Intelligent time-coordinated guidance methods, systems and equipment for multiple hypersonic vehicles

Abstract

This invention discloses an intelligent time-coordinated guidance method, system, and device for multiple hypersonic vehicles, relating to the field of hypersonic vehicle cooperative guidance. The method includes: determining the cooperative time planning objective, flight process constraints, and terminal constraints for multiple hypersonic vehicles based on the dynamic and kinematic models of the hypersonic vehicles; determining the heading angle corridor and roll angle flip logic based on the aforementioned objective and constraints; calculating the lateral cooperative time adjustment factor in lateral guidance based on the heading angle corridor and roll angle flip logic; calculating the roll angle profile in longitudinal guidance based on the lateral cooperative time adjustment factor; and determining a deep learning framework based on a Transformer network based on the lateral cooperative time adjustment factor and the roll angle profile. This invention enables online planning of hypersonic vehicle trajectories, achieving intelligent time-coordinated guidance for multiple hypersonic vehicles.

Description

Technical Field

[0001] This invention relates to the field of cooperative guidance for hypersonic vehicles, and in particular to an intelligent time-coordinated guidance method, system and equipment for multiple hypersonic vehicles. Background Technology

[0002] Hypersonic vehicles, flying in near-space, can reach speeds exceeding Mach 5. They possess high speed, long range, and strong penetration capabilities, making them suitable for rapid long-range strikes and holding significant strategic deterrence value. In recent years, coordinated operations involving multiple hypersonic vehicles have become a hot research topic. Reentry guidance for multiple hypersonic vehicles involves guiding multiple vehicles to coordinate a coordinated, time-limited attack on a target under various typical constraints, thereby achieving a devastating strike. With the development of artificial intelligence, deep learning has become an effective way to address computational latency issues. How to utilize deep learning algorithms to solve the online trajectory planning problem for hypersonic vehicles and shorten the online planning time is also a noteworthy research area.

[0003] Cooperative guidance has been well-established in multi-missile missions, with mature theoretical research and applications. However, for hypersonic vehicles, due to their complex nonlinear models, extremely high flight speeds, and rapidly changing flight environments, traditional cooperative guidance control methods cannot be directly applied to the reentry phase. In recent years, some scholars have also studied the cooperative guidance problem of multiple hypersonic vehicles. Yu Jianglong et al. proposed a cooperative guidance strategy for multiple hypersonic vehicles that considers attack time and attack angle. This strategy divides the planning problem into two stages: in the gliding reentry phase, the coordination of the angle of attack is achieved by controlling the bank angle; in the terminal guidance phase, the coordination of the attack time is achieved by controlling the angle of attack. However, the gliding reentry phase does not consider the limitation of the coordinated attack time, and the computing power of the onboard computer seriously affects the planning time. Li Zhenhua et al. analyzed the time range of reentry flight of multiple hypersonic vehicles and studied the time-cooperative reentry guidance algorithm for multiple hypersonic vehicles. However, this method essentially calculates a coordination time based on the capabilities of each hypersonic vehicle, and then each hypersonic vehicle separately implements the coordination time constraint. If a hypersonic vehicle encounters an unknown disturbance during its mission, multiple hypersonic vehicles cannot guarantee the same arrival time. Furthermore, this method involves a large computational load and suffers from insurmountable computational latency issues. Therefore, researching the intelligent coordination time planning problem for hypersonic vehicles and conducting simulation verification remains a challenging technical problem. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent time-coordinated guidance method, system, and device for multiple hypersonic vehicles, which can realize online trajectory planning and enable multiple hypersonic vehicles to arrive at the terminal area simultaneously.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] In a first aspect, the present invention provides an intelligent time-coordinated guidance method for multiple hypersonic vehicles, comprising:

[0007] Establish dynamic and kinematic models for hypersonic vehicles;

[0008] Based on the dynamic and kinematic models of hypersonic vehicles, the objectives of cooperative time planning for multiple hypersonic vehicles, as well as the constraints during flight and the terminal constraints, are determined.

