Aircraft hover throttle prediction method, system, aircraft, medium, and program product

Abstract

This application provides a method, system, aircraft, medium, and program product for predicting hover throttle of an aircraft, relating to the field of aircraft technology. The method includes: acquiring the current operating data of the aircraft; wherein the current operating data includes the current vertical motion throttle, current flight attitude, current vertical velocity, and current state of charge; inputting the current vertical motion throttle, current flight attitude, and current vertical velocity into a pre-built payload prediction model to obtain the predicted payload weight of the aircraft; and inputting the current state of charge and the predicted payload weight into a pre-built hover throttle prediction model to obtain the predicted hover throttle of the aircraft. This application embodiment can achieve the technical effect of quickly and accurately predicting the hover throttle of an aircraft.

Description

Technical Field

[0001] This application relates to the field of aircraft technology, and more specifically, to an aircraft hovering throttle prediction method, system, aircraft, medium, and program product. Background Technology

[0002] In the flight control process of aircraft, it is often necessary to predict the hover throttle of the aircraft based on its payload weight. However, currently, most methods rely on external sensors (such as force gauges or weighing platforms) to detect the payload weight of the aircraft, or on technicians manually calibrating the payload weight based on their experience. This has problems such as high cost, complex integration, and inability to adaptively respond to load changes in real time. It is difficult to adaptively estimate the payload weight of the aircraft in real-world application scenarios to accurately predict the hover throttle of the aircraft. Summary of the Invention

[0003] The purpose of this application is to provide a method, system, aircraft, medium, and program product for predicting hover throttle of an aircraft, so as to achieve the technical effect of quickly and accurately predicting the hover throttle of an aircraft.

[0004] In a first aspect, embodiments of this application provide a method for predicting the hovering throttle of an aircraft, including:

[0005] Acquire the current operating data of the aircraft; wherein, the current operating data includes the current vertical motion throttle, current flight attitude, current vertical speed, and current state of charge;

[0006] The current vertical motion throttle, the current flight attitude, and the current vertical speed are input into a pre-built payload prediction model to obtain the predicted payload of the aircraft.

[0007] The current state of charge and the predicted payload weight are input into a pre-built hover throttle prediction model to obtain the predicted hover throttle of the aircraft.

[0008] In the above implementation process, by pre-constructing a payload prediction model and a hover throttle prediction model, a series of current operating data of the aircraft, such as the current vertical motion throttle, current flight attitude, current vertical speed, and current state of charge, are obtained. The current vertical motion throttle, current flight attitude, and current vertical speed are input into the payload prediction model to obtain the predicted payload weight of the aircraft. The current state of charge and the predicted payload weight are input into the hover throttle prediction model to obtain the predicted hover throttle of the aircraft. It is possible to directly and adaptively estimate the payload weight of the aircraft based on the current operating data of the aircraft without relying on external sensors or manual calibration, thereby quickly and accurately predicting the hover throttle of the aircraft.

[0009] Furthermore, the load weight prediction model is used to perform the following operations:

[0010] Based on the vertical motion dynamics model of the aircraft, the throttle command parameters of the aircraft are determined according to the current vertical motion throttle and the current flight attitude, as well as the no-load hover throttle and no-load weight of the aircraft.

[0011] Based on the vertical velocity model and segmented drag model of the aircraft, and according to the throttle command parameters and the current vertical velocity, an adaptive update law for the load estimation parameters in the vertical velocity observer is constructed.

[0012] The adaptive update law is used to iteratively update the load estimation parameters until the error converges and the latest load estimation parameters are obtained.

[0013] The predicted load weight is obtained by back-calculating based on the latest load estimation parameters.

[0014] Furthermore, based on the vertical motion dynamics model of the aircraft, the throttle command parameters of the aircraft are determined according to the current vertical motion throttle and the current flight attitude, as well as the aircraft's no-load hover throttle and no-load weight, including:

[0015] Based on the current vertical motion throttle and the current flight attitude, determine the vertical component of the current vertical motion throttle;

[0016] Substituting the vertical component of the current vertical motion throttle, the no-load hover throttle, and the no-load weight into the vertical motion dynamics model, the throttle command parameters are obtained.

[0017] Furthermore, the no-load hover throttle is obtained by fitting the charged state of the aircraft in the no-load state with the hover throttle.

[0018] In the above implementation process, the following steps are taken using a payload prediction model: Based on the aircraft's vertical motion dynamics model, the throttle command parameters of the aircraft are determined according to the current vertical motion throttle, current flight attitude, and the aircraft's empty hover throttle and empty weight; Based on the aircraft's vertical velocity model and segmented drag model, an adaptive update law for the payload estimation parameters in the vertical velocity observer is constructed according to the throttle command parameters and the current vertical velocity; The payload estimation parameters are iteratively updated using the adaptive update law until the error converges to obtain the latest payload estimation parameters; The predicted payload weight is obtained by back-calculation based on the latest payload estimation parameters, which can directly and adaptively estimate the aircraft's payload weight based on the aircraft's current operating data.

[0019] Furthermore, the hovering throttle prediction model is constructed by performing the following operations:

[0020] Obtain a training dataset; wherein the training dataset includes multiple sets of training data, and each set of training data includes a calibrated state of charge, a calibrated load weight, and a calibrated hover throttle.

[0021] The hovering throttle prediction model is obtained by training a pre-established neural network model using a training dataset.

