[go: up one dir, main page]

CN109213203B - An automatic landing control method for carrier-based aircraft based on predictive control - Google Patents

An automatic landing control method for carrier-based aircraft based on predictive control Download PDF

Info

Publication number
CN109213203B
CN109213203B CN201811066281.5A CN201811066281A CN109213203B CN 109213203 B CN109213203 B CN 109213203B CN 201811066281 A CN201811066281 A CN 201811066281A CN 109213203 B CN109213203 B CN 109213203B
Authority
CN
China
Prior art keywords
time
lon
lat
matrix
carrier
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.)
Active
Application number
CN201811066281.5A
Other languages
Chinese (zh)
Other versions
CN109213203A (en
Inventor
蒋烁莹
徐文萤
郑亚龙
江驹
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.)
Nanjing University of Aeronautics and Astronautics
Original Assignee
Nanjing University of Aeronautics and Astronautics
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 Nanjing University of Aeronautics and Astronautics filed Critical Nanjing University of Aeronautics and Astronautics
Priority to CN201811066281.5A priority Critical patent/CN109213203B/en
Publication of CN109213203A publication Critical patent/CN109213203A/en
Application granted granted Critical
Publication of CN109213203B publication Critical patent/CN109213203B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/12—Target-seeking control

Landscapes

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses an automatic carrier landing control method of a carrier-based aircraft based on anticipation control, and belongs to the technical field of automatic carrier landing, guidance and flight control. The invention discloses an automatic carrier landing method of a carrier-based aircraft based on predictive control by utilizing known lower slideway information and predictable longitudinal deck movement information in the automatic carrier landing process of the carrier-based aircraft. The lower slideway information is obtained by calculating in advance according to the relative position and relative motion relation of the carrier-based aircraft and the ship, and the deck motion information is obtained by estimating through a deck motion estimator based on particle filtering. Meanwhile, the interference of the wake flow of the warship is considered in the design process of the controller. The method can ensure landing accuracy and robustness under the interference of deck motion and wake flow of the carrier-based aircraft during landing.

