CN110442149A - A kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm - Google Patents
A kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm Download PDFInfo
- Publication number
- CN110442149A CN110442149A CN201910693128.3A CN201910693128A CN110442149A CN 110442149 A CN110442149 A CN 110442149A CN 201910693128 A CN201910693128 A CN 201910693128A CN 110442149 A CN110442149 A CN 110442149A
- Authority
- CN
- China
- Prior art keywords
- differential evolution
- unmanned plane
- individual
- evolution algorithm
- aerodynamic parameter
- 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
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
Abstract
The unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm that the invention discloses a kind of, for realizing the accurate identification to aerodynamic parameter.This method is under the frame of traditional differential evolution algorithm, the phenomenon that being easily trapped into local minimum for differential evolution algorithm introduces self adaptive pantographic factor F and crossover probability CR, when evolutionary generation is less, zoom factor F is maximum, bigger disturbance is generated to entire population, is conducive to the diversification of population, is avoided enter into precocity, later period, F became smaller, be conducive to local optimal searching, while dynamically increasing crossover probability CR, the convergence rate in algorithm later period is promoted.The identification result of DE algorithm is improved, compared with not improving, for mean value closer to true value, variance is smaller, it was demonstrated that improve having higher success rate for DE algorithm identification, robustness is more preferable, engineering application value with higher.
Description
Technical field
The present invention and flying vehicles control field more particularly to a kind of unmanned plane based on improved differential evolution algorithm are pneumatically joined
Number discrimination method.
Background technique
During aircraft development, whether it is dynamic to be required to flight for flight characteristics prediction, design of control law, flight simulation etc.
Mechanical model.Obtaining flight dynamics model at present, there are mainly three types of approach: wind tunnel test, CFD is calculated and flight mechanics model
Identification.Wind tunnel test is the main source of rudders pneumatic power parameter, but it has can not simulate live flying environment, period completely
The defects of length, somewhat expensive.CFD, which is calculated, has the advantages such as relative efficiency, low cost, but the precision of its calculated result often needs
Test verification.Using data are output and input, using identification technology, the gas of reflection aircraft essence dynamic characteristic is established
Dynamic mathematical model, and the undetermined coefficient in mathematical model is picked out, it can be more nearly actual physical system, and to wind-tunnel
The important means of test and CFD verification.
Differential evolution algorithm is to simulate nature biotechnology population with " survival of the fittest, the survival of the fittest " as the evolutionary development of principle
A kind of regular and formation random illumination searching algorithm, it remains the global search strategy based on population, while having drop
The complexity of low genetic manipulation, dynamically track search situation simultaneously adjust search strategy, there is stronger global convergence ability and robust
The advantages such as property.Differential evolution algorithm is widely used in the parameter identification of complication system.
Differential evolution algorithm performance is better than particle swarm algorithm and other evolution algorithms.Relative to least square method constant gradient class
Algorithm, it is able to carry out global optimizing and insensitive to initial value.
Summary of the invention
The present invention provides a kind of based on the unmanned plane Aerodynamic Parameter Identification method for improving difference algorithm, improves pneumatic ginseng
The accuracy of number identification.
Concrete mode of the present invention are as follows: differential evolution algorithm applies to unmanned plane Aerodynamic Parameter Identification, substantially
It is to be formed by solution space to carry out optimizing in aerodynamic parameter to be identified, to obtain the minimum of objective function (fitness function)
Value.Its key step is divided into four initialization, variation, intersection, selection parts.Population is initialized first, it is empty in solution
Between at random generate include certain amount individual population, variation be population generation new individual method, by random two to
The difference of amount is added on third vector multiplied by certain weight generates variation individual.Crossover operation is by variation individual and original individual
The intersection on each component is carried out with certain probability.Last selection operation is the fitness value and original of the individual after intersecting
The fitness value of body is compared, and is preferentially chosen and is entered the next generation.