[0022] Furthermore, before inputting the current state of charge and the predicted payload weight into the pre-built hover throttle prediction model to obtain the predicted hover throttle of the aircraft, the method further includes:

[0023] The current state of charge is normalized based on the maximum state of charge of the aircraft;

[0024] The current payload weight is normalized based on the maximum payload weight of the aircraft.

[0025] In the above implementation process, by acquiring a training dataset including multiple sets of training data such as "calibrated state of charge - calibrated load weight - calibrated hover throttle", a pre-established neural network model is trained using the training dataset to obtain a hover throttle prediction model, which can ensure that the hover throttle prediction model can accurately predict the hover throttle based on the aircraft's state of charge and load weight.

[0026] Furthermore, the method also includes:

[0027] The statistical prediction of the hover throttle dataset corresponds to the operational quality index; wherein, each predicted hover throttle in the predicted hover throttle dataset is the predicted hover throttle obtained at a sampling moment within the sampling period;

[0028] If the operational quality indicators meet the indicator value conditions, the hovering throttle prediction model is optimized based on the predicted hovering throttle dataset.

[0029] Furthermore, the operational quality indicators include the aircraft's maximum flight altitude difference, maximum flight speed, and standard deviation of predicted hover throttle during the sampling period;

[0030] The conditions for determining the value of the indicator include:

[0031] The maximum flight altitude difference is less than or equal to the altitude deviation threshold;

[0032] The maximum flight speed is less than or equal to the flight speed threshold; and...

[0033] The predicted hover throttle standard deviation is less than or equal to the throttle standard deviation threshold.

[0034] In the above implementation process, the predicted hovering throttle obtained at each sampling time during the sampling period is collected as the predicted hovering throttle dataset. The corresponding operation quality indicators of the predicted hovering throttle dataset are statistically analyzed. When the operation quality indicators meet the indicator value conditions, the hovering throttle prediction model is optimized based on the predicted hovering throttle dataset. The hovering throttle prediction model can be continuously optimized based on high-quality predicted hovering throttle related data, further ensuring that the hovering throttle prediction model can accurately predict the hovering throttle based on the aircraft's state of charge and payload weight.

[0035] Secondly, embodiments of this application provide an aircraft hovering throttle prediction system, comprising:

[0036] The operational data acquisition module is used to acquire the current operational data of the aircraft; wherein, the current operational data includes the current vertical motion throttle, current flight attitude, current vertical speed, and current state of charge;

[0037] An adaptive load prediction module is used to input the current vertical motion throttle, the current flight attitude, and the current vertical speed into a pre-built load prediction model to obtain the predicted load weight of the aircraft.

[0038] The hovering throttle prediction module is used to input the current state of charge and the predicted load weight into a pre-built hovering throttle prediction model to obtain the predicted hovering throttle of the aircraft.

[0039] Thirdly, embodiments of this application provide an aircraft including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor; the processor executes the computer program to implement the method described above.

[0040] Fourthly, embodiments of this application provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program; wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the method described above.

[0041] Fifthly, embodiments of this application provide a computer program product, the computer program product including instructions, which, when executed by a computer, cause the computer to perform the method described above. Attached Figure Description

[0042] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 A flowchart illustrating an aircraft hovering throttle prediction method provided in the first embodiment of this application;

[0044] Figure 2 A schematic diagram of the structure of an aircraft hovering throttle prediction system provided in the second embodiment of this application;

[0045] Figure 3 This is a schematic diagram of the structure of an aircraft provided in the third embodiment of this application. Detailed Implementation

[0046] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0047] It should be noted that in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance. Furthermore, the step numbers in the text are only for the convenience of explaining the embodiments of this application and are not intended to limit the order in which the steps are performed.

[0048] In related technologies, during the flight control process of aircraft, it is often necessary to predict the hovering throttle of the aircraft based on its payload weight. However, currently, most methods rely on external sensors (such as force gauges or weighing platforms) to detect the payload weight of the aircraft, or on technicians manually calibrating the payload weight based on their experience. This approach suffers from problems such as high cost, complex integration, and inability to adaptively respond to load changes in real time. Consequently, it is difficult to adaptively estimate the payload weight of the aircraft in practical application scenarios to accurately predict the hovering throttle of the aircraft.

[0049] To address this, this application proposes a method for predicting hover throttle for aircraft. By pre-constructing a payload prediction model and a hover throttle prediction model, a series of current operational data of the aircraft, including the current vertical motion throttle, current flight attitude, current vertical velocity, and current state of charge, are obtained. The current vertical motion throttle, current flight attitude, and current vertical velocity are input into the payload prediction model to obtain the predicted payload weight of the aircraft. The current state of charge and the predicted payload weight are input into the hover throttle prediction model to obtain the predicted hover throttle of the aircraft. This method can adaptively estimate the payload weight of the aircraft directly based on the current operational data without relying on external sensors or manual calibration, thereby quickly and accurately predicting the hover throttle of the aircraft.

[0050] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0051] The methods provided in this application can be executed by relevant terminal devices, and the following description uses a server as the execution subject.

[0052] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a hovering throttle prediction method for an aircraft, as provided in the first embodiment of this application. The hovering throttle prediction method includes steps S101-S103:

[0053] S101. Obtain the current operating data of the aircraft; wherein, the current operating data includes the current vertical motion throttle, current flight attitude, current vertical speed and current state of charge.

[0054] As an example, the sampling period can be preset according to the actual application requirements.

