An Adaptive Atmospheric Parameter Reconstruction Method for Aircraft Based on Coarse-Fine Consistency
By combining nonlinear least squares and extended Kalman filtering to improve the Sage-Husa algorithm, a coarse-fine consistency detection architecture is constructed. This solves the problems of traditional atmospheric data sensors being susceptible to interference and the decrease in accuracy of filtering algorithms, and achieves high-precision and robust atmospheric parameter estimation.
Patent Information
- Application Number
- CN202611402991.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-10
- Publication Date
- 2026-11-10
AI Technical Summary
Traditional atmospheric data sensors are susceptible to interference from the external environment, leading to measurement failures. Existing filtering algorithms suffer from decreased accuracy or divergence in complex environments, making it difficult to balance convergence speed, smoothness, and fast tracking capability.
An adaptive atmospheric parameter reconstruction method based on coarse-fine consistency is adopted. The coarse wind speed is obtained by nonlinear least squares method. Combined with extended Kalman filter and improved Sage-Husa algorithm, the covariance expansion coefficient is dynamically adjusted to construct a coarse-fine consistency detection architecture and improve the estimation accuracy.
Under complex wind fields and sensor noise interference, it significantly improves the estimation accuracy of atmospheric parameters and system robustness, fast tracking capability and smoothness, and solves the divergence problem of traditional filtering algorithms.
Smart Images

Figure CN123067533A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of atmospheric environment perception and multi-source information fusion computing technology for aircraft, and in particular to an adaptive atmospheric parameter reconstruction method for aircraft based on coarse-fine consistency. Background Technology
[0002] Angle of attack, sideslip angle, and vacuum velocity are important atmospheric parameters characterizing the airflow environment around an aircraft, and are crucial for its flight efficiency and safe handling. Traditional atmospheric data sensors (such as pitot tubes) are susceptible to external environmental interference, and physical damage can cause measurement errors, affecting flight safety. Virtual Atmospheric Data Systems (VADS) use mathematical models to replace physical hardware, do not rely on traditional atmospheric data measurement devices, and have significant advantages in applications such as high angle of attack, high-speed maneuvering, and transatmospheric flight of aerospace vehicles.
[0003] In complex environments such as the reentry phase of aerospace vehicles, wind disturbances (such as wind shear, gusts, and turbulence) and high-dynamic maneuvers of the aircraft cause wind speed state system noise and navigation system measurement noise to exhibit strong time-varying characteristics. Traditional extended Kalman filters (EKFs) often suffer from decreased filtering accuracy or even divergence in atmospheric parameter estimation due to unknown or abrupt changes in noise statistical characteristics. While existing Sage-Husa adaptive filters can estimate noise covariance online, they are prone to divergence and loss of positive definiteness of the covariance matrix during severe aircraft maneuvers or sudden wind field changes due to significant fluctuations in the innovation data. Furthermore, existing techniques often suffer from slow convergence speed and difficulty in balancing smoothness and fast tracking capability when calculating virtual atmospheric parameters without initial wind speed information. Summary of the Invention
[0004] The purpose of this invention is to provide an adaptive atmospheric parameter reconstruction method for aircraft based on coarse-fine consistency, which solves the problems of slow convergence speed and difficulty in balancing smoothness and fast tracking capability in atmospheric parameter estimation by Kalman and adaptive filtering algorithms.
[0005] To achieve the above objectives, this invention provides a method for reconstructing adaptive atmospheric parameters for aircraft based on coarse-fine consistency, comprising the following steps: S1. Obtain the aircraft's navigation and flight control parameters; S2. Based on the aerodynamic and dynamic equations, a measurement model is established. The instantaneous wind speed at the current moment is solved by the nonlinear least squares method. The instantaneous wind speed is then smoothed by a sliding window to obtain a smoothed coarse estimate of the wind speed, which is then used as a reference wind speed. S3. Establish a state equation with the horizontal component of wind speed as the state variable, perform one-step state prediction based on extended Kalman filtering, obtain the prior state estimate of wind speed and the uncorrected prior error covariance matrix, and calculate measurement information in combination with navigation parameters. S4. Calculate the deviation index between the coarsely estimated wind speed and the prior state estimate, and construct a dynamic divergence criterion threshold function based on the deviation index. S5. Use the dynamic divergence criterion threshold function to determine whether the filtering divergence condition is met. If it is met, calculate the covariance inflation coefficient and perform inflation correction on the uncorrected prior error covariance matrix. S6. Solve for the optimal wind speed estimate and perform adaptive bounded update for process noise; S7. Based on the optimal wind speed state estimate and ground speed, atmospheric parameters are obtained by solving the velocity vector triangle relationship.