Further, the initialization procedure are as follows:
xji,0=Lj+rand(0,1)(Uj-Lj)
J=1,2 ..., D, i=1,2 ..., Np
UjFor the upper limit of solution space, LjFor the lower limit of solution space.Np individual is randomly generated in solution space, each individual is one
The vector of a D dimension, D is the number of aerodynamic parameter to be identified, that is, produces Np parameter vectors to be identified.The bound of solution space
Depending on the priori knowledge to aircraft correlation aerodynamic parameter.
Further, the mutation process are as follows:
vi,G+1=xr1,G+F·(xr2,G-xr3,G)
In the G times iteration, to i-th of individual xi,GMutation operation is carried out, three individuals is randomly selected and (is different from xi,G), wherein
Two are subtracted each other multiplied by being added on third individual after a zoom factor F, as xi,GVariation individual vi,G+1.It introduces simultaneously
Self adaptive pantographic factor F:
F=F0·2λ
Wherein G is current iteration algebra, GmFor greatest iteration algebra.
Further, the crossover process are as follows:
Variation individual and former individual are intersected, CR is crossover probability, and rand (0,1) is the random number between 0 to 1, rand
It (D) is 1,2, a random number between D guarantees that the individual at least one dimension parameter to be identified after intersecting comes from
In variation individual.Crossover probability CR
Wherein G is current iteration algebra, GmFor greatest iteration algebra.
Further, the selection course are as follows:
The criterion of selection operation is greedy criterion, chooses the small individual of objective function and enters the next generation, so as to be identified is pneumatic
Parameter is gradually close to true value.
Further, above-mentioned variation, intersection, the continuous iteration of selection operation, until reaching termination condition.
Further, fitness function f chooses as follows:
F=(∫ | Z-Y | dt)T(∫|Z-Y|dt)
Y is actual output response, and Z is model output response to be identified.
Further, the minimum value of the objective function (fitness function) are as follows:
Further, the F0Take 0.3.
Further, the CRmin=0.3, CRmax=0.9, K take 4.
Compared with prior art, the invention has the advantages that:
1. relative to the most common least square method constant gradient class algorithm of Aerodynamic Parameter Identification, differential evolution algorithm is able to carry out entirely
Office's optimizing and insensitive to initial value.
2. the phenomenon that being easily trapped into local minimum for differential evolution algorithm introduces self adaptive pantographic factor F and intersects general
Rate CR, when evolutionary generation is less, zoom factor F is maximum, and bigger disturbance is generated to entire population, is conducive to population
Diversification, avoid enter into precocity, the later period, F became smaller, be conducive to local optimal searching, while dynamically increasing crossover probability CR so that
The convergence rate in algorithm later period can be promoted.
3. improving the identification result of DE algorithm, compared with not improving, for mean value closer to true value, variance is smaller, it was demonstrated that improves DE and calculates
Method identification has higher success rate, and robustness is more preferable.
Detailed description of the invention
Attached drawing illustrates the illustrative embodiments of the disclosure, and it is bright together for explaining the principles of this disclosure,
Which includes these attached drawings to provide further understanding of the disclosure, and attached drawing includes in the description and constituting this theory
A part of bright book.
Fig. 1 is the target Aerodynamic Parameter Identification process according at least one embodiment of the disclosure
Fig. 2 is the process of improved differential evolution algorithm
Fig. 3 is the desired input signals design according at least one embodiment of the disclosure
Fig. 4 is to pass through improved DE algorithm and the identification of the DE algorithm of standard according to the target of the disclosure at least one embodiment
As a result comparison diagram.
Specific embodiment
The disclosure is described further with reference to the accompanying drawings and detailed description.It is understood that this place is retouched
The specific embodiment stated is only used for explaining related content, rather than the restriction to the disclosure.It also should be noted that in order to
Convenient for description, part relevant to the disclosure is only illustrated in attached drawing.