[0055] After the sampling time arrives, the current operating data of the aircraft is acquired, including the current vertical motion throttle, current flight attitude, current vertical speed, and current state of charge.

[0056] It should be noted that the current vertical motion throttle refers to the actual output throttle used to drive vertical motion at the sampling time, which is the actual output throttle of each motor on the aircraft. The sum of all values. Current flight attitude refers to the aircraft's attitude data at the sampling moment, including pitch angle. and roll angle Current vertical velocity refers to the aircraft's vertical velocity at the sampling time, which is the vertical component of the flight velocity. Current state of charge refers to the aircraft's battery state of charge (SOC) at the sampling time.

[0057] In practical applications, aircraft can include multi-rotor drones.

[0058] S102. Input the current vertical motion throttle, current flight attitude, and current vertical speed into the pre-built payload prediction model to obtain the predicted payload of the aircraft.

[0059] As an example, after obtaining the current vertical motion throttle, current flight attitude, and current vertical speed, a pre-built payload prediction model is invoked. The current vertical motion throttle, current flight attitude, and current vertical speed are input into the payload prediction model, and the payload prediction model is used to predict the payload of the aircraft based on the current vertical motion throttle, current flight attitude, and current vertical speed, thereby obtaining the predicted payload of the aircraft.

[0060] S103. Input the current state of charge and the predicted payload weight into the pre-built hover throttle prediction model to obtain the predicted hover throttle of the aircraft.

[0061] As an example, after obtaining the current state of charge and the predicted payload weight, a pre-built hover throttle prediction model is invoked. The current state of charge and the predicted payload weight are input into the hover throttle prediction model, and the hover throttle prediction model is used to predict the hover throttle of the aircraft based on the current state of charge and the predicted payload weight, thereby obtaining the predicted hover throttle of the aircraft.

[0062] This application embodiment pre-constructs a payload prediction model and a hover throttle prediction model to obtain a series of current operating data of the aircraft, including the current vertical motion throttle, current flight attitude, current vertical speed, and current state of charge. The current vertical motion throttle, current flight attitude, and current vertical speed are input into the payload prediction model to obtain the predicted payload weight of the aircraft. The current state of charge and the predicted payload weight are input into the hover throttle prediction model to obtain the predicted hover throttle of the aircraft. This allows for the adaptive estimation of the aircraft's payload weight directly based on the aircraft's current operating data without relying on external sensors or manual calibration, thereby quickly and accurately predicting the hover throttle of the aircraft.

[0063] In an optional embodiment, the payload weight prediction model is used to perform the following operations: based on the aircraft's vertical motion dynamics model, determine the aircraft's throttle command parameters according to the current vertical motion throttle and current flight attitude, as well as the aircraft's empty hover throttle and empty weight; based on the aircraft's vertical velocity model and segmented drag model, construct a vertical velocity observer and an adaptive update law for the payload estimation parameters in the vertical velocity observer according to the throttle command parameters and the current vertical velocity; use the adaptive update law to iteratively update the payload estimation parameters until the error converges to obtain the latest payload estimation parameters; and back-calculate based on the latest payload estimation parameters to obtain the predicted payload weight.

[0064] As an example, after obtaining the current vertical motion throttle, current flight attitude, and current vertical speed, the empty hover throttle and empty weight of the aircraft are obtained.

[0065] It should be noted that the no-load hover throttle of an aircraft refers to the actual output throttle used by the aircraft to maintain hovering when it is unloaded.

[0066] Obtain the vertical motion dynamics model of the aircraft. Based on the vertical motion dynamics model, determine the throttle command parameters of the aircraft according to the current vertical motion throttle, current flight attitude, and the aircraft's no-load hover throttle and no-load weight.

[0067] In an optional implementation of this embodiment, the determination of the throttle command parameters of the aircraft based on the vertical motion dynamics model of the aircraft, according to the current vertical motion throttle and current flight attitude, as well as the aircraft's no-load hovering throttle and no-load weight, includes: determining the vertical component of the current vertical motion throttle based on the current vertical motion throttle and current flight attitude; and substituting the vertical component of the current vertical motion throttle, the no-load hovering throttle, and the no-load weight into the vertical motion dynamics model to obtain the throttle command parameters.

[0068] Determine the vertical component of the current vertical motion throttle based on the current vertical motion throttle and the current flight attitude. For example, the vertical component of the current vertical motion throttle... , These are various motors i Actual output throttle The normalized value; and These are the pitch and roll angles of the aircraft at the sampling time, respectively.

[0069] In an optional embodiment of this example, the no-load hover throttle is obtained by fitting the charged state of the aircraft when it is in a no-load state with the hover throttle.

[0070] The no-load hovering throttle is obtained by fitting the state of charge of the aircraft in an unloaded state with the hovering throttle. For example, the no-load hovering throttle. ; SOC It refers to the battery charge state of the aircraft.

[0071] Understandable It is the vertical component of the current vertical motion throttle, and similarly, It is also a vertical component.

[0072] Obtain the vertical motion dynamics model of the aircraft, for example, the vertical motion dynamics model is shown in equation (1):

[0073] (1);

[0074] In equation (1), These are throttle command parameters; This is the unloaded weight; It is gravitational acceleration.

[0075] The vertical component of the current vertical motion throttle Unloaded hovering throttle and unloaded weight Substituting the parameters into the vertical motion dynamics model shown in equation (1), the throttle command parameters can be obtained. Understandably, the throttle command parameters This indicates the pulling force provided by the motor in the vertical direction.

