Deformable aircraft prediction correction guidance method based on deep learning

Abstract

The invention provides a deformation aircraft prediction correction guidance method based on deep learning, and belongs to the field of aircraft guidance and control. The method comprises the steps that firstly, a deformation aircraft flight performance prediction model is trained, input of the model is speed, guidance parameters and deformation parameters of a deformation aircraft, and output of the model is a range capability value and a transverse range capability value; and in each guidance cycle, based on the prediction model, constructing a nonlinear optimization problem taking the transverse range capability as an optimization performance index and taking the required range as a constraint, and solving the nonlinear optimization problem to obtain a current guidance parameter and a deformation parameter of the deformable aircraft, so as to obtain a current attack angle amplitude and a current heeling angle amplitude, thereby realizing prediction correction guidance of the deformable aircraft. The method can make a guidance decision giving consideration to the deformation instruction, the attack angle instruction and the heeling angle instruction in real time in the flight process of the deformable aircraft, enables the transverse stroke of the deformable aircraft to be maximum on the premise that the deformable aircraft reaches a target point, and can effectively improve the transverse maneuverability of the deformable aircraft.

Description

Technical Field

[0001] The present invention belongs to the technical field of aircraft guidance and control, and in particular relates to a deformation aircraft prediction and correction guidance method based on deep learning. Background Art

[0002] A shape-shifting aircraft is a type of aircraft that changes its wing shape, wingspan and other structures during flight. Unlike traditional aircraft that can only be used for a single purpose in a specific environment, a shape-shifting aircraft can change its shape according to the needs of multiple factors such as the environment, profile and combat mission. Through deformation, the aircraft can effectively increase its range and reduce fuel consumption, and can adapt to flight missions with a wide speed range and long range, and has better mission adaptability. Reentry flight refers to the process of an aircraft returning to the Earth's atmosphere and accurately arriving at the predetermined location, facing strict mechanical and thermal load environmental conditions. In order to better exert the flight performance of the shape-shifting aircraft and meet flight constraints, it is necessary to design a reentry guidance law suitable for the shape-shifting aircraft.

[0003] Most existing methods focus on studying trajectory planning methods for deformable aircraft, while there are few predictive correction guidance methods for reentry of deformable aircraft. The predictive correction guidance method for conventional aircraft does not consider the impact of deformation on aerodynamic performance and cannot be directly transferred and applied to the reentry guidance of deformable aircraft. In order to comprehensively consider the impact of deformation factors and guidance instructions on flight missions, it is very important to design a predictive correction guidance method for deformable aircraft. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the existing technology and propose a prediction and correction guidance method for a deformable aircraft based on deep learning. The present invention can make guidance decisions that take into account deformation instructions, angle of attack and roll angle instructions in real time during the flight of the deformable aircraft, so that the lateral travel of the deformable aircraft is maximized on the premise of reaching the target point, which can effectively improve the lateral maneuverability of the deformable aircraft, thereby improving the maneuverability and anti-interception capability of the deformable aircraft.

[0005] The embodiment of the present invention provides a prediction and correction guidance method for a deformable aircraft based on deep learning, comprising:

[0006] Training a flight performance prediction model for a deformable aircraft, wherein the input of the model is the speed, guidance parameters, and deformation parameters of the deformable aircraft, and the output is the range capability value and the lateral range capability value of the deformable aircraft;

[0007] In each guidance cycle, based on the flight performance prediction model of the deformable aircraft, a nonlinear optimization problem is constructed with the lateral range capability as the optimization performance index and the required range as the constraint;

[0008] The nonlinear optimization problem is solved to obtain the current guidance parameters and deformation parameters of the deformable aircraft, and then the current attack angle and roll angle amplitude of the deformable aircraft are obtained to achieve predictive correction guidance for the deformable aircraft.

[0009] In a specific embodiment of the present invention, the deformable aircraft flight performance prediction model adopts a neural network model with four fully connected layers, the first layer of the model adopts the elu function, and the second and third layers adopt the tanh function.

[0010] In a specific embodiment of the present invention, the method further comprises:

[0011] Based on the relationship between the attack angle and speed of the deformable aircraft, the roll angle corridor of the deformable aircraft in each aerodynamic configuration is established; the specific steps are as follows:

[0012] 1) Establish the relationship between the attack angle and speed of the deformable aircraft;

[0013] Among them, the expression of the angle of attack of the deformable aircraft with respect to the speed during flight is:

[0014]

[0015] Among them, α is the attack angle of the deformable aircraft, V is the speed of the deformable aircraft; V b ,V a are the preset speed segmentation points, V b is smaller than the initial velocity of the deformed aircraft, V a Greater than the reentry terminal velocity V of the deformed vehicle f ; is the angle of attack with the maximum lift-to-drag ratio; α m is the initial value of the angle of attack;

[0016] 2) Establish the roll angle corridor of the shape-shifting aircraft in each aerodynamic configuration;

[0017] The reentry flight of the shape-shifting aircraft in each aerodynamic configuration is divided into the initial descent phase, the balanced flight phase, and the terminal capability management phase. For any aerodynamic configuration:

[0018] The initial descent segment includes the velocity range from the reentry point to the first valley point of the trajectory, V ≥ V b ; Upper limit of the initial descending section tilt angle corridor ν 1,max and the lower bound ν 1,min The expression is:

[0019]

[0020] Where k is the deformation coefficient, which indicates the deformation of the deformable aircraft; ν initis the constant roll angle adopted in the initial descent;

[0021] The minimum tilt angle of the balanced flight stage is 0, and the speed range is [V a ,V b ], the balanced flight stage includes three process constraints: overload constraint, dynamic pressure constraint and thermal flow constraint, and the expression is as follows:

[0022]

[0023] in, They represent the heat flux, overload and dynamic pressure borne by the deformed aircraft at time t, They represent the maximum heat flux, maximum dynamic pressure, and maximum overload constraints that the deformable aircraft can withstand respectively; S ref , m0, g0 represent the reference area, mass and gravitational acceleration of the deformable aircraft respectively; C L (α,V,k),C D (α, V, k) represent the lift coefficient and drag coefficient of the deformable aircraft respectively; ρ is the atmospheric density at the current altitude of the deformable aircraft; C1 is the heat flow calculation constant; V c is the first cosmic velocity; ρ0 is the atmospheric density at sea level; R d represents the radius of curvature of the leading edge of the deformable aircraft;

[0024] Then the air density of the minimum distance from the center of the earth corresponding to each constraint in formula (3) is:

[0025]

[0026] Among them, r Q ,r p ,r n They represent the minimum geocentric distances corresponding to the heat flow constraint, overload constraint, and dynamic pressure constraint respectively;