Description

Automatic carrier landing control method of carrier-based aircraft based on anticipation control
Technical Field
The invention relates to an automatic carrier landing control method of a carrier-based aircraft based on anticipation control, and belongs to the technical fields of automatic carrier landing technology, guiding technology and flight control.
Background
The basic principle of the automatic carrier landing control system of the carrier-based aircraft is that a tracking radar on an aircraft carrier measures the actual position of the carrier-based aircraft, a deck motion sensor measures the motion condition of a flight deck of the aircraft carrier, data are transmitted to a compensation computer to compensate the deck motion, then the ideal position of the carrier-based aircraft is calculated, the ideal position and the actual position of the carrier-based aircraft are input into an instruction computer and are compared to obtain an error signal, a control instruction of the carrier-based aircraft is obtained through calculation of a guide control law according to the error signal, the control instruction is sent to the carrier-based aircraft through a radio data link, an autopilot on the carrier-based aircraft operates the carrier-based aircraft to eliminate errors according to the error signal received by a receiving device, and the carrier landing is safe at a preset position.
The landing of the carrier-based aircraft generally adopts a lower slideway to track the landing. The so-called gliding path tracking landing (equiangular gliding of a carrier-based aircraft) is that at the final stage of the carrier landing, after the carrier-based aircraft intercepts and captures a proper gliding path, the same gliding track angle, pitch angle, speed and sinking rate are always kept until the carrier-based aircraft collides with an aircraft carrier flight deck, and the impact type landing is realized. Due to the influence of deck motion, the whole process of the carrier-based aircraft glideslope tracking landing can be divided into two stages, namely a glideslope tracking stage and a deck motion compensation stage. The deck movement of the carrier-based aircraft is generally added into an automatic carrier landing control system 12.5s before carrier landing, so that the carrier-based aircraft can track the deck movement simultaneously in the process of tracking the gliding carrier landing. In the actual glide tracking stage, the traditional PID (proportion-integration-differentiation) controller is difficult to enable the carrier-based aircraft to quickly track the glide track; in the actual deck compensation stage, the traditional PID controller is difficult to enable the carrier-based aircraft to completely track deck movement in the final landing stage, so that the landing success rate is reduced. Therefore, the design of the automatic carrier landing control method of the carrier-based aircraft is particularly important for safe carrier landing of the carrier-based aircraft on the aircraft carrier, and is directly related to the success rate and the safety of the automatic carrier landing of the carrier-based aircraft.
At present, aiming at the research of automatic landing of a carrier-based aircraft, the key point of the research of scholars at home and abroad is to design a control law only by considering the current information of the movement of a glide slope and a deck. The glide track information and the deck movement information of the carrier-based aircraft in the glide process are foreseeable information, but scholars at home and abroad do not utilize the foreseeable future information to control the carrier-based aircraft.
Disclosure of Invention
The invention provides an automatic carrier landing control method of a carrier-based aircraft based on forecast control, which is used for carrying out forecast control on the carrier-based aircraft by utilizing future information and current information of a glide slope track and deck movement (sinking and floating and swaying), thereby realizing the tracking of the glide slope and the compensation of the deck movement, effectively inhibiting and eliminating lateral deviation and keeping the carrier-based aircraft stable.
The invention adopts the following technical scheme for solving the technical problems:
an automatic carrier landing control method of a carrier-based aircraft based on anticipation control comprises the following steps:
(1) deck motion prediction calculation
Discrete model with deck motion:
Figure BDA0001798386090000021
in the formula xkIs tkMoment of deck motion state quantity, xk-1Is tk-1Moment of deck motion state quantity, vkTo observe noise,. phik,k-1For the state vector x from tk-1Time shift to tkThe transition matrix of the time of day,
Figure BDA0001798386090000022
a is a system matrix of deck movements, TsFor the sampling time, Γk,k-1Is tk-1Noise vector w of time instantsk-1For tkState vector x of time of daykThe matrix of the noise coefficients of influence,
Figure BDA0001798386090000023
its variance matrix is Qk-1B is the input matrix of deck movements, wk-1Is system dynamic noise, zkIs tkObservation of deck movement at time HkIs an observation coefficient matrix with a variance matrix of Rk；
According to the discrete model (1), the deck motion estimation method based on particle filter design comprises the following processes:
1) from a priori probability P (x)0) Generating a population of particles
Figure BDA0001798386090000024
The weight of all particles of the particle swarm is 1/N;
2) and (3) prediction: from particles at time k-1
Figure BDA0001798386090000025
Obtaining predicted particles at the k moment by using a system state equation
Figure BDA0001798386090000026
3) Updating the weight value: predicting particles based on observed vector machine at time k, using
Figure BDA0001798386090000031
Updating the weight of each particle; and further carrying out normalization processing on the weight:
Figure BDA0001798386090000032
wherein:
Figure BDA0001798386090000033
the weight of the jth particle at time k,
Figure BDA0001798386090000034
the weight of the jth particle at time k-1,
Figure BDA0001798386090000035
in order to be a priori at all,
Figure BDA0001798386090000036
the weight of the jth particle at the k moment after normalization processing is carried out;
4) resampling: according to
Figure BDA0001798386090000037
Weight of (2)
Figure BDA0001798386090000038
Resampling to obtain particles
Figure BDA0001798386090000039
And reset the weight
Figure BDA00017983860900000310
Are all 1/N;
5) and (3) state estimation at the k moment:
Figure BDA00017983860900000311
simultaneously enabling k to be k +1, if k is smaller than a set threshold value, returning to the step 2), and otherwise, returning to the step 6);
6) deck movement information xkThe expression for the optimal estimate at the future time τ is
Figure BDA00017983860900000312
Where m is τ/TsAnd tau is the time of the future,
Figure BDA00017983860900000313
for state estimation at time k, state transition matrix
Figure BDA00017983860900000314
(2) Calculating the corrected slip reference track of the shipboard aircraft
First, the carrier-based aircraft captures the glidepath, knowing the initial glide height-ZEA0Angle of glide gammacVelocity of sliding down VcCalculating the landing time td
Figure BDA00017983860900000315
And length R of lower chuteA
Figure BDA00017983860900000316
Secondly, calculating a three-dimensional gliding reference track under a ground coordinate system with the ideal carrier landing point as an origin
Figure BDA00017983860900000317
Wherein: t is time, XEATDc(t) is the forward coordinate position of the carrier-based aircraft, YEATDc(t) is the lateral coordinate position of the carrier-based aircraft, ZEATDc(t) is the height coordinate position of the carrier-based aircraft, HEATDc(t) is the height value of the carrier-based aircraft, and Z is the height coordinate of the coordinate system which is positive downwardsEATDc(t)＝-HEATDc(t)，(ψS+λac) For ships and warshipsAzimuth angle of runway or glidepath, whereinSIs ship azimuth angle, λacAn included angle of the oblique angle deck is formed;
thirdly, superposing the estimated sinking and floating height and the swaying distance of the deck motion output by the deck estimator to obtain a corrected ship-based aircraft glide reference track;
(3) calculating longitudinal flight control law and transverse and lateral flight control law
The first step is to calculate the longitudinal flight control law, and the known longitudinal discretization model of the airplane is as follows:
Δxlon(k+1)＝AlonΔxlon(k)+BwlonΔw(k)+BlonΔulon(k)
