CN108828957A - Aircraft overall situation finite time neural network control method based on handover mechanism - Google Patents

Aircraft overall situation finite time neural network control method based on handover mechanism Download PDF

Info

Publication number
CN108828957A
CN108828957A CN201810948465.8A CN201810948465A CN108828957A CN 108828957 A CN108828957 A CN 108828957A CN 201810948465 A CN201810948465 A CN 201810948465A CN 108828957 A CN108828957 A CN 108828957A
Authority
CN
China
Prior art keywords
indicate
formula
design
neural network
design 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.)
Granted
Application number
CN201810948465.8A
Other languages
Chinese (zh)
Other versions
CN108828957B (en
Inventor
许斌
王霞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northwestern Polytechnical University
Original Assignee
Northwestern Polytechnical University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Northwestern Polytechnical University filed Critical Northwestern Polytechnical University
Priority to CN201810948465.8A priority Critical patent/CN108828957B/en
Publication of CN108828957A publication Critical patent/CN108828957A/en
Application granted granted Critical
Publication of CN108828957B publication Critical patent/CN108828957B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

The present invention relates to a kind of aircraft overall situation finite time neural network control method based on handover mechanism, belongs to flying vehicles control field, for solving the problems, such as aircraft overall situation ANN Control.Aircraft longitudinal direction model decoupling is first height subsystem and speed subsystem by this method, is controlled for height subsystem using Backstepping, is used PID control for speed subsystem.To height subsystem, handover mechanism is introduced to realize ANN Control in efficient detected field and approach the switching between overseas robust control, neural network weight is updated based on tracking error and modeling error simultaneously, improve the learning performance of neural network, provide robust designs scheme on this basis, it can be achieved that system tracking error finite time convergence control.The present invention guarantees that aircraft ANN Control works in efficient detected field always, realizes closed-loop system global stability, guarantees the performance requirement of practical engineering application.