[0027] Take r in formula (4) Q ,r n ,r p The minimum value of is taken as the minimum flight geocenter distance, and the expression is as follows:

[0028] r min (V) = min{r Q ,r q ,r n} (5)

[0029] Then calculate the maximum roll angle of the balanced flight segment according to the quasi-balanced flight conditions:

[0030]

[0031] Where g is the gravitational acceleration at the height of the deformable aircraft, ν2,max (V,k) represents the maximum roll angle calculated according to the quasi-balanced gliding condition;

[0032] The upper limit of the roll angle corridor in the balanced flight phase is ν 2,max and the lower bound ν 2,min It is expressed as:

[0033]

[0034] The terminal energy management section satisfies the terminal state constraint, and the speed range is V∈[V f ,V a ];

[0035] Combined with the quasi-balanced flight condition, the terminal balanced flight roll angle ν QEGC,f The calculation expression is:

[0036]

[0037] Among them, r f ,v f They represent the terminal geocentric distance and terminal velocity of the reentry flight respectively; the upper limit of the roll angle corridor of the terminal energy management phase ν 3,max (V,k) and the lower bound ν 3,min The expression of (V,k) is:

[0038]

[0039] Among them, k ν is the shrinkage coefficient.

[0040] In a specific embodiment of the present invention, the method further comprises:

[0041] Based on the roll angle corridor of the deformable aircraft in each aerodynamic configuration, a training set for training the flight performance prediction model of the deformable aircraft is established, and the specific steps are as follows:

[0042] 1) Based on the value range of the guidance parameter ω and the deformation coefficient k, the flight trajectory simulation results of the deformable aircraft are obtained under different combinations of the guidance parameter ω and the deformation coefficient k, and the ballistic information of the flight trajectory is sampled;

[0043] In the simulation of each flight trajectory, according to the guidance parameter ω and the roll angle corridor, the roll angle ν = |ν cmd |:

[0044] |ν cmd |=(1-ω)ν min (k)+ων max (k) (10)

[0045] Among them, the value range of ω is [0,1], ν max (k),ν min (k) are the upper and lower limits of the roll angle corridor corresponding to the aerodynamic configuration with deformation coefficient k;

[0046] According to the flight speed and formula (1), the angle of attack α is calculated;

[0047] The simulation is performed using the roll angle and the angle of attack to obtain the flight trajectory simulation result under the corresponding combination of ω and k; then, the basic ballistic information sequence of the flight trajectory is recorded by sampling, and the ballistic information includes: the height r of each sampling point i , longitude λ i , latitude φ i , speed V i 、Ballistic inclination angle θ i , trajectory deviation angle σ i and flight time t i , where the subscript i represents the i-th sampling point;

[0048] 2) Based on the trajectory information obtained in step 1), the flight performance index of each sampling point on each flight trajectory is calculated, including: the aircraft range capability and the aircraft lateral range capability;

[0049] The calculation expression of the aircraft range capability at the i-th sampling point is as follows:

[0050]

[0051] Among them, DR i represents the range capability of the aircraft at the i-th sampling point, S i represents the flight range of the aircraft at the i-th sampling point, S f Indicates the distance the aircraft has flown along the entire trajectory;

[0052] Let the coordinate difference between the i-th sampling point and the terminal point in the trajectory coordinates be:

[0053]

[0054] Among them, λ f ,φ f Indicates the longitude and latitude of the terminal point of this trajectory, Δ xi ,Δ yi Respectively represent the longitude difference and latitude difference between the i-th sampling point and the terminal point in the trajectory coordinates;

[0055] Then the aircraft range capability at the i-th sampling point is:

[0056] CR i =|Δ yi cos(σ i )-Δxi sin(σ i )| (13)

[0057] 3) Using the results of steps 1) and 2), construct a training set;

[0058] In the set speed range [V f , V0] to obtain a uniformly distributed velocity sequence V uniform , where V f is the reentry flight termination speed, V0 is the reentry flight initial speed;

[0059] On each flight trajectory obtained in step 1), according to the original V i ,CR i ,DR i Data sequence, interpolation to obtain the aircraft range capability and aircraft cross-range capability in the speed sequence V uniform The corresponding values ​​on the interpolation result are denoted as CR uniform ,DR uniform ;

[0060] Using V uniform ,CR uniform ,DR uniform Construct a training set; wherein the input of a single training sample of the training set is the number of times the deformed aircraft in each flight trajectory is in V uniform The output is the aircraft range capability value and the aircraft lateral range capability value corresponding to the speed value under the flight trajectory, respectively from CR uniform ,DR uniform Get from

[0061] The input and output of the training set are recorded as:

[0062] X={x j,s |1≤j≤42000,j∈N,s∈{1,2,3}},Y={y j,s |1≤j≤42000,j∈N,s∈{1,2}},

[0063] Among them, X represents the input data set of the training set, Y represents the output data set of the training set, j represents the jth training sample, s represents the sth dimension, and x j,s Represents the value of the sth dimension of the input data in the jth training sample, y j,s Represents the value of the sth dimension of the output data in the jth training sample.

[0064] In a specific embodiment of the present invention, the method further comprises:

[0065] Before using the training set to train the deformable aircraft flight performance prediction model, the data of each dimension of the training set is normalized, and the calculation expression is as follows:

[0066]

[0067] in, represents the normalized value of the input data of the sth dimension of the jth training sample, represents the normalized value of the output data of the sth dimension of the jth training sample; x mean,s ,x std,s Respectively represent the mean and variance of the input data of the sth dimension of the training sample; y mean,s ,y std,s They represent the mean and variance of the output data of the sth dimension of the training sample respectively.

[0068] In a specific embodiment of the present invention, the construction of a nonlinear optimization problem with the horizontal range capability as the optimization performance index and the required range as the constraint includes:

[0069] Calculate the required range d based on the position of the transformable aircraft and the handover point desired , the calculation expression is as follows:

[0070] d desired = R arccos[cos(φ msl )·cos(φ tgt )·cos(λ msl -λ tgt )+sin(φ msl )·sin(φ tgt )] (16)

[0071] Among them, λ msl ,λ tgt are the longitudes of the deformed aircraft and the target point, respectively, msl ,φ tgt are the latitudes of the deformed aircraft and the target point, respectively;

[0072] The predicted value output by the deformable aircraft flight performance prediction model is CR net (V,ω,k),DR net (V,ω,k), then the nonlinear optimization problem with equality constraints is constructed as follows:

[0073]

[0074] Among them, ω min ,ω max Respectively represent the minimum and maximum values ​​of the guidance parameters; k min ,kmax Represent the minimum and maximum values ​​of the deformation coefficient respectively.

