An atmospheric data fusion voting algorithm for reentry vehicles and a reentry vehicle

Abstract

This invention discloses an atmospheric data fusion voting algorithm and a reentry vehicle, comprising: Step 1, calculating the overload reference values ​​of the atmospheric data system, including: normal overload reference values ​​and lateral overload reference values; Step 2, calculating the error criteria of the atmospheric data system, the error criteria including: normal overload tolerance error, lateral overload tolerance error, and airspeed tolerance error; Step 3, based on monitoring the current working mode of the atmospheric data system, and combining the overload reference values ​​and error criteria, reconstructing, fusing, or directly outputting the output of the atmospheric data system. The technical solution provided by this invention addresses the difficulty in clearly determining which aspect of performance or safety issues should be addressed for various performance enhancement requirements of atmospheric data systems in reentry vehicles.

Description

Technical Field

[0001] This invention relates to, but is not limited to, the field of automatic flight control technology, and particularly to an atmospheric data fusion voting algorithm for reentry vehicles and a reentry vehicle. Background Technology

[0002] Reentry vehicles are characterized by a wide speed range and a broad altitude range. Their speed range extends from subsonic to hypersonic speeds, and their altitude range extends from the vacuum environment of space to the complex atmosphere at low altitudes. Their reentry and recovery missions also require high angle of attack and high maneuverability.

[0003] In the complex operating environment of reentry vehicles, atmospheric data systems face severe challenges: on the one hand, in terms of performance, the dynamic response of atmospheric data systems is poor under high angle-of-attack and high-maneuverability flight conditions, and the measurement of airflow angles also produces large errors, which leads to a serious deterioration in the measurement performance of atmospheric data systems; on the other hand, in terms of safety, due to cost and space constraints, hardware redundancy configuration cannot be improved, and the flight control system needs to estimate and back up airflow angle parameters from the perspective of analytical redundancy to avoid loss of control of the aircraft caused by atmospheric failure. Given the various performance enhancement requirements for atmospheric data systems in reentry vehicles, it is difficult to clearly determine which aspect of performance or safety issues should be addressed. Summary of the Invention

[0004] The purpose of this invention is to solve the above-mentioned technical problems. This invention provides an atmospheric data fusion voting algorithm and a reentry vehicle to address the difficulty in clearly determining which aspect of performance or safety issues should be addressed for various performance enhancement requirements of atmospheric data systems in reentry vehicles.

[0005] The technical solution of the present invention: In a first aspect, embodiments of the present invention provide an atmospheric data fusion voting algorithm for reentry vehicles, comprising:

[0006] Step 1: Calculate the overload reference value of the atmospheric data system, which includes: normal overload reference value and lateral overload reference value;

[0007] Step 2: Calculate the error criteria of the atmospheric data system. The error criteria include: normal overload tolerance error, lateral overload tolerance error, and airspeed tolerance error.

[0008] Step 3: Based on the current working mode of the atmospheric data monitoring system, and in conjunction with the overload reference value and the error criterion, reconstruct, fuse, or directly output the output of the atmospheric data system.

[0009] Optionally, in the atmospheric data fusion voting algorithm for reentry vehicles as described above, before step 1, the following steps are further included:

[0010] Step A involves classifying the operating modes of the atmospheric data system into the following categories:

[0011] Operating mode 1, normal mode of atmospheric data system; under this operating mode 1, the output of atmospheric data system is generated based on different maneuvering states;

[0012] Operating mode 2, atmospheric data system downgrade mode; in this operating mode 2, airspeed is measurable, but airflow angle is not measurable; by fusing information from vacuum speed, inertial navigation, and aircraft dynamics model, the airflow angle is reconstructed, which includes angle of attack and sideslip angle;

[0013] Operating mode 3, air data system failure mode; in this operating mode 3, airspeed and airflow angle are unmeasurable; by fusing inertial navigation information and aircraft dynamics model, airspeed and airflow angle are reconstructed;

[0014] In troubleshooting, determining the operating mode of the atmospheric data system includes:

[0015] If the working mode of the atmospheric data system is 1, then the atmospheric data system fault word is set to valid;

[0016] If the atmospheric data system is operating in mode 2, then set the atmospheric data system fault word to downgraded.

[0017] If the working mode of the atmospheric data system is 3, then the atmospheric data system fault word is set to complete failure.

[0018] Optionally, in the atmospheric data fusion voting algorithm for reentry vehicles as described above, the method for calculating the normal overload reference value in step 1 includes:

[0019] S11, determine the validity of the normal overload reference value, and identify whether the normal overload reference value is valid or invalid:

[0020] S12, when the valid flag of the normal overload reference value is valid, the normal overload reference value is calculated by using the aircraft longitudinal body aerodynamic model and the control surface overload compensation respectively. The aircraft longitudinal body aerodynamic model and the control surface overload compensation are both calculated by aerodynamic interpolation table.

[0021] S13, the normal overload reference values ​​calculated separately by the aircraft longitudinal aerodynamic model and the control surface overload compensation are added together to obtain the total normal overload reference value.

[0022] Optionally, in the atmospheric data fusion voting algorithm for reentry vehicles as described above, the method for calculating the lateral overload reference value in step 1 includes:

[0023] S21, determine the validity of the lateral overload reference value, and identify whether the lateral overload reference value is valid or invalid:

[0024] S22, when the valid flag of the lateral overload reference value is valid, the lateral overload reference value is calculated separately using the aircraft lateral body aerodynamic model and the control surface overload compensation; wherein, the aircraft lateral body aerodynamic model and the control surface overload compensation are both calculated by aerodynamic interpolation tables;

[0025] S23, the lateral overload reference values ​​calculated separately by the aircraft lateral aerodynamic model and the control surface overload compensation are added together to obtain the total lateral overload reference value.

[0026] Optionally, in the atmospheric data fusion voting algorithm for reentry vehicles as described above, the method for calculating the normal overload tolerance error and the lateral overload tolerance error in step 2 includes:

[0027] Step 21: Calculate the average normal force coefficient slope k_avg of the aircraft from zero lift angle of attack to stall angle of attack at each Mach number. Calculate a normal force coefficient slope k between two adjacent angles of attack. Calculate the lift sloping nonlinearity based on k_avg and k: k_non = abs(k - k_avg); and normalize the lift sloping nonlinearity to obtain:

[0028] k_non_01=k_non / (k_non_max-k_non_min);

[0029] Step 22, calculate the maximum error criterion based on the nonlinearity of the lift line slope obtained by normalization, including:

[0030] The maximum error criterion for calculating normal overload, Sigma1, is:

[0031] Sigma1=k_non_01*(Sigma_nz_worst-Sigma_nz_best)+Sigma_nz_best;

[0032] The Sigma2 criterion for calculating the maximum error of lateral overload is:

[0033] Sigma2=k_non_01*(Sigma_ny_worst-Sigma_ny_best)+Sigma_ny_best;

[0034] Wherein, Sigma_nz_best is the maximum absolute error of the normal force coefficient in the linear lift region, and Sigma_nz_worst is the maximum absolute error of the normal force coefficient in the stall region; Sigma_ny_best is the maximum absolute error of the lateral force coefficient in the linear lift region, and Sigma_ny_worst is the maximum absolute error of the lateral force coefficient in the stall region.

