CN116911004A - Trajectory drop point correction method based on neural network - Google Patents
Trajectory drop point correction method based on neural network Download PDFInfo
- Publication number
- CN116911004A CN116911004A CN202310824507.8A CN202310824507A CN116911004A CN 116911004 A CN116911004 A CN 116911004A CN 202310824507 A CN202310824507 A CN 202310824507A CN 116911004 A CN116911004 A CN 116911004A
- Authority
- CN
- China
- Prior art keywords
- neural network
- trajectory
- drop point
- training
- ballistic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000012937 correction Methods 0.000 title claims abstract description 15
- 238000004364 calculation method Methods 0.000 claims abstract description 16
- 238000004088 simulation Methods 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 26
- 239000002245 particle Substances 0.000 claims description 12
- 238000003062 neural network model Methods 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 6
- 230000001133 acceleration Effects 0.000 claims description 5
- 230000001419 dependent effect Effects 0.000 claims description 3
- 238000002474 experimental method Methods 0.000 claims description 3
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 claims description 3
- 239000010931 gold Substances 0.000 claims description 3
- 229910052737 gold Inorganic materials 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000010304 firing Methods 0.000 claims description 2
- 238000000638 solvent extraction Methods 0.000 claims 2
- 238000005192 partition Methods 0.000 claims 1
- 238000010200 validation analysis Methods 0.000 claims 1
- 238000013178 mathematical model Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000000611 regression analysis Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000012897 Levenberg–Marquardt algorithm Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/15—Correlation function computation including computation of convolution operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Linguistics (AREA)
- Mathematical Analysis (AREA)
- Molecular Biology (AREA)
- Computational Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Optimization (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computer Hardware Design (AREA)
- Databases & Information Systems (AREA)
- Geometry (AREA)
- Algebra (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of gun trajectory, in particular to a trajectory drop point correction method based on a neural network. The invention utilizes the characteristic of infinite approximation of the neural network to the nonlinear function, realizes the correction of the drop point of the trajectory, greatly eliminates errors generated by the limitation of a mathematical model under the influence of various factors such as different trajectory conditions, meteorological conditions, geographic conditions and the like of the self-propelled gun, improves the drop point precision of the calculation trajectory, and increases the reliability of simulation.
Description
Technical Field
The invention relates to the technical field of gun trajectory, in particular to a trajectory drop point correction method based on a neural network.
Background
The accurate prediction of the falling point of the cannon is an important problem to be solved in the modern military field, and the prediction of the falling point of the self-propelled cannon becomes very complex under the action of multiple random factors such as different ballistic conditions, meteorological conditions, geographical conditions and the like. In the prior art, the traditional external trajectory calculation is equivalent to the particle calculation, and although a plurality of disturbance factors are added to carry out numerical calculation, uncertain or difficult disturbance factors such as movement of the projectile around the mass center exist, so that the traditional external trajectory calculation has a certain limitation.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a ballistic drop point correction method based on a neural network, which improves the drop point precision of a solution trajectory and increases the reliability of simulation by adding drop point correction.
In order to achieve the above purpose, the technical scheme of the invention is as follows:
a ballistic drop point correction method based on a neural network comprises the following steps:
step 1, according to ballistic conditions, meteorological conditions and geographical conditions, a ballistic differential equation set is listed, and calculation of the outside-particle ballistic drop points is carried out through Matlab;
step 2, making a difference between the falling point coordinate obtained through calculation and the falling point coordinate of the shell launching experimental data to obtain an error sample;
step 3, determining the number of samples and parameters of the neural network;
step 4, training a model by using Matlab;
and 5, saving the trained neural network model, performing off-line simulation by applying the trained neural network, and correcting the calculated falling point coordinates by using the obtained error value.
Further, in step 1, the system of ballistic differential equations is:
in the formula ,the force of gravity is applied to the acceleration vector,Cas a function of the coefficient of trajectory,Has a function of the density of the air,Gas a function of the resistance force,v x 、 v y 、v z respectively isx、y、zThe velocity of the projectile in the direction is,ω x 、ω z respectively isx、zThe wind speed in the direction of the wind,vfor the theoretical projectile velocity,v r in order to achieve the actual velocity of the projectile,αin a ground coordinate systemxThe included angle between the axis and the north direction,Rfor the radius of the earth,Λis latitude, omega is the earth rotation angular velocity vector, and the initial parameters of the selected shell aretWhen the value of the sum is =0,θthe angle of inclination of the projectilev x =v 0 cosθ 0 ,v y =v 0 cosθ 0 ,v z =0,x=y=z=0, and the falling point coordinates of the outside-particle trajectory were calculated by Matlab using the range-Kutta method and cubic spline interpolation.