[0075] In a specific embodiment of the present invention, solving the nonlinear optimization problem to obtain the current guidance parameters and deformation parameters of the deformable aircraft includes:

[0076] 1) Based on the optimization problem of formula (17), the augmented Lagrangian function is constructed:

[0077] L=CR net (V,ω,k)+λ(DR net (V,ω,k)-d desired )+ρ||DR net (V,ω,k)-d desired || 2 (18)

[0078] Among them, λ,ρ are Lagrange multipliers, ρ is a positive number;

[0079] 2) Set the initial iteration values ​​ω0, k0, λ0;

[0080] 3) Combined with the trained deformable aircraft flight performance prediction model, the gradient ascent method is used to optimize the variables, and the variable obtained in the nth iteration is set as ω n ,k n ,λ n ,but:

[0081]

[0082] in, All are calculated by the trained deformation aircraft flight performance prediction model;

[0083] 4) Update the augmentation coefficient λ n+1 =λ n +ρ(DR net (V,ω,k)-d desired );

[0084] 5) Judgment: If |ω is satisfied n+1 -ω n |≤∈ ω ,|k n+1 -k n |≤∈ k , where ∈ ω ,∈ k is the convergence threshold, then go to step 6); otherwise, set n=n+1 and return to step 3);

[0085] 6) Determine ω*=ω n ,k*=k n For the root of seeking;

[0086] The obtained k* is used as the deformation parameter instruction k output by the guidance law. cmd .

[0087] In a specific embodiment of the present invention, obtaining the current attack angle and roll angle amplitude of the deformable aircraft includes:

[0088] 1) According to the current speed of the deformable aircraft and formula (1), the angle of attack command α is calculated cmd ;

[0089] 2) Calculate the roll angle amplitude according to the guidance parameter ω*. The calculation expression is:

[0090] ν cmd |=(1-ω*)ν min (k)+ω*ν max (k) (20).

[0091] In a specific embodiment of the present invention, the method further comprises:

[0092] Set the cross-range threshold to δCR thre , the range error contraction coefficient is K, then in each subsequent guidance cycle, the sign of the roll angle is obtained by the following method:

[0093] 1) Assume the current heading angle of the aircraft is σ, calculate the lateral range CR of the target point relative to the current heading of the aircraft tgt , the expression is as follows:

[0094] CR tgt =(φ tgt -φ)sinσ-(λ tgt -λ)cosσ (21)

[0095] 2) Let the roll angle command obtained in the previous guidance cycle be ν last , calculate the horizontal error, the expression is as follows:

[0096]

[0097] Where, sign function is the sign function; if the aircraft is in the first guidance cycle, then ν last The sign of is positive, and the cross-travel error record value δCR is set record = δCR;

[0098] 3) Calculate the roll angle flip flag, the expression is as follows:

[0099]

[0100] 4) Calculate the roll angle command and update the traverse record value;

[0101] If FLAG = True, the cross-distance error obtained in Step 2 is used as the new cross-distance record value δCR record =δCR, the tilt angle command is ν cmd =-sign(ν last )|ν cmd |;

[0102] If FLAG = False, the traverse record value is not updated and the roll angle command is ν cmd =sign(ν last )|ν cmd |.

[0103] The characteristics and beneficial effects of the present invention are:

[0104] (1) The present invention is based on the typical aerodynamic configuration of a deformable aircraft and, according to the flight process constraints, generates a roll angle corridor for the deformable aircraft with a fixed angle of attack profile, thereby facilitating parameterization of the control profile of the deformable aircraft and simplifying the prediction correction guidance decision space.

[0105] (2) The present invention constructs a data set of flight performance indicators of deformable aircraft and trains a deep learning model for flight performance prediction. When making online decisions, it can take into account the impact of deformation decisions and control decisions on the overall flight mission. It has good real-time performance, high guidance accuracy, and high practical value.

[0106] (3) The method of the present invention can be widely used in various mission scenarios of shape-shifting aircraft, and has important reference value for the precise guidance and control of shape-shifting aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] Figure 1 A flowchart of a method for predicting, correcting and guiding a deformable aircraft based on deep learning according to an embodiment of the present invention;

[0108] Figure 2 It is a schematic diagram of a downward-tilted side angle corridor of a typical configuration of a deformable aircraft in a specific embodiment of the present invention;

[0109] Figure 3 It is a Loss diagram of flight performance prediction model training in a specific embodiment of the present invention;

[0110] Figure 4 It is a flight trajectory curve diagram of a deformable aircraft in a specific embodiment of the present invention.

[0111] Figure 5 It is a graph showing the change in height of a deformable aircraft over time in a specific embodiment of the present invention.

[0112] Figure 6It is a graph showing the change of the angle of attack command of a deformable aircraft over time in a specific embodiment of the present invention.

[0113] Figure 7 It is a graph showing the variation of the roll angle command of a deformable aircraft with time in a specific embodiment of the present invention.

[0114] Figure 8 It is a curve diagram of the deformation coefficient of a deformable aircraft changing with time in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0115] The present invention proposes a prediction, correction and guidance method for a deformable aircraft based on deep learning. The present invention is described in detail below in conjunction with the accompanying drawings and embodiments.

[0116] The embodiment of the present invention provides a prediction and correction guidance method for a deformable aircraft based on deep learning, comprising:

[0117] Training a flight performance prediction model for a deformable aircraft, wherein the input of the model is the speed, guidance parameters, and deformation parameters of the deformable aircraft, and the output is the range capability value and the lateral range capability value of the deformable aircraft;

[0118] In each guidance cycle, based on the flight performance prediction model of the deformable aircraft, a nonlinear optimization problem is constructed with the lateral range capability as the optimization performance index and the required range as the constraint;

[0119] The nonlinear optimization problem is solved to obtain the current guidance parameters and deformation parameters of the deformable aircraft, and then the current attack angle and roll angle amplitude of the deformable aircraft are obtained to achieve predictive correction guidance for the deformable aircraft.

[0120] In a specific implementation of the present invention, the prediction and correction guidance method for a deformable aircraft based on deep learning has the following overall process: Figure 1 As shown, the following steps are included:

[0121] 1) Based on the relationship between the angle of attack and speed of the deformable aircraft, establish the roll angle corridor of the deformable aircraft in each aerodynamic configuration; the specific steps are as follows:

[0122] 1-1) Establish the relationship between the attack angle and speed of the deformable aircraft.