wherein A islonSystem matrix being longitudinal equation of state of aircraft, BwlonAs an interference matrix, BlonFor the input matrix, Δ xlon(k) Is [ Δ V (k) Δ α (k) Δ q (k) Δ θ (k) Δ H (k)]TΔ v (k) is a speed variation at time k, Δ α (k) is an attack angle variation at time k, Δ q (k) is a pitch angle rate variation at time k, Δ θ (k) is a pitch angle variation at time k, and Δ h (k) is a height variation at time k; Δ ulon(k) Is [ Delta delta ] ofT(k) Δδe(k)]T，ΔδTDelta throttle change at time keDelta w (k) is the elevator variation at the moment k, and delta w (k) is the wake disturbance of the ship;
adding error amount, known trajectory information and generalized output, and expanding the equation into the following form:
Figure BDA0001798386090000041
wherein Xlon(k)＝[Her(k) xlon(k) HR(k) vlon(k)]T,Her(k) Is the height error, x, corrected by adding deck motion prediction at time klon(k) Is the longitudinal state quantity of the airplane,
Figure BDA0001798386090000045
is a height difference vector whichIn
Figure BDA0001798386090000046
Is time k and k + MrlonDifference between predicted values of height of glidepath, H, added with deck movement prediction and corrected at momentr(k) Is a forecast value of the height of the lower slideway at the moment k after the prediction and correction of deck movement is added, MrlonIs a longitudinal look ahead step;
Figure BDA0001798386090000043
is time k +1 and k + MrlonAdding the difference between the predicted height values of the lower slideway after the deck motion prediction correction at the moment,
Figure BDA0001798386090000044
is k + MrlonTime and k + MrlonAdding the difference between the predicted height values of the lower slideway after the deck motion prediction correction at the moment,
Figure BDA0001798386090000042
is the integral of the error, vlon(0) To adjust the initial error value, ulon(k) For longitudinal control of the aircraft, Her(j) To add the corrected height error of deck motion prediction at time j,
Figure BDA0001798386090000051
Figure BDA0001798386090000052
W1lon,W2lon,W3lonis a regulating parameter, Z1lon(k),Z2lon(k),Z3lon(k) For three generalized outputs, GlonTo expand the output matrix of the equation, FlonTo expand the state matrix of the equation, HwlonAn interference matrix which is an extended equation;
looking for a control signal DeltaublonSo as to control the performance index JlonThe size of the particles is minimized and,
Figure BDA0001798386090000053
the control law obtained is:
Figure BDA0001798386090000054
wherein: u. ofblon(k) Control input quantity at time k, kelon，kxlon，kvlon，kwlon，
Figure BDA0001798386090000056
To control law gain, xlon *Is a state quantity at an equilibrium point, ulon *As a control quantity at the balance point, Her(s) is the height error after the deck motion prediction correction is added at the moment s,
Figure BDA0001798386090000055
time k + i and k + MrlonAnd (k) adding the difference between the predicted values of the height of the lower slipway after the deck motion prediction correction at the moment, wherein w (k) is the wake disturbance of the ship.
Figure BDA0001798386090000057
kwlon＝-Rlon -1(GlonPlonHwlon),
Rlon＝W2lon TW2lon+Glon TPlonGlon
Wherein: rlonIntermediate calculation variables.
PlonIs a steady state solution of the following discrete algebraic Riccati equation:
Figure BDA0001798386090000061
wherein
Figure BDA0001798386090000069
Is a control law gain matrix;
and secondly, calculating a transverse and lateral flight control law, wherein a known transverse and lateral discretization model of the airplane is as follows:
Δxlat(k+1)＝AlatΔxlat(k)+BwlatΔw(k)+BlatΔulat(k)
wherein A islatSystem matrix being the transverse equation of state of the aircraft, BwlatAs an interference matrix, BlatFor the input matrix, Δ xlat(k) Is [ Delta beta (k) Delta p (k) Delta r (k) Delta phi (k) Delta psi (k) Delta y (k)]TΔ β (k) is a yaw angle variation at time k, Δ p (k) is a roll angle rate variation at time k, Δ r (k) is a yaw angle rate variation at time k, Δ φ (k) is a roll angle variation at time k, Δ ψ (k) is a yaw angle variation at time k, Δ y (k) is a yaw angle variation at time k, Δ u (k) is a yaw angle variation at time klon(k) Is [ Delta delta ] ofa(k) Δδr(k)]T，Δδa(k) Deltar(k) The rudder deflection variation at the moment k, and delta w (k) is the wake flow interference of the ship;
adding error amount, known trajectory information and generalized output, and expanding the equation into the following form:
Figure BDA0001798386090000063
wherein Xlat(k)＝[yer(k) xlat(k) yR(k) vlat(k)]T,yer(k) Is the lateral deviation, x, of the corrected deck motion prediction added at time klat(k) Is the transverse state quantity of the airplane,
Figure BDA0001798386090000066
is a lateral deviation vector, wherein
Figure BDA0001798386090000067
Is time k and k + MrlonTemporal reference cross with deck motion prediction correctionDifference between directional deviation information, yr(k) Reference lateral deviation information at time k, M, corrected by adding deck motion predictionrlatIs a look-ahead step size in the lateral direction,
Figure BDA0001798386090000068
is time k +1 and k + MrlonThe difference between the reference lateral deviation information added with the deck motion forecast correction at the moment,
Figure BDA0001798386090000064
is k + MrlonTime and k + MrlonThe difference between the reference lateral deviation information added with the deck motion forecast correction at the moment,
Figure BDA0001798386090000065
is the integral of the lateral offset, vlat(0) Is the initial lateral deviation, ulat(k) Is the lateral control input;
Figure BDA0001798386090000071
Figure BDA0001798386090000072
W1lat,W2lat,W3latis an adjustment parameter; z1lat(k)，Z2lat(k)，Z3lat(k) For three generalized outputs, GlatTo expand the output matrix of the equation, FlatTo expand the state matrix of the equation, HwlatAn interference matrix which is an extended equation;
looking for a control signal DeltaublatSo as to control the performance index JlatMinimization
Figure BDA0001798386090000073
The control law obtained is:
Figure BDA0001798386090000074
wherein: u. ofblat(k) Control input quantity at time k, kelat，kxlat，kvlat，kwlat，
Figure BDA0001798386090000077
To control law gain, xlat *Is a state quantity at an equilibrium point, ulat *As a control quantity at the balance point, yer(s) adding the corrected lateral deviation error of the deck motion prediction for the moment s,
Figure BDA0001798386090000076
time k + i and k + MrlonAdding the lateral deviation difference after the deck motion prediction correction at the moment, wherein w (k) is the wake disturbance of the ship,
Figure BDA0001798386090000078
kwlat＝-Rlat -1(GlatPlatHwlat),
Rlat＝W2lat TW2lat+Glat TPlatGlat,
wherein: rlatCalculating variables for the intermediate;
Platis a steady state solution of the following discrete algebraic Riccati equation:
Figure BDA0001798386090000075
wherein
Figure BDA0001798386090000081
Is a control law gain matrix.
The invention has the following beneficial effects:
(1) according to the relative position and the absolute position information of the ship and the carrier aircraft, a landing instruction signal is calculated on line, a carrier aircraft gliding reference track is generated, and the carrier aircraft is controlled to track the reference track through a flight control system.
(2) Under the interference of deck movement, the interference is effectively inhibited through the feedforward control of the anticipating controller by utilizing a deck movement predicted value given by the deck movement predictor, and the track tracking error is reduced.
(3) The method has the advantages that the future information is utilized for feedforward control, the current information is utilized for feedback control, the control plane and the throttle of the carrier-based aircraft can be operated averagely in advance to achieve the purpose of tracking compensation, the instantaneous energy is reduced, the response speed is accelerated, and the carrier-based aircraft can be ensured to land on the aircraft carrier safely.
Drawings
Fig. 1 shows a functional block diagram of an automatic carrier landing control method of a carrier-based aircraft based on model reference adaptive control.
Fig. 2 shows a height trajectory tracking effect diagram in the carrier aircraft landing process.
Fig. 3 shows an effect diagram of height trajectory tracking error in the carrier aircraft landing process.
Fig. 4 shows a transverse lateral deviation correction effect diagram in the carrier aircraft landing process.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings.