Description

Aircraft overall situation finite time neural network control method based on handover mechanism
Technical field
The present invention relates to a kind of flying vehicles control methods, have more particularly to a kind of aircraft overall situation based on handover mechanism Neural network control method between in limited time belongs to flying vehicles control field.
Background technique
The new demand proposed in face of dual-use field to vehicle technology, contemporary aircraft flight envelope constantly expand Greatly, the configuration design of aircraft innovation and the flight environment of vehicle of complexity, cause vehicle dynamics have complex nonlinear and it is strong not The features such as certainty.Neural network can approach unknown dynamics and model uncertainty, be widely used in aircraft Control, but most methods are assumed to be controlled premised on neural network can be approached effectively always in whole region at present Device design, this, which allows for closed-loop system, can only guarantee semi-global stability, be difficult to ensure the premise in practical applications. 《Global neural dynamic surface tracking control of strict-feedback systems with application to hypersonic flight vehicle》(Bin Xu,Chenguang Yang,Yongping Pan,《IEEE Transactions on Neural Networks and Learning Systems》,2015,26(10): 2563-2575) text is directed to hypersonic aircraft vertical passage modelling dynamic surface control, realizes mind based on handover mechanism Switching through network-control and robust control, it is ensured that closed-loop system global stability, but the design be based only upon tracking error into Row neural network weight updates, and can not achieve the rapid finite time Convergence of tracking error.
Summary of the invention
Technical problems to be solved
For current aircraft neural network control method it is less consideration in control process neural network approach whether one Straight effective problem, the present invention devise a kind of aircraft overall situation finite time ANN Control side based on handover mechanism Method, this method are realized in efficient detected field between ANN Control and the outer robust control of efficient detected field using handover mechanism Switching guarantees that neural network works in efficient detected field, realizes the global stability of closed-loop system, while being based on tracking error Neural network weight is updated with modeling error, the learning performance of neural network is improved, provides robust on this basis and set Meter scheme, it can be achieved that system tracking error finite time convergence control.
Technical solution
A kind of aircraft overall situation finite time neural network control method based on handover mechanism, it is characterised in that step is such as Under：
Step 1：Consider aircraft vertical passage kinetic model：
The kinetic model is by five quantity of state X=[V, h, γ, α, q]TU=[δ is inputted with two controlse,Φ]T Composition；Wherein, V indicates speed, and γ indicates that flight path angle, h indicate height, and α indicates that the angle of attack, q indicate rate of pitch, δeIt indicates Angle of rudder reflection, Φ indicate throttle valve opening；T, D, L and MyyRespectively indicate thrust, resistance, lift and pitch rotation torque；m,IyyWith G respectively indicate quality, pitch axis rotary inertia and gravity caused by acceleration；
The expression formula of power, torque and each coefficient is respectively：
T=TΦ(α)Φ+T0(α)≈(β1Φ+β2)α3+(β3Φ+β4)α2+(β5Φ+β6)α+(β7Φ+β8),
Wherein,Indicating dynamic pressure, ρ indicates atmospheric density,Indicate mean aerodynamic chord, zTIndicate moment of thrust brachium, S table Show pneumatic area of reference, And β(·)Indicate aerodynamic parameter；
Step 2：Definition height tracing error is eh=h-hd, design flight-path angle instruction γdFor：
Wherein, hdIndicate elevation references instruction,Indicate the first differential of elevation references instruction, kh>0 and ki>0 is design Parameter；
According to time-scale separation, regard speed as slow dynamics, the first differential of design flight-path angle instruction is：
Wherein,Indicate the second-order differential of elevation references instruction；
Take x1=γ, x2=θ, x3=q, wherein θ=α+γ, because Tsin α is far smaller than L, during controller design Approximation is ignored；Posture subsystem (3)-(5) are written as following Strict-feedback form：
Wherein, fi, i=1,3 be the unknown smoothed non-linearity function obtained according to dummy vehicle, is metIts InIt is known function；gi=ωgiθgi, i=1,3 be the unknown smoothed non-linearity function obtained according to dummy vehicle, ωgiIt is unknown, θgiIt is known that meetingWherein gi>0 HeIt is known constant；
Designing switching function is：
Wherein,
Wherein, λi2>λi1>0, i=1,2,3 expression neural networks effectively approach unknown nonlinear function fiCompact subset boundary It is given by designer, b>0 and τk>0 is design parameter；
Step 3：Defining track angle tracking error is：
e1=x1-γd (11)
Designing pitch angle virtual controlling amount is：
In formula, Indicate g1Estimated value,It indicatesEstimated value,Indicate the switching function according to (9) design, k1>0, l1>0 and 0<υ1<1 is design parameter；Design adaptive neural network ControlAnd robust controlFor：
In formula,Indicate f1Estimated value,Indicate the estimated value of neural network optimal weights vector, θf1Indicate nerve net Network basis function vector,For design parameter；
Designing firstorder filter is：
In formula,It indicatesThe signal obtained afterwards by filter (15),For the signal obtained after filteringDifferential Signal, α2>0 is design parameter；
Defining modeling error is：
WhereinIt is obtained by following formula：
In formula, B1>0 is design parameter；
DesignAdaptive law is：
In formula, γ1>0, γz1>0 and δf1>0 is design parameter；
DesignAdaptive law is：
In formula, Γ1>0 HeFor design parameter；
Defining pitching angle tracking error is：
Designing pitch rate virtual controlling amount is：
In formula,k2>0, l2>0 and 0<υ2<1 is design parameter；