[0123] In this embodiment, the expression of the angle of attack of the deformable aircraft in flight with respect to the speed is set as:

[0124]

[0125] Among them, α is the attack angle of the deformable aircraft, V is the speed of the deformable aircraft; V b ,V aare the preset speed segmentation points, V b It is required to be smaller than the initial velocity of the deformed aircraft, V a Required to be greater than the reentry terminal velocity V of the deformed aircraft f ; This embodiment uses the same value for each aerodynamic configuration of the deformable aircraft, taking V b =4500m / s,V a =3000m / s. is the angle of attack with the maximum lift-to-drag ratio; α m is the initial value of the angle of attack. In order to adapt to the low air density at high altitudes in the initial stage of flight, a larger value is usually taken. In this embodiment, the value is 16 degrees.

[0126] 1-2) Establish a roll angle corridor for the deformable aircraft in each aerodynamic configuration.

[0127] In this embodiment, the reentry flight of the shape-shifting aircraft in each aerodynamic configuration is divided into an initial descent phase, a balanced flight phase, and a terminal capability management phase; wherein, for any aerodynamic configuration:

[0128] a) The initial descent section includes the section from the reentry point to the first valley point of the trajectory. In this section, the aerodynamic force is very insufficient. The main task of this section is to ensure that the maximum thermal peak value that appears in this section is less than the set allowable value. In view of the very limited maneuverability of the aircraft in the initial descent section, the aircraft's roll angle is not adjusted in this section, and a constant roll angle is adopted. The speed range of this section is V ≥ V b The upper limit of the initial descending section tilt angle corridor ν 1,max and the lower bound ν 1,min The expression is:

[0129]

[0130] Wherein, k is the deformation coefficient, which represents the deformation amount of the deformable aircraft, such as the different sweep angles of the variable-sweep wing aircraft. In a specific embodiment of the present invention, the value range of k is [1,4], ν init The constant tilt angle taken in the initial descent section is 30 degrees in a specific embodiment of the present invention.

[0131] b) The design goal of the balanced flight stage is to reasonably dissipate energy while meeting strict process constraints.

[0132] In this embodiment, the minimum tilt angle of the balanced flight stage is 0, and the speed range is [V a ,V b The balanced flight stage mainly considers three process constraints: overload constraint, dynamic pressure constraint and thermal flow constraint. The expressions are as follows:

[0133]

[0134] in, They represent the heat flux, overload and dynamic pressure borne by the deformed aircraft at time t, They represent the maximum heat flux, maximum dynamic pressure, and maximum overload constraints that the deformable aircraft can withstand respectively; S ref , m0, g0 represent the reference area, mass and gravitational acceleration of the deformable aircraft respectively; C L (α,V,k),C D (α, V, k) represent the lift coefficient and drag coefficient of the deformable aircraft, respectively, which are related to the angle of attack α, velocity V, and deformation coefficient k, where α is calculated by formula (1); ρ is the atmospheric density at the current altitude of the aircraft; C1 is the heat flow calculation constant, which is 110311.7; V c is the first cosmic velocity, which is 7900 m / s; ρ0 is the atmospheric density at sea level, which is 1.225 kg / m 3 ; R d It represents the curvature radius of the leading edge of the deformed aircraft, which is related to the aircraft's own parameters. In this embodiment, it is 0.002.

[0135] Then the air density of the minimum distance from the center of the earth corresponding to each constraint in formula (3) is:

[0136]

[0137] Among them, r Q ,r p ,r n They represent the minimum geocentric distance corresponding to the heat flow constraint, overload constraint, and dynamic pressure constraint, respectively, and can be obtained by the expression or interpolation table of the change of air density with height. From formula (4), it can be seen that when the relationship between the angle of attack and the speed is determined and the deformation parameter k remains unchanged, r Q ,r p ,r n It only depends on the speed V.

[0138] In this embodiment, r in formula (4) is Q ,r n ,r p The minimum value of is taken as the minimum flight geocenter distance, and the expression is as follows:

[0139] r min (V) = min{r Q ,r q ,r n} (5)

[0140] Then, the maximum roll angle of the balanced flight segment is calculated according to the quasi-balanced flight condition (QEGC):

[0141]

[0142] Where g is the gravitational acceleration at the height of the deformable aircraft, ν 2,max (V,k) represents the maximum roll angle calculated according to the quasi-balanced gliding condition.

[0143] The upper limit of the roll angle corridor in the balanced flight phase is ν 2,max and the lower bound ν 2,min It is expressed as:

[0144]

[0145] c) The terminal energy management section needs to satisfy the terminal state constraint. The speed range of this section is V∈[V f ,V a ].

[0146] In this embodiment, in combination with the quasi-balanced flight condition (QEGC), the terminal balanced flight roll angle ν QEGC,f The calculation expression is:

[0147]

[0148] Among them, r f ,v f They represent the end-of-reentry distance from the Earth and the end-of-reentry speed, respectively. In this embodiment, r f =30km,V f =2000m / s. Upper limit of the roll angle corridor of the terminal energy management section ν 3,max (V,k) and the lower bound ν 3,min The expression of (V,k) is:

[0149]

[0150] Among them, k ν is the contraction coefficient, which is 4 in this embodiment. The above formula can ensure that a At this speed, the roll angle corridors of the terminal energy management section and the balanced flight section can be effectively connected.

[0151] A schematic diagram of a typical configuration downward tilt angle corridor of a deformable aircraft in a specific embodiment of the present invention is shown in FIG. Figure 2 The figure shows the schematic diagram of the tilt angle corridor when the deformation coefficient k = 1. The upper limit of the tilt angle corridor is ν 1,max (V,k),ν 2,max (V,k),ν 3,max (V, k) is spliced ​​along the velocity direction, denoted as ν max (V,k); the lower limit of the tilt angle corridor is ν 1,min (V,k),ν 2,min (V,k),ν 3,min(V, k) is spliced ​​along the velocity direction, denoted as ν min (V,k).

[0152] 2) Based on the results of step 1), establish a training set; the specific steps are as follows:

[0153] 2-1) Based on the value range of the guidance parameter ω and the deformation coefficient k, the flight trajectory simulation results of the deformable aircraft are obtained under different combinations of the guidance parameter ω and the deformation coefficient k, and the ballistic information of the flight trajectory is sampled.

[0154] In this embodiment, the initial state of the aircraft reentry process is kept unchanged, and the guidance parameter ω and the deformation coefficient k are uniformly taken within the value range. Then, the simulation results of the corresponding flight trajectory are obtained under each combination of the guidance parameter ω and the deformation coefficient k. It is possible to obtain a number of trajectories with different ranges and different lateral maneuvering distances (wherein each trajectory has the same initial state and the guidance parameters and deformation coefficients remain unchanged during the flight).