The principle of the automatic landing control method of the shipboard aircraft based on the anticipation control is shown in fig. 1, when a ship runs on the sea, the actually measured sinking and floating height and swaying distance of the deck motion are predicted by a deck motion predictor based on particle filtering to obtain the predicted sinking and floating height and the predicted swaying distance of the deck motion in the next time period. The prediction information is input into a landing instruction and glide reference trajectory generation module to obtain corrected glide reference trajectory information. And finally, inputting the corrected reference track information into a flight controller based on anticipation control. The method can realize the tracking of the height and the lateral deviation of the track of the down sliding rail after the carrier-based aircraft enters the down sliding stage, and can realize the estimation and compensation of the deck motion at the beginning of 12.5 seconds before landing, thereby effectively inhibiting the track deviation caused by the deck motion, keeping the carrier-based aircraft stable, and ensuring the success rate and the safety of the automatic landing of the carrier-based aircraft.
Deck movement predictor
The input signal includes: actually measured sinking and floating height and swaying distance of deck movement. The output signal includes: and (4) estimating the sinking and floating height and the swaying distance of the deck movement. And (4) sending the actual deck movement to an estimation module to obtain deck movement estimation information, and introducing the deck movement estimation information into an automatic carrier landing control system of the carrier-based aircraft 12.5 seconds before carrier landing.
The calculation method of the deck motion estimation module comprises the following steps:
and for deck motion in a known transfer function expression form, designing a predictor by adopting a particle filtering method. Firstly, a state space equation of deck motion is obtained by a transfer function:
Figure BDA0001798386090000091
wherein x is a deck motion state variable; omega is system dynamic noise; z is an observed signal; v is the observed noise and is the noise,
Figure BDA0001798386090000092
the differential of the state variable of the deck movement is shown, A is a system matrix of the deck movement, B is an input matrix of the deck movement, and C is an output matrix of the deck movement.
Discretizing the above formula to obtain a discrete model of deck motion:
Figure BDA0001798386090000093
in the formula xkIs tkMoment of deck motion state quantity, xk-1Is tk-1Moment of deck motion state quantity, wk-1Is the dynamic noise of the system, vkTo observe noise,. phik,k-1For the state vector x from tk-1Time shift to tkThe transition matrix of the time of day,
Figure BDA0001798386090000097
wherein A is a system matrix of deck movements, gammak,k-1Is tk-1Noise vector w of time instantsk-1For tkState vector x of time of daykThe matrix of the noise coefficients of influence,
Figure BDA0001798386090000094
its variance matrix is Qk-1Where B is the input matrix for deck movements, zkIs tkObservation of deck movement at time HkIs an observation coefficient matrix with a variance matrix of Rk，TsIs the sampling time.
According to the discrete model (2), the deck motion estimation algorithm flow is designed based on particle filtering as follows:
(1) from a priori probability P (x)0) Generating a population of particles
Figure BDA0001798386090000095
The weight of all particles of the particle swarm is 1/N;
(2) and (3) prediction: from particles at time k-1
Figure BDA0001798386090000096
Obtaining predicted particles at the k moment by using a system state equation
Figure BDA0001798386090000101
(3) Updating the weight value: predicting particles based on observed vector machine at time k, using
Figure BDA0001798386090000102
Updating the weight of each particle; and further carrying out normalization processing on the weight:
Figure BDA0001798386090000103
wherein:
Figure BDA0001798386090000104
the weight of the jth particle at time k,
Figure BDA0001798386090000105
the weight of the jth particle at time k-1,
Figure BDA0001798386090000106
in order to be a priori at all,
Figure BDA0001798386090000107
to normalize the weight of the jth particle at time k after processing,
(4) resampling: according to
Figure BDA0001798386090000108
Weight of (2)
Figure BDA0001798386090000109
Resampling to obtain particles
Figure BDA00017983860900001010
And reset the weight
Figure BDA00017983860900001011
Are all 1/N;
(5) and (3) state estimation at the k moment:
Figure BDA00017983860900001012
and simultaneously, if k is equal to k +1, returning to the step (2) if k is smaller than the set threshold, and otherwise, exiting the loop.
(6) Deck movement information xkThe expression for the optimal estimate at the future time τ is
Figure BDA00017983860900001013
Where m is τ/TsAnd tau is the time of the future,
Figure BDA00017983860900001014
state transition matrix for the estimated value of the state at time k in (5)
Figure BDA00017983860900001015
Carrier landing instruction and gliding reference track generation module
The input signal includes: azimuth angle (psi) of the ship runway or glidepathS+λac) Wherein ψSIs ship azimuth angle, λacThe included angle of the oblique angle deck and the estimated sinking and floating height and swaying distance of the deck motion output by the deck predictor are adopted. The output signal comprises a three-dimensional gliding reference track signal XEATDc(t),YEATDc(t),ZEATDc(t) and a speed command signal Vc. And the gliding reference track signal and the speed command signal are output to a flight control module based on anticipation control.
First, the carrier-based aircraft captures the glidepath, knowing the initial glide height-ZEA0Angle of glide gammacVelocity of sliding down VcCalculating the landing time td
Figure BDA00017983860900001016
And length R of lower chuteA
Figure BDA00017983860900001017
Secondly, calculating a three-dimensional gliding reference track under a ground coordinate system with the ideal carrier landing point as an origin
Figure BDA0001798386090000111
Wherein: xEATDc(t) is the forward coordinate position of the carrier-based aircraft, YEATDc(t) is the lateral coordinate position of the carrier-based aircraft, ZEATDc(t) is the height coordinate position of the carrier-based aircraft, HEATDc(t) is the height value of the carrier-based aircraft, and Z is the height coordinate of the coordinate system which is positive downwardsEATDc(t)＝-HEATDc(t)；
And thirdly, superposing the estimated sinking and floating height and the swaying distance of the deck motion output by the deck predictor to obtain a corrected gliding reference track.
Flight controller based on predictive control
The input signal includes: four longitudinal state quantities of the carrier-based aircraft fed back by the sensor, namely flight speed V, attack angle alpha, pitch angle rate q and pitch angle theta; five transverse and lateral state quantities fed back by the sensor, namely a sideslip angle beta, a roll angle rate p, a yaw angle rate r, a roll angle phi and a yaw angle psi; speed instruction V output by carrier landing instruction and gliding reference track generation modulecCorrected glide reference track signal XEATDc(t),YEATDc(t),ZEATDc(t), and wake disturbances w.
The output signal includes: throttle opening deltaTAngle delta of elevatoreAileron declination angle deltaaRudder deflection angle deltar. And the signals are sent to an actuating mechanism so as to control the carrier-based aircraft to fly.
The specific process is as follows: firstly, longitudinal flight control laws are calculated, and secondly, transverse and lateral flight control laws are calculated.
Longitudinal flight control law:
firstly, discretizing a longitudinal state equation of the airplane to obtain a longitudinal discretization model of the airplane:
Δxlon(k+1)＝AlonΔxlon(k)+BwlonΔw(k)+BlonΔulon(k)
wherein A islonSystem matrix being longitudinal equation of state of aircraft, BwlonAs an interference matrix, BlonFor the input matrix, Δ xlon(k) Is [ Δ V (k) Δ α (k) Δ q (k) Δ θ (k)]T，Δulon(k) Is [ Delta delta ] ofT(k) Δδe(k)]TWhere Δ v (k) is a speed variation at time k, Δ α (k) is an attack angle variation at time k, Δ q (k) is a pitch angle rate variation at time k, Δ θ (k) is a pitch angle variation at time k, and Δ h (k) is an altitude variation at time k; delta deltaTDelta throttle change at time keAnd delta w (k) is the elevator variation at the moment k, and delta w (k) is the wake disturbance.
Adding error amount, known trajectory information and generalized output, and expanding the equation into the following form:
Figure BDA0001798386090000121
wherein Xlon(k)＝[Her(k) xlon(k) HR(k) vlon(k)]T,Her(k) Is the height error, x, corrected by adding deck motion prediction at time klon(k) Is the longitudinal state quantity of the airplane,
Figure BDA0001798386090000122
is a height difference vector, wherein
Figure BDA0001798386090000123