Designing firstorder filter is：
In formula,It indicatesThe signal obtained afterwards by filter (22),For the signal obtained after filteringDifferential Signal, α3>0 is design parameter；
Defining pitch rate tracking error is：
It is as follows to design angle of rudder reflection：
In formula, Indicate g3Estimated value,It indicatesEstimated value,Indicate the switching function according to (9) design, k3>0, l3>0 and 0<υ3<1 is design parameter；Design adaptive neural network ControlAnd robust controlFor：
In formula,Indicate f3Estimated value,Indicate the estimated value of neural network optimal weights vector,Indicate nerve Network basis function vector,For design parameter；
Defining modeling error is：
WhereinIt is obtained by following formula：
In formula, B3>0 is design parameter；
DesignAdaptive law is：
In formula, γ3>0, γz3>0 HeFor design parameter；
DesignAdaptive law is：
In formula, Γ3>0 HeFor design parameter；
Step 4：Defining speed tracing error is：
In formula, VdFor speed command；Designing throttle valve opening Φ is：
In formula, kpV>0, kiV>0 and kdV>0 is design parameter；
Step 5：According to obtained angle of rudder reflection δeWith throttle valve opening Φ, vehicle dynamics model (1)-(5) are returned to, Tracing control is carried out to height and speed.
Beneficial effect
A kind of aircraft overall situation finite time neural network control method based on handover mechanism proposed by the present invention, and it is existing There is technology to compare to have the beneficial effect that：
(1) the present invention is based on handover mechanisms to realize ANN Control and robust control outside efficient detected field in efficient detected field Switching between system makes neural network work in efficient detected field always, it is ensured that closed-loop system global stability.
(2) handover mechanism proposed by the invention fully consider neural network input information, it can be achieved that switching function 0~ 1 smoothly switch, reduces the consumption of control energy.
(3) present invention provides the Hybrid Learning Method combined based on tracking error and modeling error, can effectively improve and force The learning performance of neural network near field.
(4) present invention provide robust designs scheme, it can be achieved that system tracking error finite time convergence control.
Detailed description of the invention
Fig. 1 is the flow chart of the aircraft overall situation finite time neural network control method the present invention is based on handover mechanism.
Specific embodiment
Now in conjunction with embodiment, attached drawing, the invention will be further described：
The technical solution used to solve the technical problems of the present invention is that：A kind of aircraft overall situation based on handover mechanism is limited Time neural network control method, is realized by following steps：
(a) consider aircraft vertical passage kinetic model：
The kinematics model is by five quantity of state X=[V, h, γ, α, q]TU=[δ is inputted with two controlse,Φ]T Composition；Wherein, V indicates speed, and γ indicates that flight path angle, h indicate height, and α indicates that the angle of attack, q indicate rate of pitch, δeIt indicates Angle of rudder reflection, Φ indicate throttle valve opening；T, D, L and MyyRespectively indicate thrust, resistance, lift and pitch rotation torque；m,IyyWith G respectively indicate quality, pitch axis rotary inertia and gravity caused by acceleration.
(b) defining height tracing error is eh=h-hd, design flight-path angle instruction γdFor：
Wherein, hdIndicate elevation references instruction,Indicate the first differential of elevation references instruction, kh>0 and ki>0 is design Parameter.
According to time-scale separation, regard speed as slow dynamics, the first differential of design flight-path angle instruction is：
Wherein,Indicate the second-order differential of elevation references instruction.
Take x1=γ, x2=θ, x3=q, wherein θ=α+γ, because Tsin α is far smaller than L, during controller design Approximation is ignored.Posture subsystem (3)-(5) are written as following Strict-feedback form：
Wherein, fi, i=1,3 be the unknown smoothed non-linearity function obtained according to dummy vehicle, is metIts InIt is known function；gi=ωgiθgi, i=1,3 be the unknown smoothed non-linearity function obtained according to dummy vehicle, ωgiIt is unknown, θgiIt is known that meetingWhereing i>0 HeIt is known constant.
Designing switching function is：
Wherein,
Wherein, λi2>λi1>0, i=1,2,3 expression neural networks effectively approach unknown nonlinear function fiCompact subset boundary It is given by designer, b>0 and τk>0 is design parameter.
(c) defining track angle tracking error is：
e1=x1-γd (11)
Designing pitch angle virtual controlling amount is：
In formula, Indicate g>Estimated value,It indicatesEstimated value,Indicate the switching function according to (9) design, k1>0, l1>0 and 0<υ1<1 is design parameter.Design adaptive neural network ControlAnd robust controlFor：
In formula,Indicate f1Estimated value,Indicate the estimated value of neural network optimal weights vector, θf1Indicate nerve net Network basis function vector,For design parameter.
Designing firstorder filter is：
In formula,It indicatesThe signal obtained afterwards by filter (15),For the signal obtained after filteringDifferential Signal, α2>0 is design parameter.
Defining modeling error is：
WhereinIt is obtained by following formula：
In formula, B1>0 is design parameter.
DesignAdaptive law is：
In formula, γ1>0, γz1>0 and δf1>0 is design parameter.
DesignAdaptive law is：
In formula, Γ1>0 HeFor design parameter.
Defining pitching angle tracking error is：
Designing pitch rate virtual controlling amount is：
In formula,k2>0, l2>0 and 0<υ2<1 is design parameter.
Designing firstorder filter is：
In formula,It indicatesThe signal obtained afterwards by filter (22),For the signal obtained after filteringDifferential Signal, α3>0 is design parameter.
Defining pitch rate tracking error is：
It is as follows to design angle of rudder reflection：