[0155] In the simulation of each flight trajectory, according to the guidance parameter ω and the roll angle corridor obtained in step 1), the roll angle ν = |ν cmd |:

[0156] |ν cmd |=(1-ω)ν min (V,k)+ων max (V,k) (10)

[0157] In a specific embodiment of the present invention, the value range of ω is [0,1], and the value range of k is [1,4], which represents the deformation coefficient, that is, different configurations of the aircraft. max (V,k),ν min (V,k) are the upper and lower limits of the roll angle corridor corresponding to the aerodynamic configuration with deformation coefficient k.

[0158] According to the flight speed and the relationship between the angle of attack and the speed of the deformed aircraft established in step 1), namely, formula (1), the angle of attack α can be obtained. The above-calculated roll angle and angle of attack are used in the aircraft simulation process to obtain the flight trajectory simulation results under the corresponding combination of ω and k. In this embodiment, by traversing different deformation coefficients k and control section weight values ​​ω, several flight trajectories can be obtained.

[0159] Furthermore, the basic trajectory information sequence of each flight trajectory is recorded by sampling, and the trajectory information includes: the height r of each sampling point i , longitude λ i , latitude φ i , speed V i 、Ballistic inclination angle θ i , trajectory deviation angle σi and flight time t i (The sampling frequency of this embodiment is 1 second), where the subscript i represents the i-th sampling point. In a specific embodiment of the present invention, 21 values ​​are taken in the value range of ω, 4 values ​​are taken in the value range of k, and a total of 84 simulation results of trajectories are obtained. In the simulation of each trajectory, the simulation step length is 1 second, the starting speed is 5000 m / s, and the ending speed is 2000 m / s.

[0160] 2-2) Based on the trajectory information obtained in step 2-1), the flight performance index of each sampling point on each flight trajectory is calculated.

[0161] In this embodiment, the aircraft range capability and the aircraft lateral range capability at each sampling point of each flight trajectory can be obtained according to the ballistic information of the sampling point.

[0162] The aircraft range capability refers to the flight distance of the aircraft from the current moment to the point where the handover speed and altitude requirements are met under a certain guidance and aerodynamic configuration selection strategy. In this embodiment, the aircraft range capability calculation expression at the i-th sampling point is as follows:

[0163]

[0164] Among them, DR i represents the range capability of the aircraft at the i-th sampling point, S i represents the flight range of the aircraft at the i-th sampling point, S f Indicates the distance the aircraft flies along the entire trajectory.

[0165] The lateral range capability of an aircraft refers to the lateral flight distance of the aircraft from the current moment to the moment when the handover speed and altitude requirements are met under a certain guidance and aerodynamic configuration selection strategy, that is, the distance between the terminal handover point and the current trajectory of the aircraft.

[0166] In this embodiment, the coordinate difference between the i-th sampling point and the terminal point in the trajectory coordinates is assumed to be:

[0167]

[0168] Among them, λ f ,φ f Indicates the longitude and latitude of the terminal point of this trajectory, Δ xi ,Δ yi They represent the longitude difference and latitude difference between the i-th sampling point and the terminal point in the trajectory coordinates respectively.

[0169] Then the aircraft range capability at the i-th sampling point is:

[0170] CR i =|Δ yicos(σ i )-Δ xi sin(σ i )| (13)

[0171] 2-3) Using the results of steps 2-1) and 2-2), construct a training set.

[0172] In the embodiment of the present invention, within the set speed range [V f ,V0](V f is the reentry flight termination velocity, V0 is the reentry flight initial velocity, and a specific embodiment of the present invention is [2000, 5000] m / s) uniformly sampling multiple points (a specific embodiment of the present invention is 500 points) to obtain a uniformly distributed velocity sequence V uniform . (There is no requirement for the value range, try to use as many as possible)

[0173] On each flight trajectory obtained in step 2-1), according to the original data sequence V i ,CR i ,DR i Interpolation results in the aircraft range capability and aircraft range capability in the speed sequence V uniform The corresponding values ​​on the interpolation result are denoted as CR uniform ,DR uniform .

[0174] Then use V uniform ,CR uniform ,DR uniform Construct the training set:

[0175] In this embodiment, the input of a single training sample of the training set is the number of times the deformed aircraft in each flight trajectory is V uniform The output is the aircraft range capability value and the aircraft lateral range capability value corresponding to the speed value under the flight trajectory, respectively from CR uniform ,DR uniform In a specific embodiment of the present invention, a total of 84*500=42000 training samples are generated (84 trajectories, each trajectory has 500 sampling points, corresponding to 500 speed values), and the input dimension of each training sample is 3 and the output dimension is 2.

[0176] In this embodiment, the input and output of the training set are recorded as:

[0177] X={x j,s |1≤j≤42000,j∈N,s∈{1,2,3}},Y={y j,s|1≤j≤42000,j∈N,s∈{1,2}},

[0178] Among them, X represents the input data set of the training set, Y represents the output data set of the training set, j represents the jth training sample, s represents the sth dimension, and x j,s Represents the value of the sth dimension of the input data in the jth training sample, y j,s Represents the value of the sth dimension of the output data in the jth training sample.

[0179] 3) Using the training set obtained in step 2) to train the deformable aircraft flight performance prediction model, a trained deformable aircraft flight performance prediction model is obtained; the specific steps are as follows:

[0180] 3-1) Preprocess the training set data obtained in step 2).

[0181] In this embodiment, the data of each dimension of the training set obtained in step 2) is normalized, and the calculation expression is as follows:

[0182]

[0183] in, represents the normalized value of the input data of the sth dimension of the jth training sample, represents the normalized value of the output data of the sth dimension of the jth training sample; x mean,s ,x std,s Respectively represent the mean and variance of the input data of the sth dimension of the training sample; y mean,s ,y std,s They represent the mean and variance of the output data of the sth dimension of the training sample respectively.

[0184] 3-2) Construct a flight performance prediction model for deformable aircraft.

[0185] This embodiment uses a neural network model with four fully connected layers as a deformation aircraft flight performance prediction model. The input dimension of the model is 3 (speed V, guidance parameter ω and deformation parameter k), the hidden layer dimension is 128, and the output dimension is 2 (range capability value and lateral range capability value). In the forward propagation process, the elu function and tanh function are used to introduce the nonlinear properties of the neural network. Among them, the elu function is used for the first layer, the tanh function is used for the second and third layers, and the output layer has no activation function. The training loss function is as follows:

[0186]

[0187] Among them, Y pre represents the output data predicted by the model, Represents the true value of the training set output data.