Is time k and k + MrlonDifference between predicted values of height of glidepath, H, added with deck movement prediction and corrected at momentr(k) Is a forecast value of the height of the lower slideway at the moment k after the prediction and correction of deck movement is added, MrlonIs a longitudinal look ahead step;
Figure BDA0001798386090000124
is time k +1 and k + MrlonAdding the difference between the predicted height values of the lower slideway after the deck motion prediction correction at the moment,
Figure BDA0001798386090000125
is k + MrlonTime and k + MrlonAdding the difference between the predicted height values of the lower slideway after the deck motion prediction correction at the moment,
Figure BDA0001798386090000126
is the integral of the error, vlon(0) To adjust the initial error value, ulon(k) For longitudinal control of the aircraft, Her(j) To add the corrected height error of deck motion prediction at time j,
Figure BDA0001798386090000127
Figure BDA0001798386090000128
W1lon,W2lon,W3lonis a regulating parameter, Z1lon(k),Z2lon(k),Z3lon(k) For three generalized outputs, GlonTo expand the output matrix of the equation, FlonTo expand the state matrix of the equation, HwlonAn interference matrix which is an extended equation;
the target is to find the control signal DeltaublonSo as to control the performance index JlonAnd (4) minimizing.
Figure BDA0001798386090000131
The demand-satisfying predictive controller can be obtained when the system satisfies the following two conditions:
1)(Flon Glon) Is stable;
2)W2lon'W2lon＞0；
the longitudinal control law module adopts a predictive control method to design a controller, and consists of a feedback control component and a feedforward control component. In the glide-slope tracking phase, the height error H is determinederAnd longitudinal state quantity error DeltaxlonFeeding a feedback control component to obtain ideal glide height information H known in advanceRAnd the interference signal w is sent into a feedforward control component, the two parts are combined to carry out predictive control, the height tracking of the lower slide is realized, and the control law is as follows:
Figure BDA0001798386090000132
wherein: u. ofblon(k) Control input quantity at time k, kelon，kxlon，kvlon，kwlon，
Figure BDA0001798386090000136
To control law gain, xlon *Is a state quantity at an equilibrium point, ulon *In order to control the amount at the balance point,Her(s) is the height error after the deck motion prediction correction is added at the moment s,
Figure BDA0001798386090000133
time k + i and k + MrlonAnd (k) adding the difference between the predicted values of the height of the lower slipway after the deck motion prediction correction at the moment, wherein w (k) is the wake disturbance of the ship.
Figure BDA0001798386090000137
kwlon＝-Rlon -1(GlonPlonHwlon),
Rlon＝W2lon TW2lon+Glon TPlonGlon
Wherein: rlonIntermediate calculation variables.
PlonIs a steady state solution of the following discrete algebraic Riccati equation:
Figure BDA0001798386090000134
wherein
Figure BDA0001798386090000135
Is a control law gain matrix;
lateral-lateral flight control law:
firstly, discretizing a transverse state equation of the airplane to obtain a transverse discretization model of the airplane:
Δxlat(k+1)＝AlatΔxlat(k)+BwlatΔw(k)+BlatΔulat(k)
wherein A islatSystem matrix being the transverse equation of state of the aircraft, BwlatAs an interference matrix, BlatFor the input matrix, Δ xlat(k) Is [ Delta beta (k) Delta p (k) Delta r (k) Delta phi (k) Delta psi (k) Delta y (k)]TΔ β (k) is the change in sideslip angle at time kAmount, Δ p (k) is the roll rate change at time k, Δ r (k) is the yaw rate change at time k, Δ φ (k) is the roll angle change at time k, Δ ψ (k) is the yaw angle change at time k, Δ y (k) is the yaw amount at time k, Δ u (k) is the yaw amount at time klon(k) Is [ Delta delta ] ofa(k) Δδr(k)]T，Δδa(k) Deltar(k) The rudder deflection variation at the moment k, and delta w (k) is the wake flow interference of the ship;
adding error amount, known trajectory information and generalized output, and expanding the equation into the following form:
Figure BDA0001798386090000141
wherein Xlat(k)＝[yer(k) xlat(k) yR(k) vlat(k)]T,yer(k) Is the lateral deviation, x, of the corrected deck motion prediction added at time klat(k) Is the transverse state quantity of the airplane,
Figure BDA0001798386090000142
is a lateral deviation vector, wherein
Figure BDA0001798386090000143
Is time k and k + MrlonDifference between the reference lateral deviation information, y, of the moment added to the deck motion prediction correctionr(k) Reference lateral deviation information at time k, M, corrected by adding deck motion predictionrlatIs a look-ahead step size in the lateral direction,
Figure BDA0001798386090000144
is time k +1 and k + MrlonThe difference between the reference lateral deviation information added with the deck motion forecast correction at the moment,
Figure BDA0001798386090000145
is k + MrlonTime and k + MrlonTemporal reference lateral deviation information with deck motion prediction correctionThe difference between the difference of the two phases,
Figure BDA0001798386090000146
is the integral of the lateral offset, vlat(0) Is the initial lateral deviation, ulat(k) Is the lateral control input;
Figure BDA0001798386090000151
Figure BDA0001798386090000152
W1lat,W2lat,W3latis an adjustment parameter; z1lat(k)，Z2lat(k)，Z3lat(k) For three generalized outputs, GlatTo expand the output matrix of the equation, FlatTo expand the state matrix of the equation, HwlatAn interference matrix which is an extended equation;
the target is to find the control signal DeltaublatSo as to control the performance index JlatAnd (4) minimizing.
Figure BDA0001798386090000153
The demand-satisfying predictive controller can be obtained when the system satisfies the following two conditions:
1)(Flat Glat) Is stable;
2)W2lat'W2lat＞0；
the lateral control law module adopts a predictive control method to design a controller, and consists of a feedback control component and a feedforward control component. In the glide-slope tracking phase, the lateral deviation y is measurederAnd state quantity error DeltaxblatFeeding a feedback control component to obtain a corrected reference lateral deviation y known in advanceRAnd the interference signal w is sent into a feedforward control component, the two parts are combined to carry out predictive control, the correction of the lateral deviation of the lower slide way is realized, and the control law is as follows:
Figure BDA0001798386090000154
wherein: u. ofblat(k) Control input quantity at time k, kelat，kxlat，kvlat，kwlat，kyr(i) To control law gain, xlat *Is a state quantity at an equilibrium point, ulat *As a control quantity at the balance point, yer(s) adding the corrected lateral deviation error of the deck motion prediction for the moment s,
Figure BDA0001798386090000155
time k + i and k + MrlonAnd (k) adding the lateral deviation difference after the deck motion prediction correction at the moment, wherein w (k) is the wake disturbance.
Figure BDA0001798386090000164
kwlat＝-Rlat -1(GlatPlatHwlat),
Rlat＝W2lat TW2lat+Glat TPlatGlat,
Wherein: rlatIntermediate calculation variables.
PlatIs a steady state solution of the following discrete algebraic Riccati equation:
Figure BDA0001798386090000161
wherein
Figure BDA0001798386090000165
Is a control law gain matrix.
In order to verify the automatic carrier landing control method of the carrier-based aircraft provided by the invention, taking a dynamics and kinematics model of a certain unmanned aerial vehicle as an example, deck motion compensation is added in the last 12.5 seconds of a reference track, and main simulation parameters are set as follows:
Figure BDA0001798386090000163
the numerical simulation verification under the MATLAB software platform shows that the automatic carrier landing control method of the carrier-based aircraft can enable the carrier-based aircraft to track the glide reference track with high precision, so that the carrier landing task is successfully completed.
Fig. 2 is a comparison graph of the height trajectory tracking effect of the predictive control and the PID control, and fig. 3 is a comparison graph of the height trajectory tracking error of the predictive control and the PID control, and it can be seen from these two graphs that the response time of the predictive control is faster, the tracking accuracy is higher, and the compensation effect on the deck movement is better compared with the PID control.
Fig. 4 is a comparison graph of lateral offset correction of the predictive control and the PID control, and it can be seen that the predictive control can accurately correct the lateral offset after about 15 seconds, while the PID control always has the lateral offset. Compared with PID control, the predictive control tracking precision is high, and the control effect is better.

