Flight path prediction method of aircraft

Abstract

The invention relates to a flight trajectory prediction method of an aircraft, and relates to the technical field of hypersonic aircrafts, and the method comprises the steps: determining an original trajectory database of a hypersonic aircraft under different initial states and different flight control laws based on a motion equation; training a to-be-trained network model based on the original trajectory database to obtain a current trajectory prediction model; determining a to-be-migrated trajectory prediction model from the current trajectory prediction model according to the target flight trajectory of the to-be-predicted aircraft and the original flight trajectory in the original trajectory database, and migrating model parameters in the to-be-migrated trajectory prediction model to the to-be-corrected trajectory prediction model; and performing parameter correction on the to-be-corrected trajectory prediction model based on the target flight trajectory to obtain a target trajectory prediction model, and predicting the flight trajectory of the to-be-predicted aircraft based on the target trajectory prediction model. According to the invention, the accuracy of flight estimation obtained through prediction is improved.

Description

Background Art

[0002] At present, trajectory prediction based on machine learning uses offline training to train the model when the training set and test set data are in the same distribution. At the same time, in the actual application process, the fitting error of the model can be used as the evaluation standard, and the neural network prediction model can be trained through a large amount of observation data. By using different neural networks to superimpose on each other in the test set, better prediction results can be obtained.

[0003] However, since the observed target (that is, the aircraft to be predicted) is often non-cooperative, the defender cannot obtain important information such as the mass, volume, flight control law, etc. of the aircraft to be predicted. Conventional machine learning trajectory prediction algorithms need to train a model from scratch, which requires a large amount of observation data, which makes the model training efficiency low; at the same time, the accuracy of the obtained model trajectory prediction results is also low.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The purpose of the present disclosure is to provide a method for predicting the flight trajectory of an aircraft, thereby overcoming, at least to a certain extent, the problems of low model training efficiency and low accuracy of model trajectory prediction results caused by limitations and defects of related technologies.

[0006] According to one aspect of the present disclosure, a method for predicting a flight trajectory of an aircraft is provided, comprising:

[0007] Establishing a motion equation of a hypersonic aircraft, and determining an original trajectory database of the hypersonic aircraft under different initial states and different flight control laws based on the motion equation;

[0008] Training the network model to be trained based on the original trajectory database to obtain a current trajectory prediction model;

[0009] According to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, a trajectory prediction model to be migrated is determined from the current trajectory prediction model, and model parameters in the trajectory prediction model to be migrated are migrated to the trajectory prediction model to be corrected;

[0010] The trajectory prediction model to be corrected is parameterized based on the target flight trajectory to obtain a target trajectory prediction model, and the flight trajectory of the aircraft to be predicted is predicted based on the target trajectory prediction model.

[0011] In an exemplary embodiment of the present disclosure, establishing the motion equation of a hypersonic vehicle includes:

[0012] Obtain the flight speed, ballistic inclination, ballistic deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude of the hypersonic aircraft;

[0013] In the ballistic coordinate system, the three-degree-of-freedom motion equation of the hypersonic aircraft is established according to the flight speed, ballistic inclination, ballistic deviation, the distance from the hypersonic aircraft to the center of the earth, the longitude and the latitude.

[0014] In an exemplary embodiment of the present disclosure, an original trajectory database of a hypersonic vehicle under different initial states and different flight control laws is determined based on the equation of motion, including:

[0015] Determining different initial state variables of the hypersonic aircraft based on flight speed, trajectory inclination, trajectory deviation, distance from the hypersonic aircraft to the center of the earth, longitude, and latitude;

[0016] Determining different flight control laws of the hypersonic aircraft according to the angle of attack control command and the roll angle control command of the hypersonic aircraft;

[0017] Inputting the different initial state variables and different flight control laws into the motion equation of three degrees of freedom to obtain original flight trajectories under different initial state variables and different flight control laws;

[0018] The original trajectory database is generated according to the original flight trajectory and the initial state variables and control laws corresponding to the original flight trajectory.

[0019] In an exemplary embodiment of the present disclosure, the network model to be trained is trained based on the original trajectory database to obtain a current trajectory prediction model, including:

[0020] Inputting the initial state variables and control laws corresponding to the original flight trajectory in the original trajectory database into the network model to be trained to obtain the current prediction result;

[0021] A current loss function is constructed according to the current prediction result and the original flight trajectory, and parameters in the network model to be trained are adjusted based on the current loss function to obtain a current trajectory prediction model.

[0022] In an exemplary embodiment of the present disclosure, according to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, determining the trajectory prediction model to be migrated from the current trajectory prediction model includes:

[0023] Acquiring a target flight trajectory of the aircraft to be predicted, and performing noise reduction processing on the target flight trajectory to obtain a flight trajectory after noise reduction processing;

[0024] Determine a first trajectory sequence of the flight trajectory after the noise reduction process, and determine a second trajectory sequence of the original flight trajectory in the original trajectory database;

[0025] The similarity between the first trajectory sequence and the second trajectory sequence is calculated, and a trajectory prediction model to be migrated is determined from the current trajectory prediction model according to the similarity.

[0026] In an exemplary embodiment of the present disclosure, calculating the similarity between the first trajectory sequence and the second trajectory sequence, and determining the trajectory prediction model to be migrated from the current trajectory prediction model according to the similarity, includes:

[0027] Calculating the Euclidean distance between the first trajectory sequence and the second trajectory sequence, and determining the similarity between the first trajectory sequence and the second trajectory sequence according to the Euclidean distance;

[0028] sorting the original flight trajectories based on the similarity between the first trajectory sequence and the second trajectory sequence, and extracting the original flight trajectory with the smallest similarity from the original flight trajectories based on the trajectory sorting result;

[0029] The current trajectory prediction model corresponding to the original flight trajectory with the smallest similarity is used as the trajectory prediction model to be migrated.

[0030] In an exemplary embodiment of the present disclosure, migrating model parameters in a trajectory prediction model to be migrated to a trajectory prediction model to be corrected includes:

[0031] The model parameters in the trajectory prediction model to be migrated are extracted, and the model parameters in the trajectory prediction model to be corrected are initialized based on the extracted model parameters to achieve parameter migration.

[0032] In an exemplary embodiment of the present disclosure, parameters of the trajectory prediction model to be corrected are corrected based on the target flight trajectory to obtain the target trajectory prediction model, including:

[0033] Obtaining sample data of the observed aircraft to be predicted, and inputting the sample data into the trajectory prediction model to be corrected to obtain a target prediction result;

[0034] A target loss function is constructed according to the target prediction result and the target flight trajectory, and parameters of the trajectory prediction model to be corrected are corrected based on the target loss function to obtain a target trajectory prediction model.