[0009] Based on the objectives of multi-hypersonic vehicle cooperative time planning, as well as the flight process constraints and terminal constraints, the heading angle corridor and roll angle flip logic are determined.

[0010] Based on the heading angle corridor and the roll angle flip logic, the lateral coordination time adjustment factor in lateral guidance is calculated;

[0011] Calculate the tilt angle profile in longitudinal guidance based on the lateral coordination time adjustment factor;

[0012] Based on the lateral coordination time adjustment factor in lateral guidance and the tilt angle profile in longitudinal guidance, a deep learning framework based on Transformer networks is determined; the deep learning framework based on Transformer networks is used to plan the trajectory of hypersonic vehicles online to achieve intelligent time-coordinated guidance for multiple hypersonic vehicles.

[0013] Secondly, the present invention provides an intelligent time-coordinated guidance system for multiple hypersonic vehicles, comprising:

[0014] The dynamics and kinematics model building module is used to build dynamics and kinematics models for hypersonic vehicles.

[0015] The objective and constraint determination module is used to determine the objectives of multi-hypersonic vehicle cooperative time planning, as well as the flight process constraints and terminal constraints, based on the dynamic and kinematic models of the hypersonic vehicles.

[0016] The heading angle corridor and roll angle flip logic determination module is used to determine the heading angle corridor and roll angle flip logic based on the multi-hypersonic vehicle cooperative time planning objective, as well as the flight process constraints and terminal constraints.

[0017] The lateral coordination time adjustment factor calculation module is used to calculate the lateral coordination time adjustment factor in lateral guidance based on the heading angle corridor and the roll angle flip logic.

[0018] The tilt angle profile calculation module is used to calculate the tilt angle profile in longitudinal guidance based on the lateral coordination time adjustment factor.

[0019] The Transformer-based deep learning framework determination module is used to determine the Transformer-based deep learning framework based on the lateral coordination time adjustment factor in lateral guidance and the tilt angle profile in longitudinal guidance. The Transformer-based deep learning framework is used to plan the trajectory of hypersonic vehicles online to achieve intelligent time-coordinated guidance for multiple hypersonic vehicles.

[0020] Thirdly, the present invention provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the intelligent time-coordinated guidance method for multiple hypersonic vehicles according to the first aspect.

[0021] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0022] This invention provides an intelligent time-coordinated guidance method, system, and device for multiple hypersonic vehicles. Its purpose is to achieve coordinated reentry times for multiple hypersonic vehicles through longitudinal guidance, based on a predictive correction guidance method, by coarsely adjusting reentry time via lateral guidance. Using traditional predictive correction algorithms as a training set, a Transformer-based neural network is designed, enabling multiple hypersonic vehicles to generate roll angle control quantities online and achieve online trajectory planning. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 A flowchart illustrating the intelligent time-coordinated guidance method for multiple hypersonic vehicles provided in an embodiment of the present invention;

[0025] Figure 2 Ground trajectory diagrams of three hypersonic vehicles provided in this embodiment of the invention;

[0026] Figure 3 The altitude versus dimensionless energy curves of three hypersonic vehicles during flight, provided in an embodiment of the present invention.

[0027] Figure 4 The plot angle and dimensionless energy curves of three hypersonic vehicles during flight, provided for embodiments of the present invention;

[0028] Figure 5 Three-dimensional trajectory diagrams of three hypersonic vehicles during flight, provided in an embodiment of the present invention;

[0029] Figure 6 A structural diagram of a deep learning framework based on Transformer networks provided in an embodiment of the present invention;

[0030] Figure 7 A schematic diagram of the loss function for the training process provided in an embodiment of the present invention;

[0031] Figure 8 A schematic diagram of the verification process loss function provided in an embodiment of the present invention;

[0032] Figure 9 Ground trajectory map for online trajectory planning of multiple hypersonic vehicles using a trained neural network (i.e., a deep learning framework based on Transformer networks) provided in an embodiment of the present invention;

[0033] Figure 10 The altitude and dimensionless energy curves of three hypersonic vehicles during the online planning process provided in this embodiment of the invention;

[0034] Figure 11 The results of the tilt angle control during the online trajectory planning process of the first hypersonic vehicle provided in this embodiment of the invention, and the comparison results with the traditional prediction and correction method;

[0035] Figure 12 The figure shows the results of the tilt angle control during the online trajectory planning process of the second hypersonic vehicle provided in this embodiment of the invention, and the comparison results with the traditional prediction and correction method.