[0035] Step 23, calculate the tolerable error of normal overload and the tolerable error of lateral overload:

[0036] Sigma_nz = q * S * Sigma1 / mg;

[0037] Sigma_ny = q * S * Sigma² / mg;

[0038] Where Sigma_nz is the tolerable error of normal overload, Sigma_ny is the tolerable error of lateral overload, S is the aircraft reference area, m is the aircraft mass, g is the gravitational acceleration, and q is the measured dynamic pressure.

[0039] Optionally, in the atmospheric data fusion voting algorithm for reentry vehicles as described above, step 2, calculating the airspeed tolerance error, includes:

[0040] Step 24, construct the vector relationship between the ground speed, vacuum speed, and wind speed of the reentry vehicle as follows:

[0041]

[0042] Among them, wind speed V w V is the velocity of air relative to the geographic coordinate system, or the velocity in a vacuum. T The ground speed V is the velocity of the reentry vehicle relative to the air. g S represents the velocity of the reentry vehicle relative to the geographic coordinate system; the superscript B indicates the vector in the body coordinate system. be It is represented as the attitude transformation matrix from the navigation coordinate system to the body coordinate system, and the superscript N indicates the vector in the geographic coordinate system;

[0043] Step 25: Using the attitude angles of the reentry vehicle, the attitude transformation matrix is ​​obtained as follows:

[0044]

[0045] Among them, ψ, θ, These are the aircraft's yaw angle, pitch angle, and roll angle, respectively.

[0046] Step 26, based on the relationships in steps 24 and 25, through the ground speed V g and vacuum velocity V TThe wind speed V is calculated from the vector difference. w By statistically analyzing the average wind speed of the reentry trajectory, a constant wind speed curve related to altitude and season is obtained; the tolerable airspeed error is twice that of the constant wind speed curve.

[0047] Optionally, in the atmospheric data fusion voting algorithm for reentry vehicles as described above, step 3 includes:

[0048] When the atmospheric data system fault word is completely failed, a reconstruction process is performed, and the reconstructed angle of attack, reconstructed sideslip angle, and reconstructed vacuum velocity are output.

[0049] When the fault word of the atmospheric data system is degraded, vacuum velocity information is added, and reconstruction processing is performed to output the reconstructed angle of attack, reconstructed sideslip angle, and reconstructed vacuum velocity.

[0050] If the atmospheric data system fault word is valid, then perform the following judgment:

[0051] If the absolute value of the error between the normal overload signal and the normal overload reference value of the inertial navigation system is greater than or equal to the tolerable error of the normal overload, and the absolute value of the error between the lateral overload signal and the lateral overload reference value is greater than or equal to the tolerable error of the lateral overload, and the absolute value of the error between the vacuum velocity and the ground velocity of the atmospheric data system is greater than or equal to the tolerable error of the airspeed, then the reconstruction process is performed, and the reconstructed angle of attack, reconstructed sideslip angle, and reconstructed vacuum velocity are output.

[0052] Otherwise, if the absolute value of the error between the inertial navigation normal overload signal and the normal overload reference value is less than the tolerable error of normal overload, and the absolute value of the error between the lateral overload signal and the lateral overload reference value is less than the tolerable error of lateral overload, and the absolute value of the error between the vacuum velocity and the ground velocity of the atmospheric data system is less than the tolerable error of airspeed, then the following logical judgment is executed: when the angle of attack is less than the preset value, the airflow angle of the atmospheric data system is directly output; otherwise, fusion processing is performed, and the airflow angle after information fusion is output.

[0053] Otherwise, directly output the airflow angle of the atmospheric data system.

[0054] Optionally, in the atmospheric data fusion voting algorithm for reentry vehicles described above, the reconstruction process in step 3 is performed as follows:

[0055] By utilizing the control outputs of the inertial navigation system and flight control system actuators, along with vacuum speed information, and combining them with the aircraft dynamics model, a Kalman filter is constructed to reconstruct the airflow angle and vacuum speed. Specifically, the pitch angle, roll angle, roll rate, pitch rate, yaw rate, and three-axis accelerations of the airframe provided by the inertial navigation system are used as measurement quantities. The pitch angle, roll angle, roll rate, pitch rate, yaw rate, angle of attack, sideslip angle, and vacuum speed of the aircraft are selected as state quantities, and the reconstructed angle of attack, sideslip angle, and vacuum speed are output.

[0056] The fusion process in step 3 is performed as follows:

[0057] By utilizing an inertial navigation system and an air data system, combined with an aircraft dynamics model, a Kalman filter is constructed to compensate for airflow angles during high angles of attack and high-maneuverability flight. Specifically, the three-axis ground speed, pitch angle, roll angle, yaw angle, three-axis angular velocity, and three-axis angular acceleration provided by the inertial navigation system, and the angle of attack, sideslip angle, and vacuum speed provided by the air data system are used as measurement quantities. The three-axis airspeed, pitch angle, roll angle, yaw angle, angle of attack, sideslip angle, vacuum speed, and three-axis wind speed of the aircraft are used as state quantities, and the reconstructed angle of attack and sideslip angle are output.

[0058] Optionally, in the atmospheric data fusion voting algorithm for reentry vehicles as described above, step 3 further includes:

[0059] Step 4: Verify the atmospheric data output by the atmospheric data system after applying the atmospheric data fusion voting algorithm, including:

[0060] S41, Parameter Settings, including: Wind Field Settings, Kalman Filter Parameter Settings, Error Adjustment Settings;

[0061] S42, Algorithm verification, including: verifying the accuracy and latency of the output results of the fusion processing and reconstruction processing from the perspectives of accuracy and latency.

[0062] Secondly, embodiments of the present invention also provide a reentry vehicle, the reentry vehicle being equipped with an atmospheric data fusion voting processor, a memory, as well as an inertial navigation system, a flight control system, and an atmospheric data system;

[0063] The inertial navigation system is configured to provide inertial navigation information to the atmospheric data fusion voting processor;

[0064] The flight control system is configured to provide the atmospheric data fusion voting processor with control outputs of its actuators;

[0065] The atmospheric data system is configured to provide vacuum velocity, angle of attack, and sideslip angle to the atmospheric data fusion voting processor;

[0066] The memory is configured to store executable instructions;

[0067] The atmospheric data fusion voting processor is specifically configured to implement the atmospheric data fusion voting algorithm for reentry vehicles as described above when executing the executable instructions stored in the memory.