[0188] The initial learning rate used in the training is 0.002, and the learning rate decay method is adopted; the optimizer is the Adam optimizer, and its parameters are the default parameters of the torch library. During the training process, the learning rate lr is reduced to 0.8 times the original every two epochs, and a total of 200 epochs are trained, and the batch size of the batch training data is 32.

[0189] The trained deformable aircraft flight performance prediction model is denoted as net, and the output calculated by the model based on the input V, ω, and k values ​​is denoted as CR net (V,ω,k) and DR net (V,ω,k). Since this model is composed of a neural network and has the characteristics of back-propagation derivation, the partial derivative of flight performance with respect to ω,k can be calculated based on the input V,ω,k value. At this point, the offline part of this embodiment is completely completed.

[0190] 4) During the online guidance process, in each guidance cycle, a nonlinear optimization problem is first constructed with the traverse capability as the optimization performance index and the required range as the constraint; specifically, as follows:

[0191] Calculate the required range d based on the position of the transformable aircraft and the handover point desired , the calculation expression is as follows:

[0192] d desired = R arccos[cos(φ msl )·cos(φ tgt )·cos(λ msl -λ tgt )+sin(φ msl )·sin(φ tgt )] (16)

[0193] Among them, λ msl ,λ tgt are the longitudes of the deformed aircraft and the target point, respectively, msl ,φ tgt are the latitudes of the deformed aircraft and the target point respectively.

[0194] Note that the prediction value of the prediction model trained in step 3) is CR net (V,ω,k),DR net (V,ω,k), then the nonlinear optimization problem with equality constraints is constructed as follows:

[0195]

[0196] Among them, ω min ,ωmax They represent the minimum and maximum values ​​of the guidance parameters respectively, and in this embodiment, they are 0 and 1 respectively; k min ,k max Respectively represent the minimum and maximum values ​​of the deformation coefficient, and in this embodiment, they are 1 and 4 respectively. 5) Solve the nonlinear optimization problem established in step 4).

[0197] This embodiment uses the augmented Lagrangian function method to solve the nonlinear optimization problem in step 4), as follows:

[0198] 5-1) Based on the optimization problem shown in formula (17), the augmented Lagrangian function is constructed:

[0199] L=CR net (V,ω,k)+λ(DR net (V,ω,k)-d desired )+ρ||DR net (V,ω,k)-d desired || 2 (18)

[0200] Among them, λ,ρ are Lagrange multipliers, and ρ is a positive number.

[0201] 5-2) Set the initial iteration values ​​ω0, k0, λ0. In this embodiment, ω0 = 0.5, k0 = 2.5, λ0 = 0.5.

[0202] 5-3) Combined with the trained model in step 3), use the gradient ascent method to optimize the variables. Let the variable obtained in the nth iteration be ω n ,k n ,λ n .

[0203]

[0204] in, They are all calculated by the model trained in step 3).

[0205] 5-4) Update the augmentation coefficient λ n+1 =λ n +ρ(DR net (V,ω,k)-d desired ).

[0206] 5-5) Judgment: If |ω is satisfied n+1 -ω n |≤∈ ω ,|k n+1 -k n |≤∈ k , where ∈ ω ,∈ kis the convergence threshold, proceed to step 5-6); otherwise, set n=n+1 and return to step 5-3).

[0207] 5-6) Determine ω*=ω n ,k*=k n To seek the root.

[0208] The obtained k* is used as the deformation parameter command k output by the guidance law cmd , then go to step 6).

[0209] 6) Based on the result of step 5), calculate the angle of attack command and the roll angle command; the specific steps are as follows:

[0210] 6-1) The angle of attack command α is calculated based on the current speed of the deformed aircraft and formula (1): cmd .

[0211] 6-2) Calculate the roll angle amplitude according to the guidance parameter ω* obtained in step 5), and the calculation expression is:

[0212] |ν cmd |=(1-ω*)ν min (k)+ω*ν max (k) (20)

[0213] Furthermore, the method described in this embodiment also includes:

[0214] 7) Set the horizontal range threshold to δCR thre , the range error contraction coefficient is K, then in each subsequent guidance cycle, the sign of the roll angle is obtained by the following method:

[0215] 7-1) Let the current heading angle of the aircraft be σ, and calculate the lateral range CR of the target point relative to the current heading of the aircraft tgt , the expression is as follows:

[0216] CR tgt =(φ tgt -φ)sinσ-(λ tgt -λ)cosσ (21)

[0217] 7-2) Let the roll angle command obtained in the previous guidance cycle be ν last , calculate the horizontal error, the expression is as follows:

[0218]

[0219] Where, sign function is the sign function. If the aircraft is in the first guidance cycle, then ν last The sign of is positive by default, and the cross-range error record value δCR is set record =δCR.

[0220] 7-3) Calculate the roll angle flip flag, the expression is as follows:

[0221]

[0222] 7-4) Calculate the roll angle command and update the lateral travel record value.

[0223] If FLAG = True, the cross-distance error obtained in Step 2 is used as the new cross-distance record value δCR record =δCR, the tilt angle command is ν cmd =-sign(ν last )|ν cmd |.

[0224] If FLAG = False, the traverse record value is not updated and the roll angle command is ν cmd =sign(ν last )|ν cmd |.

[0225] At this point, the angle of attack command (step 6), roll angle command (step 6) and deformation parameter command (step 5) in each guidance cycle are obtained. In this guidance cycle, the aircraft executes the obtained angle of attack command, roll angle command and deformation parameter command. When the next guidance cycle arrives, steps 4), 5) and 6) are repeated to perform a new round of command calculations. The above process is repeated until the aircraft reaches the predetermined area and completes the flight mission.

[0226] Furthermore, the process of the deformation aircraft prediction correction guidance method based on deep learning of the present invention is explained below through python simulation. Figure 3 : is a Loss diagram for training a flight performance prediction model in a specific embodiment of the present invention. Figure 3 As shown in the figure, a total of 100 epochs were conducted during the training process, and the loss function of a single epoch decreased from 0.003 to below 0.0001, indicating that the fitting effect of the flight performance prediction model is good.