[0036] Figure 13 The figure shows the results of the tilt angle control during the online trajectory planning process of the third hypersonic vehicle provided in this embodiment of the invention, and the comparison results with the traditional prediction and correction method. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] This invention addresses the problem of online trajectory planning for the glide phase of hypersonic vehicles with cooperative time constraints, providing an intelligent time-coordinated guidance method, system, and device for multiple hypersonic vehicles. Considering the coordinated arrival time of multiple hypersonic vehicles at the terminal region, and the need to satisfy thermal flux constraints, overload constraints, dynamic pressure constraints, and quasi-equilibrium gliding constraints during flight, an online trajectory planning method based on a Transformer neural network is proposed. This method effectively shortens the online planning time, is more robust, better suited to practical application scenarios, and can be applied to online real-time planning. Simulation verification experiments using Matlab software demonstrate the effectiveness of this invention.

[0039] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] Example 1

[0041] like Figure 1 As shown in the figure, the main implementation steps of the intelligent time-coordinated guidance method for multiple hypersonic vehicles provided in this embodiment are as follows.

[0042] Step 101: Establish the dynamic and kinematic models of the hypersonic vehicle.

[0043] This embodiment selects the CAV-H model, using energy variables as a quasi-energy variable. As the independent variable, a dynamic model of the hypersonic vehicle is established, as shown below:

[0044]

[0045] Where θ and φ represent the longitude and latitude of the hypersonic vehicle, V represents the dimensionless velocity of the hypersonic vehicle, R represents the dimensionless distance of the hypersonic vehicle from the Earth's center, γ represents the trajectory inclination angle, ψ represents the heading angle, σ represents the roll angle of the hypersonic vehicle, and Ω represents the dimensionless angular velocity of Earth's rotation. This represents the dimensionless acceleration of a hypersonic vehicle. This represents the dimensionless drag acceleration of a hypersonic vehicle.

[0046] The kinematic model of a hypersonic vehicle is defined as a linear function of its angle of attack and velocity, and its expression is as follows:

[0047]

[0048] Where α is the angle of attack during the flight of the hypersonic vehicle.

[0049] To ensure integration accuracy, the dimensionless operations for variables are shown in Table 1.

[0050] Table 1. Dimensionless operation table for variables

[0051]

[0052]

[0053] Wherein, the Earth's radius R0 = 6378140m, and the gravitational acceleration at the Earth's surface g0 = 9.807m / s². 2 Since variables such as longitude, latitude, heading angle, and ballistic inclination angle are all in radians, there is no need to perform dimensionless processing.

[0054] Step 102: Based on the dynamic and kinematic models of the hypersonic vehicles, determine the objectives of the coordinated time planning of multiple hypersonic vehicles, as well as the flight process constraints and terminal constraints.

[0055] For the cooperative time planning objective of multiple hypersonic vehicles, the definition is as follows: Considering M hypersonic vehicles, for any i-th hypersonic vehicle, if the time of arrival of the i-th hypersonic vehicle in the terminal region is... It is assumed that M hypersonic vehicles achieve the goal of time coordination.

[0056] Meanwhile, hypersonic vehicles also need to meet flight process constraints during flight. Specifically, they also need to meet overload constraints, heat flux density constraints, dynamic pressure constraints, and quasi-equilibrium gliding constraints, which can be expressed as:

[0057]

[0058]

[0059] q=0.5ρV 2 ≤q max .

[0060]

[0061] in, ρ is the heat flux density, n is the overload during flight, q is the dynamic pressure during flight, and ρ is the atmospheric density at the current location of the hypersonic vehicle.