[0076] In practical applications, considering that the relationship between throttle and motor pull is usually nonlinear in real-world scenarios, it can be approximated as linear or determined by the pull curve. The throttle command parameters are obtained by fitting, namely: .

[0077] Obtain the vertical velocity model and segmented drag model of the aircraft. Based on the vertical velocity model and segmented drag model, construct the vertical velocity observer and the adaptive update law of the load estimation parameters in the vertical velocity observer according to the throttle command parameters and the current vertical velocity.

[0078] An adaptive update law is used to iteratively update the load estimation parameters until the error converges, at which point the latest load estimation parameters are obtained.

[0079] The predicted load weight is obtained by back-calculating using the latest load estimation parameters.

[0080] In practical applications, the specific process is as follows:

[0081] The vertical velocity model of the aircraft is shown in equation (2):

[0082] (2);

[0083] In equation (2), It is the vertical velocity of the aircraft in the NED coordinate system; yes The derivative of represents the vertical acceleration; It is gravitational acceleration; It is a symbolic function; These are throttle command parameters; This is the unloaded weight; It is the predicted load capacity to be calculated; It is the air drag coefficient; It is air density; It is the windward area of ​​the aircraft.

[0084] Discretize the vertical velocity model shown in equation (2):

[0085] (3);

[0087] In equation (3), let air resistance , and Indicates a walk, and sign indicates a sign function. dt Indicates the sampling interval.

[0088] because These are known values; the parameters are implemented. The estimation of the vertical velocity in the NED coordinate system can indirectly estimate the payload weight of the aircraft. Therefore, an adaptive vertical velocity observer can be constructed using the current vertical velocity and throttle command parameters to estimate the payload weight of the aircraft.

[0089] Define vertical velocity tracking error:

[0090] (4);

[0091] In equation (4), yes The estimated value.

[0092] Construct a vertical velocity observer based on the vertical velocity model shown in equation (2):

[0093] (5);

[0094] In equation (5), the actual parameters are taken into account. Since it is unknown, adaptive parameters are designed. right Make an estimate, yes The estimated value; It is the observation error gain value.

[0095] To ensure the convergence of estimation errors, a Lyapunov function is constructed:

[0096] (6);

[0097] In equation (6), It is adaptive gain. .

[0098] For equation (6) V Differentiation yields .

[0099] First, ignoring the drag effect of the vertical velocity model, we can obtain:

[0100] (7).

[0101] Combining equations (5) and (7), we can obtain:

[0102] (8).

[0103] According to Lyapunov theory, in order to make Design an adaptive update law:

[0104] (9);

[0105] In equations (8)-(9), yes The derivative of .

[0106] To eliminate load estimation errors caused by drag, a drag model is introduced, and the vertical velocity observer and its adaptive update law are modified as follows:

[0107] (10);

[0108] (11);

[0109] In equations (10)-(11), This is the drag model for the aircraft. Since the drag experienced by the aircraft during ascent and descent is relatively small, a linear drag model can be used to approximate the drag experienced by the aircraft during the low-speed phase. During the high-speed phase, a piecewise drag model based on the quadratic drag model is used to approximate the drag experienced by the aircraft. The piecewise drag model is described as follows:

[0110] Low speed phase ( <V thr ):

[0111] (12);

[0112] High-speed phase ( ≥V thr ):

[0113] (13);

[0114] In equations (12)-(13), and These are different drag coefficients, V thr It is a pre-set speed threshold. ≠ This allows for consideration of drag characteristics at different flight speeds.

[0115] Discretize the vertical velocity observer:

[0116] (14);

[0117] (15);

[0118] (16).

[0119] To ensure the validity of the estimated values, the projection operator is used to adapt the parameters. This estimate is within a reasonable range:

[0120] (17);

[0121] In equation (17), It is a projection operator; , , It is the maximum payload of the aircraft.

[0122] The final calculated predicted load capacity is: .

[0123] Furthermore, when the thrust or excitation (T) normalized +F aero_comp When the thrust is very small, the adaptive term does not generate effective excitation and may cause estimation drift; therefore, it is paused under zero or minimal thrust conditions. The update is equivalent to a zero-thrust protection mechanism.

[0124] This application embodiment utilizes a payload weight prediction model to perform the following: Based on the aircraft's vertical motion dynamics model, the throttle command parameters of the aircraft are determined according to the current vertical motion throttle, current flight attitude, and the aircraft's empty hover throttle and empty weight; based on the aircraft's vertical velocity model and segmented drag model, an adaptive update law for the payload estimation parameters in the vertical velocity observer is constructed according to the throttle command parameters and the current vertical velocity; the payload estimation parameters are iteratively updated using the adaptive update law until the error converges to obtain the latest payload estimation parameters; the predicted payload weight is obtained by back-calculation based on the latest payload estimation parameters, which can directly and adaptively estimate the aircraft's payload weight based on the aircraft's current operating data.

[0125] In an optional embodiment, the hover throttle prediction model is constructed by performing the following operations: obtaining a training dataset; wherein the training dataset includes multiple sets of training data, each set of training data including a calibrated state of charge, a calibrated load weight, and a calibrated hover throttle; and training a pre-established neural network model using the training dataset to obtain the hover throttle prediction model.

[0126] As an example, a training dataset is obtained, wherein the training dataset includes multiple sets of training data, each of the multiple sets of training data including a calibrated state of charge, a calibrated load weight and a calibrated hover throttle.