Concrete mode of the present invention are as follows: differential evolution algorithm applies to unmanned plane Aerodynamic Parameter Identification, substantially
It is to be formed by solution space to carry out optimizing in aerodynamic parameter to be identified, to obtain the minimum of objective function (fitness function)
Value.Its key step is divided into four initialization, variation, intersection, selection parts.Population is initialized first, it is empty in solution
Between at random generate include certain amount individual population, variation be population generation new individual method, by random two to
The difference of amount is added on third vector multiplied by certain weight generates variation individual.Crossover operation is by variation individual and original individual
The intersection on each component is carried out with certain probability.Last selection operation is the fitness value and original of the individual after intersecting
The fitness value of body is compared, and is preferentially chosen and is entered the next generation.
Further, the initialization procedure are as follows:
xji,0=Lj+rand(0,1)(Uj-Lj)
J=1,2 ..., D, i=1,2 ..., Np
UjFor the upper limit of solution space, LjFor the lower limit of solution space.Np individual is randomly generated in solution space, each individual is one
The vector of a D dimension, D is the number of aerodynamic parameter to be identified, that is, produces Np parameter vectors to be identified.The bound of solution space
Depending on the priori knowledge to aircraft correlation aerodynamic parameter.
Further, the mutation process are as follows:
vi,G+1=xr1,G+F·(xr2,G-xr3,G)
In the G times iteration, to i-th of individual xi,GMutation operation is carried out, three individuals is randomly selected and (is different from xi,G), wherein
Two are subtracted each other multiplied by being added on third individual after a zoom factor F, as xi,GVariation individual vi,G+1.It introduces simultaneously
Self adaptive pantographic factor F:
F=F0·2λ
Wherein G is current iteration algebra, GmFor greatest iteration algebra.
Further, the crossover process are as follows:
Variation individual and former individual are intersected, CR is crossover probability, and rand (0,1) is the random number between 0 to 1, rand
It (D) is 1,2, a random number between D guarantees that the individual at least one dimension parameter to be identified after intersecting comes from
In variation individual.Crossover probability CR
Wherein G is current iteration algebra, GmFor greatest iteration algebra.
Further, the selection course are as follows:
The criterion of selection operation is greedy criterion, chooses the small individual of objective function and enters the next generation, so as to be identified is pneumatic
Parameter is gradually close to true value.
Further, above-mentioned variation, intersection, the continuous iteration of selection operation, until reaching termination condition.
Further, fitness function f (target component) chooses as follows:
F=(∫ | Z-Y | dt)T(∫|Z-Y|dt)
Y is actual output response, and Z is model output response to be identified.Target component identification process is as shown in Figure 1.
Further, the minimum value of the objective function (fitness function) are as follows:
Further, the F0Take 0.3.
Further, the CRmin=0.3, CRmax=0.9, K take 4.Fig. 2 is the stream of improved differential evolution algorithm
Journey.
A possibility that verify improved differential evolution algorithm, establish the aircraft six-degree-of-freedom dynamics mould of the ring of light 1800
Type therefrom isolates the state equation of longitudinal short-period mode, as model to be identified.
α is the angle of attack, and q is pitch rate, and δ e is lifting angle of rudder reflection.Wherein, For ginseng to be identified
Number.The value of above five aerodynamic derivatives is calculated by Tornado aerodynamic prediction software, respectively (4.885,0.298 ,-
1.0785, -9.6198, -0.9105), in this, as true value.
Further, using the excitation of elevator as input, true unmanned plane is given respectively and is moved with mathematical model
L-G simulation test, the input/output signal of gathering simulation test are reverse using differential evolution algorithm and improved differential evolution algorithm
The aerodynamic parameter for picking out the aircraft compares the identifier of two kinds of algorithms with true value respectively, according to it close to journey
Spend the superiority-inferiority to judge identification algorithm;" 3211 " signal of the input a length of 2s when being shown in Fig. 3.