[0062] For hypersonic vehicles, terminal constraints must also be met. Specifically, these constraints include terminal flight range constraints, latitude and longitude constraints, time constraints, altitude constraints, and heading angle constraints.

[0063]

[0064] Where the subscript f,i represents the terminal state of the i-th hypersonic vehicle, and the superscript * indicates the value of the terminal state variable of the i-th hypersonic vehicle, for example, h i (e f,i ) represents the terminal energy e of the i-th hypersonic vehicle. f,i The expected value of the height below is Similarly, θ and φ represent longitude and latitude, respectively. t, ψ, s togo These represent time, heading angle, and range to be flown, respectively. The coordinated time constraint value is given by the formula... Provided.

[0065] Step 103: Based on the multi-hypersonic vehicle cooperative time planning objective, as well as the flight process constraints and terminal constraints, determine the heading angle corridor and the roll angle flip logic.

[0066] Based on the multi-hypersonic vehicle cooperative time planning objective described in step 102, as well as the flight process constraints and terminal constraints, the following heading angle corridor and roll angle flip logic are designed:

[0067]

[0068] Wherein, sgn(σ i ) represents the sign of the panning angle in the current guidance round, Δψ i =ψ i -ψ LOS,i Let ψ be the heading angle deviation of the i-th hypersonic vehicle. i Let ψ be the heading angle of the i-th hypersonic vehicle. LOS,i Let be the line-of-sight angle of the i-th hypersonic vehicle observing the target; The line-of-sight angle for a hypersonic vehicle to observe a target, Δψ max,i Let Δψ be the upper boundary of the heading angle corridor for the i-th hypersonic vehicle. min,i Let σ be the lower boundary of the heading angle corridor for the i-th hypersonic vehicle. p,iθ represents the bank angle of the i-th hypersonic vehicle in the previous guidance round; the superscript * represents the value of the terminal state variable of the i-th hypersonic vehicle. i ,φ i These are the longitude and latitude of the i-th hypersonic vehicle, respectively.

[0069] The following is a design method for a heading angle corridor.

[0070] Step 1: Determine the initial boundary value Δψ that can reach the terminal region based on the initial state of the hypersonic vehicle. init,i and the turning point of the heading angle corridor (θ) m,i ,|Δψ m,i |).

[0071] Step 2: Determine the magnitude of the terminal heading angle of the corridor based on the terminal heading angle deviation, and take |Δψ(e f,i )|=0.8Δψ error,i ;Δψ(e f,i ) represents the desired terminal heading angle of the i-th hypersonic vehicle, Δψ error,i It represents 0.8 times the expected terminal heading angle of the i-th hypersonic vehicle.

[0072] Step 3: If the current longitude θ of the i-th hypersonic vehicle i Less than the turning point of the corridor (θ) m,i ,|Δψ m,i |), that is, θ i <θ m,i The corrected heading angle deviation of the heading angle corridor is then:

[0073]

[0074] If the longitude of the hypersonic vehicle is greater than the turning point of the heading angle corridor (θ) m,i ,|Δψ m,i |), that is, θ i >θ m,i The corrected heading angle deviation of the heading angle corridor is then:

[0075] |Δψ i |=K ρ,i / θ(e i )+|Δψ(e f,i )|-K ρ / θ(e f,i ).

[0076] Wherein, Δψ(e f,i K is set to 0.8 times the desired terminal heading angle. ρ,i It is the lateral coordination time adjustment factor of the i-th hypersonic vehicle, which is designed in step 104.

[0077] Step 104: Calculate the lateral coordination time adjustment factor in lateral guidance based on the heading angle corridor and the tilt angle flip logic.

[0078] Based on the heading angle corridor and roll angle flipping logic described in step 103, the lateral coordination time adjustment factor K in lateral guidance is calculated using Newton's iteration method. ρ,i The details are as follows.

[0079] In lateral guidance, using only time as the iterative variable, the established tilt angle amplitude profile is composed of three points {σ 0,i (e 0,i ),σ mid,i (e mid,i ),σ f,i (e f,i A linear function consisting of σ and σ', where σ' is the σ' ... 0,i Let σ be the initial energy value of the i-th hypersonic vehicle. f,i Let σ be the terminal energy value of the i-th hypersonic vehicle. mid,i =(σ 0,i +σ f,i ) / 2. σ i (e i Let σ be the amplitude of the downtilt angle of the i-th hypersonic vehicle at energy e. 0,i (e 0,i For example, let represent the initial roll angle amplitude of the i-th hypersonic vehicle at the initial energy. The lateral coordination time error of the i-th hypersonic vehicle is defined as...

[0080] The Newton-Raphson iteration method is used to search for the lateral time adjustment factor K that approximately satisfies the cooperative time constraint for the i-th hypersonic vehicle. ρ,i The termination condition for the iteration is The iteration method is as follows:

[0081]

[0082] Where, η k =1 / 2 λ ,(λ=0,1,2…) is to satisfy The iterative adjustment factor.

[0083] Step 105: Calculate the tilt angle profile in longitudinal guidance based on the lateral coordination time adjustment factor.

[0084] In longitudinal guidance, the waiting range, coordination time, and heading angle are used as iterative variables. The tilt angle amplitude profile established for the i-th hypersonic vehicle is composed of five points {σ... 0,i (e 0,i),σ 1,i (e 1,i ),σ 2,i (e 2,i ),σ 3,i (e 3,i ),σ f,i (e f,i The linear function formed by )}, with σ chosen as the iteration variable. 1,i ,σ 2,i ,σ 3,i ,in j = 1, 2, 3 represent three different energy values ​​during the flight of the i-th hypersonic vehicle, σ i (e i Let σ be the amplitude of the downtilt angle of the i-th hypersonic vehicle at energy e. 0,i (e 0,i For example, let represent the initial roll angle amplitude of the i-th hypersonic vehicle at the initial energy. The pre-flight range error of the i-th hypersonic vehicle is defined as... Define the heading angle error of the i-th hypersonic vehicle as G. i (σ 1,i ,σ 2,i ,σ 3,i )=ψ i (e f,i )-ψ i * Define the cooperative time error of the i-th hypersonic vehicle as

[0085] The Newton-Raphson iteration method is used to search for the bank angle profile variable σ of the i-th hypersonic vehicle that satisfies the range constraint before flight. 1,i ,σ 2,i ,σ 3,i The termination condition for the iteration is F. i (σ 1,i ,σ 2,i ,σ 3,i )<t error,i J i (σ 1,i ,σ 2,i ,σ 3,i )<s togoerror,i G i (σ 1,i ,σ 2,i ,σ 3,i )<ψ error,i Where s togoerror,i ψ error,i t error,i The maximum allowable waiting range error, heading angle error, and coordination time error for the i-th hypersonic vehicle are determined by the following iterative method:

[0086]

[0087] Wherein, η k =1 / 2 λ , (λ=0,1,2…) is the iteration adjustment factor to ensure that the error after iteration is less than that before iteration.

[0088] Step 106: Determine the deep learning framework based on the Transformer network based on the lateral coordination time adjustment factor in lateral guidance and the tilt angle profile in longitudinal guidance; the deep learning framework based on the Transformer network is used to plan the trajectory of hypersonic vehicles online to achieve intelligent time coordination guidance for multiple hypersonic vehicles.

[0089] Using the lateral coordination time adjustment factor in lateral guidance and the trajectory generated by the tilt angle profile in longitudinal guidance as the training set, a deep learning network based on the Transformer network is trained to obtain a deep learning framework based on the Transformer network, thereby realizing intelligent online trajectory rapid planning.

[0090] In this embodiment, based on the deep learning framework of Transformer network, the effectiveness of the method is verified by numerical simulation using Matlab, and its planning speed is compared with that of traditional methods.