[0227] In the online stage of the method described in this embodiment, the initial altitude, longitude, latitude, speed, trajectory inclination, and trajectory deviation of the aircraft are set to 56 km, 0 deg, 0 deg, 5000 m / s, 0 deg, and 90 deg, respectively. The longitude and latitude of the target point are 28 deg and 5 deg, respectively. The altitude range of the target point is [29, 31] km. The aircraft termination speed is 2000 m / s, the guidance period is 1 s, and the lateral range threshold is δCR. thre =10km, the cross-distance error contraction coefficient is K=2. Figure 4 FIG. 1 is a graph showing the flight trajectory of a deformed aircraft in a specific embodiment of the present invention. Figure 4As shown, by using the method described in this embodiment, the deformable aircraft can reach the predetermined target point. Figure 5 FIG. 1 is a graph showing the change in height of a deformable aircraft over time in a specific embodiment of the present invention. Figure 5 As shown, the height at which the deformed aircraft reaches the predetermined target is 30.233 km, which meets the target point height requirement. Figure 6 is a graph showing the change of the angle of attack command of a deformable aircraft over time in a specific embodiment of the present invention; Figure 7 is a graph showing the change of the roll angle command of a deformable aircraft over time in a specific embodiment of the present invention; Figure 8 This is a curve diagram of the deformation coefficient of a deformable aircraft over time in a specific embodiment of the present invention. During the flight of this embodiment, the aircraft performed a total of 7 roll angle flips.

Claims

1. A prediction and correction guidance method for a deformable aircraft based on deep learning, characterized in that: include: Training a flight performance prediction model for a deformable aircraft, wherein the input of the model is the speed, guidance parameters, and deformation parameters of the deformable aircraft, and the output is the range capability value and the lateral range capability value of the deformable aircraft; In each guidance cycle, based on the flight performance prediction model of the deformable aircraft, a nonlinear optimization problem is constructed with the lateral range capability as the optimization performance index and the required range as the constraint; The nonlinear optimization problem is solved to obtain the current guidance parameters and deformation parameters of the deformable aircraft, and then the current attack angle and roll angle amplitude of the deformable aircraft are obtained to achieve predictive correction guidance for the deformable aircraft.

2. The method according to claim 1, characterized in that The deformable aircraft flight performance prediction model adopts a neural network model with four fully connected layers, the first layer of the model adopts the elu function, and the second and third layers adopt the tanh function.

3. The method according to claim 1, characterized in that The method further comprises: Based on the relationship between the attack angle and speed of the deformable aircraft, the roll angle corridor of the deformable aircraft in each aerodynamic configuration is established; the specific steps are as follows: 1) Establish the relationship between the attack angle and speed of the deformable aircraft; Among them, the expression of the angle of attack of the deformable aircraft with respect to the speed during flight is: Among them, α is the attack angle of the deformable aircraft, V is the speed of the deformable aircraft; V b ,V a are the preset speed segmentation points, V b is smaller than the initial velocity of the deformed aircraft, V a Greater than the reentry terminal velocity V of the deformed vehicle f ; α L / Dmax is the angle of attack with the maximum lift-to-drag ratio; α m is the initial value of the angle of attack; 2) Establish the roll angle corridor of the shape-shifting aircraft in each aerodynamic configuration; The reentry flight of the shape-shifting aircraft in each aerodynamic configuration is divided into the initial descent phase, the balanced flight phase, and the terminal capability management phase. For any aerodynamic configuration: The initial descent segment includes the velocity range from the reentry point to the first valley point of the trajectory, V ≥ V b ; Upper limit of the initial descending section tilt angle corridor ν 1,max and the lower bound ν 1,min The expression is: Where k is the deformation coefficient, which indicates the deformation of the deformable aircraft; ν init is the constant roll angle adopted in the initial descent; The minimum tilt angle of the balanced flight stage is 0, and the speed range is [V a ,V b ], the balanced flight stage includes three process constraints: overload constraint, dynamic pressure constraint and thermal flow constraint, and the expression is as follows: in, They represent the heat flux, overload and dynamic pressure borne by the deformed aircraft at time t, They represent the maximum heat flux, maximum dynamic pressure, and maximum overload constraints that the deformable aircraft can withstand respectively; S ref , m0, g0 represent the reference area, mass and gravitational acceleration of the deformable aircraft respectively; C L (α,V,k),C D (α, V, k) represent the lift coefficient and drag coefficient of the deformable aircraft respectively; ρ is the atmospheric density at the current altitude of the deformable aircraft; C1 is the heat flow calculation constant; V c is the first cosmic velocity; ρ0 is the atmospheric density at sea level; R d represents the radius of curvature of the leading edge of the deformable aircraft; Then the air density of the minimum distance from the center of the earth corresponding to each constraint in formula (3) is: Among them, r Q ,r p ,r n They represent the minimum geocentric distances corresponding to the heat flow constraint, overload constraint, and dynamic pressure constraint respectively; Take r in formula (4) Q ,r n ,r p The minimum value of is taken as the minimum flight geocenter distance, and the expression is as follows: r min (V)=min{r Q ,r q ,r n } (5) Then calculate the maximum roll angle of the balanced flight segment according to the quasi-balanced flight conditions: Where g is the gravitational acceleration at the height of the deformable aircraft, ν 2,max (V,k) represents the maximum roll angle calculated according to the quasi-balanced gliding condition; The upper limit of the roll angle corridor in the balanced flight phase is ν 2,max and the lower bound ν 2,min It is expressed as: The terminal energy management section satisfies the terminal state constraint, and the speed range is V∈[V f ,V a ]; Combined with the quasi-balanced flight condition, the terminal balanced flight roll angle ν QEGC,f The calculation expression is: Among them, r f ,v f They represent the terminal geocentric distance and terminal velocity of the reentry flight respectively; the upper limit of the roll angle corridor of the terminal energy management phase ν 3,max (V,k) and the lower bound ν 3,min The expression of (V,k) is: Among them, k ν is the shrinkage coefficient.