Advancing oneIn step 2, the coordinates of the falling point are calculatedxValue sumzA value; the error sample is the difference value.
Further, in step 3, the neural network parameters include input layer, output layer, hidden layer, algorithm, data set division and error optimization parameters.
Further, the data of the nodes of the input layer corresponds to the data of the initial parameters of the projectile launching experiment in the step 1, and is the initial projectile velocityv 0 Angle of emissionθ 0 Two parameters are used as independent variables.
Further, the output layer is the error sample in step 2, and is used as a dependent variable.
Further, the hidden layer setting method comprises the following steps: in the field function in Matlab, hiddenaersize is set as a 3-layer hidden layer, and the number of hidden layer nodes is selectedM:
,
wherein ,nandmthe number of the input and output layers is respectively,ataking a constant between 0 and 10, thereforeMFor constants between 2 and 12, selecting the number of hidden layer nodes to be 2 to 12, training, and selecting a constant capable of obtaining a minimum mean square error value as the constantMValues.
Further, the algorithm uses a tranfcn function.
Further, the data set division uses a divideFcn function to divide the sample into a training data sample, a verification data sample and a test data sample; the error optimization parameter is performFcn, the training step number epoch is 10000 steps, the training precision gold is 0.01, namely, the training is completed when the training reaches 10000 steps or less than the training precision 0.01.
Further, the specific steps of the step 5 are as follows:
using a function RFModelSavePath to store the trained model under a computer directory;
inputting initial parameters in shell launching experimental data into a neural network model to obtain corresponding errors, and inputting the same initial parameters into Matlab to calculate out the outside trajectory drop points of particles;
and the calculated trajectory coordinates and the error value are summed to obtain a more accurate trajectory drop point.
The beneficial effects of the invention are as follows:
1. according to the trajectory drop point prediction method based on the neural network, aiming at the limitation of the existing numerical value solution of the trajectory outside the particles, the influence of disturbance parameters which are not considered by a numerical value solution on the solution result is greatly eliminated, and the accuracy of the method exceeds that of the traditional numerical value solution and is more approximate to real experimental data;
2. the method combines classical numerical value calculation and intelligent algorithm, has innovative thought, and uses a large number of samples, thereby increasing the accuracy of the neural network model and ensuring that the obtained error has higher universality;
3. the method is simple and convenient, can optimize the drop point coordinates of the numerical settlement by only storing the offline neural network model, and has the characteristics of high calculation speed, small occupied space, high calculation precision and high reliability.
Drawings
FIG. 1 is a schematic flow chart of a method for correcting a ballistic drop point based on a neural network;
fig. 2 is a schematic diagram of a Matlab neural network used in the present invention.
Detailed Description
The invention will now be described in detail by way of example with reference to the accompanying drawings.
Examples:
as shown in fig. 1, the embodiment relates to a method for predicting a drop point of a trajectory based on a neural network, which adds drop point correction to the existing calculation accuracy, improves the drop point accuracy of the calculated trajectory, and increases the reliability of simulation, and comprises the following specific steps:
step 1, according to ballistic conditions, meteorological conditions and geographical conditions, a ballistic differential equation set is listed, and the calculation of the outside-particle ballistic drop point is carried out through Matlab, wherein the method comprises the following specific steps:
meanwhile, the change of the earth curvature, the change of the gravity acceleration along with the change of the altitude, the Coriolis force generated by the rotation of the earth, the air resistance of the crosswind and the longitudinal wind to the shell and the change of the air resistance acceleration are considered, and the external trajectory equation set of the particles is obtained as follows:
in the formula ,for the initial gravitational acceleration vector,Cas a function of the coefficient of trajectory,Has a function of the density of the air,Gas a function of the resistance force,v x 、v y 、v z respectively isx、y、zThe velocity of the projectile in the direction is,ω x 、ω z is thatx、zThe wind speed in the direction of the wind,vfor the theoretical projectile velocity,v r in order to achieve the actual velocity of the projectile,αin a ground coordinate systemxThe included angle between the axis and the north direction,Rfor the radius of the earth,Λis latitude, omega is the earth rotation angular velocity vector, and the initial parameters of the selected shell aretWhen the value of the sum is =0,θthe angle of inclination of the projectilev x =v 0 cosθ 0 ,v y =v 0 cosθ 0 ,v z = 0,x=y=z=0, using the range-Kutta method and cubic spline interpolation, calculating the falling point coordinates of the outside trajectory of the particle by Matlab;
step 2, making a difference between the falling point coordinate obtained through calculation and the falling point coordinate of the shell launching experimental data to obtain an error sample; wherein the coordinates of the drop point are calculatedxValue sumzA value; the error sample is a value obtained by making a difference;