[0068] The beneficial effects of this invention are as follows: This invention provides an atmospheric data fusion voting algorithm and a reentry vehicle. By analyzing the characteristics of the reentry vehicle, a fusion voting algorithm based on overload reference values ​​and error criteria of the atmospheric data system is designed and applicable to this type of aircraft. The fusion voting algorithm first calculates the overload reference value and error criteria of the atmospheric data system. The overload reference value includes normal overload reference value and lateral overload reference value, and the corresponding error criteria include normal overload tolerance error, lateral overload tolerance error, and airspeed tolerance error. The processing method of the fusion voting algorithm is as follows: based on the monitoring of the current working mode of the atmospheric data system (including three working modes), combined with the calculated overload reference value and error criteria, the output of the atmospheric data system is formed into three output forms according to the fusion voting logic: output after reconstruction processing, output after fusion processing, or direct output. By using the atmospheric data fusion voting algorithm provided by this invention, the computational load of the aircraft aerodynamic model is effectively reduced, and the airborne feasibility of the fusion voting algorithm is enhanced. In particular, the technical solution provided by the embodiments of the present invention can vote on the output of the atmospheric data system based on the working mode of the atmospheric data system, and output the optimal atmospheric data under various conditions. Attached Figure Description

[0069] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of the present invention and do not constitute a limitation on the technical solutions of the present invention.

[0070] Figure 1 This invention provides a flowchart of an atmospheric data fusion voting algorithm for reentry vehicles.

[0071] Figure 2 A schematic diagram illustrating the principle of the atmospheric data fusion voting algorithm for reentry vehicles provided in an embodiment of the present invention;

[0072] Figure 3 A schematic diagram illustrating the principle of airflow angle and airspeed reconstruction processing in the atmospheric data fusion voting algorithm for reentry vehicles provided in an embodiment of the present invention;

[0073] Figure 4This is a schematic diagram illustrating the principle of airflow angle fusion processing in the atmospheric data fusion voting algorithm for reentry vehicles provided in an embodiment of the present invention.

[0074] Figure 5 This is a schematic diagram of the constant wind field setting in the atmospheric data fusion voting algorithm for reentry vehicles provided in an embodiment of the present invention;

[0075] Figure 6 This is a schematic diagram of the simulation accuracy results of Monte Carlo target shooting in Example 1 of the simulation implementation of the present invention; Figure 6 Figure a shows the simulated structure at the angle of attack, and Figure b shows the simulated results at the sideslip angle.

[0076] Figures 7a to 7c This is a schematic diagram of the simulation results for frequency domain time delay evaluation of airflow angle fusion in Simulation Implementation Example 2 of the present invention; Figure 7a , Figure 7b and Figure 7c The frequency domain delay evaluation curves for angle-of-attack reconstruction at 1Hz, 2Hz, and 3Hz are shown respectively.

[0077] Figure 8 This is a schematic diagram illustrating the vector relationship between ground speed, vacuum speed, and wind speed related to the reentry vehicle in an embodiment of the present invention. Detailed Implementation

[0078] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

[0079] As explained in the background section, reentry vehicles are characterized by a wide velocity and altitude range, and their reentry and recovery missions require high angle-of-attack and high maneuverability. Furthermore, due to the complex operating environment of reentry vehicles, their atmospheric data systems present design challenges in terms of both performance and safety. Current regulations address various performance enhancement requirements for atmospheric data systems in reentry vehicles, making it difficult to definitively determine which aspect of performance or safety should be addressed.

[0080] To integrate data from several modes, including resolution redundancy, performance enhancement, and state judgment, a data fusion voting algorithm needs to be designed. To address the aforementioned problems and requirements, this invention provides an atmospheric data fusion voting algorithm and a reentry vehicle. The atmospheric data fusion voting algorithm provided in this invention designs switching logic under different fault modes, is compatible with different operating modes of the embedded atmosphere, and utilizes information fusion methods to design an airflow angle fusion algorithm.

[0081] The present invention provides the following specific embodiments, which can be combined with each other. For the same or similar concepts or processes, they may not be described again in some embodiments.

[0082] Figure 1 This invention provides a flowchart of an atmospheric data fusion voting algorithm for reentry vehicles. The atmospheric data fusion voting algorithm for reentry vehicles provided by this invention includes the following steps:

[0083] Step 1: Calculate the overload reference value of the atmospheric data system, which includes: normal overload reference value and lateral overload reference value;

[0084] Step 2: Calculate the error criteria of the atmospheric data system. The error criteria include: normal overload tolerance error, lateral overload tolerance error, and airspeed tolerance error.

[0085] Step 3: Based on the current working mode of the atmospheric data monitoring system, and in conjunction with the overload reference value and error criteria, reconstruct, fuse, or directly output the output of the atmospheric data system.

[0086] like Figure 2 The diagram shown is a schematic representation of the atmospheric data fusion voting algorithm for reentry vehicles provided in an embodiment of the present invention. Combined with... Figure 1 and Figure 2 As shown, the implementation steps of the atmospheric data fusion voting algorithm for reentry vehicles provided by this invention will be described in detail. Before performing step 1 above, it is necessary to classify the working modes of the atmospheric data system. It should be noted that the atmospheric data system in this embodiment is an embedded atmospheric data system. This embodiment includes performing the following steps:

[0087] Step one: Classify the working modes of the embedded atmospheric data system, specifically into the following three categories:

[0088] Operating mode 1, normal mode for atmospheric data system;

[0089] In the normal mode of this atmospheric data system, the system outputs atmospheric data based on different maneuvering states. In specific implementation, under high maneuvering and high dynamic conditions, it is necessary to compensate and fuse the airflow angle; under low dynamic conditions, it is necessary to output the atmospheric sensor data normally.

[0090] Operating mode 2, atmospheric data system degradation mode;

[0091] In the downgraded mode of this atmospheric data system, airspeed is measurable but airflow angle is not. In this case, the airflow angle is reconstructed by fusing information from vacuum speed, inertial navigation, and aircraft dynamics model. It should be noted that the airflow angle includes angle of attack and sideslip angle.

[0092] Operating mode 3, atmospheric data system failure mode;

[0093] In this atmospheric data system failure mode, airspeed and airflow angle are unmeasurable; under these circumstances, airspeed and airflow angle are reconstructed by fusing inertial navigation information and aircraft dynamics model.

[0094] It should be noted that the following judgments need to be made during fault handling: if the working mode of the atmospheric data system is 1, then the atmospheric data system fault word is set to valid; if the working mode of the atmospheric data system is 2, then the atmospheric data system fault word is set to downgraded; if the working mode of the atmospheric data system is 3, then the atmospheric data system fault word is set to completely invalid.

[0095] Step 2: Calculate the overload reference value for the atmospheric data system;

[0096] To prevent accidental switching to reconfiguration mode during normal flight, it is necessary to calculate overload reference values. The following explains the specific method for calculating the normal and lateral overload reference values ​​in this step.

[0097] (1) Calculate the normal overload reference value:

[0098] First, the validity of the normal overload reference value is determined. If any of the following conditions are met, the validity flag of the normal overload reference value is set to invalid:

[0099] a) The input angle of attack or Mach number exceeds the data range provided by the aerodynamics department;

[0100] b) The aircraft is in the non-maneuvering mission segment in the longitudinal direction, and the absolute value of the aircraft's pitch rate is greater than the performance value maneuvering pitch rate for three consecutive flight control cycles.

[0101] c) The aircraft is in the maneuvering phase in the longitudinal direction;

[0102] Otherwise, the valid flag for the normal overload reference value is set to valid.