4. The method according to claim 3, characterized in that: The method further comprises: Based on the roll angle corridor of the deformable aircraft in each aerodynamic configuration, a training set for training the flight performance prediction model of the deformable aircraft is established, and the specific steps are as follows: 1) Based on the value range of the guidance parameter ω and the deformation coefficient k, the flight trajectory simulation results of the deformable aircraft are obtained under different combinations of the guidance parameter ω and the deformation coefficient k, and the ballistic information of the flight trajectory is sampled; In the simulation of each flight trajectory, according to the guidance parameter ω and the roll angle corridor, the roll angle ν = |ν cmd |: |n cmd |=(1-ω)ν min (k)+ων max (k) (10) Among them, the value range of ω is [0,1], ν max (k),ν min (k) are the upper and lower limits of the roll angle corridor corresponding to the aerodynamic configuration with deformation coefficient k; According to the flight speed and formula (1), the angle of attack α is calculated; The simulation is performed using the roll angle and the angle of attack to obtain the flight trajectory simulation result under the corresponding combination of ω and k; then, the basic ballistic information sequence of the flight trajectory is recorded by sampling, and the ballistic information includes: the height r of each sampling point i , longitude λ i , latitude φ i , speed V i 、Ballistic inclination angle θ i , trajectory deviation angle σ i and flight time t i , where the subscript i represents the i-th sampling point; 2) Based on the trajectory information obtained in step 1), the flight performance index of each sampling point on each flight trajectory is calculated, including: the aircraft range capability and the aircraft lateral range capability; The calculation expression of the aircraft range capability at the i-th sampling point is as follows: Among them, DR i represents the range capability of the aircraft at the i-th sampling point, S i represents the flight range of the aircraft at the i-th sampling point, S f Indicates the distance the aircraft has flown along the entire trajectory; Let the coordinate difference between the i-th sampling point and the terminal point in the trajectory coordinates be: Among them, λ f ,φ f Indicates the longitude and latitude of the terminal point of this trajectory, Δ xi ,Δ yi Respectively represent the longitude difference and latitude difference between the i-th sampling point and the terminal point in the trajectory coordinates; Then the aircraft range capability at the i-th sampling point is: CR i =|D yi cos(σ i )-D xi sin(σ i )| (13) 3) Using the results of steps 1) and 2), construct a training set; In the set speed range [V f , V0] to obtain a uniformly distributed velocity sequence V uniform , where V f is the reentry flight termination speed, V0 is the reentry flight initial speed; On each flight trajectory obtained in step 1), according to the original V i ,CR i ,DR i Data sequence, interpolation to obtain the aircraft range capability and aircraft cross-range capability in the speed sequence V uniform The corresponding values ​​on the interpolation result are denoted as CR uniform ,DR uniform ; Using V uniform ,CR uniform ,DR uniform Construct a training set; wherein the input of a single training sample of the training set is the number of times the deformed aircraft in each flight trajectory is in V uniform The output is the aircraft range capability value and the aircraft lateral range capability value corresponding to the speed value under the flight trajectory, respectively from CR uniform ,DR uniform Get from The input and output of the training set are recorded as: X={x j,s |1≤j≤42000,j∈N,s∈{1,2,3}},Y={y j,s |1≤j≤42000,j∈N,s∈{1,2}}, Among them, X represents the input data set of the training set, Y represents the output data set of the training set, j represents the jth training sample, s represents the sth dimension, and x j,s Represents the value of the sth dimension of the input data in the jth training sample, y j,s Represents the value of the sth dimension of the output data in the jth training sample.

5. The method according to claim 4, characterized in that The method further comprises: Before using the training set to train the deformable aircraft flight performance prediction model, the data of each dimension of the training set is normalized, and the calculation expression is as follows: in, represents the normalized value of the input data of the sth dimension of the jth training sample, represents the normalized value of the output data of the sth dimension of the jth training sample; x mean,s ,x std,s Respectively represent the mean and variance of the input data of the sth dimension of the training sample; y mean,s ,y std,s They represent the mean and variance of the output data of the sth dimension of the training sample respectively.

6. The method according to claim 5, characterized in that The construction of a nonlinear optimization problem with the horizontal range capability as the optimization performance index and the required range as the constraint includes: Calculate the required range d based on the transformable aircraft's own position and the handover point position desired , the calculation expression is as follows: d desired =R arccos[cos(φ msl )·cos(φ tgt )·cos(λ msl -l tgt )+sin(φ msl )·sin(φ tgt )](16) Among them, λ msl ,λ tgt are the longitudes of the deformed aircraft and the target point, respectively, msl ,φ tgt are the latitudes of the deformed aircraft and the target point, respectively; The predicted value output by the deformable aircraft flight performance prediction model is CR net (V,ω,k),DR net (V,ω,k), then the nonlinear optimization problem with equality constraints is constructed as follows: Among them, ω min ,ω max Respectively represent the minimum and maximum values ​​of the guidance parameters; k min ,k max Represent the minimum and maximum values ​​of the deformation coefficient respectively.

7. The method according to claim 6, characterized in that The step of solving the nonlinear optimization problem to obtain the current guidance parameters and the deformation parameters of the deformable aircraft includes: 1) Based on the optimization problem of formula (17), the augmented Lagrangian function is constructed: L=CR net (V,ω,k)+λ(DR net (V,ω,k)-d desired )+ρ||DR net (V,ω,k)-d desired || 2 (18) Among them, λ,ρ are Lagrange multipliers, ρ is a positive number; 2) Set the initial iteration values ​​ω0, k0, λ0; 3) Combined with the trained deformable aircraft flight performance prediction model, the gradient ascent method is used to optimize the variables, and the variable obtained in the nth iteration is set as ω n ,k n ,λ n ,but: in, All are calculated by the trained deformation aircraft flight performance prediction model; 4) Update the augmentation coefficient λ n+1 =λ n +ρ(DR net (V,ω,k)-d desired ); 5) Judgment: If |ω is satisfied n+1 -ω n |≤∈ ω ,|k n+1 -k n |≤∈ k , where ∈ ω ,∈ k is the convergence threshold, then go to step 6); otherwise, set n=n+1 and return to step 3); 6) Determine ω*=ω n ,k*=k n For the root of seeking; The obtained k* is used as the deformation parameter instruction k output by the guidance law. cmd .

8. The method according to claim 7, characterized in that The obtaining of the current attack angle and roll angle amplitude of the deformable aircraft comprises: 1) According to the current speed of the deformable aircraft and formula (1), the angle of attack command α is calculated cmd ; 2) Calculate the roll angle amplitude according to the guidance parameter ω*. The calculation expression is: |n cmd |=(1-ω*)n min (k)+ω*ν max (k) (20).

9. The method according to claim 8, characterized in that The method further comprises: Set the cross-range threshold to δCR thre , the range error contraction coefficient is K, then in each subsequent guidance cycle, the sign of the roll angle is obtained by the following method: 1) Assume the current heading angle of the aircraft is σ, calculate the lateral range CR of the target point relative to the current heading of the aircraft tgt , the expression is as follows: CR tgt =(φ tgt -φ)sinσ-(λ tgt -λ)cosσ (21) 2) Let the roll angle command obtained in the previous guidance cycle be ν last , calculate the horizontal error, the expression is as follows: Where, sign function is the sign function; if the aircraft is in the first guidance cycle, then ν last The sign of is positive, and the cross-travel error record value δCR is set record = δCR; 3) Calculate the roll angle flip flag, the expression is as follows: 4) Calculate the roll angle command and update the traverse record value; If FLAG = True, the cross-distance error obtained in Step 2 is used as the new cross-distance record value δCR record =δCR, the tilt angle command is ν cmd =-sign(ν last )|ν cmd |; If FLAG = False, the traverse record value is not updated and the roll angle command is ν cmd =sign(ν last )|ν cmd |.

Images