[0006] Preferably, the navigation parameters in S1 include ground speed, attitude angle, angular velocity, and specific force acceleration, and the flight control parameters include control surface deflection angle and engine thrust.
[0007] Preferably, the step S2, which uses the nonlinear least squares method to calculate the instantaneous wind speed at the current moment, includes: The nonlinear functional relationship is established based on the aerodynamic and dynamic models, as shown in the following equation: ; In the formula, Linear acceleration, Angular acceleration, Let the horizontal component of the wind speed be the unknown. Vacuum speed, For the angle of attack, Sideslip angle; A rough estimate of wind speed is obtained by minimizing the sum of squared residuals between the measured acceleration and the model-calculated acceleration.
[0008] Preferably, the sliding window smoothing process for instantaneous wind speed in S2 includes: In length Instantaneous wind speed obtained by nonlinear least squares method within a sliding window Mean filtering is performed to obtain a smoothed coarse estimate of the wind speed. This is for reference wind speed information.
[0009] Preferably, S3 includes: S31. Set the state vector to the horizontal component of wind speed. The measurement vector is The state equation and measurement equation are established as follows: ; In the formula, Here is the state transition matrix. Here is the system noise matrix. For measuring the noise matrix; S32. The extended Kalman filter obtains the prior state estimate and the prior error covariance matrix through the one-step state prediction formula, as shown in the following equation: ; In the formula, These are the prior state estimates. This is the state transition matrix from the previous time step to the current time step. This is the posterior state estimate from the previous time step. This is the process noise vector from the previous time step. Let be the prior error covariance matrix. For posterior covariance, Let be the process noise covariance matrix of the previous time step. This is the transpose of the matrix; S33, Combined with the actual measurement values output by the navigation system Calculation of measurement information As shown in the following formula: ; In the formula, Input parameters for the current navigation and flight control. It is a nonlinear measurement function. This is the measurement noise vector from the previous time step.
[0010] Preferably, S4 includes: S41. Calculate the Euclidean distance between the smoothed coarse estimate of wind speed and the prior state estimate. ; S42. Construct a dynamic divergence criterion threshold function, as shown in the following formula: ; In the formula, This serves as a basic fixed threshold for suppressing false noise triggering during stable flight. This is the sensitivity adjustment coefficient. To ensure the minimum protection threshold for system stability and ; S43, when When it increases, When the value decreases, the divergence criterion is triggered.
[0011] Preferably, S5 includes: S51. Calculate the new information statistics matrix. traces The trace of the theoretical covariance matrix is shown in the following equation: ; In the formula, The trace of the matrix, Let be the first-order Jacobian matrix with respect to the state vector. The uncorrected prior error covariance matrix. The measurement noise covariance matrix at the previous time step. This is the transpose of the matrix; S52. Check whether the divergence condition is met, including: If the divergence condition is not met, then retain the original... constant; If the criterion is met If the filter diverges, the expansion coefficient is calculated. As shown in the following formula: ; In the formula, This is the process noise from the previous moment. Here is the state transition matrix. Let be the posterior error covariance matrix of the previous time step. This is the transpose of the matrix; S53, limited ,like Then forced The prior error covariance matrix is corrected using the expansion coefficient, as shown in the following equation: ; In the formula, This is the transpose of the matrix.
[0012] Preferably, S6 includes: S61. Using the corrected prior error covariance matrix and measurement information, calculate the Kalman gain to update the wind speed state estimate and the posterior covariance matrix, and obtain the optimal wind speed estimate. S62. Adaptive update of process noise covariance matrix based on improved Sage-Husa algorithm.
[0013] Preferably, the update rule for adaptively updating the process noise covariance matrix in S62 includes: According to the Sage-Husa algorithm formula, the process noise update term for the current step is estimated from the Kalman gain and the current innovation vector, combined with the time decay coefficient. The process noise covariance calculation matrix is obtained, and then the diagonal elements of the matrix are limited to be between the lower protection limit and the upper protection limit representing the maximum physical wind variance. By imposing engineering quantity constraints, the online algorithm is prevented from causing the matrix to become negatively definite or the calculation to diverge.
[0014] Preferably, the mathematical mapping rules for solving the atmospheric parameters in S7 include: Based on the vector triangle relationship between vacuum velocity, ground velocity, and wind speed, the three components of the corresponding vacuum velocity in the body coordinate system are obtained. ; The vector triangle relationship is shown in the following formula: ; In the formula, This represents the attitude transformation matrix from the geographic system to the machine system. Indicates the local geographic coordinate system; Indicates the body coordinate system; Indicates ground speed, Indicates wind speed. Vacuum speed; The true atmospheric parameters, including vacuum velocity, angle of attack, and sideslip angle, are obtained by inverse kinematics, as shown below: ; In the formula, Vacuum speed, For the angle of attack, It is the sideslip angle.
[0015] Therefore, the above-mentioned adaptive atmospheric parameter reconstruction method for aircraft based on coarse-fine consistency has the following beneficial effects: This method innovatively constructs a consistency detection architecture that integrates coarse and fine estimation. On one hand, the coarsely estimated wind speed obtained using nonlinear least squares provides a physical benchmark (reference wind speed) for filtering; on the other hand, it introduces a coarse-fine error-based approach into the EKF. Dynamically Adjusted Adaptive Divergence Criterion When encountering sudden and severe changes in the external wind field, the dynamic threshold automatically decreases to quickly trigger covariance expansion correction, increasing the confidence in the current measurement data; during stable flight, the threshold is automatically raised to avoid false triggering by noise. Combined with an improved Sage-Husa noise estimation method with engineering boundary protection, this method improves the estimation accuracy of atmospheric parameters and the robustness of the system, solving the technical challenges of traditional filtering algorithms easily diverging and struggling to balance smoothness and dynamic tracking capability under complex wind fields and time-varying sensor noise interference.
[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0017] Figure 1 This is an overall architecture diagram of an adaptive atmospheric parameter reconstruction method for aircraft based on coarse-fine consistency according to the present invention. Figure 2 The flowchart of the improved adaptive Sage-Husa extended Kalman filter algorithm that integrates coarse and fine consistency criteria proposed in this invention is shown below. Figure 3 The figure shows the simulation comparison results of the atmospheric parameter estimation error curves under the superposition of strong measurement noise and complex wind field environment of the present invention. Detailed Implementation
[0018] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0019] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.
[0020] Example Please see Figures 1-3 This invention provides a method for reconstructing adaptive atmospheric parameters for aircraft based on coarse-fine consistency, comprising the following steps: S1. Multi-source navigation and flight control parameter acquisition. During flight, the ground speed in the aircraft's body coordinate system output by the navigation system (INS / GPS, etc.) is acquired in real time. angular velocity acceleration angular acceleration Attitude angle and the control surface deflection angle output by the flight control system. Throttle opening related to (ailerons, elevators, rudder) and engine thrust. .
[0021] S2. Coarse estimation of the channel. A measurement model is established based on aerodynamic and dynamic equations. The instantaneous wind speed at the current moment is calculated using the nonlinear least squares method. The instantaneous wind speed is then smoothed using a sliding window to obtain a smoothed coarse estimate of the wind speed, which is used as a reference wind speed.
[0022] The instantaneous wind speed at the current moment is calculated using the nonlinear least squares method. Navigation information and atmospheric parameters are obtained using the aircraft's dynamic equations and aerodynamic model. The nonlinear functional relationship is shown in the following equation: ; Combining the velocity vector triangle relationship: ; In the formula, Indicates the local geographic coordinate system (northeast). Indicates the body coordinate system; Indicates ground speed, Indicates wind speed. Indicates vacuum velocity, The vertical component of wind speed is very small compared to the horizontal component, and can be considered as zero. ; This represents the attitude transformation matrix.
[0023] Atmospheric parameters can also be represented by ground speed and horizontal wind speed: ; Therefore, a direct relationship is established between navigation measurement parameters and wind speed, as shown in the following formula: ; In the formula, For linear acceleration, Angular acceleration, Let be the horizontal component of the wind speed to be determined. The vertical component of the wind speed is very small compared to the horizontal component and can be considered as zero. Vacuum speed, For the angle of attack, It is the sideslip angle.
[0024] The above formula is obtained Viewed as a nonlinear least squares problem, i.e., minimizing the objective function, as shown in the following equation: ; In the formula, , The minimum value is obtained quickly and stably using the nonlinear least squares LM method, and its iterative process is as follows: ; In the formula, , For the number of iterations, The damping factor, for exist Jacobian matrix at the location.
[0025] Solve for the instantaneous wind speed at the current moment. Then, introduce a length of The instantaneous wind speed within the sliding window is smoothed by mean smoothing to obtain a smoothed coarse estimate of the wind speed. This is for reference wind speed information.
[0026] S3. Prior Prediction using Adaptive EKF. A state equation is established with the horizontal component of wind speed as the state variable. One-step state prediction is performed based on the Extended Kalman Filter (EKF) to obtain the prior state estimate of wind speed and the uncorrected prior error covariance matrix. Measurement information is then calculated in conjunction with navigation parameters. This includes: S31. Set the state vector to the horizontal component of wind speed. The measurement vector is The state equation and measurement equation are established as follows: ; In the formula, This is the state transition matrix, here it is the identity matrix. Here is the system noise matrix. This is the measurement noise matrix. The two are consistent: , .
[0027] S32. The extended Kalman filter obtains the prior state estimate and the prior error covariance matrix through the one-step state prediction formula, as shown in the following equation: ; In the formula, These are the prior state estimates. This is the state transition matrix from the previous time step to the current time step. This is the posterior state estimate from the previous time step. This is the process noise vector from the previous time step. Let be the prior error covariance matrix. For posterior covariance, Let be the process noise covariance matrix of the previous time step. This is the transpose of the matrix.
[0028] S33, Combined with the actual measurement values output by the navigation system Calculation of measurement information As shown in the following formula: ; In the formula, Input parameters for the current navigation and flight control. It is a nonlinear measurement function. This is the measurement noise vector from the previous time step.
[0029] S4. Calculate the coarse and fine deviation indices and dynamic thresholds. Calculate the coarse estimate of wind speed. Compared with prior state estimates Euclidean distance between As a deviation indicator, a dynamic divergence criterion threshold function is constructed based on the deviation indicator. This includes: S41. Calculate the Euclidean distance between the smoothed coarse estimate of wind speed and the prior state estimate. As shown in the following formula: ; S42. Construct a dynamic divergence criterion threshold function, as shown in the following formula: ; In the formula, This serves as a basic fixed threshold for suppressing false noise triggering during stable flight. This is the sensitivity adjustment coefficient. To ensure the minimum protection threshold for system stability and .
[0030] S43, when When it increases, When the value decreases, the divergence criterion is triggered.
[0031] When the wind field changes suddenly, causing When it increases dramatically, Rapidly descending and approaching This makes the system highly susceptible to triggering correction mechanisms in volatile environments; under stable conditions, Stay To avoid misjudgment caused by measurement noise in the vicinity.
[0032] S5. Coarse-to-fine consistency detection and covariance dilation. A dynamic divergence criterion threshold function is used to determine whether the filter divergence condition is met. This includes: S51. Calculate the new information statistics matrix. traces The trace of the theoretical covariance matrix is shown in the following equation: ; In the formula, The trace of the matrix, Let be the first-order Jacobian matrix with respect to the state vector. The uncorrected prior error covariance matrix. The measurement noise covariance matrix at the previous time step. This is the transpose of the matrix.
[0033] Among them, the new information statistics matrix Through the new information sequence The calculation yields the following result: ; In the formula, The forgetting factor of the new information takes values in the range of 1-2. between.
[0034] S52. Introduce a filter divergence criterion based on covariance matching technology to detect whether the divergence condition is met.
[0035] If the divergence condition is not met, then retain the original... constant.
[0036] If the criterion is met If the filter diverges or the tracking is severely lagging, then the filter is determined to be diverging, and the expansion coefficient is calculated. As shown in the following formula: ; In the formula, This is the process noise from the previous moment. Here is the state transition matrix. Let be the posterior error covariance matrix of the previous time step. This is the transpose of the matrix.
[0037] S53, limited ;like Then forced The prior covariance is inflated using an inflation coefficient, as shown in the following equation: ; In the formula, This is the transpose of the matrix.
[0038] S6. Solve for the optimal wind speed estimate, and based on the improved Sage-Husa adaptive Kalman filter algorithm, complete the state update and improved Sage-Husa noise update calculations, including: S61. Using the corrected prior error covariance matrix and measurement information, calculate the Kalman gain to update the wind speed state estimate and the posterior covariance matrix, and obtain the optimal wind speed estimate.
[0039] Calculate Kalman gain And complete the optimal wind speed estimation and posterior covariance. The update is shown in the following formula: ; In the formula, It is the Jacobian matrix of the measurement equation.
[0040] S62. Adaptive update of the process noise covariance matrix based on the improved Sage-Husa algorithm. The update rule includes: estimating the process noise update term for the current step using the Kalman gain and the current innovation vector according to the Sage-Husa algorithm formula, and combining it with the time decay coefficient. The process noise covariance calculation matrix is obtained, and then the diagonal elements of the matrix are limited to be between the lower protection limit and the upper protection limit representing the maximum physical wind variance. By imposing engineering quantity constraints, the online algorithm is prevented from causing the matrix to become negatively definite or the calculation to diverge.
[0041] Combined with time decay coefficient The noise update process is performed, in which... This is the forgetting factor. The process is as follows: ; This embodiment calculates the process noise covariance. Apply protective constraints to diagonal elements: Force setting ,in, This represents the upper limit of variance when the aircraft encounters the maximum possible wind shear physical limit. This represents the lower limit of variance when the aircraft encounters the physical limit of the maximum possible wind shear.
[0042] S7. Atmospheric Parameter Trigonometric Calculation. Atmospheric parameters are obtained by calculating the optimal wind speed estimate and ground speed using the velocity vector triangle relationship. The mathematical mapping rules for atmospheric parameter calculation include: obtaining the three components of the corresponding vacuum speed in the body coordinate system based on the vector triangle relationship between vacuum speed, ground speed, and wind speed. As shown in the following formula: ; In the formula, This represents the attitude transformation matrix from the geographic system to the machine system. Indicates the local geographic coordinate system; Indicates the body coordinate system; Indicates ground speed, Indicates wind speed. It is the vacuum velocity.
[0043] The true atmospheric parameters, including vacuum velocity, angle of attack, and sideslip angle, are obtained by inverse kinematics, as shown below: ; In the formula, Vacuum speed, For the angle of attack, It is the sideslip angle.
[0044] To verify the effectiveness of the proposed technique for estimating atmospheric parameters under complex wind conditions, and to compare the estimation accuracy of airspeed, angle of attack, and sideslip angle between the traditional EKF method and this method, this embodiment sets acceleration... The measured noise is angular acceleration The measured noise is Comparative simulations were conducted in complex wind field environments including gusts, wind shear, and turbulence. The simulation results for the estimation errors of airspeed, angle of attack, and sideslip angle are as follows: Figure 3 As shown in Table 1, the root mean square error (RMSE) between the estimated and actual values of the statistical atmospheric parameters is as follows.
[0045] Table 1. Root mean square error between estimated and actual atmospheric parameters.
[0046] Experimental results show that the proposed method achieves a significant leap in estimation accuracy of key atmospheric parameters compared to the traditional EKF method. As shown in Table 1, the proposed method improves the reconstruction accuracy of vacuum velocity by 66.44%, angle of attack by 21.16%, and sideslip angle by 58.32%.
[0047] like Figure 3 As shown in the estimation error curve, this method not only exhibits high smoothness filtering characteristics in the steady segment, eliminating numerical jitter caused by sensor Gaussian white noise (compared to the blue coarse estimation curve), but also demonstrates significantly better agile tracking capability than EKF (green curve) at multiple sudden wind disturbance moments of 10s, 20s, 40s, and 50s. At the same time, using the coarse estimation result as the initial value for filtering at the initial moment also significantly accelerates the convergence speed, greatly improving the robustness and accuracy of the aircraft's atmospheric environment perception.
[0048] Therefore, the present invention adopts the above-mentioned adaptive atmospheric parameter reconstruction method for aircraft based on coarse and fine consistency, which effectively solves the problem that traditional EKF is difficult to balance smoothness and fast tracking capability under complex wind fields and time-varying sensor noise interference, and greatly improves the estimation accuracy of atmospheric parameters and system robustness.
[0049] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.
Claims
1. A method for reconstructing adaptive atmospheric parameters of an aircraft based on coarse-fine consistency, characterized in that, Includes the following steps: S1. Obtain the aircraft's navigation and flight control parameters; S2. Based on the aerodynamic and dynamic equations, a measurement model is established. The instantaneous wind speed at the current moment is solved by the nonlinear least squares method. The instantaneous wind speed is then smoothed by a sliding window to obtain a smoothed coarse estimate of the wind speed, which is then used as a reference wind speed. S3. Establish a state equation with the horizontal component of wind speed as the state variable, perform one-step state prediction based on extended Kalman filtering, obtain the prior state estimate of wind speed and the uncorrected prior error covariance matrix, and calculate measurement information in combination with navigation parameters. S4. Calculate the deviation index between the coarsely estimated wind speed and the prior state estimate, and construct a dynamic divergence criterion threshold function based on the deviation index. S5. Use the dynamic divergence criterion threshold function to determine whether the filtering divergence condition is met. If it is met, calculate the covariance inflation coefficient and perform inflation correction on the uncorrected prior error covariance matrix. S6. Solve for the optimal wind speed estimate and perform adaptive bounded update for process noise; S7. Based on the optimal wind speed state estimate and ground speed, atmospheric parameters are obtained by solving the velocity vector triangle relationship.
2. The method for reconstructing adaptive atmospheric parameters of an aircraft based on coarse-fine consistency according to claim 1, characterized in that: The navigation parameters in S1 include ground speed, attitude angle, angular velocity, and specific force acceleration, while the flight control parameters include control surface deflection angle and engine thrust.
3. The method for reconstructing adaptive atmospheric parameters of an aircraft based on coarse-fine consistency according to claim 2, characterized in that, The calculation of the instantaneous wind speed at the current moment using the nonlinear least squares method in S2 includes: The nonlinear functional relationship is established based on the aerodynamic and dynamic models, as shown in the following equation: ; In the formula, Linear acceleration, Angular acceleration, Let the horizontal component of the wind speed be the unknown. Vacuum speed, For the angle of attack, Sideslip angle; A rough estimate of wind speed is obtained by minimizing the sum of squared residuals between the measured acceleration and the model-calculated acceleration.
4. The method for reconstructing adaptive atmospheric parameters of an aircraft based on coarse-fine consistency according to claim 3, characterized in that, The sliding window smoothing process for instantaneous wind speed in S2 includes: In length of Instantaneous wind speed obtained by nonlinear least squares method within a sliding window Mean filtering is performed to obtain a smoothed coarse estimate of the wind speed. This is for reference wind speed information.
5. The method for reconstructing adaptive atmospheric parameters of an aircraft based on coarse-fine consistency according to claim 4, characterized in that, S3 includes: S31, S31, Set the state vector to the horizontal component of wind speed. The measurement vector is The state equation and measurement equation are established as follows: ; In the formula, Here is the state transition matrix. Here is the system noise matrix. For measuring the noise matrix; S32. The extended Kalman filter obtains the prior state estimate and the prior error covariance matrix through the one-step state prediction formula, as shown in the following equation: ; In the formula, These are the prior state estimates. This is the state transition matrix from the previous time step to the current time step. This is the posterior state estimate from the previous time step. This is the process noise vector from the previous time step. Let be the prior error covariance matrix. For posterior covariance, Let be the process noise covariance matrix of the previous time step. This is the transpose of the matrix; S33, Combined with the actual measurement values output by the navigation system Calculation of measurement information As shown in the following formula: ; In the formula, Input parameters for the current navigation and flight control. It is a nonlinear measurement function. This is the measurement noise vector from the previous time step.
6. The method for reconstructing adaptive atmospheric parameters of an aircraft based on coarse-fine consistency according to claim 5, characterized in that, S4 includes: S41. Calculate the Euclidean distance between the smoothed coarse estimate of wind speed and the prior state estimate. ; S42. Construct a dynamic divergence criterion threshold function, as shown in the following formula: ; In the formula, This serves as a basic fixed threshold for suppressing false noise triggering during stable flight. This is the sensitivity adjustment coefficient. To ensure the minimum protection threshold for system stability and ; S43, when When it increases, When the value decreases, the divergence criterion is triggered.
7. The method for reconstructing adaptive atmospheric parameters of an aircraft based on coarse-fine consistency according to claim 6, characterized in that, S5 includes: S51. Calculate the new information statistics matrix. traces The trace of the theoretical covariance matrix is shown in the following equation: ; In the formula, The trace of the matrix, Let be the first-order Jacobian matrix with respect to the state vector. The uncorrected prior error covariance matrix. The measurement noise covariance matrix at the previous time step. This is the transpose of the matrix; S52. Check whether the divergence condition is met, including: If the divergence condition is not met, then retain the original... constant; If the criterion is met If the filter diverges, the expansion coefficient is calculated. As shown in the following formula: ; In the formula, This is the process noise from the previous moment. Here is the state transition matrix. Let be the posterior error covariance matrix of the previous time step. This is the transpose of the matrix; S53, limited ,like Then forced The prior error covariance matrix is corrected using the expansion coefficient, as shown in the following equation: ; In the formula, This is the transpose of the matrix.
8. The method for reconstructing adaptive atmospheric parameters of an aircraft based on coarse-fine consistency according to claim 7, characterized in that, S6 includes: S61. Using the corrected prior error covariance matrix and measurement information, calculate the Kalman gain to update the wind speed state estimate and the posterior covariance matrix, and obtain the optimal wind speed estimate. S62. Adaptive update of process noise covariance matrix based on improved Sage-Husa algorithm.
9. The method for reconstructing adaptive atmospheric parameters of an aircraft based on coarse-fine consistency according to claim 8, characterized in that, The update rules for adaptively updating the process noise covariance matrix in S62 include: According to the Sage-Husa algorithm formula, the process noise update term for the current step is estimated from the Kalman gain and the current innovation vector, combined with the time decay coefficient. The process noise covariance calculation matrix is obtained, and then the diagonal elements of the matrix are limited to be between the lower protection limit and the upper protection limit representing the maximum physical wind variance. By imposing engineering quantity constraints, the online algorithm is prevented from causing the matrix to become negatively definite or the calculation to diverge.
10. The method for reconstructing adaptive atmospheric parameters of an aircraft based on coarse-fine consistency according to claim 9, characterized in that, The mathematical mapping rules for solving the atmospheric parameters in S7 include: Based on the vector triangle relationship between vacuum velocity, ground velocity, and wind speed, the three components of the corresponding vacuum velocity in the body coordinate system are obtained. ; The vector triangle relationship is shown in the following formula: ; In the formula, This represents the attitude transformation matrix from the geographic system to the machine system. Indicates the local geographic coordinate system; Indicates the body coordinate system; Indicates ground speed, Indicates wind speed. Vacuum speed; The true atmospheric parameters, including vacuum velocity, angle of attack, and sideslip angle, are obtained by inverse kinematics, as shown below: ; In the formula, Vacuum speed, For the angle of attack, It is the sideslip angle.