step 3, determining the number of samples and parameters of the neural network; wherein the number of the samples is 500, namely the initial parameters of 500 groups of projectile firing experimental data and the errors obtained in the step 2A sample; the neural network parameters comprise input layer, output layer, hidden layer, algorithm, data set division and error optimization parameters, the data of the nodes of the input layer correspond to the data of the initial parameters of the projectile launching experiment in the step 1, and the data is the initial speed of the projectilev 0 Angle of emissionθ 0 Two parameters as independent variables; the output layer is an error sample in the step 2 and is used as a dependent variable; the hidden layer setting method comprises the following steps: in the fitnet function in Matlab, hiddenaersize is set to [M MM]I.e. 3 hidden layers, the invention selects the number of hidden layer nodes according to an empirical formulaM:
,
wherein ,nandmthe number of the input and output layers is respectively,ataking a constant between 0 and 10, thereforeMThe constant is between 2 and 12, the number of hidden layer nodes is selected to be between 2 and 12, training is carried out, and the constant which can obtain the minimum mean square error value is selected asMA value; in this embodiment, through trial calculation, the number of the finally selected hidden layer nodes is 10, the mean square error is minimum, and the neural network structure schematic diagram is shown in fig. 2; the algorithm uses a tranFcn function, and inputs a tranlm' namely a Levenberg-Marquardt algorithm; the data set division uses a divideFcn function, 350 groups of samples are selected as training data samples, 75 groups of samples are selected as verification data samples, and 75 groups of samples are selected as test data samples; the error optimization parameter is performFcn, the training step number epoch is 10000 steps, and the training precision gold is 0.01, namely the training is completed when the training reaches 10000 steps or less than the training precision of 0.01; in this embodiment, the correlation of regression analysis, which is the regression analysis obtained after training is completed, is 1, which indicates that the training result has reliability;
step 4, training a model by using Matlab;
and 5, saving a trained neural network model, performing off-line simulation by applying the trained neural network, and correcting the calculated falling point coordinates by using the obtained error value, wherein the method comprises the following specific steps of:
the function RFModelSavePath is used for storing a trained model under a computer directory, the model is input into a neural network model according to initial parameters in shell emission experimental data, corresponding errors can be obtained, the same initial parameters are input into Matlab for resolving out-of-particle trajectory drop points, and the resolved trajectory coordinates and error values are summed to obtain more accurate trajectory drop points.
The invention utilizes the characteristic of infinite approximation of the neural network to the nonlinear function, realizes the correction of the ballistic drop point, and greatly eliminates the errors of the self-propelled gun due to the limitation of the mathematical model under the influence of various factors such as different ballistic conditions, meteorological conditions, geographical conditions and the like.
Alterations, modifications, substitutions and variations of the embodiments herein will be apparent to those of ordinary skill in the art in light of the teachings of the present invention without departing from the spirit and principles of the invention.
Claims (10)
1. The ballistic drop point correction method based on the neural network is characterized by comprising the following steps of:
step 1, according to ballistic conditions, meteorological conditions and geographical conditions, a ballistic differential equation set is listed, and calculation of the outside-particle ballistic drop points is carried out through Matlab;
step 2, making a difference between the falling point coordinate obtained through calculation and the falling point coordinate of the shell launching experimental data to obtain an error sample;
step 3, determining the number of samples and parameters of the neural network;
step 4, training a model by using Matlab;
and 5, saving the trained neural network model, performing off-line simulation by applying the trained neural network, and correcting the calculated falling point coordinates by using the obtained error value.
2. The method for correcting a trajectory drop point based on a neural network according to claim 1, wherein in step 1, the system of trajectory differential equations is:
in the formula ,the force of gravity is applied to the acceleration vector,Cas a function of the coefficient of trajectory,Has a function of the density of the air,Gas a function of the resistance force,v x 、v y 、v z respectively isx、y、zThe velocity of the projectile in the direction is,ω x 、ω z respectively isx、zThe wind speed in the direction of the wind,vfor the theoretical projectile velocity,v r in order to achieve the actual velocity of the projectile,αin a ground coordinate systemxThe included angle between the axis and the north direction,Rfor the radius of the earth,Λis latitude, omega is the earth rotation angular velocity vector, and the initial parameters of the selected shell aretWhen the value of the sum is =0,θthe angle of inclination of the projectilev x =v 0 cosθ 0 ,v y =v 0 cosθ 0 ,v z =0,x=y =z=0, and the falling point coordinates of the outside-particle trajectory were calculated by Matlab using the range-Kutta method and cubic spline interpolation.