[0127] For example, the training dataset D includes N sets of training data, each set of training data is , It consists of a set of training data consisting of a calibrated state of charge and a calibrated load weight. This is a calibrated hovering throttle corresponding to this set of training data, representing the ideal hovering throttle for a manually calibrated aircraft under this calibrated state of charge and calibrated load weight. N is a positive integer.

[0128] A pre-established neural network model is trained using a training dataset to obtain a hover throttle prediction model. The current state of charge and the predicted payload weight are then input into the hover throttle prediction model to obtain the predicted hover throttle of the aircraft.

[0129] In practical applications, a lightweight neural network model can be pre-built, and then trained offline on the server using a training set. The objective loss function is as follows:

[0130] (18);

[0131] In equation (18), These are the network parameters of the neural network model; It is the size of the training dataset; Indicates will The actual hovering throttle output after inputting into the neural network model; It is the regularization coefficient. This represents the square of the L2 norm.

[0132] In an optional embodiment of this example, before inputting the current state of charge and the predicted payload weight into the pre-built hover throttle prediction model to obtain the predicted hover throttle of the aircraft, the method further includes: normalizing the current state of charge based on the maximum state of charge of the aircraft; and normalizing the current payload weight based on the maximum payload weight of the aircraft.

[0133] By inputting the normalized values ​​of the current state of charge and the predicted load weight into the hover throttle prediction model, the numerical differences between different data can be avoided from affecting the running results of the hover throttle prediction model.

[0134] This application embodiment obtains a training dataset including multiple sets of training data such as "calibrated state of charge - calibrated load weight - calibrated hover throttle", and uses the training dataset to train a pre-established neural network model to obtain a hover throttle prediction model. This ensures that the hover throttle prediction model can accurately predict the hover throttle based on the aircraft's state of charge and load weight.

[0135] In an optional embodiment, the method further includes steps S104-S105:

[0136] S104. Statistical analysis of the operational quality metrics corresponding to the predicted hover throttle dataset; wherein, each predicted hover throttle in the predicted hover throttle dataset is the predicted hover throttle obtained at a sampling moment within the sampling period.

[0137] As an example, the predicted hover throttle obtained at each sampling moment within the sampling period is collected as the predicted hover throttle data.

[0138] Statistical prediction of the running quality indicators corresponding to the hovering throttle dataset.

[0139] In an optional implementation of this embodiment, the operational quality indicators include the aircraft's maximum flight altitude difference, maximum flight speed, and standard deviation of predicted hover throttle during the sampling period.

[0140] In practical applications, for each predicted hovering throttle in the predicted hovering throttle dataset, the sampling time t corresponding to that predicted hovering throttle can be determined, the flight altitude h(t) of the aircraft at that sampling time can be obtained, and the flight altitude h(t) and the flight altitude level value can be calculated. The difference between them is taken as the flight altitude difference corresponding to the predicted hover throttle. This yields the flight altitude difference corresponding to each predicted hover throttle position. The maximum value among all predicted hover throttle positions can then be selected as the maximum flight altitude difference for the aircraft during the sampling period. .

[0141] For each predicted hover throttle in the predicted hover throttle dataset, determine the sampling time corresponding to that predicted hover throttle and obtain the aircraft's flight speed at that sampling time. This yields the flight speed corresponding to each predicted hover throttle position. The maximum value from all predicted hover throttle positions can then be selected as the aircraft's maximum flight speed during the sampling period. .

[0142] And based on the statistical prediction of all predicted hover throttle, the standard deviation of the predicted hover throttle is also calculated. .

[0143] Thus, a series of operational quality metrics corresponding to the predicted hover throttle dataset were obtained.

[0144] In another optional embodiment of this example, the operational quality indicators may also include the aircraft's maximum attitude deviation, maximum flight speed, and predicted hover throttle standard deviation during the sampling period.

[0145] In practical applications, for each predicted hovering throttle in the predicted hovering throttle dataset, the sampling time corresponding to that predicted hovering throttle can be determined, and the flight attitude of the aircraft at that sampling time can be obtained. Calculate the flight attitude Compared with flight attitude reference value The difference between them is taken as the flight attitude deviation corresponding to the predicted hover throttle. This yields the flight attitude deviation corresponding to each predicted hover throttle position. The maximum value can then be selected from all predicted hover throttle positions as the maximum flight attitude deviation of the aircraft during the sampling period. .

[0146] S105. If the operating quality indicators meet the indicator value conditions, optimize the hovering throttle prediction model based on the predicted hovering throttle dataset.

[0147] As an example, after obtaining the operational quality index corresponding to the predicted hover throttle dataset, it is determined whether the operational quality index meets the pre-set index value conditions. If it does, the predicted hover throttle dataset is considered a high-quality dataset, and the hover throttle prediction model is optimized based on the predicted hover throttle dataset. Otherwise, the predicted hover throttle dataset is considered a low-quality dataset, and the hover throttle prediction model is not optimized for the time being.

[0148] In an optional implementation of this embodiment, the index value conditions include: the maximum flight altitude difference is less than or equal to the altitude deviation threshold; the maximum flight speed is less than or equal to the flight speed threshold; and the predicted hovering throttle standard deviation is less than or equal to the throttle standard deviation threshold.