[0091] This invention proposes an intelligent time-coordinated guidance method for multiple hypersonic vehicles, enabling them to achieve rapid online trajectory planning under constraints of overload, heat flux density, dynamic pressure, quasi-equilibrium gliding conditions, and terminal flight range constraints. The main advantages of this method are as follows: 1) This method can coordinate the reentry time of multiple hypersonic vehicles while satisfying the terminal attack angle constraint. 2) This method can plan reasonable hypersonic vehicle gliding trajectories while satisfying the constraints of overload, heat flux density, dynamic pressure, and quasi-equilibrium gliding conditions, and meeting the terminal flight range constraint. 3) This method utilizes artificial intelligence deep learning networks, effectively improving the trajectory planning speed, solving the computational delay problem of hypersonic vehicles, and exhibiting good online planning capabilities and strong robustness.

[0092] The effectiveness of the proposed method is verified through a specific example of time-coordinated guidance planning for multiple hypersonic vehicles. The specific implementation steps of this example are as follows:

[0093] (1) Hypersonic vehicle system setup

[0094] Consider a hypersonic vehicle system consisting of three hypersonic vehicles with different initial conditions. The hypersonic vehicle model adopts the CAV-H model, with a mass of 907 kg and an aerodynamic reference area of ​​0.484 m². 2 They need to reach the terminal area while simultaneously meeting the flight process constraints and the terminal waiting range constraints. The initial conditions for the hypersonic vehicle are shown in Table 2. The constraints on overload, heat flux density, dynamic pressure, and quasi-equilibrium gliding conditions during flight are shown in Table 3. To facilitate the planning of the tilt angle profile, the constraints are transformed into constraints on the tilt angle, and the maximum tilt angle constraint is found to be 80°, which is more stringent than the constraints before the transformation. The latitude and longitude of the terminal area are (θ...). f ,φ f )=(94°,0°), the maximum allowable error of the terminal waiting flight range is s togoerror =3km, the maximum allowable heading angle error is ψ error =3°, the maximum allowable time error is t error =5s.

[0095] Table 2 Initial Conditions for Hypersonic Vehicles

[0096]

[0097]

[0098] Table 3. Constraints on Overload, Heat Flux Density, Dynamic Pressure, and Quasi-equilibrium Gliding Conditions During Flight

[0099]

[0100] (2) Simulation analysis of coordinated time prediction correction guidance

[0101] In this example, let the Earth's radius R0 = 6378140m and the gravitational acceleration at the Earth's surface g0 = 9.807m / s². 2 The ground trajectory diagrams of the three hypersonic vehicles are as follows: Figure 2 As shown. Figure 3 This is a graph showing the altitude and dimensionless energy curves of three hypersonic vehicles during their flight. Figure 4 The graph shows the tilt angle and dimensionless energy curves of three hypersonic vehicles during their flight. Figure 5 The diagram shows the three-dimensional trajectories of three hypersonic vehicles during their flight. It can be seen that, under the algorithm proposed in this invention, the hypersonic vehicles reached the terminal region while satisfying flight constraints. The iteratively obtained coordinated time is... The flight range error, heading angle error, and coordination time error are shown in Table 4, and they meet the requirements for maximum flight range error, heading angle error, and coordination time error. This example verifies the effectiveness of the proposed method.

[0102] Table 4. Errors in flight range, heading angle, and coordination time.

[0103] Δψ(°) Δt(s) <![CDATA[Δs togo (km)]]> Δh(m) Aircraft 1 -2.31 0.03 0.45 -84.42 Aircraft 2 1.34 0.04 -0.53 -63.34 Aircraft 3 2.75 0.07 -1.54 -91.02