[0103] If the validity flag for the normal overload reference value is valid, the normal overload reference value is calculated as follows:

[0104] The reference value of normal overload is calculated by using the longitudinal aerodynamic model of the aircraft (overload generated by the aircraft body) and the control surface overload compensation (overload generated by the control surface). The longitudinal aerodynamic model of the aircraft body and the control surface overload compensation are both calculated by aerodynamic interpolation tables.

[0105] In this implementation, to reduce airborne computational complexity and storage space, the aerodynamic interpolation table is simplified as follows:

[0106] a) Analysis shows that the sideslip angle has a relatively small impact on the normal overload of the reentry vehicle. To reduce the computational complexity of the airborne system, the normal overload coefficient generated by the longitudinal aerodynamic model of the aircraft can ignore the influence of the sideslip angle and can be obtained directly by interpolation of the angle of attack and Mach number.

[0107] b) Analysis shows that the elevator of the reentry vehicle has strong nonlinearity in normal overload increment when it is positively deflected, but better linearity when it is negatively deflected. To reduce computational complexity, the negative deflection data is simplified to two points: the maximum negative deflection and zero deflection.

[0108] c) Analysis shows that the normal overload coefficient increment of the elevator of the reentry vehicle is similar when Ma is below 0.8. Therefore, the data when Ma is below 0.8 can be combined into an average value.

[0109] The normal overload reference values ​​for the aircraft body and control surfaces are calculated using the above method. The total normal overload reference value is obtained by adding the two values ​​together.

[0110] (2) Calculate the lateral overload reference value:

[0111] First, the validity of the lateral overload reference value is determined. If any of the following conditions are met, the validity flag of the lateral overload reference value is set to invalid:

[0112] a) The input angle of attack, sideslip angle, or Mach number exceeds the data range provided by the aerodynamics department;

[0113] b) The aircraft is in the non-maneuvering mission segment in the lateral direction, and the absolute value of the aircraft's roll rate is greater than the maneuvering roll rate for three consecutive flight control cycles.

[0114] c) The aircraft is in the maneuvering phase laterally.

[0115] Otherwise, the lateral overload reference value validity flag will be set to valid.

[0116] If the validity flag for the lateral overload reference value is valid, the lateral overload reference value is calculated as follows:

[0117] The lateral overload reference value is calculated using the aircraft's lateral aerodynamic model and control surface overload compensation.

[0118] In the specific calculation method, to reduce computational complexity, the lateral aerodynamic model of the aircraft is processed as follows:

[0119] Analysis shows that the lateral overload of the reentry vehicle has a very good linearity with the sideslip angle. Therefore, the change of lateral force with the sideslip angle is simplified to a linear change: CY = CYb * beta. CYb is stored as an interpolation table of angle of attack and Mach number, which can be obtained directly from the interpolation of angle of attack and Mach number. Here, CY is the lateral force aerodynamic coefficient, CYb is the derivative of the lateral force aerodynamic coefficient with the sideslip angle, and beta is the aircraft's sideslip angle.

[0120] The lateral overload reference values ​​of the aircraft body and control surfaces are calculated using the above method. The total lateral overload reference value is obtained by adding the two values ​​together.

[0121] Step 3: Calculate the error criteria for the atmospheric data system;

[0122] This step includes: calculating the tolerable error for normal overload, calculating the tolerable error for lateral overload, and calculating the tolerable error for airspeed. The calculation methods for the above three tolerable errors are explained below.

[0123] (1) Calculate the tolerable error of normal overload;

[0124] In the nonlinear lift region caused by high angles of attack and large control surfaces, misjudgments of the three operating modes mentioned above may occur. Therefore, it is necessary to amplify the absolute value of the tolerable error in the nonlinear lift region. The calculation method for the tolerable error of normal overload is as follows:

[0125] Let Sigma_nz_best be the maximum absolute error of the normal force coefficient in the linear lift region, and Sigma_nz_worst be the maximum absolute error of the normal force coefficient in the stall region. Then, first calculate the average normal force coefficient slope k_avg between the zero lift angle of attack and the stall angle of attack at each Mach number. Then, calculate a normal force coefficient slope k between two adjacent angles of attack. Based on k_avg and k, calculate the nonlinearity of the lift line slope k_non = abs(k - k_avg).

[0126] The nonlinearity of the lift line slope is normalized according to the following formula:

[0127] k_non_01=k_non / (k_non_max-k_non_min);

[0128] After normalization, the maximum error criterion Sigma1 for calculating the normal overload is based on the nonlinearity of the lift line slope obtained after normalization.

[0129] Sigma1=k_non_01*(Sigma_nz_worst-Sigma_nz_best)+Sigma_nz_best;

[0130] The maximum allowable error of normal overload (i.e., the tolerable error of normal overload) Sigma_nz is calculated as follows:

[0131] Sigma_nz = q * S * Sigma1 / mg;

[0132] Where Sigma_nz is the tolerable error of normal overload, S is the aircraft reference area, m is the aircraft mass, g is the gravitational acceleration, and q is the measured dynamic pressure.

[0133] (2) Calculate the tolerable error of lateral overload;

[0134] The lateral overload of a reentry vehicle is about 10 times smaller than the normal overload and is more easily masked by noise. Therefore, it is considered a secondary condition. The calculation method for the tolerable error of the lateral overload is similar to that for the tolerable error of the normal overload, and the calculation method is as follows:

[0135] Let Sigma_ny_best be the maximum absolute error of the lateral force coefficient in the linear lift region, and Sigma_ny_worst be the maximum absolute error of the lateral force coefficient in the stall region. Then, first calculate the average normal force coefficient slope k_avg between the zero lift angle of attack and the stall angle of attack at each Mach number. Then, calculate a normal force coefficient slope k between two adjacent angles of attack. Based on k_avg and k, calculate the nonlinearity of the lift line slope k_non = abs(k - k_avg).

[0136] The nonlinearity of the lift line slope is normalized according to the following formula:

[0137] k_non_01=k_non / (k_non_max-k_non_min);

[0138] After normalization, the maximum error criterion for lateral overload, Sigma2, is calculated based on the nonlinearity of the lift line slope obtained after normalization.

[0139] Sigma2=k_non_01*(Sigma_ny_worst-Sigma_ny_best)+Sigma_ny_best;

[0140] The maximum permissible lateral overload error (i.e., the tolerable lateral overload error) Sigma_ny is calculated as follows:

[0141] Sigma_ny = q * S * Sigma² / mg;

[0142] Where Sigma_ny is the lateral overload tolerance error, S is the aircraft reference area, m is the aircraft mass, g is the gravitational acceleration, and q is the measured dynamic pressure.