[0149] The maximum flight altitude difference of the aircraft during the sampling period was obtained based on the predicted hover throttle dataset. Maximum flight speed and the standard deviation of predicted hovering throttle After this series of operational quality indicators, the maximum flight altitude difference will be... With respect to the preset height deviation threshold Comparison, with maximum flight speed Compared to the preset flight speed threshold Compare and calculate the predicted hover throttle standard deviation. Compared with the preset throttle standard deviation threshold If a comparison is made, ≤ , ≤ ,and ≤ If the condition is met, the indicator value is determined to be satisfied; otherwise, the indicator value is determined not to be satisfied.

[0150] In another optional implementation of this embodiment, the index value conditions include: the maximum flight attitude deviation is less than or equal to the attitude deviation threshold; the maximum flight speed is less than or equal to the flight speed threshold; and the predicted hovering throttle standard deviation is less than or equal to the throttle standard deviation threshold.

[0151] The maximum flight attitude deviation of the aircraft during the sampling period was obtained based on the predicted hover throttle dataset. Maximum flight speed and the standard deviation of predicted hovering throttle After this series of operational quality indicators, the maximum flight attitude deviation will be... Compared with the preset attitude deviation threshold Comparison, with maximum flight speed Compared to the preset flight speed threshold Compare and calculate the predicted hover throttle standard deviation. Compared with the preset throttle standard deviation threshold If a comparison is made, ≤ , ≤ , ≤ ,and ≤ If the condition is met, the indicator value is determined to be satisfied; otherwise, the indicator value is determined not to be satisfied.

[0152] In practical applications, if it is determined that the hover throttle prediction model should be optimized based on the predicted hover throttle dataset, then the indicator function Igood(t) = 1 can be defined; otherwise, it is 0.

[0153] Assign quality weights to each predicted hover throttle in the predicted hover throttle dataset:

[0154] (19);

[0155] In equation (19), These are the quality weights of each predicted hover throttle position; It is an exponential function; Is Predicted hover throttle value at any given moment. This is the hovering throttle reference value; , , These are weighting coefficients, used to adjust the influence of the speed, attitude, and throttle deviation terms on the mass weights. They are usually positive real numbers and can be optimized according to actual conditions. and It is the flight attitude deviation threshold; It is the throttle deviation threshold; It is the flight speed threshold; , yes At any given moment, the flight attitude (pitch angle) and roll angle ).

[0156] Based on the predicted hover throttle dataset, obtain a new training dataset D uploaded by the flight terminal. new , the new training dataset D new After merging with the original training dataset D, the hover throttle prediction model is retrained or fine-tuned:

[0157] (20);

[0158] In equation (20), These are the original model parameters. These are the updated model parameters; Represents the argument of a complex number; weight That is, the weight is equal to the indicator function multiplied by the quality weight; It is the regularization coefficient. These are the current model parameters. Predicted hover throttle value.

[0159] If the decrease in validation set loss satisfies:

[0160] (twenty one);

[0161] In equation (21), It is the decrease in validation set loss. The larger the value, the more significant the improvement. It is the loss on the validation set, used to determine whether the new model is truly better; It is a pre-set minimum improvement threshold.

[0162] This yields a new hover throttle prediction model, which is then sent to the aircraft.

[0163] This application embodiment collects the predicted hovering throttle at each sampling moment within the sampling period as a predicted hovering throttle dataset, statistically analyzes the operational quality indicators corresponding to the predicted hovering throttle dataset, and optimizes the hovering throttle prediction model based on the predicted hovering throttle dataset when the operational quality indicators meet the indicator value conditions. This allows for continuous optimization of the hovering throttle prediction model based on high-quality predicted hovering throttle related data, further ensuring that the hovering throttle prediction model can accurately predict the hovering throttle based on the aircraft's state of charge and payload weight.

[0164] Please refer to Figure 2 , Figure 2 This is a schematic diagram of a hovering throttle prediction device for an aircraft, provided in the second embodiment of this application. The second embodiment of this application provides an aircraft hovering throttle prediction system, including: an operation data acquisition module 201, used to acquire the current operation data of the aircraft; wherein the current operation data includes the current vertical motion throttle, current flight attitude, current vertical speed, and current state of charge; an adaptive load prediction module 202, used to input the current vertical motion throttle, current flight attitude, and current vertical speed into a pre-built load prediction model to obtain the predicted load weight of the aircraft; and a hovering throttle prediction module 203, used to input the current state of charge and the predicted load weight into a pre-built hovering throttle prediction model to obtain the predicted hovering throttle of the aircraft.

[0165] In an optional embodiment, the payload weight prediction model is used to perform the following operations: based on the aircraft's vertical motion dynamics model, determine the aircraft's throttle command parameters according to the current vertical motion throttle and current flight attitude, as well as the aircraft's empty hover throttle and empty weight; based on the aircraft's vertical velocity model and segmented drag model, construct a vertical velocity observer and an adaptive update law for the payload estimation parameters in the vertical velocity observer according to the throttle command parameters and the current vertical velocity; use the adaptive update law to iteratively update the payload estimation parameters until the error converges to obtain the latest payload estimation parameters; and back-calculate based on the latest payload estimation parameters to obtain the predicted payload weight.

[0166] In an optional embodiment, the determination of the throttle command parameters of the aircraft based on the vertical motion dynamics model of the aircraft, according to the current vertical motion throttle and current flight attitude, as well as the aircraft's no-load hovering throttle and no-load weight, includes: determining the vertical component of the current vertical motion throttle based on the current vertical motion throttle and current flight attitude; and substituting the vertical component of the current vertical motion throttle, the no-load hovering throttle, and the no-load weight into the vertical motion dynamics model to obtain the throttle command parameters.