Further, the setting of DE algorithm are as follows: maximum number of iterations Gm=1500, total individual number Np=50, zoom factor F
=0.6, crossover probability CR=0.4, fitness function and the objective function to be minimized are
F=(∫ | Z-Y | dt)T(∫|Z-Y|dt)
Y is actual output response, and Z is model output response to be identified.The boundary of solution room is set as
Xmin=[1,0, -3, -15, -3];
Xmax=[11,1.2,0, -3,0];
DE algorithm is improved in addition to self adaptive pantographic factor F and crossover probability CR, remaining parameter setting is consistent with standard DE algorithm.
Two kinds of algorithms carry out 20 identifications.
As shown in figure 4, improving the identification result of DE algorithm, compared with not improving, for mean value closer to true value, variance is smaller,
It proves to improve having higher success rate for DE algorithm identification, robustness is more preferable.
Claims (10)
1. a kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm, which is characterized in that including walking as follows
Rapid: initialization variation, intersects, four parts of selection, is formed by solution space in aerodynamic parameter to be identified and carries out optimizing, with
Obtain the minimum value of objective function (fitness function).
2. a kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm according to claim 1,
It is characterized in that, the initialization procedure are as follows:
xji,0=Lj+rand(0,1)(Uj-Lj)
J=1,2 ..., D, i=1,2 ..., Np
UjFor the upper limit of solution space, LjFor the lower limit of solution space.Np individual is randomly generated in solution space, each individual is one
The vector of a D dimension, D is the number of aerodynamic parameter to be identified, that is, produces Np parameter vectors to be identified.The bound of solution space
Depending on the priori knowledge to aircraft correlation aerodynamic parameter.
3. a kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm according to claim 1,
It is characterized in that, the mutation process are as follows:
vi,G+1=xr1,G+F·(xr2,G-xr3,G)
In the G times iteration, to i-th of individual xi,GMutation operation is carried out, three individuals is randomly selected and (is different from xi,G), wherein
Two are subtracted each other multiplied by being added on third individual after a zoom factor F, as xi,GVariation individual vi,G+1.It introduces simultaneously
Self adaptive pantographic factor F:
F=F0·2λ
Wherein G is current iteration algebra, GmFor greatest iteration algebra.
4. a kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm according to claim 1,
It is characterized in that, the crossover process are as follows:
Variation individual and former individual are intersected, CR is crossover probability, and rand (0,1) is the random number between 0 to 1, rand
It (D) is 1,2, a random number between D guarantees that the individual at least one dimension parameter to be identified after intersecting comes from
In variation individual.Crossover probability CR
Wherein G is current iteration algebra, GmFor greatest iteration algebra.
5. a kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm according to claim 1,
It is characterized in that, the selection course are as follows:
The criterion of selection operation is greedy criterion, chooses the small individual of objective function and enters the next generation, so as to be identified is pneumatic
Parameter is gradually close to true value.
6. a kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm according to claim 1,
It is characterized in that: above-mentioned variation, intersection, the continuous iteration of selection operation, until reaching termination condition.
7. a kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm according to claim 1,
It is characterized in that, for unmanned plane Pneumatic Identification problem, fitness function f chooses as follows:
F=(∫ | Z-Y | dt)T(∫|Z-Y|dt)
Y is actual output response, and Z is model output response to be identified.
8. a kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm according to claim 1,
It is characterized in that, the minimum value of the objective function (fitness function) are as follows:
min f(x1,x2,…,xD)
s.t.xj∈[Lj,Uj], j=1,2 ..., D.
9. a kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm according to claim 3,
It is characterized in that: the F0Take 0.3.