3. The method for correcting a ballistic drop point based on a neural network according to claim 1, wherein in step 2, the drop point coordinates are calculatedxValue sumzA value; the error sample is the difference value.
4. A method of correcting ballistic drop points based on neural networks according to claim 1, 2 or 3, wherein in step 3, the neural network parameters include input layer, output layer, hidden layer, algorithm, data set partitioning and error optimization parameters.
5. The neural network-based trajectory landing point correction method of claim 4, wherein the data of the nodes of the input layer corresponds to the data of the initial parameters of the projectile firing experiment in step 1, and is the initial projectile velocityv 0 Angle of emissionθ 0 Two parameters are used as independent variables.
6. The neural network-based trajectory drop correction method of claim 4, wherein the output layer is the error sample in step 2 as a dependent variable.
7. The method for correcting the trajectory drop point based on the neural network according to claim 4, wherein the hidden layer setting method is as follows: in the field function in Matlab, hiddenaersize is set as a 3-layer hidden layer, and the number of hidden layer nodes is selectedM:
,
wherein ,nandmthe number of the input and output layers is respectively,ataking a constant between 0 and 10, thereforeMFor constants between 2 and 12, selecting the number of hidden layer nodes to be 2 to 12, training, and selecting a constant capable of obtaining a minimum mean square error value as the constantMValues.
8. The neural network-based trajectory drop correction method of claim 4, wherein the algorithm uses a tranfcn function.
9. The neural network-based ballistic drop point correction method of claim 4, wherein the data set partitioning uses a divideFcn function to partition the samples into training data samples, validation data samples, and test data samples; the error optimization parameter is performFcn, the training step number epoch is 10000 steps, the training precision gold is 0.01, namely, the training is completed when the training reaches 10000 steps or less than the training precision 0.01.
10. The method for correcting the trajectory drop point based on the neural network according to claim 1, wherein the specific steps of the step 5 are as follows:
using a function RFModelSavePath to store the trained model under a computer directory;
inputting initial parameters in shell launching experimental data into a neural network model to obtain corresponding errors, and inputting the same initial parameters into Matlab to calculate out the outside trajectory drop points of particles;
and the calculated trajectory coordinates and the error value are summed to obtain a more accurate trajectory drop point.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310824507.8A CN116911004A (en) | 2023-07-06 | 2023-07-06 | Trajectory drop point correction method based on neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310824507.8A CN116911004A (en) | 2023-07-06 | 2023-07-06 | Trajectory drop point correction method based on neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116911004A true CN116911004A (en) | 2023-10-20 |
Family
ID=88362123
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310824507.8A Pending CN116911004A (en) | 2023-07-06 | 2023-07-06 | Trajectory drop point correction method based on neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116911004A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117272873A (en) * | 2023-11-22 | 2023-12-22 | 山东建筑大学 | Projection drop point prediction algorithm based on mathematical series enhancement deep learning method |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20130087307A (en) * | 2012-01-27 | 2013-08-06 | 주식회사 한화 | Trajectory correction method for artillery projectiles |
CN112114521A (en) * | 2020-07-30 | 2020-12-22 | 南京航空航天大学 | Intelligent prediction control entry guidance method for spacecraft |
CN112597700A (en) * | 2020-12-15 | 2021-04-02 | 北京理工大学 | Aircraft trajectory simulation method based on neural network |
CN114299106A (en) * | 2021-09-13 | 2022-04-08 | 武汉理工大学 | High-altitude parabolic early warning system and method based on visual sensing and track prediction |
CN114580301A (en) * | 2022-03-16 | 2022-06-03 | 中国人民解放军空军预警学院 | Hypersonic aircraft trajectory online prediction method based on parameter extrapolation |
CN115775472A (en) * | 2022-10-28 | 2023-03-10 | 中急管(北京)网络科技有限公司 | Intelligent pre-judging system and algorithm for low-lost motion target disposal drop point |