[0143] (3) Calculate the tolerable airspeed error;

[0144] Ground speed, vacuum speed, and wind speed can theoretically constitute a combination such as Figure 8 vector triangle, Figure 8 The diagram shown illustrates the vector relationship between ground speed, vacuum speed, and wind speed related to the reentry vehicle in this embodiment of the invention. The air velocity relative to the geographic coordinate system is wind speed V. w The reentry vehicle's speed relative to the air is the vacuum speed V. T The reentry vehicle's velocity relative to the geographic coordinate system is the ground speed V. g Then the vector relationship between ground speed, vacuum speed and wind speed satisfies:

[0145]

[0146] In the above formula, the superscript B represents the vector in the body coordinate system, the superscript N represents the vector in the geographic coordinate system, and S... be The attitude transformation matrix, representing the transformation from the navigation coordinate system to the body coordinate system, can be expressed using the attitude angles of the reentry vehicle as follows:

[0147]

[0148] In the above formula, ψ, θ, These are the aircraft's yaw angle, pitch angle, and roll angle, respectively. According to the above formula, the ground speed V... g and vacuum velocity V T The wind speed V is calculated from the vector difference. w By statistically analyzing the average wind speed of the reentry trajectory, a constant wind speed curve related to altitude and season can be obtained. The tolerable airspeed error is twice the dynamic value (i.e., the constant wind speed curve).

[0149] It should be noted that we assume the ground speed V g Accurate, if the ground speed V g and vacuum velocity V T If the difference between the values ​​is greater than or equal to the tolerable airspeed error, then the aircraft's airspeed is considered to be incorrect, and reconstruction is required. The judgment logic is explained in detail below.

[0150] Step four: Monitor the processing method of the voting logic;

[0151] The monitoring and voting logic for this step includes the following scenarios:

[0152] If the atmospheric data system fault word is "complete failure," which corresponds to operating mode 3 above, then it enters reconstruction mode, i.e., performs reconstruction processing and outputs the reconstructed angle of attack, reconstructed sideslip angle, and reconstructed vacuum velocity; if... Figure 3The diagram shown is a schematic representation of the principle of airflow angle and airspeed reconstruction processing in the atmospheric data fusion voting algorithm for reentry vehicles provided in an embodiment of the present invention.

[0153] Otherwise, if the atmospheric data system fault word is degraded, i.e., operating mode 2 as described above, then vacuum velocity information is added, and the system enters reconstruction mode, i.e., reconstruction processing is performed, and the reconstructed angle of attack, reconstructed sideslip angle, and reconstructed vacuum velocity are output; Figure 3 The reconstruction principle is shown.

[0154] Otherwise, if the atmospheric data system fault word is valid, i.e., it is operating mode 1 as described above, and processing is performed based on the following judgment method:

[0155] If the absolute value of the error between the inertial navigation system's overload signal (including normal and lateral overload signals) and the overload reference value (including normal and lateral overload reference values) is greater than or equal to the corresponding tolerable overload error (including normal and lateral overload tolerable errors), and the absolute value of the error between the vacuum velocity and ground velocity of the atmospheric data system is greater than or equal to the tolerable airspeed error, then the system enters reconstruction mode, i.e., performs reconstruction processing, and outputs the reconstructed angle of attack, reconstructed sideslip angle, and reconstructed vacuum velocity. It should be noted that the criteria for entering reconstruction mode in this step include: the absolute value of the error between the normal overload signal and the normal overload reference value is greater than or equal to the tolerable normal overload error; the absolute value of the error between the lateral overload signal and the lateral overload reference value is greater than or equal to the tolerable lateral overload error; and the absolute value of the error between the vacuum velocity and ground velocity of the atmospheric data system is greater than or equal to the tolerable airspeed error.

[0156] Otherwise, if the absolute value of the error between the inertial navigation overload signal (including normal overload signal and lateral overload signal) and the overload reference value (including normal overload reference value and lateral overload reference value) is less than the corresponding tolerable overload error (including tolerable normal overload error and tolerable lateral overload error), and the absolute value of the error between the vacuum velocity and ground velocity of the atmospheric data system is less than the tolerable airspeed error; the judgment in this step includes: the absolute value of the error between the normal overload signal and the normal overload reference value is less than the tolerable normal overload error, and the absolute value of the error between the lateral overload signal and the lateral overload reference value is less than the tolerable lateral overload error, and the absolute value of the error between the vacuum velocity and ground velocity of the atmospheric data system is less than the tolerable airspeed error; then the judgment is executed according to the following logic: when the angle of attack is less than a certain value, the airflow angle of the atmospheric data system is directly output (i.e., the output mode under low dynamic conditions); otherwise, the fusion mode is entered, that is, fusion processing is performed, and the airflow angle (including angle of attack and sideslip angle) after information fusion is output.

[0157] Otherwise, directly output the airflow angle of the atmospheric data system. For example... Figure 4The diagram shown illustrates the principle of airflow angle fusion processing in the atmospheric data fusion voting algorithm for reentry vehicles provided in this embodiment of the invention. It should be noted that this logic means that if the aircraft is in a normal, stable mission, the airflow angle is not fused; however, when the aircraft is performing high-maneuver missions, the airflow angle is fused to improve its accuracy.

[0158] The following explains how the reconstruction and fusion processes are performed in step four.

[0159] 4.1 Reconstruction of airflow angle and airspeed;

[0160] like Figure 3 As shown, the airflow angle reconstruction processing method comprehensively utilizes the control outputs of the inertial navigation system and flight control system actuators (such as throttle opening and deflection angle of each control surface) and vacuum speed information, and combines them with the aircraft dynamics model to construct a Kalman filter to achieve the reconstruction processing of airflow angle and vacuum speed.

[0161] The Kalman filter algorithm has a relatively mature solution process; its key lies in the selection of state and measurement variables. Based on the analysis of the sensor performance of this type of aircraft, using the pitch angle, roll angle, roll rate, pitch rate, yaw rate, and the three-axis accelerations of the airframe provided by the inertial navigation system as measurement variables, the measurement variable is Z=[θφp qra x a y a z ] T Based on the principle of closure of the state equation, the pitch angle, roll angle, roll rate, pitch rate, yaw rate, angle of attack, sideslip angle, and vacuum speed of the aircraft are selected as state variables, i.e., X=[θ,φ,p,q,r,α,β,V], which are the reconstructed data, and the reconstructed angle of attack, sideslip angle, and vacuum speed are output.

[0162] 4.2, Methods to improve airflow angle fusion performance, i.e., airflow angle fusion processing;

[0163] like Figure 4 As shown, a Kalman filter is constructed using an inertial navigation system and an atmospheric data system, combined with the aircraft's kinematic equations, to compensate for airflow angles during high angles of attack and high-maneuverability flight. The output and input definitions of the state equation and measurement equation in the Kalman filter are as follows:

[0164] The measured quantities are the three-axis ground speed, pitch angle, roll angle, yaw angle, three-axis angular velocity, and three-axis angular acceleration provided by the inertial navigation system, and the angle of attack, sideslip angle, and airspeed provided by the air data system. The measured quantities are the aircraft's three-axis airspeed, pitch angle, roll angle, yaw angle, angle of attack, sideslip angle, airspeed, and three-axis wind speed. The angle of attack and sideslip angle are taken as outputs. The signal flow graph is as follows: Figure 4 As shown.

[0165] Step 5: Verify the atmospheric data fusion voting algorithm;

[0166] To verify the performance of airflow angle fusion and the reconstructed airflow angle and vacuum velocity, it is necessary to perform flight trajectory simulation in conjunction with the aircraft model, generate ideal navigation, flight control, and atmospheric data, set sensor noise intensity, and implement and analyze the performance of the airflow angle fusion algorithm under different flight conditions.