10. a kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm according to claim 4,
It is characterized in that: the CRmin=0.3, CRmax=0.9, K take 4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910693128.3A CN110442149A (en) | 2019-07-30 | 2019-07-30 | A kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910693128.3A CN110442149A (en) | 2019-07-30 | 2019-07-30 | A kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110442149A true CN110442149A (en) | 2019-11-12 |
Family
ID=68432094
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910693128.3A Pending CN110442149A (en) | 2019-07-30 | 2019-07-30 | A kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110442149A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111786601A (en) * | 2020-06-24 | 2020-10-16 | 佛山科学技术学院 | Fractional order model identification method and device for permanent magnet synchronous motor |
CN111859803A (en) * | 2020-07-17 | 2020-10-30 | 东莞理工学院 | Super capacitor model parameter identification method based on differential evolution |
CN112564126A (en) * | 2020-12-14 | 2021-03-26 | 辽宁电能发展股份有限公司 | Power system network loss minimum reactive power optimization method based on improved differential evolution algorithm |
CN115129069A (en) * | 2022-08-31 | 2022-09-30 | 福思(杭州)智能科技有限公司 | Method and device for controlling traveling device, storage medium, and electronic device |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004280792A (en) * | 2003-02-14 | 2004-10-07 | United Technol Corp <Utc> | System and method of search for accelerated active set with respect to second programming for predictive control of real time model |
US20100223029A1 (en) * | 2008-06-24 | 2010-09-02 | Airbus Uk Limited | Method of designing a composite panel |
CN102298329A (en) * | 2011-08-19 | 2011-12-28 | 北京航空航天大学 | Small-size unmanned rotary wing aircraft dynamic model identification method based on adaptive genetic algorithm |
CN103020456A (en) * | 2012-12-11 | 2013-04-03 | 中国航天空气动力技术研究院 | Method for estimating identification error intervals of flight vehicle pneumatic parameters |
CN103927451A (en) * | 2014-04-21 | 2014-07-16 | 西北工业大学 | Ontrack identification method for system parameters of space robot |
CN104991997A (en) * | 2015-06-11 | 2015-10-21 | 北京航空航天大学 | Generalized rate related P-I hysteresis model-establishing method of adaptive difference evolutionary algorithm optimization |
CN106786524A (en) * | 2016-12-19 | 2017-05-31 | 清华大学 | Load model parameters discrimination method based on noise-like signal and improved differential evolution |
CN109284579A (en) * | 2018-11-12 | 2019-01-29 | 东南大学 | Nonlinear system parameter discrimination method based on improved differential evolution algorithm |
-
2019
- 2019-07-30 CN CN201910693128.3A patent/CN110442149A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004280792A (en) * | 2003-02-14 | 2004-10-07 | United Technol Corp <Utc> | System and method of search for accelerated active set with respect to second programming for predictive control of real time model |
US20100223029A1 (en) * | 2008-06-24 | 2010-09-02 | Airbus Uk Limited | Method of designing a composite panel |
CN102298329A (en) * | 2011-08-19 | 2011-12-28 | 北京航空航天大学 | Small-size unmanned rotary wing aircraft dynamic model identification method based on adaptive genetic algorithm |
CN103020456A (en) * | 2012-12-11 | 2013-04-03 | 中国航天空气动力技术研究院 | Method for estimating identification error intervals of flight vehicle pneumatic parameters |
CN103927451A (en) * | 2014-04-21 | 2014-07-16 | 西北工业大学 | Ontrack identification method for system parameters of space robot |
CN104991997A (en) * | 2015-06-11 | 2015-10-21 | 北京航空航天大学 | Generalized rate related P-I hysteresis model-establishing method of adaptive difference evolutionary algorithm optimization |
CN106786524A (en) * | 2016-12-19 | 2017-05-31 | 清华大学 | Load model parameters discrimination method based on noise-like signal and improved differential evolution |
CN109284579A (en) * | 2018-11-12 | 2019-01-29 | 东南大学 | Nonlinear system parameter discrimination method based on improved differential evolution algorithm |
Non-Patent Citations (4)
Title |
---|
AJIT KUMAR: "Aerodynamic parameter estimation using neuro-fuzzy model based method", 《2017 FIRST INTERNATIONAL CONFERENCE ON RECENT ADVANCES IN AEROSPACE ENGINEERING》 * |
冯长辉: "基于遗传算法的四旋翼无人机系统参数辨识", 《计算机测量与控制》 * |
曾玉龙: "基于改进遗传算法的四旋翼无人机参数辨识", 《基于改进遗传算法的四旋翼无人机参数辨识》 * |
邵干等: "小型无人机气动参数辨识的新型HGAPSO算法", 《航空学报》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111786601A (en) * | 2020-06-24 | 2020-10-16 | 佛山科学技术学院 | Fractional order model identification method and device for permanent magnet synchronous motor |
CN111786601B (en) * | 2020-06-24 | 2021-09-28 | 佛山科学技术学院 | Fractional order model identification method and device for permanent magnet synchronous motor |
CN111859803A (en) * | 2020-07-17 | 2020-10-30 | 东莞理工学院 | Super capacitor model parameter identification method based on differential evolution |
CN112564126A (en) * | 2020-12-14 | 2021-03-26 | 辽宁电能发展股份有限公司 | Power system network loss minimum reactive power optimization method based on improved differential evolution algorithm |
CN115129069A (en) * | 2022-08-31 | 2022-09-30 | 福思(杭州)智能科技有限公司 | Method and device for controlling traveling device, storage medium, and electronic device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110442149A (en) | A kind of unmanned plane Aerodynamic Parameter Identification method based on improved differential evolution algorithm | |
US8195345B2 (en) | Method of generating an integrated fuzzy-based guidance law for aerodynamic missiles | |
US20190050734A1 (en) | Compression method of deep neural networks | |
Menon et al. | Clearance of nonlinear flight control laws using hybrid evolutionary optimization | |
CN111813146A (en) | Reentry prediction-correction guidance method based on BP neural network prediction voyage | |
CN111723983A (en) | Time parameterization route planning method and system for unmanned aerial vehicle in unknown environment | |
CN110765706B (en) | Aerofoil unsteady stall aerodynamic coefficient modeling method based on OHNGBM (1, 1) | |
Cayiroglu et al. | Wing aerodynamic optimization by using genetic algoritm and Ansys | |
CN108646554A (en) | A kind of quick anti-interference longitudinal guidance method of the aircraft based on specified performance | |
CN113627075B (en) | Projectile pneumatic coefficient identification method based on adaptive particle swarm optimization extreme learning | |
CN113341696A (en) | Intelligent setting method for attitude control parameters of carrier rocket | |
CN110096798B (en) | Multi-state finite element model correction method | |
CN116339373A (en) | Monte Carlo self-adaptive dynamic programming unmanned aerial vehicle control method and system | |
CN115796021A (en) | Adaptive PID control parameter setting method for variant aircraft based on neural network | |
CN110532614B (en) | Pneumatic optimization method for rotating speed characteristic of rotating missile | |
CN112698666B (en) | Aircraft route optimization method based on meteorological grid | |
Trad et al. | Airfoils generation using neural networks, CST curves and aerodynamic coefficients | |
CN115016248A (en) | Motor PID control method for optimizing RBF neural network based on PSO algorithm | |
CN114021308A (en) | Pneumatic parameter identification method based on adaptive genetic algorithm and neural network | |
CN110717260A (en) | Unmanned aerial vehicle maneuvering capability model establishing method | |
Menon et al. | Improved clearance of flight control laws using hybrid optimisation | |
Shi et al. | An adaptive sequential sampling strategy-based multi-objective optimization of aerodynamic configuration for a tandem-wing UAV via a surrogate model | |
CN115951585B (en) | Hypersonic aircraft reentry guidance method based on deep neural network | |
KR102484249B1 (en) | Method and apparatus for improving the accuracy of 6-dof flight simulation using flight test data | |
CN113821057B (en) | Planetary soft landing control method and system based on reinforcement learning and storage medium |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191112 |