[0104] (3) Simulation Analysis of Intelligent Online Trajectory Planning Method

[0105] In this example, a Transformer-based online trajectory planning deep learning neural network was designed. The network's structural framework is as follows: Figure 6 As shown. The network input consists of the state variables X of three hypersonic vehicles. input ={R,θ,φ,γ,ψ}, the network includes an input position encoding layer, six encoding layers, and an output layer. The output is the tilt angle control quantity for three hypersonic vehicles. The loss function for the training process and the loss function for the validation process are as follows: Figure 7 , Figure 8 As shown, the loss function for the training set converges to 0.05, and the loss function for the test set converges to 0.06. Figure 9 Ground trajectory diagrams for online trajectory planning of multiple hypersonic vehicles using trained neural networks (i.e., a deep learning framework based on Transformer networks) are presented. Figure 10 This graph shows the altitude versus dimensionless energy curves of three hypersonic vehicles during the online planning process. Figures 11-13 The results of the roll angle control parameters during the online trajectory planning process for three hypersonic vehicles are presented and compared with those of the traditional predictive correction method. Table 4 shows a comparison of the planning time between the intelligent online trajectory planning method and the traditional predictive correction trajectory planning method. It can be seen that the intelligent method can effectively shorten the planning time. This verifies the effectiveness and real-time performance of the proposed Transformer-based online planning method.

[0106] Table 5. Comparison of planning time between intelligent online trajectory planning method and traditional predictive correction trajectory planning method.

[0107] Planning average time Longest planning time Predictive correction method 4.79s 7.96s Transformer-based online planning methods <0.01s 0.01s

[0108] Example 2

[0109] In order to implement the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, an intelligent time-coordinated guidance system for multiple hypersonic vehicles is provided below.

[0110] This embodiment provides an intelligent time-coordinated guidance system for multiple hypersonic vehicles, comprising:

[0111] The dynamics and kinematics model building module is used to build dynamics and kinematics models for hypersonic vehicles.

[0112] The objective and constraint determination module is used to determine the objectives of multi-hypersonic vehicle cooperative time planning, as well as the flight process constraints and terminal constraints, based on the dynamic and kinematic models of the hypersonic vehicles.

[0113] The heading angle corridor and roll angle flip logic determination module is used to determine the heading angle corridor and roll angle flip logic based on the multi-hypersonic vehicle cooperative time planning objective, as well as the flight process constraints and terminal constraints.

[0114] The lateral coordination time adjustment factor calculation module is used to calculate the lateral coordination time adjustment factor in lateral guidance based on the heading angle corridor and the roll angle flip logic.

[0115] The tilt angle profile calculation module is used to calculate the tilt angle profile in longitudinal guidance based on the lateral coordination time adjustment factor.

[0116] The Transformer-based deep learning framework determination module is used to determine the Transformer-based deep learning framework based on the lateral coordination time adjustment factor in lateral guidance and the tilt angle profile in longitudinal guidance. The Transformer-based deep learning framework is used to plan the trajectory of hypersonic vehicles online to achieve intelligent time-coordinated guidance for multiple hypersonic vehicles.

[0117] Example 3

[0118] This invention provides an electronic device including a memory and a processor. The memory stores a computer program, and the processor runs the computer program to enable the electronic device to execute the intelligent time-coordinated guidance method for multiple hypersonic vehicles as described in Embodiment 1.

[0119] Alternatively, the aforementioned electronic device may be a server.

[0120] In addition, this embodiment of the invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the intelligent time-coordinated guidance method for multiple hypersonic vehicles of Embodiment 1.

[0121] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0122] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for intelligent time-coordinated guidance of multiple hypersonic vehicles, characterized in that, include: Establish dynamic and kinematic models for hypersonic vehicles; Based on the dynamic and kinematic models of hypersonic vehicles, the objectives of cooperative time planning for multiple hypersonic vehicles, as well as the constraints during flight and the terminal constraints, are determined. Based on the objectives of multi-hypersonic vehicle cooperative time planning, as well as the flight process constraints and terminal constraints, the heading angle corridor and roll angle flip logic are determined. Based on the heading angle corridor and roll angle flip logic, the lateral coordination time adjustment factor in lateral guidance is calculated; the lateral coordination time adjustment factor is: ;in, ; This refers to the horizontal coordination time error; Calculate the tilt angle profile in longitudinal guidance based on the lateral coordination time adjustment factor; Based on the lateral coordination time adjustment factor in lateral guidance and the tilt angle profile in longitudinal guidance, a deep learning framework based on Transformer networks is determined; the deep learning framework based on Transformer networks is used to plan the trajectory of hypersonic vehicles online to achieve intelligent time-coordinated guidance for multiple hypersonic vehicles.