Claims (1)

1.一种基于预见控制的舰载机自动着舰控制方法，其特征在于，包括如下步骤：1. a carrier-based aircraft automatic landing control method based on foreseeing control, is characterized in that, comprises the steps: (1)甲板运动预测值计算(1) Calculation of the predicted value of deck motion 有甲板运动的离散模型：Discrete model with deck motion:
Figure FDA0002933469220000011
Figure FDA0002933469220000011
 式中xk为tk时刻的甲板运动状态量，xk-1为tk-1时刻的甲板运动状态量，vk为观测噪声，Φk,k-1为状态向量x从tk-1时刻转移到tk时刻的转移矩阵，
Figure FDA0002933469220000012
A为甲板运动的系统矩阵，Ts为采样时间，Γk,k-1为tk-1时刻的噪声向量wk-1对tk时刻的状态向量xk影响的噪声系数矩阵，
Figure FDA0002933469220000013
其方差阵为Qk-1，B为甲板运动的输入矩阵，wk-1为系统动态噪声，zk为tk时刻的甲板运动观测量，Hk为观测系数矩阵，其方差阵为Rk；where x k is the deck motion state quantity at time t k , x k-1 is the deck motion state quantity at time t k-1 , v k is the observation noise, Φ k,k-1 is the state vector x from t k- The transition matrix from time 1 to time t k ,
Figure FDA0002933469220000012
A is the system matrix of deck motion, T s is the sampling time, Γ k,k-1 is the noise coefficient matrix of the influence of the noise vector w k-1 at time t k -1 on the state vector x k at time t k ,
Figure FDA0002933469220000013
The variance matrix is Q k-1 , B is the input matrix of deck motion, w k-1 is the system dynamic noise, z k is the observed amount of deck motion at time t k , H k is the observation coefficient matrix, and its variance matrix is R k ; 根据离散模型(1)，基于粒子滤波设计甲板运动预估方法流程如下：According to the discrete model (1), the process of designing the deck motion prediction method based on particle filter is as follows: 1)由先验概率P(x0)产生粒子群
Figure FDA0002933469220000014
粒子群的所有粒子权值为1/N；1) Generate the particle swarm from the prior probability P(x 0 )
Figure FDA0002933469220000014
The weight of all particles in the particle swarm is 1/N; 2)预测：由k-1时刻的粒子
Figure FDA0002933469220000015
利用系统状态方程，得到k时刻的预测粒子
Figure FDA0002933469220000016
2) Prediction: by the particle at time k-1
Figure FDA0002933469220000015
Using the equation of state of the system, the predicted particle at time k is obtained
Figure FDA0002933469220000016
 3)权值更新：根据k时刻的观测向量机预测粒子，利用
Figure FDA0002933469220000017
更新每个粒子的权值；进而对权值进行归一化处理：
Figure FDA0002933469220000018
其中：
Figure FDA0002933469220000019
为k时刻第j个粒子的权值，
Figure FDA00029334692200000110
为k-1时刻第j个粒子的权值，
Figure FDA00029334692200000111
为先验概率，
Figure FDA00029334692200000112
为归一化处理后k时刻第j个粒子的权值；3) Weight update: According to the observation vector machine at time k to predict the particle, use
Figure FDA0002933469220000017
Update the weights of each particle; then normalize the weights:
Figure FDA0002933469220000018
in:
Figure FDA0002933469220000019
is the weight of the jth particle at time k,
Figure FDA00029334692200000110
is the weight of the jth particle at time k-1,
Figure FDA00029334692200000111
is the prior probability,
Figure FDA00029334692200000112
is the weight of the jth particle at time k after normalization; 4)重采样：根据
Figure FDA00029334692200000113
的权值
Figure FDA00029334692200000114
重新采样得到粒子
Figure FDA00029334692200000115
并重置权值
Figure FDA00029334692200000116
均为1/N；4) Resampling: according to
Figure FDA00029334692200000113
weight of
Figure FDA00029334692200000114
Resample to get particles
Figure FDA00029334692200000115
and reset the weights
Figure FDA00029334692200000116
Both are 1/N; 5)k时刻状态估计：
Figure FDA0002933469220000021
同时令k＝k+1，若k小于设定阈值，返回到第2)步，否则到第6)步；5) State estimation at time k:
Figure FDA0002933469220000021
At the same time, make k=k+1, if k is less than the set threshold, return to the 2) step, otherwise go to the 6) step; 6)甲板运动信息xk在未来τ时刻的最优估计值的表达式为
Figure FDA0002933469220000022
其中m＝τ/Ts，τ为未来时刻，
Figure FDA0002933469220000023
为k时刻状态估计值，状态转移阵
Figure FDA0002933469220000024
6) The expression of the optimal estimated value of deck motion information x k at time τ in the future is:
Figure FDA0002933469220000022
where m=τ/T s , τ is the future time,
Figure FDA0002933469220000023
is the estimated value of the state at time k, the state transition matrix
Figure FDA0002933469220000024
 (2)计算修正后的舰载机下滑基准轨迹(2) Calculate the corrected carrier-based aircraft glide reference trajectory 第一步，舰载机捕获下滑道，已知初始下滑高度-ZEA0、下滑角γc、下滑速度Vc，计算着舰时间td In the first step, the carrier-based aircraft captures the glide path, the initial glide height -Z EA0 , the glide angle γ c , and the glide speed V c are known, and the landing time t d is calculated.
Figure FDA0002933469220000025
Figure FDA0002933469220000025
 和下滑道长度RA and glide path length R A
Figure FDA0002933469220000026
Figure FDA0002933469220000026
 第二步，计算以理想着舰点为原点的地面坐标系下的三维下滑基准轨迹The second step is to calculate the three-dimensional glide reference trajectory in the ground coordinate system with the ideal landing point as the origin
Figure FDA0002933469220000027
Figure FDA0002933469220000027
 其中：t为时间，td为着舰时间，XEATDc(t)为舰载机的前向坐标位置，YEATDc(t)为舰载机的侧向坐标位置，ZEATDc(t)为舰载机的高度坐标位置，HEATDc(t)为舰载机的高度值，因为坐标系的高度坐标向下为正，所以ZEATDc(t)＝-HEATDc(t)，(ψS+λac)为舰船跑道或下滑道的方位角，其中ψS为舰船方位角，λac为斜角甲板夹角；Among them: t is the time, t d is the landing time, X EATDc (t) is the forward coordinate position of the carrier aircraft, Y EATDc (t) is the lateral coordinate position of the carrier aircraft, and Z EATDc (t) is the ship The height coordinate position of the carrier aircraft, H EATDc (t) is the height value of the carrier aircraft, because the height coordinate of the coordinate system is positive downward, so Z EATDc (t)=-H EATDc (t), (ψ S +λ ac ) is the azimuth angle of the ship’s runway or glide path, where ψ S is the ship’s azimuth angle, and λ ac is the angle between the inclined deck; 第三步，叠加甲板预估器输出的预估的甲板运动的沉浮高度与横荡距离，得到修正后的舰载机下滑基准轨迹；The third step is to superimpose the estimated deck movement height and sway distance output by the deck predictor to obtain the corrected carrier-based aircraft gliding reference trajectory; (3)计算纵向飞行控制律和横侧向飞行控制律(3) Calculate the longitudinal flight control law and the lateral flight control law 第一步计算纵向飞行控制律，已知飞机的纵向离散化模型：The first step is to calculate the longitudinal flight control law, and the longitudinal discretization model of the aircraft is known: Δxlon(k+1)＝AlonΔxlon(k)+BwlonΔwlon(k)+BlonΔulon(k)Δx lon (k+1)=A lon Δx lon (k)+B wlon Δw lon (k)+B lon Δu lon (k) 其中，Alon为飞机的纵向状态方程的系统矩阵，Bwlon为干扰矩阵，Blon为输入矩阵，Δxlon(k)为[ΔV(k) Δα(k) Δq(k) Δθ(k) ΔH(k)]T，ΔV(k)为k时刻的速度变化量，Δα(k)为k时刻的迎角变化量，Δq(k)为k时刻的俯仰角速率变化量，Δθ(k)为k时刻的俯仰角变化量，ΔH(k)为k时刻的高度变化量；Δulon(k)为[ΔδT(k) Δδe(k)]T，ΔδT为k时刻的油门变化量，Δδe为k时刻的升降舵变化量，Δwlon(k)为纵向舰尾流干扰变化量；Among them, A lon is the system matrix of the longitudinal state equation of the aircraft, B wlon is the interference matrix, B lon is the input matrix, and Δx lon (k) is [ΔV(k) Δα(k) Δq(k) Δθ(k) ΔH (k)] T , ΔV(k) is the velocity change at time k, Δα(k) is the angle of attack change at time k, Δq(k) is the pitch rate change at time k, and Δθ(k) is The pitch angle change at time k, ΔH(k) is the height change at time k; Δu lon (k) is [Δδ T (k) Δδ e (k)] T , Δδ T is the throttle change at time k, Δδ e is the elevator variation at time k, Δw lon (k) is the longitudinal ship wake disturbance variation; 加入误差量、已知轨迹信息和广义输出，将方程扩展成以下形式：Adding in the amount of error, known trajectory information, and generalized output, the equation is expanded into the following form:
Figure FDA0002933469220000031
Figure FDA0002933469220000031
 其中Xlon(k)＝[Her(k) xlon(k) HR(k) vlon(k)]T,Her(k)是k时刻加入甲板运动预测修正后的高度误差,xlon(k)为飞机纵向状态量，
Figure FDA0002933469220000032
是高度差向量，其中
Figure FDA0002933469220000033
为k时刻与k+Mrlon时刻的加入甲板运动预测修正后的下滑道高度预见值之差，Hr(k)是k时刻的加入甲板运动预测修正后的下滑道高度预见值，Mrlon是纵向的预见步长；
Figure FDA0002933469220000034
为k+1时刻与k+Mrlon时刻的加入甲板运动预测修正后的下滑道高度预见值之差，
Figure FDA0002933469220000035
为k+Mrlon时刻与k+Mrlon时刻的加入甲板运动预测修正后的下滑道高度预见值之差，
Figure FDA0002933469220000036
是误差积分，vlon(0)为可调初始误差值，wlon(k)是k时刻飞机受到的纵向舰尾流干扰，ulon(k)为飞机的纵向控制量，Her(j)为是j时刻加入甲板运动预测修正后的高度误差，
Figure FDA0002933469220000037
Figure FDA0002933469220000038
W1lon,W2lon,W3lon是调节参数,Z1lon(k),Z2lon(k),Z3lon(k)为三个广义输出，Glon为扩展方程的输出矩阵，Flon为扩展方程的状态矩阵，Hwlon为扩展方程的干扰矩阵，Alon为飞机的纵向状态方程的系统矩阵，Bwlon为干扰矩阵，Blon为输入矩阵；where X lon (k)=[H er (k) x lon (k) H R (k) v lon (k)] T , Her (k) is the height error after adding the deck motion prediction correction at time k, x lon (k) is the longitudinal state quantity of the aircraft,
Figure FDA0002933469220000032
is the height difference vector, where
Figure FDA0002933469220000033
is the difference between the predicted value of the glide path height after adding the deck motion prediction and correction at the time k and k+M rlon , H r (k) is the predicted value of the glide path height after adding the deck motion prediction and correction at the time k, and M rlon is Longitudinal look-ahead step;
Figure FDA0002933469220000034
is the difference between the predicted value of the glide path height after adding the deck motion prediction and correction at the time k+1 and the time k+M rlon ,
Figure FDA0002933469220000035