[0035] In an exemplary embodiment of the present disclosure, the trajectory prediction model to be corrected includes a convolutional neural network model and a bidirectional long short-term memory network model;

[0036] Among them, performing parameter correction on the trajectory prediction model to be corrected based on the target loss function to obtain the target trajectory prediction model includes:

[0037] The weight value of the convolutional neural network is fixed, and the weight value and bias of the bidirectional long short-term memory network model are fine-tuned based on the target loss function to obtain a target trajectory prediction model;

[0038] In the process of fine-tuning the weight value and bias of the bidirectional long short-term memory network model, the selected optimizer includes an Adam optimizer that adaptively adjusts the learning rate.

[0039] In an exemplary embodiment of the present disclosure, predicting the flight trajectory of the aircraft to be predicted based on the target trajectory prediction model includes:

[0040] Obtaining the flight speed, trajectory inclination, trajectory deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude of the aircraft to be predicted;

[0041] Constructing a model input vector according to the flight speed, trajectory inclination, trajectory deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude of the aircraft to be predicted;

[0042] The model input vector is input into the target trajectory prediction model to obtain the flight trajectory prediction result of the aircraft to be predicted.

[0043] The flight trajectory prediction method of the aircraft provided by the embodiment of the present disclosure, on the one hand, can establish the motion equation of the hypersonic aircraft, and determine the original trajectory database of the hypersonic aircraft under different initial states and different flight control laws based on the motion equation; then train the network model to be trained based on the original trajectory database to obtain the current trajectory prediction model; then determine the trajectory prediction model to be migrated from the current trajectory prediction model according to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, and migrate the model parameters in the trajectory prediction model to be migrated to the trajectory prediction model to be corrected; finally, based on The target flight trajectory performs parameter correction on the trajectory prediction model to be corrected to obtain the target trajectory prediction model, and predicts the flight trajectory of the aircraft to be predicted based on the target trajectory prediction model, thereby improving the accuracy of the predicted flight trajectory and solving the problem of low accuracy of the model trajectory prediction results in the prior art; on the other hand, since the model parameters in the trajectory prediction model to be migrated can be migrated to the trajectory prediction model to be corrected; finally, the trajectory prediction model to be corrected is corrected based on the target flight trajectory to obtain the target trajectory prediction model, there is no need to train the trajectory prediction model to be corrected from scratch, thereby improving the training efficiency of the model.

[0044] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0046] Figure 1 The following schematically shows a flow chart of a method for predicting a flight trajectory of an aircraft according to an exemplary embodiment of the present disclosure.

[0047] Figure 2 A diagram schematically shows an example of a model structure of a CNN-BiLSTM according to an example embodiment of the present disclosure.

[0048] Figure 3 A framework example diagram schematically illustrates an implementation principle of a CNN-BiLSTM according to an example embodiment of the present disclosure.

[0049] Figure 4 A structural example diagram of a 1DCNN according to an example embodiment of the present disclosure is schematically shown.

[0050] Figure 5 A structural example diagram of an LSTM according to an exemplary embodiment of the present disclosure is schematically shown.

[0051] Figure 6 A diagram schematically shows a structural example of a BiLSTM according to an exemplary embodiment of the present disclosure.

[0052] Figure 7 An example diagram schematically illustrates a specific generation process of an original trajectory database according to an example embodiment of the present disclosure.

[0053] Figure 8 An example diagram schematically illustrates aircraft trajectory prediction based on a 1DCNN-BiLSTM model according to an example embodiment of the present disclosure.

[0054] Fig. 9 A block diagram schematically shows a flight trajectory prediction device for an aircraft according to an exemplary embodiment of the present disclosure.

[0055] Fig.10 An electronic device for implementing a method for predicting a flight trajectory of an aircraft according to an exemplary embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0056] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as being limited to the examples set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concepts of the example embodiments are fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0057] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0058] In recent years, a lot of research has been done on trajectory prediction of hypersonic vehicles. At present, trajectory prediction based on machine learning mainly considers the situation where the training set and the test aircraft data are in the same distribution. The model adopts offline training, takes the model fitting error as the evaluation criterion, trains the neural network prediction model through a large amount of observation data, and obtains better prediction results by superimposing different neural networks on the test set. However, since the observed targets are often non-cooperative, the defender cannot obtain important information such as the mass, volume, and flight control law of the target aircraft. Conventional machine learning trajectory prediction algorithms need to train a model from scratch, which requires a large amount of observation data, and the prediction is not real-time.

[0059] In view of this situation, this paper proposes a trajectory prediction model based on 1DCNN-BiLSTM transfer learning. For the trajectory prediction problem of non-cooperative targets, the 1DCNN-BiLSTM prediction model library is trained by establishing a trajectory database, and fast and high-precision prediction is achieved by fine-tuning similar trajectory network parameters under small sample observation data conditions. The flight trajectory prediction method of the aircraft provided in the exemplary embodiment of the present disclosure can maximize the value of the offline trajectory database, and can quickly predict the trajectory of the target with high precision under small sample observation conditions, so as to improve the response speed of the anti-missile defense system.

[0060] In an exemplary embodiment, a method for predicting the flight trajectory of an aircraft is first provided in this exemplary embodiment. The method can be run on a terminal device, a server, a server cluster or a cloud server, etc. Of course, those skilled in the art can also run the method disclosed in this disclosure on other platforms as required, and this exemplary embodiment does not specifically limit this. Specifically, refer to Figure 1 As shown, the flight trajectory prediction method of the aircraft may include the following steps:

[0061] Step S110. Establishing a motion equation of a hypersonic aircraft, and determining an original trajectory database of the hypersonic aircraft under different initial states and different flight control laws based on the motion equation;

[0062] Step S120. Training the network model to be trained based on the original trajectory database to obtain a current trajectory prediction model;

[0063] Step S130. According to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, determine the trajectory prediction model to be migrated from the current trajectory prediction model, and migrate the model parameters in the trajectory prediction model to be migrated to the trajectory prediction model to be corrected;

[0064] Step S140: Based on the target flight trajectory, the trajectory prediction model to be corrected is parameterized to obtain a target trajectory prediction model, and the flight trajectory of the aircraft to be predicted is predicted based on the target trajectory prediction model.

[0065] In the above-mentioned flight trajectory prediction method of the aircraft, on the one hand, since the motion equation of the hypersonic aircraft can be established, and the original trajectory database of the hypersonic aircraft under different initial states and different flight control laws can be determined based on the motion equation; then the network model to be trained is trained based on the original trajectory database to obtain the current trajectory prediction model; then according to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, the trajectory prediction model to be migrated is determined from the current trajectory prediction model, and the model parameters in the trajectory prediction model to be migrated are migrated to the trajectory prediction model to be corrected; finally, based on the target flight trajectory of the aircraft to be predicted, the trajectory prediction model to be migrated is transferred to the trajectory prediction model to be corrected. The method performs parameter correction on the trajectory prediction model to be corrected to obtain a target trajectory prediction model, and predicts the flight trajectory of the aircraft to be predicted based on the target trajectory prediction model, thereby improving the accuracy of the predicted flight trajectory and solving the problem of low accuracy of the model trajectory prediction results in the prior art. On the other hand, since the model parameters in the trajectory prediction model to be migrated can be migrated to the trajectory prediction model to be corrected; finally, the parameters of the trajectory prediction model to be corrected are corrected based on the target flight trajectory to obtain the target trajectory prediction model. There is no need to train the trajectory prediction model to be corrected from scratch, thereby improving the training efficiency of the model.