[0167] In an optional embodiment, the no-load hover throttle is obtained by fitting the state of charge of the aircraft when it is in a no-load state with the hover throttle.

[0168] In an optional embodiment, the hover throttle prediction model is constructed by performing the following operations: obtaining a training dataset; wherein the training dataset includes multiple sets of training data, each set of training data including a calibrated state of charge, a calibrated load weight, and a calibrated hover throttle; and training a pre-established neural network model using the training dataset to obtain the hover throttle prediction model.

[0169] In an optional embodiment, the adaptive load prediction module 202 is further configured to normalize the current state of charge based on the maximum state of charge of the aircraft and normalize the current load weight based on the maximum load weight of the aircraft before inputting the current state of charge and the predicted load weight into the pre-built hover throttle prediction model to obtain the predicted hover throttle of the aircraft.

[0170] In an optional embodiment, the system further includes a model optimization module, used to: statistically analyze the operational quality indicators corresponding to the predicted hover throttle dataset; wherein each predicted hover throttle in the predicted hover throttle dataset is a predicted hover throttle obtained at a sampling moment within the sampling period; and optimize the hover throttle prediction model based on the predicted hover throttle dataset when the operational quality indicators meet the indicator value conditions.

[0171] In an optional embodiment, the operational quality indicators include the maximum flight altitude difference, maximum flight speed, and predicted hover throttle standard deviation of the aircraft during the sampling period; the indicator value conditions include: the maximum flight altitude difference is less than or equal to the altitude deviation threshold; the maximum flight speed is less than or equal to the flight speed threshold; and the predicted hover throttle standard deviation is less than or equal to the throttle standard deviation threshold.

[0172] The implementation process of the functions and roles of each module in the above system 20 is detailed in the implementation process of the corresponding steps in the method described in the first embodiment of this application, and will not be repeated here.

[0173] Please refer to Figure 3 , Figure 3 This is a schematic diagram of the structure of an aircraft according to a third embodiment of this application. In a third aspect, this third embodiment of the application provides an aircraft 30, including a processor 301, a memory 302, and a computer program stored in the memory 302 and configured to be executed by the processor 301; when the processor 301 executes the computer program, it implements the method described in the first embodiment of this application and achieves the same beneficial effects.

[0174] When the processor 301 reads a computer program from the memory 302 via the bus 303 and executes the computer program, it can implement any of the methods described in the first embodiment of this application.

[0175] Processor 301 can process digital signals and may include various computing architectures. For example, it may be a complex instruction set computer architecture, a reduced instruction set computer architecture, or an architecture that implements multiple instruction set combinations. In some examples, processor 301 may be a microprocessor.

[0176] The memory 302 can be used to store instructions executed by the processor 301 or data related to the execution of instructions. These instructions and / or data may include code for implementing some or all of the functions of one or more modules described in the embodiments of this application. The processor 301 of this disclosure embodiment can be used to execute instructions in the memory 302 to implement the method described in the first embodiment of this application. The memory 302 includes dynamic random access memory, static random access memory, flash memory, optical memory, or other memories well known to those skilled in the art.

[0177] The fourth embodiment of this application provides a computer-readable storage medium, which includes a stored computer program; wherein, when the computer program is running, it controls the device where the computer-readable storage medium is located to perform the method described in the first embodiment of this application, and can achieve the same beneficial effects.

[0178] The method described in the first embodiment of this application can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the various embodiments of this application are executed, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user equipment, a core network device, an OAM (Open Application Model), or other programmable devices.

[0179] The computer program or instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions may be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium may be any available medium that a computer can access, or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; or an optical medium, such as a digital video optical disc; or a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both volatile and non-volatile types of storage media.

[0180] The fifth embodiment of this application provides a computer program product, which includes instructions that, when executed by a computer, cause the computer to perform the method described in the first embodiment of this application and achieve the same beneficial effects.

[0181] The methods described in the first embodiment of this application can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, they can be implemented, in whole or in part, in the form of a computer program product. A computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the various embodiments of this application are performed, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user equipment, a core network device, an OAM (Open Application Model), or other programmable devices.

[0182] Computer programs or instructions can be stored in or transferred from one computer-readable storage medium to another. For example, a computer program or instructions can be transferred from one website, computer, server, or data center to another via wired or wireless means. A computer-readable storage medium can be any usable medium that a computer can access, or a data storage device such as a server or data center that integrates one or more usable media. Usable media can be magnetic media, such as floppy disks, hard disks, and magnetic tapes; optical media, such as digital video discs; or semiconductor media, such as solid-state drives. The computer-readable storage medium can be volatile or non-volatile, or may include both types.