is the difference between the predicted value of the glide path height after adding the deck motion prediction and correction at the k+M rlon time and k+M rlon time,
Figure FDA0002933469220000036
is the error integral, v lon (0) is the adjustable initial error value, w lon (k) is the longitudinal ship wake disturbance of the aircraft at time k, u lon (k) is the longitudinal control amount of the aircraft, Her (j) is the height error after adding the deck motion prediction and correction at time j,
Figure FDA0002933469220000037
Figure FDA0002933469220000038
W 1lon , W 2lon , W 3lon are adjustment parameters, Z 1lon (k), Z 2lon (k), Z 3lon (k) are three generalized outputs, G lon is the output matrix of the extended equation, and F lon is the output matrix of the extended equation State matrix, H wlon is the interference matrix of the extended equation, A lon is the system matrix of the longitudinal state equation of the aircraft, B wlon is the interference matrix, and B lon is the input matrix; 寻找控制信号Δublon，使得控制性能指标Jlon最小化，Find the control signal Δu blon to minimize the control performance index J lon ,
Figure FDA0002933469220000041
Figure FDA0002933469220000041
 所得控制律为：The resulting control law is:
Figure FDA0002933469220000042
Figure FDA0002933469220000042
 其中：ublon(k)为k时刻的控制输入量，kelon，kxlon，kvlon，kwlon，
Figure FDA0002933469220000049
为控制律增益，Mrlon是纵向的预见步长，xlon *为平衡点处的状态量，ulon *为平衡点处的控制量，Her(s)为是s时刻加入甲板运动预测修正后的高度误差，
Figure FDA0002933469220000043
为k+i时刻与k+Mrlon时刻的加入甲板运动预测修正后的下滑道高度预见值之差，wlon(k)为k时刻的纵向舰尾流干扰；Among them: u blon (k) is the control input at time k, k elon , k xlon , k vlon , k wlon ,
Figure FDA0002933469220000049
For the gain of the control law, M rlon is the longitudinal prediction step length, x lon * is the state quantity at the equilibrium point, u lon * is the control quantity at the equilibrium point, Her (s) is the addition of deck motion prediction correction at s time After the height error,
Figure FDA0002933469220000043
is the difference between the predicted value of the glide path height after adding the deck motion prediction and correction at time k+i and time k+M rlon , w lon (k) is the longitudinal ship wake disturbance at time k;
Figure FDA0002933469220000044
Figure FDA0002933469220000044
 kwlon＝-Rlon -1(GlonPlonHwlon),k wlon = -R lon -1 (G lon P lon H wlon ), Rlon＝W2lon TW2lon+Glon TPlonGlon R lon =W 2lon T W 2lon +G lon T P lon G lon 其中：
Figure FDA0002933469220000045
Rlon为中间计算变量；in:
Figure FDA0002933469220000045
R lon is an intermediate calculation variable; Plon是下面的离散代数Riccati方程的稳态解：P lon is the steady-state solution of the following discrete algebraic Riccati equation:
Figure FDA0002933469220000046
Figure FDA0002933469220000046
 其中
Figure FDA0002933469220000047
Figure FDA0002933469220000048
为控制律增益矩阵；in
Figure FDA0002933469220000047
Figure FDA0002933469220000048
is the control law gain matrix; 第二步计算横侧向飞行控制律，已知飞机的横侧向离散化模型：The second step is to calculate the lateral flight control law, and the lateral and lateral discretization model of the aircraft is known: Δxlat(k+1)＝AlatΔxlat(k)+BwlatΔwlat(k)+BlatΔulat(k)Δx lat (k+1)=A lat Δx lat (k)+B wlat Δw lat (k)+B lat Δu lat (k) 其中，Alat为飞机的横侧向状态方程的系统矩阵，Bwlat为干扰矩阵，Blat为输入矩阵，Δxlat(k)为[Δβ(k) Δp(k) Δr(k) Δφ(k) Δψ(k) Δy(k)]T，Δβ(k)为k时刻的侧滑角变化量、Δp(k)为k时刻的滚转角速率变化量、Δr(k)为k时刻的偏航角速率变化量、Δφ(k)为k时刻的滚转角变化量、Δψ(k)为k时刻的偏航角变化量，Δy(k)为k时刻的侧偏量，Δulat(k)为[Δδa(k) Δδr(k)]T，Δδa(k)为k时刻的副翼偏转变化量，Δδr(k)为k时刻的方向舵偏转变化量，Δwlat(k)为k时刻的横侧向舰尾流干扰变化量；Among them, A lat is the system matrix of the lateral state equation of the aircraft, B wlat is the interference matrix, B lat is the input matrix, and Δx lat (k) is [Δβ(k) Δp(k) Δr(k) Δφ(k ) Δψ(k) Δy(k)] T , Δβ(k) is the change in sideslip angle at time k, Δp(k) is the change in roll angle rate at time k, and Δr(k) is the yaw at time k Angular rate change, Δφ(k) is the roll angle change at time k, Δψ(k) is the yaw angle change at time k, Δy(k) is the side deflection at time k, Δu lat (k) is [Δδ a (k) Δδ r (k)] T , Δδ a (k) is the change in aileron deflection at time k, Δδ r (k) is the change in rudder deflection at time k, and Δw lat (k) is k The variation of lateral and lateral ship wake disturbance at time; 加入误差量、已知轨迹信息和广义输出，将方程扩展成以下形式：Adding in the amount of error, known trajectory information, and generalized output, the equation is expanded into the following form:
Figure FDA0002933469220000051
Figure FDA0002933469220000051
 其中Xlat(k)＝[yer(k) xlat(k) yR(k) vlat(k)]T,yer(k)是k时刻加入甲板运动预测修正后的横向偏差,xlat(k)是飞机横侧向状态量，
Figure FDA0002933469220000052
是侧偏差向量，其中
Figure FDA0002933469220000053
为k时刻与k+Mrlon时刻的加入甲板运动预测修正后的参考横向偏差信息之差，yr(k)是k时刻的加入甲板运动预测修正后的参考横向偏差信息，Mrlat是横侧向的预见步长，
Figure FDA0002933469220000054
为k+1时刻与k+Mrlon时刻的加入甲板运动预测修正后的参考横向偏差信息之差，
Figure FDA0002933469220000055
为k+Mrlon时刻与k+Mrlon时刻的加入甲板运动预测修正后的参考横向偏差信息之差，
Figure FDA0002933469220000056
是侧偏积分，vlat(0)是初始侧偏，wlat(k)为k时刻的横侧向舰尾流干扰，ulat(k)是k时刻横侧向控制输入量；where X lat (k)=[y er (k) x lat (k) y R (k) v lat (k)] T , y er (k) is the lateral deviation after adding the deck motion prediction correction at time k, x lat (k) is the lateral state quantity of the aircraft,
Figure FDA0002933469220000052
is the lateral deviation vector, where
Figure FDA0002933469220000053
is the difference between the reference lateral deviation information after adding deck motion prediction and correction at time k and k+M rlon time, y r (k) is the reference lateral deviation information after adding deck motion prediction and correction at time k, M rlat is the lateral side the predicted step size in the direction,
Figure FDA0002933469220000054
is the difference between the reference lateral deviation information after adding deck motion prediction and correction at time k+1 and time k+M rlon ,
Figure FDA0002933469220000055
is the difference between the k+M rlon time and the reference lateral deviation information after adding the deck motion prediction and correction at the k+M rlon time,
Figure FDA0002933469220000056
is the side deflection integral, v lat (0) is the initial side deflection, w lat (k) is the lateral ship wake disturbance at time k, and u lat (k) is the lateral control input at time k;
Figure FDA0002933469220000057
Figure FDA0002933469220000058
W1lat,W2lat,W3lat是调节参数；Z1lat(k)，Z2lat(k)，Z3lat(k)为三个广义输出，Glat为扩展方程的输出矩阵，Flat为扩展方程的状态矩阵，Hwlat为扩展方程的干扰矩阵，Alat为飞机的横侧向状态方程的系统矩阵，Bwlat为干扰矩阵，Blat为输入矩阵；
Figure FDA0002933469220000057
Figure FDA0002933469220000058
W 1lat , W 2lat , W 3lat are adjustment parameters; Z 1lat (k), Z 2lat (k), Z 3lat (k) are three generalized outputs, G lat is the output matrix of the extended equation, and Flat is the output matrix of the extended equation state matrix, H wlat is the interference matrix of the extended equation, A lat is the system matrix of the lateral state equation of the aircraft, B wlat is the interference matrix, and B lat is the input matrix; 寻找控制信号Δublat，使得控制性能指标Jlat最小化Find the control signal Δu blat to minimize the control performance index J lat
Figure FDA0002933469220000061
Figure FDA0002933469220000061
 所得控制律为：The resulting control law is:
Figure FDA0002933469220000062
Figure FDA0002933469220000062
 其中：ublat(k)为k时刻的控制输入量，kelat，kxlat，kvlat，kwlat，
Figure FDA0002933469220000063
为控制律增益，Mrlat是横侧向的预见步长，xlat *为平衡点处的状态量，ulat *为平衡点处的控制量，yer(s)为s时刻加入甲板运动预测修正后的侧偏误差，
Figure FDA0002933469220000064
为k+i时刻与k+Mrlon时刻的加入甲板运动预测修正后的侧偏之差，w(k)为舰尾流干扰，Among them: u blat (k) is the control input at time k, k elat , k xlat , k vlat , k wlat ,
Figure FDA0002933469220000063
is the control law gain, M rlat is the lateral prediction step size, x lat * is the state quantity at the equilibrium point, u lat * is the control quantity at the equilibrium point, and y er (s) is the prediction of the deck motion added at time s Corrected yaw error,
Figure FDA0002933469220000064
is the difference between k+i time and k+M rlon time after adding deck motion prediction and correction, w(k) is the ship wake disturbance,
Figure FDA0002933469220000065
Figure FDA0002933469220000065
 kwlat＝-Rlat -1(GlatPlatHwlat),k wlat = -R lat -1 (G lat P lat H wlat ), Rlat＝W2lat TW2lat+Glat TPlatGlat，R lat =W 2lat T W 2lat +G lat T P lat G lat , 其中：
Figure FDA0002933469220000066
Rlat为中间计算变量；in:
Figure FDA0002933469220000066
R lat is an intermediate calculation variable; Plat是下面的离散代数Riccati方程的稳态解： Plat is the steady-state solution of the following discrete algebraic Riccati equation:
Figure FDA0002933469220000067
Figure FDA0002933469220000067
 其中
Figure FDA0002933469220000068
为控制律增益矩阵。in
Figure FDA0002933469220000068
is the control law gain matrix.
CN201811066281.5A 2018-09-13 2018-09-13 An automatic landing control method for carrier-based aircraft based on predictive control Active CN109213203B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811066281.5A CN109213203B (en) 2018-09-13 2018-09-13 An automatic landing control method for carrier-based aircraft based on predictive control