In formula, Indicate g3Estimated value,It indicatesEstimated value,Indicate the switching function according to (9) design, k3>0, l3>0 and 0<υ3<1 is design parameter.Design adaptive neural network ControlAnd robust controlFor：
In formula,Indicate f3Estimated value,Indicate the estimated value of neural network optimal weights vector,Indicate nerve net Network basis function vector,For design parameter.
Defining modeling error is：
WhereinIt is obtained by following formula：
In formula, B3>0 is design parameter.
DesignAdaptive law is：
In formula, γ3>0, γz3>0 HeFor design parameter.
DesignAdaptive law is：
In formula, Γ3>0 HeFor design parameter.
(d) defining speed tracing error is：
In formula, VdFor speed command.Designing throttle valve opening Φ is：
In formula, kpV>0, kiV>0 and kdV>0 is design parameter.
(e) according to obtained angle of rudder reflection δeWith throttle valve opening Φ, vehicle dynamics model (1)-(5) are returned to, it is right Height and speed carry out tracing control.
Referring to Fig.1, the present invention is based on the aircraft overall situation finite time neural network control methods of handover mechanism to be applied to Hypersonic aircraft is realized by following steps：
(a) hypersonic aircraft vertical passage kinetic model is established：
Wherein, V indicates speed, and γ indicates that flight path angle, h indicate height, and α indicates that the angle of attack, q indicate rate of pitch, δe Indicate that angle of rudder reflection, Φ indicate throttle valve opening；T, D, L and MyyRespectively indicate thrust, resistance, lift and pitch rotation torque；m, IyyWith g respectively indicate quality, pitch axis rotary inertia and gravity caused by acceleration.
The expression formula of power, torque and each coefficient is respectively：
T=TΦ(α)Φ+T0(α)≈(β1Φ+β2)α3+(β3Φ+β4)α2+(β5Φ+β6)α+(β7Φ+β8),
β1=-3.7693e5, β2=-3.7225e4, β3=2.6814e4, β4=-1.7277e4,
β5=3.5542e4, β6=-2.4216e3, β7=6.3785e3, β8=-1.0090e2,
zT=8.36,
Wherein,Indicating dynamic pressure, ρ indicates atmospheric density,Indicate mean aerodynamic chord, zTIndicate moment of thrust brachium, S table Show pneumatic area of reference,.
(b) height tracing error e is definedh=h-hd, design flight-path angle instruction γdFor：
In formula, hdIndicate elevation references instruction,Indicate the first differential of elevation references instruction, kh=0.5 and ki=0.1.
According to time-scale separation, regard speed as slow dynamics, the first differential of design flight-path angle instruction is：
Wherein,Indicate the second-order differential of elevation references instruction.
Take x1=γ, x2=θ, x3=q, wherein θ=α+γ, because Tsin α is far smaller than L, during controller design Approximation is ignored.Write as following Strict-feedback form in posture subsystem (3)-(5)：
In formula, WhereinWithIt is unknown,With? Know.
(c) design switching function is：
Wherein
In formula, λi2>λi1>0, i=1,2,3 expression neural networks effectively approach unknown nonlinear function fiCompact subset side Boundary, b=2 and τk=1.
Defining track angle tracking error is：
e1=x1-γd (11)
Designing pitch angle virtual controlling amount is：
In formula, k1=2, l1=3 and υ1=0.3, Indicate g1Estimated value,It indicatesEstimated value,Indicate the switching function according to (9) design, compact subset boundary is λ11=0.55, λ12=1. Design neural network controlAnd robust controlFor：
In formula,Indicate f1Estimated value,Indicate the estimated value of neural network optimal weights vector, θf1Indicate nerve net Network basis function vector,
Designing firstorder filter is：
In formula,It indicatesThe signal obtained afterwards by filter (15),For the signal obtained after filteringDifferential Signal, α2=0.005.
Defining modeling error is：
WhereinIt is obtained by following formula：
In formula, B1=5.
DesignAdaptive law is：
In formula, γ1=1, γz1=1 and δf1=0.001.
DesignAdaptive law is：
In formula, Γ1=2 Hes
Defining pitching angle tracking error is：
Designing pitch rate virtual controlling amount is：
In formula, k2=3, l2=3 and υ2=0.3.
Designing firstorder filter is：
In formula,It indicatesThe signal obtained afterwards by filter (22),For the signal obtained after filteringDifferential Signal, α3=0.005.
Defining pitch rate tracking error is：
It is as follows to design angle of rudder reflection：
In formula, k3=3, l3=4 and υ3=0.3, Indicate g3Estimated value,It indicatesEstimated value,Indicate the switching function according to (9) design, compact subset boundary is λ31=0.55, λ32=1. Design neural network controlAnd robust controlFor：
In formula,Indicate f3Estimated value,Indicate the estimated value of neural network optimal weights vector,Indicate nerve net Network basis function vector,
Defining modeling error is：
WhereinIt is obtained by following formula：
In formula, B3=5.
DesignAdaptive law is：
In formula, γ3=1, γz3=1 He
DesignAdaptive law is：
In formula, Γ3=10-8With
(d) defining speed tracing error is：
In formula, VdFor speed command.Designing throttle valve opening Φ is：
In formula, kpV=5, kiV=0.001 and kdV=0.001.
(e) according to obtained angle of rudder reflection δeWith throttle valve opening Φ, back to the kinetic model of hypersonic aircraft (1)-(5) carry out tracing control to height and speed.
Aircraft longitudinal direction model decoupling is first height subsystem and speed subsystem by the present invention, for height subsystem It is controlled using Backstepping, uses PID control for speed subsystem.To height subsystem, introduces handover mechanism realization and effectively force ANN Control and the switching between overseas robust control is approached near field, while based on tracking error and modeling error to mind It is updated through network weight, improves the learning performance of neural network, provide robust designs scheme on this basis, it can be achieved that being The finite time convergence control for tracking error of uniting.The present invention guarantees that aircraft ANN Control works in efficient detected field always, It realizes closed-loop system global stability, guarantees the performance requirement of practical engineering application.

Claims (1)

1. a kind of aircraft overall situation finite time neural network control method based on handover mechanism, it is characterised in that step is such as Under：
Step 1：Consider aircraft vertical passage kinetic model：
The kinetic model is by five quantity of state X=[V, h, γ, α, q]TU=[δ is inputted with two controlse,Φ]TComposition； Wherein, V indicates speed, and γ indicates that flight path angle, h indicate height, and α indicates that the angle of attack, q indicate rate of pitch, δeIndicate that rudder is inclined Angle, Φ indicate throttle valve opening；T, D, L and MyyRespectively indicate thrust, resistance, lift and pitch rotation torque；m,IyyWith g points Not Biao Shi quality, pitch axis rotary inertia and gravity caused by acceleration；
The expression formula of power, torque and each coefficient is respectively：
T=TΦ(α)Φ+T0(α)≈(β1Φ+β2)α3+(β3Φ+β4)α2+(β5Φ+β6)α+(β7Φ+β8),
Wherein,Indicating dynamic pressure, ρ indicates atmospheric density,Indicate mean aerodynamic chord, zTIndicate that moment of thrust brachium, S indicate gas Dynamic area of reference,And β(·)Indicate aerodynamic parameter；
Step 2：Definition height tracing error is eh=h-hd, design flight-path angle instruction γdFor：
Wherein, hdIndicate elevation references instruction,Indicate the first differential of elevation references instruction, kh>0 and ki>0 is design parameter；
According to time-scale separation, regard speed as slow dynamics, the first differential of design flight-path angle instruction is：
Wherein,Indicate the second-order differential of elevation references instruction；
Take x1=γ, x2=θ, x3=q, wherein θ=α+γ, approximate during controller design because Tsin α is far smaller than L Ignore；Posture subsystem (3)-(5) are written as following Strict-feedback form：
Wherein, fi, i=1,3 be the unknown smoothed non-linearity function obtained according to dummy vehicle, is metWhereinIt is known function；gi=ωgiθgi, i=1,3 be the unknown smoothed non-linearity function obtained according to dummy vehicle, ωgi It is unknown, θgiIt is known that meetingWherein gi>0 HeIt is known constant；
Designing switching function is：
Wherein,
Wherein, λi2>λi1>0, i=1,2,3 expression neural networks effectively approach unknown nonlinear function fiCompact subset boundary by setting Meter person is given, b>0 and τk>0 is design parameter；
Step 3：Defining track angle tracking error is：
e1=x1-γd (11)
Designing pitch angle virtual controlling amount is：
In formula,Indicate g1Estimated value,It indicatesEstimated value, Indicate the switching function according to (9) design, k1>0, l1>0 and 0<υ1<1 is design parameter；Design neural network control And robust controlFor：
In formula,Indicate f1Estimated value,Indicate the estimated value of neural network optimal weights vector,Indicate neural network base Functional vector,For design parameter；
Designing firstorder filter is：
In formula,It indicatesThe signal obtained afterwards by filter (15),For the signal obtained after filteringDifferential signal, α2>0 is design parameter；
Defining modeling error is：
WhereinIt is obtained by following formula：
In formula, B1>0 is design parameter；
DesignAdaptive law is：
In formula, γ1>0, γz1>0 HeFor design parameter；
DesignAdaptive law is：
In formula, Γ1>0 HeFor design parameter；
Defining pitching angle tracking error is：
Designing pitch rate virtual controlling amount is：
In formula,k2>0, l2>0 and 0<υ2<1 is design parameter；
Designing firstorder filter is：
In formula,It indicatesThe signal obtained afterwards by filter (22),For the signal obtained after filteringDifferential signal, α3>0 is design parameter；
Defining pitch rate tracking error is：
It is as follows to design angle of rudder reflection：
In formula,Indicate g3Estimated value,It indicatesEstimated value, Indicate the switching function according to (9) design, k3>0, l3>0 and 0<υ3<1 is design parameter；Design neural network control And robust controlFor：
In formula,Indicate f3Estimated value,Indicate the estimated value of neural network optimal weights vector,Indicate neural network base Functional vector,For design parameter；
Defining modeling error is：
WhereinIt is obtained by following formula：
In formula, B3>0 is design parameter；
DesignAdaptive law is：
In formula, γ3>0, γz3>0 HeFor design parameter；
DesignAdaptive law is：
In formula, Γ3>0 HeFor design parameter；
Step 4：Defining speed tracing error is：
In formula, VdFor speed command；Designing throttle valve opening Φ is：
In formula, kpV>0, kiV>0 and kdV>0 is design parameter；
Step 5：According to obtained angle of rudder reflection δeWith throttle valve opening Φ, vehicle dynamics model (1)-(5) are returned to, to height Degree and speed carry out tracing control.
CN201810948465.8A 2018-08-20 2018-08-20 Aircraft overall situation finite time neural network control method based on handover mechanism Active CN108828957B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810948465.8A CN108828957B (en) 2018-08-20 2018-08-20 Aircraft overall situation finite time neural network control method based on handover mechanism

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810948465.8A CN108828957B (en) 2018-08-20 2018-08-20 Aircraft overall situation finite time neural network control method based on handover mechanism

Publications (2)

Publication Number Publication Date
CN108828957A true CN108828957A (en) 2018-11-16
CN108828957B CN108828957B (en) 2019-11-05

Family

ID=64150385

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810948465.8A Active CN108828957B (en) 2018-08-20 2018-08-20 Aircraft overall situation finite time neural network control method based on handover mechanism

Country Status (1)

Country Link
CN (1) CN108828957B (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110308657A (en) * 2019-07-24 2019-10-08 西北工业大学 Elastic Vehicles Global robust intelligent control method based on singular perturbation strategy
CN110320794A (en) * 2019-07-24 2019-10-11 西北工业大学 Elastic Vehicles singular perturbation Hybrid Learning control method based on disturbance-observer
CN110888323A (en) * 2019-11-26 2020-03-17 大连理工大学 Control method for intelligent optimization of switching system
CN111487866A (en) * 2020-04-09 2020-08-04 中北大学 Hypersonic aircraft nerve anti-interference control method based on mixed event trigger mechanism
CN112327627A (en) * 2020-11-14 2021-02-05 西北工业大学 Nonlinear switching system self-adaptive sliding mode control method based on composite learning
CN112379595A (en) * 2020-11-14 2021-02-19 西北工业大学 Complex interference learning control method for nonlinear strict feedback switching system
CN113093539A (en) * 2021-03-29 2021-07-09 西北工业大学 Wide-area flight robust self-adaptive switching control method based on multi-mode division
CN113110540A (en) * 2021-04-14 2021-07-13 西北工业大学 Elastomer aircraft global finite time control method based on time scale decomposition
CN115309058A (en) * 2022-06-10 2022-11-08 哈尔滨理工大学 Finite time composite learning control method for dynamic positioning ship

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6332105B1 (en) * 1999-05-21 2001-12-18 Georgia Tech Research Corporation Neural network based automatic limit prediction and avoidance system and method
US20050137724A1 (en) * 2003-10-10 2005-06-23 Georgia Tech Research Corporation Adaptive observer and related method
GB2423377A (en) * 2002-12-09 2006-08-23 Georgia Tech Res Inst Adaptive output feedback apparatuses and methods capable of controlling a non-minimum phase system.
CN103365296A (en) * 2013-06-29 2013-10-23 天津大学 Nonlinear output feedback flight control method for quad-rotor unmanned aerial vehicle
CN104765272A (en) * 2014-03-05 2015-07-08 北京航空航天大学 Four-rotor aircraft control method based on PID neural network (PIDNN) control
US9146557B1 (en) * 2014-04-23 2015-09-29 King Fahd University Of Petroleum And Minerals Adaptive control method for unmanned vehicle with slung load
CN106094860A (en) * 2016-08-29 2016-11-09 广西师范大学 Quadrotor and control method thereof
CN106647781A (en) * 2016-10-26 2017-05-10 广西师范大学 Neural-fuzzy PID control method of four-rotor aircraft based on repetitive control compensation
CN107368091A (en) * 2017-08-02 2017-11-21 华南理工大学 A kind of stabilized flight control method of more rotor unmanned aircrafts based on finite time neurodynamics
CN107450584A (en) * 2017-08-29 2017-12-08 浙江工业大学 A kind of aircraft Adaptive Attitude control method based on set time sliding formwork
CN107450324A (en) * 2017-09-05 2017-12-08 西北工业大学 Consider the hypersonic aircraft adaptive fusion method of angle of attack constraint
CN107479383A (en) * 2017-09-05 2017-12-15 西北工业大学 Hypersonic aircraft neutral net Hybrid Learning control method based on robust designs

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6332105B1 (en) * 1999-05-21 2001-12-18 Georgia Tech Research Corporation Neural network based automatic limit prediction and avoidance system and method
GB2423377A (en) * 2002-12-09 2006-08-23 Georgia Tech Res Inst Adaptive output feedback apparatuses and methods capable of controlling a non-minimum phase system.
US20050137724A1 (en) * 2003-10-10 2005-06-23 Georgia Tech Research Corporation Adaptive observer and related method
CN103365296A (en) * 2013-06-29 2013-10-23 天津大学 Nonlinear output feedback flight control method for quad-rotor unmanned aerial vehicle
CN104765272A (en) * 2014-03-05 2015-07-08 北京航空航天大学 Four-rotor aircraft control method based on PID neural network (PIDNN) control
US9146557B1 (en) * 2014-04-23 2015-09-29 King Fahd University Of Petroleum And Minerals Adaptive control method for unmanned vehicle with slung load
CN106094860A (en) * 2016-08-29 2016-11-09 广西师范大学 Quadrotor and control method thereof
CN106647781A (en) * 2016-10-26 2017-05-10 广西师范大学 Neural-fuzzy PID control method of four-rotor aircraft based on repetitive control compensation
CN107368091A (en) * 2017-08-02 2017-11-21 华南理工大学 A kind of stabilized flight control method of more rotor unmanned aircrafts based on finite time neurodynamics
CN107450584A (en) * 2017-08-29 2017-12-08 浙江工业大学 A kind of aircraft Adaptive Attitude control method based on set time sliding formwork
CN107450324A (en) * 2017-09-05 2017-12-08 西北工业大学 Consider the hypersonic aircraft adaptive fusion method of angle of attack constraint
CN107479383A (en) * 2017-09-05 2017-12-15 西北工业大学 Hypersonic aircraft neutral net Hybrid Learning control method based on robust designs

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
BIN XU: "《Neural Control for Longitudinal Dynamics of Hypersonic Aircraft》", 《2013 INTERNATIONAL CONFERENCE ON UNMANNED AIRCRAFT SYSTEMS (ICUAS)》 *
JIANMIN WANG: "《RBF Neural Network based Adaptive Sliding Mode Control for Hypersonic Flight Vehicles》", 《PROCEEDINGS OF 2016 IEEE CHINESE GUIDANCE, NAVIGATION AND CONTROL CONFERENCE》 *
JONGHO SHIN: "《Adaptive Dynamic Surface Control for a Hypersonic Aircraft Using Neural Networks》", 《IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS》 *
李砚浓: "《基于RBF 神经网络自适应PID四旋翼飞行器控制》", 《控制工程》 *
王丽: "《基于混合神经网络的鲁棒自适应飞行控制器的设计》", 《电光与控制》 *

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110308657A (en) * 2019-07-24 2019-10-08 西北工业大学 Elastic Vehicles Global robust intelligent control method based on singular perturbation strategy
CN110320794A (en) * 2019-07-24 2019-10-11 西北工业大学 Elastic Vehicles singular perturbation Hybrid Learning control method based on disturbance-observer
CN110888323A (en) * 2019-11-26 2020-03-17 大连理工大学 Control method for intelligent optimization of switching system
CN111487866A (en) * 2020-04-09 2020-08-04 中北大学 Hypersonic aircraft nerve anti-interference control method based on mixed event trigger mechanism
CN112327627B (en) * 2020-11-14 2022-06-21 西北工业大学 Nonlinear switching system self-adaptive sliding mode control method based on composite learning
CN112379595A (en) * 2020-11-14 2021-02-19 西北工业大学 Complex interference learning control method for nonlinear strict feedback switching system
CN112379595B (en) * 2020-11-14 2022-06-21 西北工业大学 Complex interference learning control method for nonlinear strict feedback switching system
CN112327627A (en) * 2020-11-14 2021-02-05 西北工业大学 Nonlinear switching system self-adaptive sliding mode control method based on composite learning
CN113093539A (en) * 2021-03-29 2021-07-09 西北工业大学 Wide-area flight robust self-adaptive switching control method based on multi-mode division
CN113093539B (en) * 2021-03-29 2022-08-09 西北工业大学 Wide-area flight robust self-adaptive switching control method based on multi-mode division
CN113110540A (en) * 2021-04-14 2021-07-13 西北工业大学 Elastomer aircraft global finite time control method based on time scale decomposition
CN113110540B (en) * 2021-04-14 2023-01-13 西北工业大学 Elastomer aircraft global finite time control method based on time scale decomposition
CN115309058A (en) * 2022-06-10 2022-11-08 哈尔滨理工大学 Finite time composite learning control method for dynamic positioning ship

Also Published As

Publication number Publication date
CN108828957B (en) 2019-11-05

Similar Documents

Publication Publication Date Title
CN108828957B (en) Aircraft overall situation finite time neural network control method based on handover mechanism
CN107450324B (en) Consider the hypersonic aircraft adaptive fusion method of angle of attack constraint
CN111665857B (en) Variant aircraft control method based on composite intelligent learning
CN109597303B (en) Full-mode flight control method of combined type rotor craft
CN107479384B (en) The non-backstepping control method of hypersonic aircraft neural network Hybrid Learning
CN107479383B (en) Hypersonic aircraft neural network Hybrid Learning control method based on robust designs
CN109426146A (en) The high-order Non-Singular Terminal Sliding Mode control method of hypersonic aircraft
CN108762098B (en) Non-minimum phase aircraft neural network control method based on Hybrid Learning
CN109062234B (en) A kind of non-minimum phase aircraft Hybrid Learning sliding-mode control
CN107526296B (en) Hypersonic aircraft neural network lea rning control method based on prediction modeling
CN110568765A (en) Asymmetric output limited control method for hypersonic aircraft facing attack angle tracking
CN102880055A (en) Method for controlling neural network of hypersonic aerocraft on basis of prediction model
CN102880052A (en) Time scale function decomposition based hypersonic aircraft actuator saturation control method
CN109283932B (en) Four-rotor aircraft attitude control method based on integral backstepping sliding mode
CN110456643A (en) Elastic Vehicles historical data learning adaptive control method based on singular perturbation
CN109466778B (en) Method for controlling pitching and rolling postures of ejection seat based on posture parameter derivation
CN110308657A (en) Elastic Vehicles Global robust intelligent control method based on singular perturbation strategy
CN109703769B (en) Air refueling docking control method based on preview strategy
CN107632518A (en) Hypersonic aircraft neutral net Hybrid Learning control method based on High-gain observer
CN111679583A (en) Adaptive control method of variant aircraft based on aerodynamic parameter estimation
CN110320794A (en) Elastic Vehicles singular perturbation Hybrid Learning control method based on disturbance-observer
CN114564045A (en) Unmanned aerial vehicle flight control law design method considering rainfall and gust conditions
CN108303889A (en) A kind of time-scale separation aircraft elasticity body controlling means based on nonlinear transformations
CN107479382A (en) Hypersonic aircraft neural network control method based on online data study
CN113051662B (en) Pneumatic modeling and performance evaluation method for folding wingtip variant aircraft based on CFD and DATCOM

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
GR01 Patent grant
GR01 Patent grant