2. The intelligent time-coordinated guidance method for multiple hypersonic vehicles according to claim 1, characterized in that, The objective of the multi-hypersonic vehicle cooperative time planning is to consider... For any hypersonic vehicle, for any... The hypersonic vehicle, if the first The moment when the hypersonic vehicle arrives at the terminal area Then it is believed The goal is to achieve time coordination between hypersonic vehicles.

3. The intelligent time-coordinated guidance method for multiple hypersonic vehicles according to claim 1, characterized in that, The flight process constraints include overload constraints, heat flux density constraints, dynamic pressure constraints, and quasi-equilibrium gliding constraints; the terminal constraints include terminal waiting flight range constraints, latitude and longitude constraints, time constraints, altitude constraints, and heading angle constraints.

4. The intelligent time-coordinated guidance method for multiple hypersonic vehicles according to claim 1, characterized in that, The logic for the heading angle corridor and heel angle flipping is as follows: ； in, Let be the heading angle deviation of the i-th hypersonic vehicle. Let be the heading angle of the i-th hypersonic vehicle. Let be the line-of-sight angle of the i-th hypersonic vehicle observing the target. The upper boundary of the heading angle corridor. The lower boundary of the heading angle corridor. The tilt angle in the previous guidance round. The symbol for the tilt angle in the current guidance round.

5. The intelligent time-coordinated guidance method for multiple hypersonic vehicles according to claim 1, characterized in that, Based on the heading angle corridor and roll angle flip logic, the lateral coordination time adjustment factor in lateral guidance is calculated, specifically including: Based on the heading angle corridor and the tilt angle flip logic, the lateral coordination time adjustment factor in lateral guidance is calculated using Newton's iteration method.

6. The intelligent time-coordinated guidance method for multiple hypersonic vehicles according to claim 1, characterized in that, Based on the lateral cooperative time adjustment factor in lateral guidance and the tilt angle profile in longitudinal guidance, a deep learning framework based on Transformer networks is determined, specifically including: Using the lateral cooperative time adjustment factor in lateral guidance and the trajectory generated by the tilt angle profile in longitudinal guidance as the training set, a deep learning network based on the Transformer network is trained to obtain a deep learning framework based on the Transformer network.

7. A smart time-coordinated guidance system for multiple hypersonic vehicles, characterized in that, include: The dynamics and kinematics model building module is used to build dynamics and kinematics models for hypersonic vehicles. The objective and constraint determination module is used to determine the objectives of multi-hypersonic vehicle cooperative time planning, as well as the flight process constraints and terminal constraints, based on the dynamic and kinematic models of the hypersonic vehicles. The heading angle corridor and roll angle flip logic determination module is used to determine the heading angle corridor and roll angle flip logic based on the multi-hypersonic vehicle cooperative time planning objective, as well as the flight process constraints and terminal constraints. The lateral coordination time adjustment factor calculation module is used to calculate the lateral coordination time adjustment factor in lateral guidance based on the heading angle corridor and roll angle flip logic; the lateral coordination time adjustment factor is... ;in, ; This refers to the horizontal coordination time error; The tilt angle profile calculation module is used to calculate the tilt angle profile in longitudinal guidance based on the lateral coordination time adjustment factor. The Transformer-based deep learning framework determination module is used to determine the Transformer-based deep learning framework based on the lateral coordination time adjustment factor in lateral guidance and the tilt angle profile in longitudinal guidance. The Transformer-based deep learning framework is used to plan the trajectory of hypersonic vehicles online to achieve intelligent time-coordinated guidance for multiple hypersonic vehicles.

8. An electronic device, characterized in that, The device includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to enable the electronic device to perform a method for intelligent time-coordinated guidance of multiple hypersonic vehicles according to any one of claims 1 to 6.

Images