-
2023
- 2023-07-06 CN CN202310824507.8A patent/CN116911004A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20130087307A (en) * | 2012-01-27 | 2013-08-06 | 주식회사 한화 | Trajectory correction method for artillery projectiles |
CN112114521A (en) * | 2020-07-30 | 2020-12-22 | 南京航空航天大学 | Intelligent prediction control entry guidance method for spacecraft |
CN112597700A (en) * | 2020-12-15 | 2021-04-02 | 北京理工大学 | Aircraft trajectory simulation method based on neural network |
CN114299106A (en) * | 2021-09-13 | 2022-04-08 | 武汉理工大学 | High-altitude parabolic early warning system and method based on visual sensing and track prediction |
CN114580301A (en) * | 2022-03-16 | 2022-06-03 | 中国人民解放军空军预警学院 | Hypersonic aircraft trajectory online prediction method based on parameter extrapolation |
CN115775472A (en) * | 2022-10-28 | 2023-03-10 | 中急管(北京)网络科技有限公司 | Intelligent pre-judging system and algorithm for low-lost motion target disposal drop point |
Non-Patent Citations (6)
Title |
---|
佟妮宸: "某型车载火炮外弹道全局敏感性分析与优化设计", 中国优秀硕士学位论文全文数据库工程科技Ⅱ辑(月刊), no. 3, 15 March 2023 (2023-03-15), pages 032 - 13 * |
吴朝峰等: "基于GA-BP算法的外弹道落点误差预测", 兵器装备工程学报, vol. 40, no. 12, 31 December 2019 (2019-12-31), pages 1 * |
赵喆等: "一种基于LSTM的弹丸长时间落点预测修正方法", 现代雷达, vol. 45, no. 2, 28 February 2023 (2023-02-28), pages 1 - 3 * |
赵江;周锐;: "基于倾侧角反馈控制的预测校正再入制导方法", 兵工学报, no. 05, 15 May 2015 (2015-05-15) * |
陈磊, 王海丽, 周伯昭, 任萱: "弹道导弹显式制导的分析与研究", 宇航学报, no. 05, 10 October 2001 (2001-10-10) * |
马新谋,樊水康编著: "火力控制技术基础", 31 August 2018, 北京理工大学出版社, pages: 99 - 105 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117272873A (en) * | 2023-11-22 | 2023-12-22 | 山东建筑大学 | Projection drop point prediction algorithm based on mathematical series enhancement deep learning method |
CN117272873B (en) * | 2023-11-22 | 2024-05-31 | 山东建筑大学 | Projection drop point prediction algorithm based on mathematical series enhancement deep learning method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111176334B (en) | Multi-unmanned aerial vehicle cooperative target searching method | |
CN111351488B (en) | Intelligent trajectory reconstruction reentry guidance method for aircraft | |
CN109740198B (en) | Analytic prediction-based three-dimensional reentry guidance method for gliding aircraft | |
CN105608251B (en) | The BNSobol methods of helicopter fire control system precision sensitivity analysis | |
CN116911004A (en) | Trajectory drop point correction method based on neural network | |
CN106547991B (en) | Method for optimizing disturbance gravitation reconstruction model along glide trajectory | |
CN110991051B (en) | Remote guidance rocket projectile drop point prediction system based on experimental design and Kriging model | |
CN107367941B (en) | Method for observing attack angle of hypersonic aircraft | |
CN108983800B (en) | Airplane attitude control method based on deep learning | |
CN110597056A (en) | Large closed-loop calibration control method for antiaircraft gun fire control system | |
CN116151102A (en) | Intelligent determination method for space target ultra-short arc initial orbit | |
CN114462293B (en) | Hypersonic speed target medium-long term track prediction method | |
JP3241742B2 (en) | Neural network orbit command controller | |
CN109211232A (en) | A kind of shell Attitude estimation method based on least squares filtering | |
CN116611160A (en) | Online real-time characteristic parameter identification and trajectory prediction method for uncontrolled aircraft based on measured trajectory parameters | |
Khalil | Study on modeling and production inaccuracies for artillery firing | |
CN115408775A (en) | Initial data calculation method for standard trajectory of spacecraft based on BP neural network | |
CN111814313A (en) | Method for designing regression orbit in high-precision gravitational field | |
CN105760573A (en) | Gravity anomaly extended interpolation method of nearspace large-range maneuverable trajectory | |
CN114935277A (en) | Online planning method for ideal trajectory of gliding extended-range guided projectile | |
CN112818496B (en) | Anti-ground-defense strategy based on ant colony algorithm | |
CN115186378A (en) | Real-time solution method for tactical control distance in air combat simulation environment | |
CN108280520B (en) | Atmosphere profile calculation method and device | |
CN115046433B (en) | Aircraft time collaborative guidance method based on deep reinforcement learning | |
CN116227340A (en) | Multiple maneuver burst prevention method in zero range line direction |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |