• <tr id="yyy80"></tr>
  • <sup id="yyy80"></sup>
  • <tfoot id="yyy80"><noscript id="yyy80"></noscript></tfoot>
  • 99热精品在线国产_美女午夜性视频免费_国产精品国产高清国产av_av欧美777_自拍偷自拍亚洲精品老妇_亚洲熟女精品中文字幕_www日本黄色视频网_国产精品野战在线观看 ?

    Continuous Sliding Mode Controller with Disturbance Observer for Hypersonic Vehicles

    2015-08-11 11:56:51ChaoxuMuQunZongBailingTianandWeiXu
    IEEE/CAA Journal of Automatica Sinica 2015年1期

    Chaoxu Mu,Qun Zong,Bailing Tian,and Wei Xu

    Continuous Sliding Mode Controller with Disturbance Observer for Hypersonic Vehicles

    Chaoxu Mu,Qun Zong,Bailing Tian,and Wei Xu

    —In this paper,a continuous sliding mode controller with disturbance observer is proposed for the tracking control of hypersonic vehicles to suppress the chattering.The finite time disturbance observer is involved to make that the continuous sliding mode controller has the property of disturbance rejection. Due to continuous terms replacing the discontinuous term of traditionalsliding mode control,switching modes of velocity and altitude firstly arrive at smallregions with respect to disturbance observation errors.Switching modes keep zero and velocity and altitude asymptotically converge to their reference commands after disturbance observation errors disappear.Simulation results have proved the proposed method can guarantee the tracking of velocity and altitude with continuous sliding mode control laws, and also has the fast convergence rate and robustness.

    Index Terms—Sliding mode control,finite time disturbance observer,chattering suppression,robustness,hypersonic vehicles.

    I.INTRODUCTION

    R ECENTLY,the research of air-breathing hypersonic vehicles(AHVs)has been an interesting issue with great practical value.It is considered as a promising transportation equipment for access to space.However,it is a very challenging problem to design the controller for AHVs as unmanageable nonlinear dynamics including the integrated airframe-propulsion,the uncertainty of atmospheric conditions,the variation of physical and aerodynamic parameters and so on[1?4].

    Due to complex dynamics of AHVs,it is difficult to obtain the real physical mode.Therefore,most research work is based on significative simplified models.For the tracking controlof AHVs in cruise phase,itmainly focuses on velocity and attitude control,where main concerns are longitudinal dynamics of AHVs and lateral movements are allowed for further curtailment.Effective controllers have been developed based on linear models at fixed points and nonlinear models in previous work.Gregory et al.used H∞andμsynthesisto a uncertain linear AHV model[5].Lind proposed the linear parameter-varying method to model and control flexible aircrafts[6].Oppenheimer etal.developed a modified dynamic inversion controller for the linear time-invariant,unstable, non-minimum phase model of AHVs[7].Sigthorsson et al. proposed robust linear output feedback control for AHVs. The control methods based on linear models offer simple and efficient ways to locally stabilize dynamical processes of AHVs[3].Nonlinear models are generally more accurate approximation of AHVs'physical models.Control strategies based on nonlinear models have been proposed for the study of flight control,including adaptive control,robust control, H∞control,sliding mode control,fuzzy logic control.Gibson et al.adopted the adaptive control for thrust and actuator uncertainties of AHVs[8].Serrani et al.introduced integrated adaptive guidance and controlfor constrained nonlinear AHV models[9].Wang et al.proposed the stochastic robust flight control based on dynamical inversion[10].Xu et al.provided the adaptive sliding mode control for the cruising control of the AHV with parameter uncertainties[11].Bowcutt adopted the multidisciplinary optimization control for AHVs[12],and intelligentcontrolalgorithms were used by Jiang etal.[13?14]. These nonlinear controllers on nonlinear vehicle models improve the controlperformance of AHVs from differentaspects.

    Sliding mode control(SMC)is one of well-known nonsmooth methods,which provides an effective and systematic approach to maintain the consistentstability with good robustness.Therefore,the method is widely accepted in the field of flight control for its robustness.Xu et al.have designed adaptive sliding mode controller for rigid and flexible AHVs, which has shown good insensitivity to uncertainties[11,15].In general,non-smooth control laws can improve robustness[16], whereas smooth control laws may not improve disturbance rejection due to Lipschitz continuity of closed loop systems. In sliding mode control,robustness is derived from the discontinuous sign function,which simultaneously leads to the fatal chattering of the traditional sliding mode control.In the practical implementation,actuators can not bear such high frequency switchings.One famous solution to obtain continuous controllers is to introduce boundary layers around sliding mode surfaces proposed by Slotine etal.[17?18],which provides an asymptotic stability to a preestablished fixed region of the origin.

    In this paper,the continuous sliding mode controller is designed to solve the tracking problem of AHVs,which involves continuous terms instead of sign functions in the traditional sliding mode control.In order to keep good robustness,the finite time observer is used to reject disturbances.Before observation errors disappear,system states arrive at small regions around sliding mode surfaces,whose scopes are rel-ated to observation errors.When observation errors converge to zero,system states reach sliding mode surfaces and then move to the origin.It means system states can asymptotically converge to the origin with continuous control laws even disturbances exist.Whereas,the boundary layer method can only guarantee the convergence to a fixed neighborhood ofthe origin.

    This paper is organized as follows.In Section II,the preliminary system description and the control-oriented modelare provided.In Section III,main results are stated and continuous control laws for velocity and altitude tracking of AHVs are designed.Simulation results are presented in Section IV to illustrate that the proposed method is effective and robust. Conclusion is given in the last section.

    II.PROBLEM FORMULATION

    A.The Longitudinal Model of Hypersonic Vehicles

    One of the most popular nonlinear models is reported inmost papers[10,19?23],which is the longitudinal model of a winged-cone AHV.Bolender et al.proposed to attach elastic characteristics of AHVs to the above longitudinal model for the study of reentry flight[24].In this paper,the classical nonlinear modelderived from NASA Langley Research Center is studied at the trim cruise condition[25].The flexible mode and the coupling between longitudinal and lateral dynamics are curtailed.

    The control-oriented model for longitudinal dynamics are described by five first order differential equations:

    where v,γ,h,αand q represent velocity,flight-path angle, altitude,angle of attack and pitch rate,R=h+REis the altitude of vehicle,REis the radius of the Earth.Iyyis moment of inertia.Left L and draft D are expressed as follows:

    Pitching moment Myyis described by

    Thrust T is defined as

    The system in(1)is linearized with the given cursing condition M a=15,v=15 060 ft/s,h=110 000 ft, γ =0o,q= 0o.The open-loop eigenvalues shown in Fig.1 are?0.895,0.784,?0.00021±0.0362j and 0.00011[26], where?0.895 and 0.784 are the short period oscillation, corresponding to q andα,?0.00021±0.0362j are the phugoid w.r.t.v andγ,0.00011 is the eigenvalue about the altitude modal.It is obvious the nonlinear aircraft in(1)is unstable.

    Fig.1.The open-loop eigenvalues of winged-cone model at the given cursing condition.

    The engine dynamic ofthe aircraftis expressed by a second order differential equation with the control inputβc

    The precise definition of every variable in the equations(1) and(2)is stated in Appendix A.

    The composite controlled system contains the equations(1) and(2).The elevator defl ection angleδeand the demand of the engine control inputβcare the control inputs,and the longitudinal velocity v and the altitude h are outputs. Considering the equations(1)and(2),v has the relation with βcandδe,which is derived as

    Similarly,the relationship is investigated among h,βcandδe. As

    then we have

    Therefore,βcandδeare explicitly contained in... v and h(4).

    The aircraft in(1)and(2)can be included by the general nonlinear formula,

    where x∈Rnis the state vector,u(t),?(x)∈Rmare inputs and outputs.f(x)and bi(x)along with the function?iare smooth functions on Rn,allowing arbitrary order derivatives to be calculated.

    As the system in(7)has m output components,it is considered to produce m subsystems if the input-outputfeedback linearization is executed,where the relative degree of the i-th subsystem is recorded as ri,i=1,···,m.

    Assumption 1.Allthe relative degree ri,i=1,···,m of m subsystems are constantand known.The system in(7)does notcontain zero dynamics,which means that the system degree?r is equal to the sum of relative degrees ri,P

    The ri-th derivative of yi=?i(x)can be expressed by Lie derivatives

    where Lf?i(x)is the Lie derivative of the function?ialong the vector field f,

    The composite controlled system in(1)and(2)reveals that the system degree is?r=7.According to(3)and(4),the relative degree of velocity subsystem to inputs is r1=3. The relative degree of altitude subsystem to inputs is r2=4 referring to(6).

    It means that the system degree equals to the total relative degrees of subsystems.In other words,the feedback linearization can be executed to reveal that... v and h(4)must be explicitly expressed by the control inputsδeandβc.The original nonlinear model is transformed into two coupling subsystems from inputs to outputs.Therefore we consider the following two subsystems for the cruising control,

    Velocity subsystem:

    Altitude subsystem:

    The ri-th derivatives of velocity and altitude dynamics havethe explicit expression with control variables,combining andas follows:

    is nonsingular over the entire flightenvelope given in[19?20]. As r1=3 and r2=4,the derivatives of v and h are calculated by the chain rule,

    The expressions of˙x,¨x,¨x0,¨α0,¨β0,¨γ,... γ,?w,?,?πandΠ refer to Appendix A[10].

    Remark 1.The feedback linearization transformation aims to explore the explicitcontrolinputs forthe outputs ofvelocity and altitude,which provides the control-oriented subsystems. However,as L3fv(x),L4fh(x)and B(x)are nonlinear,the two subsystems are still nonlinear and coupled.It is easier to design the controller with the explicit control inputs than that with implicit inputs.

    According to the formulae of(9)and(10),the generalmodel of subsystems is described as follows,

    As external disturbances d1(t)and d2(t)existing in¨βand CM(δe)are considered to be matched,the ri-th subsystem is concluded to the manifold,

    Control object:In this paper,the objective is to design a robust continuous sliding mode controller based on velocity and altitude subsystems in the manifold(13),such that the velocity v and the altitude h can track their reference vrand hr,especially existing unknown external disturbances.

    III.CONTINUOUS SLIDING MODE CONTROLLER DESIGN WITH DISTURBANCE COMPENSATION FOR AIR-BREATHING HYPERSONIC VEHICLES

    In this section,the continuous sliding mode controller is studied for the tracking control.When disturbances exist,the original feedback control law is applied to(11),

    where ˉf(x) = [ˉf1(x),ˉf2(x)]T,ˉb(x)= [ˉb1(x),ˉb2(x)]T, w=[w1,w2]Tis the auxiliary control vector.In this case, the controlled objective(13)is equivalent to the ri-th order integrator system,

    whereˉdi(t)=ˉbi(x)?d(t),i=1,2,j=1,2,···,ri?1.

    A.Disturbance Observer Design

    The purpose of disturbance observer is to efficiently estimate real-time disturbances.Observed values are used as compensation terms included in control signals.With the compensator,disturbances can be restrained to zero after a finite time,and the controller guarantees the accessibility and the stability before the free-error observation is achieved.

    Considering the tracking problem of AHVs,it is presented as ‰

    For the differential equation(16),whereσris continuous vector function,the disturbance vectorˉd(t)is bounded and has Lipshitz constants L=[L1,L2]T,where|ˉd1(t)|≤ L1and|ˉd2(t)|≤L2.Therefore,second order observers are used to observeˉd1(t)andˉd2(t)referring to Appendix B,which are expressed in the form of vectors as follows:

    whereλ0,λ1,λ2are gains of the o bserver,converges to dˉ(t)in finite time.

    Remark 2.The convergence proof of disturbance observer can refer to literatures[27?28].In addition,the l-th order observer provides more accurate derivatives than the p order observer,p≤l[29].By(17),z1can converge toˉd(t)in finite time,which guarantees thatobservation errors converge to zero in finite time.

    B.Continuous Sliding Mode Controller Design

    For the subsystem expressed in (15),define σ = [σ1,σ2,···,σri]T,the sliding surface is as follows:

    whereˉC=[0,cri?1ri?1λri?1,···,c1

    ri?1λ].If the derivative is designed as

    where edi(t)= ˉdi(t)?zi1(t),?1> 0,?2> 0,0< o1< 1, o2>1,sigoi(sri)=|sri|oisgn(sri).We can getthe following theorem.

    Theorem 1.For system(15),if the continuous controller is used

    the reachability of srihas two cases:

    1)if the observer error edi(t)=0,system states reach the sliding manifold sri=0 in finite time;

    2)ifthe observererror edi(t)/=0,system states reach theψ neighborhood of sri=0 in fi nite time and never escape from the region,where

    Proof.The Lyapunov function is selected aswhereis positive definite.

    Case 1:edi(t)=0,the time derivative of V1

    The other case is|sri|=ψ2,

    From the two cases,it can be concluded thatthe system states can reach the surface|sri|=ψin finite time.

    The region|sri|≤ ψ =min{ψ1,ψ2}is an attractive area for the states of system(15).System states do not escape from the region once reach it.To prove that,we need only to show that any system state on the boundary |sri|= ψ = min{ψ1,ψ2},never enters the region of |sri|> ψ =min{ψ1,ψ2}again.According to the above analysis,the time derivative of V1(sri)on the boundary |sri|=min{ψ1,ψ2}is always negative definite refereing to (22)and(23),namely˙V1(sri)=??,?>0.This means|sri| is monotonically decreasing,system states on the boundary enter the region|sri|<ψand never escape it.

    In sum,with the controller(21),the states of system(15) always reach the sliding mode surface sri=0 in finite time without observation errors and arrive at the region|sri|≤ψ in finite time if observation errors exist. □

    C.Continuous Sliding Mode Controller with Disturbance Compensation for The Tracking Control of AHVs

    Define the tracking error as ev(t)=v(t)?vr(t)and eh(t)=h(t)?hr(t),such that

    For disturbancesˉd1(t)andˉd2(t),the disturbance observer is used to estimate them.We have the corollary forobservation errors.

    Corollary 1.The disturbance observation errors,edi(t)= ˉdi(t)?zi1(t),i=1,2,are obviously bounded and should converge to zero after a finite time T,which is

    whereτi,i=1,2 is bounded constants.

    Theorem 2.For system(24)and(25),if the following continuous sliding mode controllerwith disturbance compensation is used,

    the velocity v and the altitude h are guaranteed to asymptotically track reference signals vrand hrwith switching modes svand sh,where z11(t)and z21(t)are compensation terms for disturbancesˉd1(t)andˉd2(t),respectively,obtained by(17).

    Proof.Via feedback linearization,two decoupled subsystems are presented.The velocity switching mode is designed in the manifold,

    whereλv= [c22λ2,c12λ,1],Ev= [ev,˙ev,¨ev]T.The time derivative of svis

    whereˉλv=[0,c22λ2,c12λ].

    The altitude switching mode is considered as

    whereˉλh=[0,c33λ3,c23λ2,c13λ].

    Combining the equations of˙svand˙sh,we have the compact expression

    The control variables in the above formula are replaced by the designed control law(27),then the derivative of V is

    Corollary 1 shows the property of observation errors edi(t), which is bounded and converges to zero after a finite time. According to Theorem 2,when edi(t)/=0,system states would reach neighborhoods of switching modes with respect to edi(t).When ed1(t)/=0,system states[ev,˙ev,¨ev]reachtheψvregion of sv=0,ψvWhen ed2(t)/=0,system statesregion of sh=0,ψh=mineventually stabilizes to zero after a finite time,system states also reach sv=0 and sh=0 in fi nite time.

    When sv=0 and sh=0,the dynamics of tracking error system on sliding modes sv=0 and sh=0 are obtained,

    It can be concluded that ev=c1e?λt,eh=c2e?λt,where c1,c2are undetermined constants,e is the natural exponent. Therefore,evand ehcan asymptotically converge to zero,v and h can track reference signals vrand hr,respectively.□

    Remark 3.The figure of the function|sri|oisgn(sri)is presented in Fig.2.It is obvious that the function is continuous.Therefore,the control law in(27)is also continuous. For the continuous controller,the chattering is eliminated. Note thatsystem states arrive ata smallneighborhood of sri, i=1,2 before observation errors edi(t)reach zero,which is considered to be the cost of eliminating the chattering. The boundary layer method keeps the width of convergence region no matter what edi(t)is,and the convergence is asymptoticalto the smallregion.The proposed method enables the asymptoticalconvergence to zero when edi(t)reaches zero and avoids chattering.

    Remark 4.The terms??1sigo1(sri)??2sigo2(sri)in the controllaw(27)can increase the convergence rate.When system states are far away from sri=0,the term??2sigo2(sri) would accelerate the convergence.When system states are very close to sri=0,the term??1sigo1(sri)help the convergence. Refer to Fig.2,|sri|oisgn(sri)with oi=0.5,oi=1.5 and oi=2.2.

    Fig.2. The continuous function|sri|oisgn(sri)with different oivalues.

    IV.SIMULATION

    In this section,simulation is executed to demonstrate the effectiveness of the proposed method.

    As a representative case study,the hypersonic vehicle is assumed to trim at v=15 060 ft/s and h=110 000 ft,and the aerodynamic coefficients and model parameters are given in Table I.

    Parameters of the sliding mode controller areλv=λh=1,?1= ?2=10,o1=0.6 and o2=1.5.Parameters of disturbance observer are designed asλ0=3,λ1=1.5, λ2=1.1,L1=15,L2=50.

    A.Tracking Control

    The reference velocity is 14 960 ft/s and the reference altitude is 110 200 ft.During the tracking process,the vehicle is perturbed by

    The continuous sliding mode controller is adopted,where the finite time disturbance observer is used as the compensator to reject disturbance.Figs.3(a)and 3(b)show that the designed observer can work effectively to exactly estimate disturbances in fi nite time,where dotted lines represent actual disturbances and solid lines depict observed values. The observed values z11and z21have well approximation to disturbancesˉd1andˉd2after a finite time.

    TABLE I AERODYNAMIC AND INERTIAL COEFFICIENTS

    Fig.3. Real disturbance values and estimated values.

    Fig.4. The positive boundaries of velocity and altitude switching modes.

    As stated in Theorem 1,before estimated errors become zero,the velocity switching mode and the altitude switching mode arrive at small regions with respect to estimated errors. In other words,if the track error of velocity ed1(t)/=0, evreaches theψvregion of sv= 0,whereψv=For the same reason,if the tracking error ed2(t)/=0 of altitude,ehalso runs into theψhregion of sh=0,whereψhOnce tracking errors edi(t)stabilizes to zero aftera finite time, system states also reach sv=0 and sh=0 in finite time.Fig.4 presents the boundaries ofthe two smallregions,whereψvand ψhwith solid lines describe positive boundaries for velocity and altitude switching modes,and

    Fig.5 further describes curves of svand sh,which locate in small regions before tracking errors reach zero,restrained

    Fig.6 provides the continuous control lawsβcandδeto avoid the chattering.Simultaneously,tracking results of velocity and altitude are also displayed in Fig.6,where velocity and altitude are both asymptotically stabilize attheir reference commands.It exhibits good robustness against disturbances with the composite controller.The boundary layer method is executed to compare with the proposed method.The sign function is replaced by the saturation function and the finite time disturbance observeris stillused as the compensator.The width of the saturation function isτ=0.5,the gain l1=20. The simulation results are illustrated in Fig.6 with dashed lines.The chattering has been eliminated,but the system responses are stillslower than thatwith the proposed method.

    Fig.5. Switching mode variables located in their regions.

    B.Robustness

    It can be observed that the composite control has finished the tracking of velocity and altitude within ten seconds in the previous simulation.The hypersonic aircraft is expected to track step commands,where vr=14 960 ft/s and hr= 110 200 ft change to vr=14 870 ft/s and hr=110 360 ft at t=20 s.The disturbances also change at t=35 s,which are designed as follows:

    The robustness ofcontrolleris checked here.Allparameters are set the same as the previous simulation.The varying disturbances are effectively estimated by the finite time observer,shown in Fig.7,where dotted lines represent disturbances, solid lines are provided by estimated values.

    Fig.6.Control laws and output responses,continuous TSMC with DO,saturation SMC with DO.

    Fig.7. Estimated values forˉd1andˉd2.

    Fig.8 presents new boundaries of svand shconsidering new varying disturbances.Switching mode variables svand share firstly restrained to small regions,and then the two switching mode variables arrive atzero when estimated errors converge to zero.

    The continuous control laws illustrated in Fig.9.v and h keep convergentto reference signals,where the robustness of the continuous controller is illustrated.Velocity and altitude are not perturbed even if new disturbances are added at t=35 s,and stillkeep welltracking to step reference signals.

    Fig.8.Switching modes and their regions.

    Fig.9.Control laws and output responses.