[0167] 1) Parameter settings, i.e., setting the input data used for verification;

[0168] a) Wind field setup

[0169] To verify the effectiveness of the designed filter, wind field simulation is required. Based on statistical data, wind can be considered a combination of three independent zero-mean stochastic processes: random constant wind, low-frequency wind, and high-frequency wind. Random constant wind is described using a random constant model, while low-frequency wind is described using the Dryden turbulence model. The wind field is approximated as a combination of constant wind and turbulence.

[0170] b) EKF (Kalman Filter) Parameter Settings

[0171] The noise characteristics of multi-source sensor information are captured, including the noise standard deviation of angular velocity, attitude angle, acceleration, airspeed, angle of attack, sideslip angle, and ground speed, and the system noise matrix Q and measurement noise matrix R of the filter are calculated.

[0172] c) Error adjustment settings

[0173] Models with aerodynamic forces, center of mass, atmospheric density, and moment of inertia are set up for deflection, and Monte Carlo target shooting is performed using uniform or normal distributions.

[0174] 2) Verification of the atmospheric data fusion voting algorithm

[0175] The verification was conducted from both accuracy and latency perspectives, specifically verifying the accuracy and latency of the output results from the fusion and reconstruction processes:

[0176] a) Conduct Monte Carlo shooting tests and set aerodynamic parameters to verify the accuracy and robustness of the algorithm.

[0177] b) Select typical 1Hz, 2Hz, and 3Hz sinusoidal overload command signals and overlay them onto the closed-loop overload command to test the time delay characteristics of the estimated airflow angle.

[0178] This invention provides an atmospheric data fusion voting algorithm for reentry vehicles. By analyzing the characteristics of reentry vehicles, a fusion voting algorithm based on overload reference values ​​and error criteria of the atmospheric data system is designed and applicable to this type of vehicle. The algorithm first calculates the overload reference values ​​and error criteria of the atmospheric data system. The overload reference values ​​include normal overload reference values ​​and lateral overload reference values, and the corresponding error criteria include tolerable normal overload error, tolerable lateral overload error, and tolerable airspeed error. The fusion voting algorithm processes the atmospheric data system's output in three ways: based on monitoring the current operating mode of the atmospheric data system (including three operating modes), combined with the calculated overload reference values ​​and error criteria, the output of the atmospheric data system is processed according to the fusion voting logic. This results in three output forms: output after reconstruction processing, output after fusion processing, or direct output. Using the atmospheric data fusion voting algorithm provided in this invention effectively reduces the computational load of the aircraft aerodynamic model and enhances the airborne feasibility of the fusion voting algorithm. In particular, the technical solution provided by the embodiments of the present invention can vote on the output of the atmospheric data system based on the working mode of the atmospheric data system, and output the optimal atmospheric data under various conditions.

[0179] Based on the atmospheric data fusion voting algorithm for reentry vehicles provided in the above embodiments of the present invention, the present invention also provides a reentry vehicle equipped with an atmospheric data fusion voting processor, a memory, an inertial navigation system, a flight control system, and an atmospheric data system.

[0180] The inertial navigation system is configured to provide inertial navigation information to the atmospheric data fusion voting processor.

[0181] The flight control system is configured to provide control outputs of its actuators to the atmospheric data fusion voting processor;

[0182] An atmospheric data system is configured to provide vacuum velocity, angle of attack, and sideslip angle to the atmospheric data fusion voting processor;

[0183] The memory is configured to hold executable instructions;

[0184] The atmospheric data fusion voting processor in this embodiment of the invention is specifically configured to implement the atmospheric data fusion voting algorithm for reentry vehicles provided in any of the above embodiments when executing executable instructions stored in the memory.

[0185] The following specific embodiments illustrate the atmospheric data fusion voting algorithm for reentry vehicles and the implementation methods of reentry vehicles provided by the present invention.

[0186] In this embodiment of the invention, taking a certain type of reentry vehicle as an example, the sensor noise is set as follows:

[0187] Sensor measurement signal Measurement noise standard deviation unit angular velocity 0.02 deg / s attitude angle 0.02 deg acceleration 0.02 <![CDATA[m / s 2 ]]> airspeed 0.1 m / s Angle of attack 0.2 deg Sideslip angle 0.2 deg Ground speed 0.05 m / s

[0188] The wind field settings are as follows: turbulence scale is set to 553.4m, turbulence intensity is set to 1.5m / s, such as... Figure 5 The diagram shown is a schematic of the constant wind field setting in the atmospheric data fusion voting algorithm for reentry vehicles provided in an embodiment of the present invention, which is set as the landing wind field in August.

[0189] Example 1

[0190] Example 1 of the monitoring voting logic at a certain moment of switching to reconstruction mode is: the atmospheric data system fault word is completely failed, so it directly switches to reconstruction mode without calculating overload reference value and other data.

[0191] Example 2

[0192] Example 2 shows the monitoring and voting logic at a certain moment when switching to reconstruction mode: the atmospheric data system is valid.

[0193] The handover criterion is calculated as follows:

[0194] The overload reference value is calculated in step two: taking the calculation at a certain moment as an example, the Mach number at a certain moment is 0.6, the angle of attack is 5 degrees, the sideslip angle is 2 degrees, the elevator surface is -3 degrees, and the rudder surface is 0.5 degrees.

[0195] The normal overload reference values ​​and lateral overload reference values ​​of the aircraft are calculated as follows:

[0196] 1) Based on the aircraft body characteristic interpolation table (inputs are angle of attack, sideslip, and Mach number), the interpolation table is simplified according to the method of ignoring sideslip angle in step 2, and the normal force coefficient is obtained as 0.62.

[0197] 2) The normal force coefficient is -0.023 obtained from the aircraft control surface characteristic interpolation table (inputs are angle of attack, elevator, and Mach number);

[0198] 3) The lateral force coefficient is 0.02 obtained from the aircraft body characteristic interpolation table (inputs are angle of attack, sideslip, and Mach number);

[0199] 4) The lateral force coefficient is 0.001 obtained from the aircraft control surface characteristic interpolation table (input is angle of attack, rudder, Mach number).

[0200] The aircraft reference area is 6.2, the dynamic pressure is 10280, and the aircraft mass is 4028kg. At this time, the calculated normal overload reference value is 0.9629, while the measured normal overload of the inertial navigation system is 1.0150. The lateral overload reference value is 0.0338, while the measured overload is -0.0112.

[0201] Step 3 calculates that the tolerable error for normal overload at this moment is 0.052, the tolerable error for lateral overload is 0.034, and the tolerable error for airspeed is 50 m / s.

[0202] Step four leads to the monitoring and voting logic. The measured vacuum speed is 162 m / s and the measured ground speed is 111 m / s. Both the overload error and the speed error are greater than the error limit, so the airflow angle reconstruction mode is entered.

[0203] Example 3

[0204] Example 3 shows the monitoring and voting logic at a certain moment when switching to fusion mode:

[0205] The atmospheric data system fault word is fully valid. The angle of attack measurement value is 19 degrees, which is greater than the threshold of 18 degrees. Therefore, it directly switches to fusion mode.

[0206] The verification examples after switching to reconstruction mode and the verification examples after switching to fusion mode in the above embodiments of the present invention are as follows:

[0207] Simulation Implementation Example 1:

[0208] This simulation implementation example 1 verifies the reconstruction processing of airflow angles. Based on logical judgment, after determining that the atmospheric data system has failed, the reconstruction process begins, such as... Figure 6 The diagram shown is a schematic representation of the simulation accuracy results of Monte Carlo target shooting in Example 1 of the simulation implementation of the present invention. Figure 6 Figure a shows the simulation results for the angle of attack, and Figure b shows the simulation results for the sideslip angle.

[0209] Simulation results show that the angle of attack estimation error of the airflow angle estimation algorithm based on extended Kalman filter is within 0.7°, the sideslip angle estimation error is within 1.1°, and the vacuum velocity estimation error is within 35 m / s. However, due to the influence of the initial state value of the extended Kalman filter algorithm, the initial estimation error is relatively large. Subsequently, after correction by the inertial navigation sensor signal used as the measurement signal, the angle of attack and sideslip angle quickly converge to within 0.3° of the true value, meeting the expected value indicators and engineering application requirements. In the frequency domain, the airflow angle delay is less than 10 ms.

[0210] Simulation Implementation Example 2:

[0211] This simulation implementation example 2 verifies the fusion processing of airflow angles, specifically performing time delay verification, such as... Figures 7a to 7cThe image shown is a schematic diagram illustrating the simulation results of frequency domain delay evaluation for airflow angle fusion in Simulation Implementation Example 2 of this invention. Figure 7a , Figure 7b and Figure 7c The frequency domain evaluation curves are for angle of attack reconstruction at 1Hz, 2Hz, and 3Hz, respectively.

[0212] Simulation results show that, under bias and interference, the angle of attack and sideslip angle estimation errors of the airflow angle compensation algorithm based on two-step extended Kalman filtering are within 0.4°, and the vacuum velocity estimation error is within 15 m / s, which meets the expected value indicators and engineering application requirements.

[0213] While the embodiments disclosed in this invention are as described above, they are merely illustrative of the embodiments to facilitate understanding of the invention and are not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and variations in the form and details of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection for this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A voting algorithm for atmospheric data fusion in reentry vehicles, characterized in that, include: Step 1: Calculate the overload reference value of the atmospheric data system, which includes: normal overload reference value and lateral overload reference value; Step 2: Calculate the error criteria of the atmospheric data system. The error criteria include: normal overload tolerance error, lateral overload tolerance error, and airspeed tolerance error. Step 3: Based on the current working mode of the atmospheric data monitoring system, and in conjunction with the overload reference value and the error criterion, reconstruct, fuse, or directly output the output of the atmospheric data system. The working modes include: Working mode 1, normal mode; Operating mode 2, degraded mode, airspeed is measurable, airflow angle is not measurable; Operating mode 3, failure mode, airspeed and airflow angle are unmeasurable; And based on the working mode, the atmospheric data system fault word is set to valid, downgraded, or completely failed; Step 2 includes: Step 21: Calculate the average normal force coefficient slope k_avg between the zero lift angle of attack and the stall angle of attack at each Mach number. Calculate a normal force coefficient slope k between two adjacent angles of attack. Calculate the nonlinearity of the lift line slope based on k_avg and k. Step 22: Calculate the maximum error criterion based on the nonlinearity of the lift line slope obtained by normalization; Step 23: Calculate the tolerable error of normal overload and the tolerable error of lateral overload; Step 24: Construct the vector relationship between the ground speed, vacuum speed and wind speed of the reentry vehicle; Step 25: Use the attitude angles of the reentry vehicle to obtain the attitude transformation matrix; Step 26, based on the relationships in steps 24 and 25, through the ground speed V g and vacuum velocity V T The wind speed V is calculated from the vector difference. w By statistically analyzing the average wind speed of the reentry trajectory, a constant wind speed curve related to altitude and season is obtained; the tolerable airspeed error is twice that of the constant wind speed curve. Step 3 includes: When the atmospheric data system fault word is completely failed, a reconstruction process is performed, and the reconstructed angle of attack, reconstructed sideslip angle, and reconstructed vacuum velocity are output. When the fault word of the atmospheric data system is degraded, vacuum velocity information is added, and reconstruction processing is performed to output the reconstructed angle of attack, reconstructed sideslip angle, and reconstructed vacuum velocity. If the atmospheric data system fault word is valid, then perform the following judgment: If the absolute value of the error between the normal overload signal and the normal overload reference value of the inertial navigation system is greater than or equal to the tolerable error of the normal overload, and the absolute value of the error between the lateral overload signal and the lateral overload reference value is greater than or equal to the tolerable error of the lateral overload, and the absolute value of the error between the vacuum velocity and the ground velocity of the atmospheric data system is greater than or equal to the tolerable error of the airspeed, then the reconstruction process is performed, and the reconstructed angle of attack, reconstructed sideslip angle, and reconstructed vacuum velocity are output. Otherwise, if the absolute value of the error between the inertial navigation normal overload signal and the normal overload reference value is less than the tolerable error of normal overload, and the absolute value of the error between the lateral overload signal and the lateral overload reference value is less than the tolerable error of lateral overload, and the absolute value of the error between the vacuum velocity and the ground velocity of the atmospheric data system is less than the tolerable error of airspeed, then the following logical judgment is executed: when the angle of attack is less than the preset value, the airflow angle of the atmospheric data system is directly output; otherwise, fusion processing is performed, and the airflow angle after information fusion is output. Otherwise, directly output the airflow angle of the atmospheric data system.

2. The atmospheric data fusion voting algorithm for reentry vehicles according to claim 1, characterized in that, Before step 1, the following are also included: Step A involves classifying the operating modes of the atmospheric data system into the following categories: Operating mode 1, normal mode of atmospheric data system; under this operating mode 1, the output of atmospheric data system is generated based on different maneuvering states; Operating mode 2, atmospheric data system downgrade mode; in this operating mode 2, airspeed is measurable, but airflow angle is not measurable; by fusing information from vacuum speed, inertial navigation, and aircraft dynamics model, the airflow angle is reconstructed, which includes angle of attack and sideslip angle; Operating mode 3, air data system failure mode; in this operating mode 3, airspeed and airflow angle are unmeasurable; by fusing inertial navigation information and aircraft dynamics model, airspeed and airflow angle are reconstructed; In troubleshooting, determining the operating mode of the atmospheric data system includes: If the working mode of the atmospheric data system is 1, then the atmospheric data system fault word is set to valid; If the atmospheric data system is operating in mode 2, then set the atmospheric data system fault word to downgraded. If the working mode of the atmospheric data system is 3, then the atmospheric data system fault word is set to complete failure.