[0066] Hereinafter, the flight trajectory prediction method of an aircraft described in an exemplary embodiment of the present disclosure will be explained and illustrated in detail with reference to the accompanying drawings.

[0067] First, the technical implementation principle of the exemplary embodiment of the present disclosure is explained and illustrated. Specifically, in the actual application process of the flight trajectory prediction method of the aircraft recorded in the exemplary embodiment of the present disclosure, in order to realize the prediction of the flight trajectory of the aircraft to be predicted; first, it is necessary to establish an offline ballistic database of the aircraft; specifically, in the process of constructing the offline ballistic database (that is, the original ballistic database or the source domain ballistic database), it is necessary to first consider the various forces acting on the aircraft, and establish the three-degree-of-freedom motion equation of the aircraft in the ballistic coordinate system; then set the angle of attack-velocity profile, divide the flight state of the aircraft into two sections, the initial dive section and the descending section, and obtain the function of the angle of attack changing with the velocity; then establish that the aircraft must meet the thermal flow constraint, overload constraint and dynamic pressure constraint during the flight process; finally, give the initial flight conditions and the angle of attack control command and the roll angle control command, and construct a simulation method for the Monte Carlo ballistic database. Secondly, establish a 1DCNN-BiLSTM network and a transfer learning framework; specifically, establish an aircraft trajectory prediction model framework, which mainly consists of two parts: (1) Generation of the source domain ballistic database. The offline ballistic database is simulated through Monte Carlo, and the trajectories in the database are trained using neural networks. After multiple iterations, a high-precision pre-trained model library is obtained; (2) Target domain trajectory prediction. The network parameters of the most similar trajectory trained in the source domain are migrated to the target domain. After fine-tuning the network, the trajectory of the target domain aircraft is predicted and the future trajectory of the target is calculated. Furthermore, trajectory prediction is performed based on 1DCNN-BiLSTM and transfer learning. Specifically, a source domain trajectory database is generated, target domain data is processed, similarity is selected, trajectory prediction is performed after fine-tuning the migration network, and the prediction model is evaluated. Based on the above-described flight trajectory prediction method for aircraft, fast and high-precision prediction can be achieved by fine-tuning the parameters of similar trajectory networks under small sample observation data conditions; at the same time, the value of the offline trajectory database can be maximized.

[0068] Secondly, the network model involved in the exemplary embodiment of the present disclosure is explained and illustrated. Specifically, the network model to be trained and the trajectory prediction model to be corrected involved in the exemplary embodiment of the present disclosure are both implemented by CNN-BiLSTM; among them, the specific model structure example diagram of CNN-BiLSTM can be referred to Figure 2 As shown; the implementation principle framework example diagram of CNN-BiLSTM can be referred to Figure 3As shown; further, the CNN recorded here refers to Convolutional Neural Networks, and the BiLSTM recorded here refers to Bidirectional Long Short Term Memory; further, the CNN recorded here can be 1DCNN, and of course 2DCNN or 3DCNN can also be selected, and this example does not impose any special restrictions on this.

[0069] In an exemplary embodiment, the 1DCNN network described herein can be used to extract feature information of an input vector input into the model; specifically, in the actual application process, the convolutional neural network CNN can map simple data to a high-dimensional space through a convolutional layer and a pooling layer to extract feature information; it should be noted here that since the CNN structure is similar to a multi-layer perceptron and can better solve the problem of low prediction accuracy, the multi-layer perceptron has a large parameter scale due to the full connection method; therefore, in order to avoid this problem, the convolutional neural network has been improved by replacing the full connection method with a sparse connection method to link two adjacent layers of neurons, and performing a pooling operation on the processed features; further, the exemplary embodiment of the present disclosure uses 1DCNN for illustration; wherein 1DCNN is a type of CNN, which is used to extract and classify features of one-dimensional data; at the same time, the 1DCNN described herein can be composed of an input layer 401, a one-dimensional convolutional layer 402, a pooling layer 403 and an output layer 404; the specific structural example diagram can be referred to Figure 4 shown.

[0070] Furthermore, in the actual application process, the input data x i When , the convolutional layer calculation formula can be shown as follows (1):

[0071]

[0072] The calculation formula of the pooling layer can be specifically shown as follows:

[0073]

[0074] Where f is the activation function, is the output of the i-th neuron in the l-th layer, The output of the i-th neuron in the l-1 layer; r is the convolution kernel compensation; j represents the sparse connection between the two layers of neurons; is the weight of the corresponding neuron; is the bias of the corresponding neuron, is the l-1th feature matrix.

[0075] In an example embodiment, the BiLSTM network described herein can be used to predict trajectories based on features extracted by CNN. Specifically, the LSTM network is improved from the recurrent neural network RNN ​​(Recurrent Neural Network). By adding a gating mechanism, historical information is recorded, solving the problem of the traditional RNN network forgetting past data. However, the single-layer LSTM network only considers forward historical information during training and lacks subsequent historical information. Therefore, its accuracy is still insufficient in long-term predictions. The BiLSTM network is composed of two LSTM networks, forward and reverse. The forward network obtains the forward historical information of the input sequence, while the backward network inputs the future information of the sequence, digging deeper into the overall timing information of the trajectory. Among them, the unit structure of LSTM can be as follows: Figure 5 Further, in Figure 5 In the example structure diagram shown, the LSTM network uses a more complex internal neural network to save and utilize sequence information, alleviating the gradient vanishing or gradient exploding problem of RNN when processing long sequences. The LSTM unit has a total of 3 gate structures:

[0076] The first gate is a forget gate structure. The forget gate reads the hidden state of the previous moment and the current input. After the simoid function, the current input of the LSTM unit and the hidden state H of the previous LSTM unit are combined. t Perform screening, selectively forget some features, and retain useful features;

[0077] The second is the input gate structure. The input gate uses the Sigmoid function to combine the current input with the hidden state H of the previous LSTM unit. t Selectively record into the cell state, and use the Tanh layer combined with the data information retained by the forget gate to update the cell information C in the LSTM unit t ;

[0078] The third is the output gate structure. The output gate uses the Sigmoid layer to t The data information is selectively output, and then the cell state information after the Tanh layer normalization is multiplied by the output of the Sigmoid layer to obtain the final output H of the cell. t , the final state is only the output part determined by the output gate.

