic model of eddy current disturbance is constructed through statistical analysis to obtain the spatiotemporal evolution law of the disturbance field. For the dynamic model, a machine learning algorithm is used to fit the relationship between the force and the change law to predict the evolution trend of the eddy disturbance.

6. The helicopter-based automatic approach control method according to claim 1, characterized in that: The S105 specifically includes: Obtain real-time heading data from the heading control system and determine the heading deviation value based on the preset heading control model; The eddy current angle disturbance observer is used to extract the eddy current disturbance angle from the heading deviation value to obtain the eddy current disturbance intensity value; According to the preset disturbance intensity range, determine whether the eddy current disturbance intensity value exceeds the preset range; If the eddy current disturbance intensity value exceeds the preset range, the execution parameters of the periodic variable pitch compensation mechanism are obtained; if it does not exceed the range, no subsequent operations are required; Through the periodic pitch compensation mechanism, a pitch compensation signal is generated to adjust the dynamic response of the heading control system; According to the pitch compensation signal, the output of the heading control model is updated to obtain the corrected heading data; The corrected heading data is used to recalculate the heading deviation value to complete the closed-loop adjustment of the heading control system.

7. The helicopter-based automatic approach control method according to claim 1, wherein: The S106 specifically includes: Obtain real-time crosswind data from the aircraft to determine whether the crosswind condition exceeds a preset threshold; If the crosswind condition exceeds the threshold, the adaptive control algorithm is used to calculate the amplitude and frequency of the periodic pitch compensation; if the crosswind condition does not exceed the preset threshold, no subsequent operation is required; Update compensation parameters according to calculation results and generate new control instructions; Obtain the aircraft's heading deflection data to determine whether there is a moment caused by ground effect; If there is a ground effect torque, the adaptive control algorithm is used to adjust the compensation parameters; if there is no ground effect torque, no subsequent operation is required; Applying the adjusted compensation parameters to the periodic variable pitch control system; Compensation action is performed through the periodic pitch control system to offset the heading deflection moment.

8. The helicopter-based automatic approach control method according to claim 1, characterized in that: The step S107 specifically comprises: obtaining the optimized heading angle estimation value and the periodic compensation result, and inputting the optimized heading angle estimation value and the periodic compensation result into the flight control system to generate a heading control instruction; Use a preset threshold to compare the deviation between the control command and the current flight status to determine whether the deviation exceeds the preset threshold; If the deviation value exceeds the preset threshold, the control parameters are recalculated according to the deviation value and the control parameter table is updated; The updated control parameters are used to regenerate heading control instructions and input them into the flight control system; Obtain the deviation between the updated heading control command and the current flight status, and determine whether the deviation is within a preset threshold; If the deviation value is within the preset threshold, the current heading control instruction is determined to be the final control instruction; The final control command is used to execute flight control and complete the heading control process.

9. The helicopter-based automatic approach control method according to claim 1, wherein: In the S108, the following steps are specifically included: Use sensors to obtain real-time operating data of the heading control system and analyze the heading control parameters in the data; Based on the preset threshold, it is determined whether the heading control parameters have abnormal jitter or deviation. If an abnormality is detected, the trigger bar is marked. For the marked trigger bar, multi-source sensor data is extracted from the data fusion module and data fusion processing is performed. The disturbance observation module analyzes the fused data to identify the source of the disturbance and its impact on heading control; According to the disturbance analysis results, the optimization algorithm is used to re-optimize the parameters of the disturbance observation module; Integrate the optimized disturbance observation module with the heading control system to generate new control instructions and adjust the channel shape; The adjusted heading control system is continuously tracked through the real-time monitoring module.

Images