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811066281.5A CN109213203B (en) 2018-09-13 2018-09-13 An automatic landing control method for carrier-based aircraft based on predictive control

Publications (2)

Publication Number Publication Date
CN109213203A CN109213203A (en) 2019-01-15
CN109213203B true CN109213203B (en) 2021-06-15

Family

ID=64983599

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811066281.5A Active CN109213203B (en) 2018-09-13 2018-09-13 An automatic landing control method for carrier-based aircraft based on predictive control

Country Status (1)

Country Link
CN (1) CN109213203B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110703799B (en) * 2019-10-28 2021-09-24 大连理工大学 Multi-carrier aircraft cooperative deck taxi trajectory planning method based on centralized optimal control
CN111813133B (en) * 2020-07-07 2022-08-05 南京航空航天大学 An autonomous landing method for UAV ships based on relatively precise single-point positioning
CN112650262A (en) * 2020-12-11 2021-04-13 南京航空航天大学 Automatic carrier-based aircraft carrier landing system based on robust predictive control under condition of uncertain parameters
CN112947121B (en) * 2021-03-10 2024-03-19 中国人民解放军92728部队 Semi-physical simulation system and method for carrier landing control of carrier-borne aircraft
CN115755788B (en) * 2022-11-02 2024-07-05 辽宁石油化工大学 Robust asynchronous predictive tracking control method for low-delay multi-stage batch process
CN117592309B (en) * 2024-01-17 2024-04-26 中国人民解放军海军航空大学 Chained landing command method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104803005A (en) * 2015-05-13 2015-07-29 南京航空航天大学 Automatic carrier landing composite control method of carrier airflow compensation-containing carrier-borne aircraft
CN107741229A (en) * 2017-10-10 2018-02-27 北京航空航天大学 A combined photoelectric/radar/inertial guidance method for carrier aircraft landing
CN107831654A (en) * 2017-10-16 2018-03-23 湘潭大学 The control method of the guaranteed cost prediction repetitive controller of uncertainty linear Discrete-time system
CN107957686A (en) * 2017-11-24 2018-04-24 南京航空航天大学 Unmanned helicopter auto landing on deck control system based on prediction control

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9547991B2 (en) * 2013-05-23 2017-01-17 Honeywell International Inc. Aircraft precision approach and shipboard landing control system and method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104803005A (en) * 2015-05-13 2015-07-29 南京航空航天大学 Automatic carrier landing composite control method of carrier airflow compensation-containing carrier-borne aircraft
CN107741229A (en) * 2017-10-10 2018-02-27 北京航空航天大学 A combined photoelectric/radar/inertial guidance method for carrier aircraft landing
CN107831654A (en) * 2017-10-16 2018-03-23 湘潭大学 The control method of the guaranteed cost prediction repetitive controller of uncertainty linear Discrete-time system
CN107957686A (en) * 2017-11-24 2018-04-24 南京航空航天大学 Unmanned helicopter auto landing on deck control system based on prediction control

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
基于H2预见控制的舰载机自动着舰控制方法;邵敏敏等;《电光与控制》;20150930;第22卷(第9期);第68-71,109页 *
基于预见控制的甲板运动跟踪控制研究;马坤等;《电光与控制》;20171130;第24卷(第11期);第74-77,99页 *
基于鲁棒预见控制的自动着舰控制研究;李蒙;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20180315;第C031-290页 *

Also Published As

Publication number Publication date
CN109213203A (en) 2019-01-15

Similar Documents

Publication Publication Date Title
CN109213203B (en) An automatic landing control method for carrier-based aircraft based on predictive control
Zhen et al. Automatic carrier landing control for unmanned aerial vehicles based on preview control and particle filtering
Zhen et al. Preview control and particle filtering for automatic carrier landing
Yu et al. ELOS-based path following control for underactuated surface vehicles with actuator dynamics
Zhen et al. Adaptive super-twisting control for automatic carrier landing of aircraft
CN105259908B (en) A kind of radar vectoring unmanned plane auto landing on deck guidance and control system and its control method
CN104880946B (en) A kind of carrier-borne aircraft auto landing on deck control method based on robust preview control
CN107807663B (en) Formation keeping control method for unmanned aerial vehicles based on adaptive control
CN105955268B (en) A kind of UUV moving-target sliding mode tracking control methods considering Local obstacle avoidance
CN106292293B (en) A kind of adaptive auto landing on deck guidance control system of the unmanned carrier-borne aircraft of fixed-wing
Wang et al. Extended state observer-based fixed-time trajectory tracking control of autonomous surface vessels with uncertainties and output constraints
CN106292294B (en) Shipborne UAV auto landing on deck control device based on model reference self-adapting control
Lungu et al. Application of H2/H∞ and dynamic inversion techniques to aircraft landing control
CN107844123A (en) A kind of Nonlinear Flight device flight tracking control method
CN106502255B (en) Design method and control method of a carrier-based aircraft automatic landing control system
CN107957686B (en) Automatic landing control system for unmanned helicopter based on predictive control
CN109782785A (en) Aircraft automatic landing control method based on direct force control
Liu et al. A hierarchical disturbance rejection depth tracking control of underactuated AUV with experimental verification
CN112731915A (en) Direct track control method for optimizing NMPC algorithm based on convolutional neural network
Lungu et al. Inverse optimal control for autonomous carrier landing with disturbances
Zhang et al. Time-optimal path planning and tracking based on nonlinear model predictive control and its application on automatic berthing
CN110262513A (en) A Design Method for Trajectory Tracking Control Structure of Marine Robot
Wang et al. Robust adaptive three-dimensional trajectory tracking control for unmanned underwater vehicles with disturbances and uncertain dynamics
Zhang et al. Research on autonomous berthing control of MASS based on real time kinematic
Sun et al. Collision avoidance guidance and control scheme for vector propulsion unmanned surface vehicle with disturbance

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