3. The atmospheric data fusion voting algorithm for reentry vehicles according to claim 2, characterized in that, The method for calculating the normal overload reference value in step 1 includes: S11, determine the validity of the normal overload reference value, and identify whether the normal overload reference value is valid or invalid: S12, when the valid flag of the normal overload reference value is valid, the normal overload reference value is calculated by using the aircraft longitudinal body aerodynamic model and the control surface overload compensation respectively. The aircraft longitudinal body aerodynamic model and the control surface overload compensation are both calculated by aerodynamic interpolation table. S13, the normal overload reference values ​​calculated separately by the aircraft longitudinal aerodynamic model and the control surface overload compensation are added together to obtain the total normal overload reference value.

4. The atmospheric data fusion voting algorithm for reentry vehicles according to claim 2, characterized in that, The method for calculating the lateral overload reference value in step 1 includes: S21, determine the validity of the lateral overload reference value, and identify whether the lateral overload reference value is valid or invalid: S22, when the valid flag of the lateral overload reference value is valid, the lateral overload reference value is calculated separately using the aircraft lateral body aerodynamic model and the control surface overload compensation; wherein, the aircraft lateral body aerodynamic model and the control surface overload compensation are both calculated by aerodynamic interpolation tables; S23, the lateral overload reference values ​​calculated separately by the aircraft lateral aerodynamic model and the control surface overload compensation are added together to obtain the total lateral overload reference value.

5. The atmospheric data fusion voting algorithm for reentry vehicles according to claim 2, characterized in that, The methods for calculating the tolerable error of normal overload and the tolerable error of lateral overload in step 2 include: Step 21: Calculate the average normal force coefficient slope k_avg of the aircraft from zero lift angle of attack to stall angle of attack at each Mach number. Calculate a normal force coefficient slope k between two adjacent angles of attack. Calculate the lift sloping nonlinearity based on k_avg and k: k_non = abs(k - k_avg); and normalize the lift sloping nonlinearity to obtain: k_non_01 = k_non / (k_non_max - k_non_min); Step 22, calculate the maximum error criterion based on the nonlinearity of the lift line slope obtained by normalization, including: The maximum error criterion for calculating normal overload, Sigma1, is: Sigma1 = k_non_01 * (Sigma_nz _worst - Sigma_nz _best) + Sigma_nz_best; The Sigma2 criterion for calculating the maximum error of lateral overload is: Sigma2 = k_non_01 * (Sigma_ny _worst - Sigma_ny _best) + Sigma_ny_best; Wherein, Sigma_nz_best is the maximum absolute error of the normal force coefficient in the linear lift region, and Sigma_nz_worst is the maximum absolute error of the normal force coefficient in the stall region; Sigma_ny_best is the maximum absolute error of the lateral force coefficient in the linear lift region, and Sigma_ny_worst is the maximum absolute error of the lateral force coefficient in the stall region. Step 23, calculate the tolerable error of normal overload and the tolerable error of lateral overload: Sigma_nz = q * S * Sigma1 / mg; Sigma_ny = q * S * Sigma2 / mg; Where Sigma_nz is the tolerable error of normal overload, Sigma_ny is the tolerable error of lateral overload, S is the aircraft reference area, m is the aircraft mass, g is the gravitational acceleration, and q is the measured dynamic pressure.

6. The atmospheric data fusion voting algorithm for reentry vehicles according to claim 5, characterized in that, Step 2, calculating the tolerable airspeed error, includes: Step 24, construct the vector relationship between the ground speed, vacuum speed, and wind speed of the reentry vehicle as follows: ； Among them, wind speed V w V is the velocity of air relative to the geographic coordinate system, or the velocity in a vacuum. T The ground speed V is the velocity of the reentry vehicle relative to the air. g S represents the velocity of the reentry vehicle relative to the geographic coordinate system; the superscript B indicates the vector in the body coordinate system. be It is represented as the attitude transformation matrix from the navigation coordinate system to the body coordinate system, and the superscript N indicates the vector in the geographic coordinate system; Step 25: Using the attitude angles of the reentry vehicle, the attitude transformation matrix is ​​obtained as follows: ； Where ψ, θ, and φ are the aircraft's yaw angle, pitch angle, and roll angle, respectively; Step 26, based on the relationships in steps 24 and 25, through the ground speed V g and vacuum velocity V T The wind speed V is calculated from the vector difference. w By statistically analyzing the average wind speed of the reentry trajectory, a constant wind speed curve related to altitude and season is obtained; the tolerable airspeed error is twice that of the constant wind speed curve.

7. The atmospheric data fusion voting algorithm for reentry vehicles according to any one of claims 1 to 6, characterized in that, The reconstruction process in step 3 is performed as follows: By utilizing the control outputs of the inertial navigation system and flight control system actuators, along with vacuum speed information, and combining them with the aircraft dynamics model, a Kalman filter is constructed to reconstruct the airflow angle and vacuum speed. Specifically, the pitch angle, roll angle, roll rate, pitch rate, yaw rate, and three-axis accelerations of the airframe provided by the inertial navigation system are used as measurement quantities. The pitch angle, roll angle, roll rate, pitch rate, yaw rate, angle of attack, sideslip angle, and vacuum speed of the aircraft are selected as state quantities, and the reconstructed angle of attack, sideslip angle, and vacuum speed are output. The fusion process in step 3 is performed as follows: By utilizing an inertial navigation system and an air data system, combined with an aircraft dynamics model, a Kalman filter is constructed to compensate for airflow angles during high angles of attack and high-maneuverability flight. Specifically, the three-axis ground speed, pitch angle, roll angle, yaw angle, three-axis angular velocity, and three-axis angular acceleration provided by the inertial navigation system, and the angle of attack, sideslip angle, and vacuum speed provided by the air data system are used as measurement quantities. The three-axis airspeed, pitch angle, roll angle, yaw angle, angle of attack, sideslip angle, vacuum speed, and three-axis wind speed of the aircraft are used as state quantities, and the reconstructed angle of attack and sideslip angle are output.

8. The atmospheric data fusion voting algorithm for reentry vehicles according to claim 1, characterized in that, Step 3 is followed by: Step 4: Verify the atmospheric data output by the atmospheric data system after applying the atmospheric data fusion voting algorithm, including: S41, Parameter Settings, including: Wind Field Settings, Kalman Filter Parameter Settings, Error Adjustment Settings; S42, Algorithm verification, including: verifying the accuracy and latency of the output results of the fusion processing and reconstruction processing from the perspectives of accuracy and latency.

9. A reentry vehicle, characterized in that, The reentry vehicle is equipped with an atmospheric data fusion voting processor, a memory, as well as an inertial navigation system, a flight control system, and an atmospheric data system; The inertial navigation system is configured to provide inertial navigation information to the atmospheric data fusion voting processor; The flight control system is configured to provide the atmospheric data fusion voting processor with control outputs of its actuators; The atmospheric data system is configured to provide vacuum velocity, angle of attack, and sideslip angle to the atmospheric data fusion voting processor; The memory is configured to store executable instructions; The atmospheric data fusion voting processor is specifically configured to implement the atmospheric data fusion voting algorithm of the reentry vehicle as described in any one of claims 1 to 8 when executing the executable instructions stored in the memory.

Images