[0079] Among them, the functions used by the LSTM network are activation functions such as Sigmoid, Tanh, and Relu. The process of network forward calculation can be shown in the following formula (3):

[0080]

[0081] Among them, i represents the input gate, f represents the forget gate, and o represents the output gate; H t , H t-1 Represents the current output and the previous time step output, C t , C t-1 is the unit information of the previous time step, is the newly generated information, W xi ,W xf ,W xo ,W hi ,W hO ,W hc and b i ,b f ,b O ,b c are the neural network weights and bias vectors corresponding to each gate structure. t and H t The updating is done by multiplying the elements in the vector. σ represents the activation function of the neural network. Selecting a suitable activation function gives the LSTM network better nonlinear fitting ability.

[0082] Furthermore, the structure of the BiLSTM network can be as follows Figure 6 shown in Figure 6 In the example structure shown in the figure, at each time step t, the output of the forward network and the reverse network are calculated simultaneously ( and ), connect the outputs of the two and perform weighted calculation to get the output y of BiLSTM t Among them, the calculation formulas of the forward network and the reverse network can be shown as follows (4):

[0083]

[0084] Among them, α represents the forward network output calculation weight, β represents the reverse network output calculation weight, and b y Indicates output bias.

[0085] The following will be combined Figure 2-Figure 6 right Figure 1 The flight trajectory prediction method of the aircraft shown in is further explained and illustrated. Specifically:

[0086] In step S110, a motion equation of the hypersonic aircraft is established, and based on the motion equation, an original trajectory database of the hypersonic aircraft under different initial states and different flight control laws is determined.

[0087] In this example embodiment, first, the motion equation of the hypersonic aircraft is established; specifically, it can be achieved in the following way: the flight speed, ballistic inclination, ballistic deviation, the distance from the hypersonic aircraft to the center of the earth, the longitude and the latitude of the hypersonic aircraft are obtained; in the ballistic coordinate system, the three-degree-of-freedom motion equation of the hypersonic aircraft is established according to the flight speed, ballistic inclination, ballistic deviation, the distance from the hypersonic aircraft to the center of the earth, the longitude and the latitude. Secondly, based on the motion equation, the original trajectory database of the hypersonic aircraft under different initial states and different flight control laws is determined; specifically, it can be achieved in the following ways: first, based on the flight speed, trajectory inclination, trajectory deviation, the distance from the hypersonic aircraft to the center of the earth, longitude and latitude, the different initial state variables of the hypersonic aircraft are determined; secondly, according to the angle of attack control instructions and the roll angle control instructions of the hypersonic aircraft, the different flight control laws of the hypersonic aircraft are determined; then, the different initial state variables and different flight control laws are input into the three-degree-of-freedom motion equation to obtain the original flight trajectories under different initial state variables and different flight control laws; finally, according to the original flight trajectory and the initial state variables and control laws corresponding to the original flight trajectory, the original trajectory database is generated.

[0088] The following will be combined Figure 6 The specific process of establishing the original ballistic database is further explained and illustrated. Specifically, in the actual application process, considering the various forces acting on the aircraft, the three-degree-of-freedom motion equation of the hypersonic aircraft is established in the ballistic coordinate system, which can be specifically shown in the following formula (5):

[0089]

[0090] Among them, V is the speed of the aircraft, θ is the ballistic inclination, δ is the ballistic deviation, r is the distance from the aircraft to the center of the earth, λ is the longitude, and φ is the latitude; They are V for the speed of the aircraft, θ for the ballistic inclination, δ for the ballistic deviation, r for the distance from the aircraft to the center of the earth, λ for the longitude, and φ for the differential of the latitude;

[0091] μ M is the Earth's gravitational constant, ω e is the angular velocity of the earth's rotation, m is the mass of the hypersonic aircraft, υ is the roll angle command, which can be used to represent the angle between the lift direction and the vertical plane containing the velocity vector, s is the equivalent area of ​​the aircraft, and ρ is the atmospheric density; X, Y, and Z are the drag, lift, and lateral force of the hypersonic aircraft, respectively. The calculation formula can be shown in the following formula (6):

[0092]

[0093] Furthermore, in the actual application process, if the aircraft side slip is not considered, β = 0; therefore, the lateral force Z = 0; q represents the dynamic pressure constraint, and the specific calculation formula can be shown in the following formula (11); C L is the lift coefficient, C D represents the drag coefficient, represents the side force coefficient. The lift coefficient, drag coefficient and side force coefficient can be obtained by inputting the angle of attack command and flight speed at the current moment and querying the interpolation table. Furthermore, the angle of attack-speed profile is set to divide the flight state of the aircraft into two sections: the initial dive section and the descending section. The function of the angle of attack changing with speed can be shown as follows:

[0094]

[0095] The roll angle command is shown in the following formula:

[0096]

[0097] It can be seen that the state variables of the system are: V, θ, δ, r, λ, φ, and the control input is the angle of attack command α and the roll angle command υ; set the initial value of the state variable, input the angle of attack command and the roll angle command into formula (5), integrate the system, and you can get the flight trajectory of the aircraft. At the same time, the hypersonic aircraft must meet the thermal flow constraints, overload constraints and dynamic pressure constraints during the flight. Therefore, it is necessary to establish thermal flux density constraints, overload constraints and dynamic pressure constraints. Among them, the thermal flux density constraint can be shown as follows in formula (9):

[0098]

[0099] Where Q represents the heat flux, represents the differential of heat flux, k Q is a constant coefficient related to the shape and appearance of the stationary point, and k is taken Q =6×10 -8 , v is the speed of the aircraft, ρ is the atmospheric density, and the simplified exponential form is shown in the following formula (10):

[0100]

[0101] Where, ρ0 = 1.225 g / L, h s =7200m; H is the flight altitude of the hypersonic vehicle.

[0102] Furthermore, the dynamic pressure constraint can be expressed as follows:

[0103]

[0104] Furthermore, the overload constraint can be expressed as follows:

[0105]

[0106] Where L and D represent the lift and drag of the aircraft respectively, α is the angle of attack, m is the mass of the aircraft, g0 is the gravitational acceleration at sea level, and n ymax Indicates the maximum overload value allowed by the aircraft.