[0183] In summary, this application provides a method, system, aircraft, medium, and program product for predicting hover throttle for an aircraft. The method for predicting hover throttle includes: acquiring the current operating data of the aircraft; wherein the current operating data includes the current vertical motion throttle, current flight attitude, current vertical speed, and current state of charge; inputting the current vertical motion throttle, current flight attitude, and current vertical speed into a pre-built load prediction model to obtain the predicted load weight of the aircraft; and inputting the current state of charge and the predicted load weight into a pre-built hover throttle prediction model to obtain the predicted hover throttle of the aircraft. This application embodiment pre-constructs a payload prediction model and a hover throttle prediction model to obtain a series of current operating data of the aircraft, including the current vertical motion throttle, current flight attitude, current vertical speed, and current state of charge. The current vertical motion throttle, current flight attitude, and current vertical speed are input into the payload prediction model to obtain the predicted payload weight of the aircraft. The current state of charge and the predicted payload weight are input into the hover throttle prediction model to obtain the predicted hover throttle of the aircraft. This allows for the adaptive estimation of the aircraft's payload weight directly based on the aircraft's current operating data without relying on external sensors or manual calibration, thereby quickly and accurately predicting the hover throttle of the aircraft.

[0184] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0185] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0186] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0187] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for predicting the hovering throttle of an aircraft, characterized in that, include: Acquire the current operating data of the aircraft; wherein, the current operating data includes the current vertical motion throttle, current flight attitude, current vertical speed, and current state of charge; The current vertical motion throttle, the current flight attitude, and the current vertical speed are input into a pre-built payload prediction model to obtain the predicted payload of the aircraft. The current state of charge and the predicted payload weight are input into a pre-built hover throttle prediction model to obtain the predicted hover throttle of the aircraft. The load weight prediction model is used to perform the following operations: Based on the vertical motion dynamics model of the aircraft, the throttle command parameters of the aircraft are determined according to the current vertical motion throttle and the current flight attitude, as well as the no-load hover throttle and no-load weight of the aircraft. Based on the vertical velocity model and segmented drag model of the aircraft, and according to the throttle command parameters and the current vertical velocity, an adaptive update law for the load estimation parameters in the vertical velocity observer is constructed. The adaptive update law is used to iteratively update the load estimation parameters until the error converges and the latest load estimation parameters are obtained. The predicted load weight is obtained by back-calculating based on the latest load estimation parameters.

2. The method according to claim 1, characterized in that, The vertical motion dynamics model of the aircraft determines the throttle command parameters of the aircraft based on the current vertical motion throttle, the current flight attitude, the aircraft's no-load hover throttle, and its no-load weight, including: Based on the current vertical motion throttle and the current flight attitude, determine the vertical component of the current vertical motion throttle; Substituting the vertical component of the current vertical motion throttle, the no-load hover throttle, and the no-load weight into the vertical motion dynamics model, the throttle command parameters are obtained.

3. The method according to claim 1, characterized in that, The no-load hover throttle is obtained by fitting the charged state of the aircraft when it is in a no-load state with the hover throttle.

4. The method according to claim 1, characterized in that, The hover throttle prediction model is constructed by performing the following operations: Obtain a training dataset; wherein the training dataset includes multiple sets of training data, and each set of training data includes a calibrated state of charge, a calibrated load weight, and a calibrated hover throttle. The hovering throttle prediction model is obtained by training a pre-established neural network model using a training dataset.

5. The method according to claim 1, characterized in that, Before inputting the current state of charge and the predicted payload weight into the pre-built hover throttle prediction model to obtain the predicted hover throttle of the aircraft, the method further includes: The current state of charge is normalized based on the maximum state of charge of the aircraft; The current payload weight is normalized based on the maximum payload weight of the aircraft.

6. The method according to any one of claims 1 to 5, characterized in that, The method further includes: The statistical prediction of the hover throttle dataset corresponds to the operational quality index; wherein, each predicted hover throttle in the predicted hover throttle dataset is the predicted hover throttle obtained at a sampling moment within the sampling period; If the operational quality indicators meet the indicator value conditions, the hovering throttle prediction model is optimized based on the predicted hovering throttle dataset.

7. The method according to claim 6, characterized in that, The operational quality indicators include the aircraft’s maximum flight altitude difference, maximum flight speed, and standard deviation of predicted hover throttle during the sampling period. The conditions for determining the value of the indicator include: The maximum flight altitude difference is less than or equal to the altitude deviation threshold; The maximum flight speed is less than or equal to the flight speed threshold; and... The predicted hover throttle standard deviation is less than or equal to the throttle standard deviation threshold.

8. An aircraft hovering throttle prediction system, characterized in that, include: The operational data acquisition module is used to acquire the current operational data of the aircraft; wherein, the current operational data includes the current vertical motion throttle, current flight attitude, current vertical speed, and current state of charge; An adaptive load prediction module is used to input the current vertical motion throttle, the current flight attitude, and the current vertical speed into a pre-built load prediction model to obtain the predicted load weight of the aircraft. The hovering throttle prediction module is used to input the current state of charge and the predicted load weight into a pre-built hovering throttle prediction model to obtain the predicted hovering throttle of the aircraft. The load weight prediction model is used to perform the following operations: Based on the vertical motion dynamics model of the aircraft, the throttle command parameters of the aircraft are determined according to the current vertical motion throttle and the current flight attitude, as well as the no-load hover throttle and no-load weight of the aircraft. Based on the vertical velocity model and segmented drag model of the aircraft, and according to the throttle command parameters and the current vertical velocity, an adaptive update law for the load estimation parameters in the vertical velocity observer is constructed. The adaptive update law is used to iteratively update the load estimation parameters until the error converges and the latest load estimation parameters are obtained. The predicted load weight is obtained by back-calculating based on the latest load estimation parameters.

9. An aircraft, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor; when the processor executes the computer program, it implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program; wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the method according to any one of claims 1 to 7.

11. A computer program product, characterized in that, The computer program product includes instructions that, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 7.

Images