[0107] Furthermore, in the process of actual application, since the flight state of a hypersonic aircraft can be analyzed by decomposing it from the longitudinal and lateral planes; and when the longitudinal plane is balanced, the aircraft is in a balanced gliding state; otherwise, the aircraft is in a jumping gliding state; at the same time, since the trajectory of the balanced gliding state is relatively straight and its trajectory prediction difficulty is relatively low, the focus of the flight trajectory prediction method of the aircraft described in the exemplary embodiment of the present disclosure is the trajectory prediction of jumping gliding aircraft. Furthermore, in the process of iterative calculation of the trajectory, the main factors affecting the longitudinal plane of the trajectory are the initial speed and altitude of the aircraft and the aerodynamic lift generated by the aircraft's angle of attack, while the main factors affecting the lateral plane movement of the aircraft are the initial speed and altitude of the aircraft and the aerodynamic lateral turning force generated by the aircraft's roll angle. Therefore, constructing a trajectory database requires giving the initial flight conditions and angle of attack control instructions and roll angle control instructions. In the process of actual application, refer to Figure 7 As shown in the figure, the specific implementation process of the Monte Carlo trajectory database simulation method is as follows:

[0108] Step 1: Given the range of the independent variable:

[0109] 0≤υ≤υ max , 20km≤h≤90km, 0≤α≤α max , 2000m / s≤V≤V max .

[0110] Step 2.1: Start the loop, set i=0, j=0, k=0; and initialize the given independent variable parameters respectively: i =υ max / n×i,h i =(h max -h min ) / n×i, α0=α max / n×i; where n is a positive integer representing the degree of trajectory differentiation; the roll angle control command adopts a constant command, expressed as follows: υ(t)=υ i ;

[0111] Step 2.2: Substitute the roll angle command formula and the angle of attack command calculation formula into the aircraft trajectory calculation formula for integral iteration, and record the longitude, latitude, and altitude data of the simulated trajectory;

[0112] Step 2.3: i=i+1, j=j+1, k=k+1, change the initial values ​​of speed, altitude and angle of attack respectively, and perform trajectory calculation;

[0113] Step 3: When i≤n, loop Step 2.1-Step 2.3 and record the input υ i , α i 、v i 、h i and output λ i ,φ i 、h i .

[0114] It should be noted that the original trajectory database needs to be generated because in the trajectory prediction problem, since the target is often non-cooperative, it is difficult to accurately determine the flight law of the target in a short time. In order to solve this problem, it is necessary to obtain the trajectory database (that is, the original trajectory database) of the gliding projectile based on Monte Carlo for different initial states and flight control laws of the aircraft, and then pre-train the 1DCNN-BiLSTM network with the simulated trajectory data to complete the pre-training of multiple neural network models.

[0115] In step S120, the network model to be trained is trained based on the original trajectory database to obtain a current trajectory prediction model.

[0116] Specifically, the network model to be trained is trained based on the original trajectory database to obtain the current trajectory prediction model, which can be achieved in the following way: the initial state variables and control laws corresponding to the original flight trajectory in the original trajectory database are input into the network model to be trained to obtain the current prediction result; the current loss function is constructed according to the current prediction result and the original flight trajectory, and the parameters in the network model to be trained are adjusted based on the current loss function to obtain the current trajectory prediction model. That is to say, in the actual application process, the original flight trajectory can be calculated based on the formula (5) recorded above, and the parameters used to calculate the original flight trajectory can be input into the network model to be trained to obtain the current prediction result; at the same time, when constructing the current loss function, the mean square error loss function can be used, and the cross entropy loss function can also be used. This example does not impose any special restrictions on this.

[0117] In step S130, according to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, the trajectory prediction model to be migrated is determined from the current trajectory prediction model, and the model parameters in the trajectory prediction model to be migrated are migrated to the trajectory prediction model to be corrected.

[0118] In this example embodiment, first, according to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, the trajectory prediction model to be migrated is determined from the current trajectory prediction model; specifically, it can be achieved in the following way: first, the target flight trajectory of the aircraft to be predicted is obtained, and the target flight trajectory is subjected to noise reduction processing to obtain the flight trajectory after noise reduction processing; secondly, a first trajectory sequence of the flight trajectory after noise reduction processing is determined, and a second trajectory sequence of the original flight trajectory in the original trajectory database is determined; then, the similarity between the first trajectory sequence and the second trajectory sequence is calculated, and the trajectory prediction model to be migrated is determined from the current trajectory prediction model according to the similarity.

[0119] In an exemplary embodiment, calculating the similarity between the first trajectory sequence and the second trajectory sequence, and determining the trajectory prediction model to be migrated from the current trajectory prediction model according to the similarity, can be implemented as follows: calculating the Euclidean distance between the first trajectory sequence and the second trajectory sequence, and determining the similarity between the first trajectory sequence and the second trajectory sequence according to the Euclidean distance; sorting the original flight trajectories based on the similarity between the first trajectory sequence and the second trajectory sequence, and extracting the original flight trajectory with the smallest similarity from the original flight trajectories based on the trajectory sorting result; and using the current trajectory prediction model corresponding to the original flight trajectory with the smallest similarity as the trajectory prediction model to be migrated.

[0120] The following will further explain and illustrate the specific determination process of the trajectory prediction model to be migrated. Specifically, in the actual application process, if it is necessary to predict the flight trajectory of the aircraft to be predicted, it is first necessary to observe the information of the aircraft to be predicted based on the monitoring system; at the same time, for the unknown information of the observed aircraft, since the observation system has a certain amount of noise, it is necessary to clean up the data for outliers and fill in the vacant values, and in addition, the position information needs to be filtered and smoothed; for example, assuming that the mean μ and standard deviation σ of two points of each adjacent three-point glide trajectory are calculated, then in the noise reduction process, the data exceeding (μ-3σ,μ+3σ) can be outliers removed, and then the vacant data after the outliers are deleted is calculated and filled with the data of the two adjacent points. The supplementary formula is shown in the following formula (13):

[0121] y(d,t)=(y(d,t-1)+y(d,t+1)) / 2; Formula (13)

[0122] Among them, y(d,t) represents the missing value filling, which is obtained by linearly averaging the data before and after.

[0123] Furthermore, after obtaining the flight trajectory after noise reduction processing, the similarity between the first trajectory sequence and the second trajectory sequence can be calculated; wherein, in the process of similarity selection, the historical sequence of the position information of the target aircraft can be respectively calculated for similarity with the trajectory information in the trajectory library at the same time step, or the similarity between the target flight trajectory and each trajectory in the trajectory library; wherein, the calculation of trajectory similarity can be characterized by the spatial Euclidean distance between the target trajectory and the database trajectory; wherein, the specific calculation process of the Euclidean distance can be shown in the following formula (14):

[0124]

[0125] Among them, d(x,y,z) is the Euclidean distance, x r ,y r 、z r is the first trajectory sequence, x p ,y p 、z p is the second trajectory sequence. Further, after obtaining the Euclidean distance, the similarity between the target trajectory and the database can be obtained. The smaller the Euclidean distance, the smaller the trajectory difference and the higher the similarity. The source domain data with the smallest trajectory similarity is selected as the migration data (the trajectory prediction model to be migrated); wherein, the specific selection process of the trajectory prediction model to be migrated can be shown in the following formula (15):

[0126]

[0127] Secondly, after obtaining the trajectory prediction model to be migrated, the model parameters in the trajectory prediction model to be migrated can be migrated to the trajectory prediction model to be corrected; specifically, it can be achieved in the following way: extract the model parameters in the trajectory prediction model to be migrated, and perform parameter initialization processing on the model parameters in the trajectory prediction model to be corrected based on the extracted model parameters to achieve parameter migration processing. Specifically, the migration process recorded here can also be called transfer learning; wherein, transfer learning is a branch of machine learning, and transfer learning mainly involves the concepts of two domains, the source domain and the target domain. The source of the migration is called the source domain (Ds), and the destination of the migration information is called the target domain (Dt). Transfer learning requires the use of the model parameters learned in the existing source domain to perform knowledge transfer on the small sample unknown model in the target domain; wherein, the source domain of this example embodiment is also the original trajectory database recorded in the previous text, and the target domain is also the target flight trajectory of the aircraft to be predicted. Furthermore, in the process of actual application, transfer learning first trains all the ballistic information in the source domain and saves the trained network model; then, the network model parameters in the source domain are copied and migrated to the target domain, and the source domain network parameters that are most similar to the target domain data set are selected for fine-tuning; finally, the target trajectory feature values ​​obtained by the CNN network are fixed, the BiLSTM network parameters are fine-tuned, and the adjusted network model is used as a prediction model for the prediction of target domain data.

[0128] In step S140, parameters of the trajectory prediction model to be corrected are corrected based on the target flight trajectory to obtain a target trajectory prediction model, and the flight trajectory of the aircraft to be predicted is predicted based on the target trajectory prediction model.

[0129] In this example embodiment, first, the trajectory prediction model to be corrected is parameterized based on the target flight trajectory to obtain the target trajectory prediction model; specifically, this can be achieved in the following manner: first, sample data of the observed aircraft to be predicted is obtained, and the sample data is input into the trajectory prediction model to be corrected to obtain a target prediction result; secondly, a target loss function is constructed according to the target prediction result and the target flight trajectory, and the trajectory prediction model to be corrected is parameterized based on the target loss function to obtain the target trajectory prediction model; wherein, the trajectory prediction model to be corrected is parameterized based on the target loss function to obtain the target trajectory prediction model, which can be achieved in the following manner: the weight value of the convolutional neural network is fixed, and the weight value and bias of the bidirectional long short-term memory network model are fine-tuned based on the target loss function to obtain the target trajectory prediction model; wherein, in the process of fine-tuning the weight value and bias of the bidirectional long short-term memory network model, the selected optimizer includes an Adam optimizer that adaptively adjusts the learning rate.

[0130] The following will further explain and illustrate the specific determination process of the target trajectory prediction model. Specifically, in the actual application process, after the parameters of the 1DCNN-BiLSTM network model with the highest similarity are migrated to the target domain prediction model as the initial parameters, the weights of the CNN network can be fixed, and the weights and biases of the Bi-LSTM network can be fine-tuned using the observed sample data; at the same time, in the process of parameter fine-tuning, the optimizer selects the Adam optimizer that can adaptively adjust the learning rate, and the initial learning rate is adjusted to 0.005. Its calculation formula can be shown in the following formula (16):

[0131]

[0132] Among them, g t represents the gradient at the tth time step, Represents the square of the gradient at the tth time step. The optimizer can accumulate historical data of each gradient parameter and adaptively adjust the learning rate as the number of learning times increases. In the process of parameter fine-tuning, the network fine-tuning parameter settings can be specifically shown in Table 1 below:

[0133] Table 1 Network fine-tuning parameter settings

[0134]

[0135] Furthermore, in order to quickly fine-tune the migration model, it is necessary to control the number of nodes in the LSTM unit and define the number of hidden nodes as 64. The specific settings of the LSTM internal unit structure can be shown in Table 2 below:

[0136] Table 2 LSTM network structure

[0137]

[0138] Finally, after obtaining the target trajectory prediction model, it is necessary to evaluate the target trajectory prediction model. Specifically, in the model evaluation, the test data can be input into the trained prediction model, and the mean square error function (RMSE) is used as the accuracy evaluation index to evaluate the model. Specifically, the mean square error function can be shown as the following formula (17):

[0139]

[0140] Among them, M is the number of data samples in the network prediction set; is the network prediction value after denormalization; y i is the test data label value.

[0141] At this point, the specific fine-tuning process of the target trajectory prediction model has been fully realized. Furthermore, after obtaining the target trajectory prediction model, the flight trajectory of the aircraft to be predicted can be predicted based on the target trajectory prediction model; specifically, it can be achieved in the following way: obtain the flight speed, ballistic inclination, ballistic deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude of the aircraft to be predicted; construct a model input vector based on the flight speed, ballistic inclination, ballistic deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude of the aircraft to be predicted; input the model input vector into the target trajectory prediction model to obtain the flight trajectory prediction result of the aircraft to be predicted; wherein, the specific prediction process can be referred to. Figure 8 shown.

[0142] It should be noted that the reason why it is necessary to determine the target trajectory prediction model and predict the flight trajectory of the aircraft to be predicted based on the target trajectory prediction model is that when observing the aircraft to be predicted, the mass and equivalent area of ​​the aircraft to be predicted cannot be accurately known, so it is impossible to accurately predict the flight trajectory of the aircraft to be predicted based on the aforementioned formula (5). The present application solves this problem by determining the target trajectory prediction model and predicting the flight trajectory of the aircraft to be predicted based on the target trajectory prediction model.

[0143] The following are embodiments of the device disclosed herein, which can be used to execute the method embodiments disclosed herein. For details not disclosed in the device embodiments disclosed herein, please refer to the method embodiments disclosed herein.

[0144] The exemplary embodiment of the present disclosure also provides a flight trajectory prediction device for an aircraft. Fig. 9 As shown, the flight trajectory prediction device of the aircraft may include an original trajectory database construction module 910, a network model training module 920, a model parameter migration module 930 and a flight trajectory prediction module 940. Among them:

[0145] The original trajectory database construction module 910 can be used to establish the motion equation of the hypersonic aircraft, and determine the original trajectory database of the hypersonic aircraft under different initial states and different flight control laws based on the motion equation;

[0146] A network model training module 920 may be used to train the network model to be trained based on the original trajectory database to obtain a current trajectory prediction model;

[0147] The model parameter migration module 930 may be used to determine the trajectory prediction model to be migrated from the current trajectory prediction model according to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, and migrate the model parameters in the trajectory prediction model to be migrated to the trajectory prediction model to be corrected;

[0148] The flight trajectory prediction module 940 may be used to perform parameter correction on the trajectory prediction model to be corrected based on the target flight trajectory to obtain a target trajectory prediction model, and predict the flight trajectory of the aircraft to be predicted based on the target trajectory prediction model.

[0149] In an exemplary embodiment of the present disclosure, a motion equation of a hypersonic aircraft is established, including: obtaining the flight speed, ballistic inclination, ballistic deviation, distance from the hypersonic aircraft to the center of the earth, longitude, and latitude of the hypersonic aircraft; in a ballistic coordinate system, establishing a three-degree-of-freedom motion equation of the hypersonic aircraft according to the flight speed, ballistic inclination, ballistic deviation, distance from the hypersonic aircraft to the center of the earth, longitude, and latitude.

[0150] In an exemplary embodiment of the present disclosure, an original trajectory database of a hypersonic aircraft under different initial states and different flight control laws is determined based on a motion equation, including: determining different initial state variables of the hypersonic aircraft based on flight speed, trajectory inclination, trajectory deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude; determining different flight control laws of the hypersonic aircraft according to an angle of attack control instruction and a roll angle control instruction of the hypersonic aircraft; inputting the different initial state variables and different flight control laws into a three-degree-of-freedom motion equation to obtain original flight trajectories under different initial state variables and different flight control laws; generating the original trajectory database according to the original flight trajectory and the initial state variables and control laws corresponding to the original flight trajectory.

[0151] In an exemplary embodiment of the present disclosure, a network model to be trained is trained based on the original trajectory database to obtain a current trajectory prediction model, including: inputting the initial state variables and control laws corresponding to the original flight trajectory in the original trajectory database into the network model to be trained to obtain a current prediction result; constructing a current loss function according to the current prediction result and the original flight trajectory, and adjusting the parameters in the network model to be trained based on the current loss function to obtain the current trajectory prediction model.

[0152] In an exemplary embodiment of the present disclosure, a trajectory prediction model to be migrated is determined from a current trajectory prediction model based on a target flight trajectory of the aircraft to be predicted and an original flight trajectory in an original trajectory database, including: acquiring the target flight trajectory of the aircraft to be predicted, and performing noise reduction processing on the target flight trajectory to obtain a flight trajectory after noise reduction processing; determining a first trajectory sequence of the flight trajectory after noise reduction processing, and determining a second trajectory sequence of the original flight trajectory in the original trajectory database; calculating the similarity between the first trajectory sequence and the second trajectory sequence, and determining the trajectory prediction model to be migrated from the current trajectory prediction model based on the similarity.

[0153] In an exemplary embodiment of the present disclosure, the similarity between the first trajectory sequence and the second trajectory sequence is calculated, and the trajectory prediction model to be migrated is determined from the current trajectory prediction model according to the similarity, including: calculating the Euclidean distance between the first trajectory sequence and the second trajectory sequence, and determining the similarity between the first trajectory sequence and the second trajectory sequence according to the Euclidean distance; sorting the original flight trajectory based on the similarity between the first trajectory sequence and the second trajectory sequence, and extracting the original flight trajectory with the smallest similarity from the original flight trajectory based on the trajectory sorting result; and using the current trajectory prediction model corresponding to the original flight trajectory with the smallest similarity as the trajectory prediction model to be migrated.

[0154] In an exemplary embodiment of the present disclosure, the model parameters in the trajectory prediction model to be migrated are migrated to the trajectory prediction model to be corrected, including: extracting the model parameters in the trajectory prediction model to be migrated, and performing parameter initialization processing on the model parameters in the trajectory prediction model to be corrected based on the extracted model parameters to realize parameter migration processing.

[0155] In an exemplary embodiment of the present disclosure, parameters of the trajectory prediction model to be corrected are performed based on the target flight trajectory to obtain the target trajectory prediction model, including: acquiring sample data of the observed aircraft to be predicted, and inputting the sample data into the trajectory prediction model to be corrected to obtain a target prediction result; constructing a target loss function according to the target prediction result and the target flight trajectory, and performing parameter correction on the trajectory prediction model to be corrected based on the target loss function to obtain the target trajectory prediction model.

[0156] In an exemplary embodiment of the present disclosure, the trajectory prediction model to be corrected includes a convolutional neural network model and a bidirectional long short-term memory network model; wherein, based on the target loss function, the parameters of the trajectory prediction model to be corrected are corrected to obtain a target trajectory prediction model, including: fixing the weight value of the convolutional neural network, and fine-tuning the weight value and bias of the bidirectional long short-term memory network model based on the target loss function to obtain the target trajectory prediction model; wherein, in the process of fine-tuning the weight value and bias of the bidirectional long short-term memory network model, the selected optimizer includes an Adam optimizer that adaptively adjusts the learning rate.

[0157] In an exemplary embodiment of the present disclosure, the flight trajectory of the aircraft to be predicted is predicted based on the target trajectory prediction model, including: obtaining the flight speed, ballistic inclination, ballistic deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude of the aircraft to be predicted; constructing a model input vector according to the flight speed, ballistic inclination, ballistic deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude of the aircraft to be predicted; inputting the model input vector into the target trajectory prediction model to obtain the flight trajectory prediction result of the aircraft to be predicted.

[0158] The specific details of each module in the above-mentioned aircraft flight trajectory prediction device have been described in detail in the corresponding aircraft flight trajectory prediction method, so they will not be repeated here.

[0159] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0160] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0161] In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.

[0162] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0163] Refer to the following Fig.10 The electronic device 1000 according to this embodiment of the present disclosure is described. Fig.10 The electronic device 1000 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0164] like Fig.10 As shown, the electronic device 1000 is in the form of a general computing device. The components of the electronic device 1000 may include, but are not limited to: the at least one processing unit 1010, the at least one storage unit 1020, a bus 1030 connecting different system components (including the storage unit 1020 and the processing unit 1010), and a display unit 1040.

[0165] The storage unit stores program codes, which can be executed by the processing unit 1010, so that the processing unit 1010 performs the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit 1010 can perform the following steps: Figure 1 The step S110 shown in the figure is as follows: establishing the motion equation of the hypersonic aircraft, and determining the original trajectory database of the hypersonic aircraft under different initial states and different flight control laws based on the motion equation; step S120: training the network model to be trained based on the original trajectory database to obtain the current trajectory prediction model; step S130: determining the trajectory prediction model to be migrated from the current trajectory prediction model according to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, and migrating the model parameters in the trajectory prediction model to be migrated to the trajectory prediction model to be corrected; step S140: performing parameter correction on the trajectory prediction model to be corrected based on the target flight trajectory to obtain the target trajectory prediction model, and predicting the flight trajectory of the aircraft to be predicted based on the target trajectory prediction model.

[0166] The storage unit 1020 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 10201 and / or a cache storage unit 10202 , and may further include a read-only storage unit (ROM) 10203 .

[0167] The storage unit 1020 may also include a program / utility 10204 having a set (at least one) of program modules 10205, such program modules 10205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0168] Bus 1030 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0169] The electronic device 1000 may also communicate with one or more external devices 1100 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 1000, and / or communicate with any device that enables the electronic device 1000 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 1050. Furthermore, the electronic device 1000 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 1060. As shown, the network adapter 1060 communicates with other modules of the electronic device 1000 via a bus 1030. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 1000, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0170] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0171] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above "Exemplary Method" section of the present specification.

[0172] According to the program product for implementing the above method in the embodiment of the present disclosure, it can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0173] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0174] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0175] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0176] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0177] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0178] Other embodiments of the present disclosure will be readily apparent to those skilled in the art after considering the specification and practicing the inventions invented herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not invented by the present disclosure. The specification and examples are to be considered merely exemplary, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A method for predicting a flight trajectory of an aircraft, characterized in that: include: Establishing a motion equation of a hypersonic aircraft, and determining an original trajectory database of the hypersonic aircraft under different initial states and different flight control laws based on the motion equation; Training the network model to be trained based on the original trajectory database to obtain a current trajectory prediction model; According to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, a trajectory prediction model to be migrated is determined from the current trajectory prediction model, and model parameters in the trajectory prediction model to be migrated are migrated to the trajectory prediction model to be corrected; The trajectory prediction model to be corrected is parameterized based on the target flight trajectory to obtain a target trajectory prediction model, and the flight trajectory of the aircraft to be predicted is predicted based on the target trajectory prediction model.

2. The method for predicting the flight trajectory of an aircraft according to claim 1, characterized in that: Establish the equations of motion for a hypersonic vehicle, including: Obtain the flight speed, ballistic inclination, ballistic deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude of the hypersonic aircraft; In the ballistic coordinate system, the three-degree-of-freedom motion equation of the hypersonic aircraft is established according to the flight speed, ballistic inclination, ballistic deviation, the distance from the hypersonic aircraft to the center of the earth, the longitude and the latitude.

3. The method for predicting the flight trajectory of an aircraft according to claim 1, characterized in that: The original trajectory database of hypersonic vehicles under different initial states and different flight control laws is determined based on the equation of motion, including: Determining different initial state variables of the hypersonic aircraft based on flight speed, trajectory inclination, trajectory deviation, distance from the hypersonic aircraft to the center of the earth, longitude, and latitude; Determining different flight control laws of the hypersonic aircraft according to the angle of attack control command and the roll angle control command of the hypersonic aircraft; Inputting the different initial state variables and different flight control laws into the motion equation of three degrees of freedom to obtain original flight trajectories under different initial state variables and different flight control laws; The original trajectory database is generated according to the original flight trajectory and the initial state variables and control laws corresponding to the original flight trajectory.

4. The method for predicting the flight trajectory of an aircraft according to claim 1, characterized in that: The network model to be trained is trained based on the original trajectory database to obtain a current trajectory prediction model, including: Inputting the initial state variables and control laws corresponding to the original flight trajectory in the original trajectory database into the network model to be trained to obtain the current prediction result; A current loss function is constructed according to the current prediction result and the original flight trajectory, and parameters in the network model to be trained are adjusted based on the current loss function to obtain a current trajectory prediction model.

5. The method for predicting the flight trajectory of an aircraft according to claim 1, characterized in that: According to the target flight trajectory of the aircraft to be predicted and the original flight trajectory in the original trajectory database, the trajectory prediction model to be migrated is determined from the current trajectory prediction model, including: Acquiring a target flight trajectory of the aircraft to be predicted, and performing noise reduction processing on the target flight trajectory to obtain a flight trajectory after noise reduction processing; Determine a first trajectory sequence of the flight trajectory after the noise reduction process, and determine a second trajectory sequence of the original flight trajectory in the original trajectory database; The similarity between the first trajectory sequence and the second trajectory sequence is calculated, and a trajectory prediction model to be migrated is determined from the current trajectory prediction model according to the similarity.

6. The method for predicting the flight trajectory of an aircraft according to claim 5, characterized in that: Calculating the similarity between the first trajectory sequence and the second trajectory sequence, and determining a trajectory prediction model to be migrated from the current trajectory prediction model according to the similarity, including: Calculating the Euclidean distance between the first trajectory sequence and the second trajectory sequence, and determining the similarity between the first trajectory sequence and the second trajectory sequence according to the Euclidean distance; sorting the original flight trajectories based on the similarity between the first trajectory sequence and the second trajectory sequence, and extracting the original flight trajectory with the smallest similarity from the original flight trajectories based on the trajectory sorting result; The current trajectory prediction model corresponding to the original flight trajectory with the smallest similarity is used as the trajectory prediction model to be migrated.

7. The method for predicting the flight trajectory of an aircraft according to claim 1, characterized in that: Migrating the model parameters in the trajectory prediction model to be migrated to the trajectory prediction model to be corrected, including: The model parameters in the trajectory prediction model to be migrated are extracted, and the model parameters in the trajectory prediction model to be corrected are initialized based on the extracted model parameters to achieve parameter migration.

8. The method for predicting the flight trajectory of an aircraft according to claim 1, characterized in that: Based on the target flight trajectory, the trajectory prediction model to be corrected is subjected to parameter correction to obtain a target trajectory prediction model, including: Obtaining sample data of the observed aircraft to be predicted, and inputting the sample data into the trajectory prediction model to be corrected to obtain a target prediction result; A target loss function is constructed according to the target prediction result and the target flight trajectory, and parameters of the trajectory prediction model to be corrected are corrected based on the target loss function to obtain a target trajectory prediction model.

9. The method for predicting the flight trajectory of an aircraft according to claim 8, characterized in that: The trajectory prediction model to be corrected includes a convolutional neural network model and a bidirectional long short-term memory network model; Among them, performing parameter correction on the trajectory prediction model to be corrected based on the target loss function to obtain the target trajectory prediction model includes: The weight value of the convolutional neural network is fixed, and the weight value and bias of the bidirectional long short-term memory network model are fine-tuned based on the target loss function to obtain a target trajectory prediction model; In the process of fine-tuning the weight value and bias of the bidirectional long short-term memory network model, the selected optimizer includes an Adam optimizer that adaptively adjusts the learning rate.

10. The method for predicting the flight trajectory of an aircraft according to claim 1, characterized in that: Predicting the flight trajectory of the aircraft to be predicted based on the target trajectory prediction model includes: Obtaining the flight speed, trajectory inclination, trajectory deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude of the aircraft to be predicted; Constructing a model input vector according to the flight speed, trajectory inclination, trajectory deviation, distance from the hypersonic aircraft to the center of the earth, longitude and latitude of the aircraft to be predicted; The model input vector is input into the target trajectory prediction model to obtain the flight trajectory prediction result of the aircraft to be predicted.

Images