    V.CONCLUSION

    In this paper,a composite continuous controlleris discussed for the tracking problem of hypersonic vehicles.The system model is linearized to velocity and altitude subsystems. The continuous controller is designed to reduce chattering. The finite time disturbance observer is introduced to reject disturbances.Because the discontinuous term is replaced by continuous terms,switching modes of velocity and altitude arrive at small regions which vary depending on disturbance observation errors.When observation errors disappear,the observer converges to disturbance signals in finite time.It turns out that restrained regions to switching modes become zero and tracking errors of velocity and altitude asymptotically converge to zero.The method has significantly improved boundary layer method.Simulations have proved the effectiveness of the proposed continuous sliding mode controlwith disturbance observer for the tracking control of AHVs and high convergence rate is also provided.

    APPENDIX A

    Nomenclature:

    v?speed of sound

    γ?flight-path angle

    h?altitude

    α?angle of attack

    q?pitch rate

    T?thrust

    D?drag

    L?lift

    CT(β)?thrust coefficient

    CD(α)?drag coefficient

    CL(α)?lift coefficient

    CM(q)?pitch rate contribution to moment

    CM(α)?angle of attack contribution to moment

    CM(δe)?elevator deflection contribution to moment

    m?mass

    RE?radius of the Earth

    R?radial distance from Earth's center

    ˉc?mean aerodynamic chord

    cβ?throttle coefficient in CT

    ce?elevator coefficient in CM(δe)

    Iyy?moment of inertia

    Myy?pitching moment

    S?reference area

    δE?elevator angular deflection

    β?throttle setting

    βc?control contribution to throttle settingβ

    wn?natural frequency for throttle settingβ

    ξ?damping ratio for throttle settingβ

    μ?gravitational constant

    ρ?density of air

    Expression:

    APPENDIX B

    Considering the system ˙x(t)=y(t)+g(t),allderivatives˙g(t), ¨g(t),···,g(p?1)(t)of the disturbance term g(t)are assumed to be bounded,such that there is a known Lipshitz constant L>0 for g(p?1)(t).

    For a continuous function x(t)defined t≥ 0,if y(t)is Lebesgue-measurable,when input noises of x(t)and y(t)are zero,the exact finite time observer for g(t)can be established as follows:with enough large parametersλi,i=0,···,p,z0,z1,···,zpcan converge to x,g(t),···,g(p?1)(t)in fi nite time.

    REFERENCES

    [1]Fidan B,Mirmirani M,Ioannou P A.Flight dynamics and control of air-breathing hypersonic vehicles:review and new directions.In:Proceedings of the 2003 AIAA International Space Planes and Hypersonic Systems and Technologies.Norfolk,USA:AIAA,2003.

    [2]Fiorentini L,Serrani A,Bolender M A,Doman D B.Nonlinear robust adaptive control of flexible air-breathing hypersonic vehicles.Journal of Guidance,Control,and Dynamics,2009,32(2):401?416

    [3]Sigthorsson D O,Jankovsky P,Serrani A,Yurkovich S,Bolender M, Doman D B.Robust linear output feedback control of an air-breathing hypersonic vehicle.Journal of Guidance,Control,and Dynamics,2008, 31(4):1052?1065

    [4]Zhu Y J,Shi Z K.Several problems of flight characteristics and flight control for hypersonic vehicles.Flight Dynamics,2005,23(3):5?8

    [5]Gregory I M,McMinn J D,Chowdhry R S,Shaughnessy J D.Hypersonic Vehicle Model and Control Law Development Using H∞and μSynthesis,NASA Technical Memorandum,USA,NASA TM-4562, 1994.

    [6]Lind R.Linear parameter-varying modeling and controlof structuraldynamics with aero-thermodynamic effects.Journal of Guidance,Control, and Dynamics,2002,25(4):733?739

    [7]Oppenheimer M W,Doman D B.Control of an unstable,nonminimum phase hypersonic vehicle model.In:Proceedings of the 2006 Aerospace Conference.Big Sky,MT:IEEE,2006.1?22

    [8]Gibson T E,Annaswamy A M.Adaptive controlof hypersonic vehicles in the presence of thrust and actuator uncertainties.In:Proceedings of the 2008 AIAA Guidance,Navigation and Control Conference and Exhibit.Hawali,USA:AIAA,2008.

    [9]Serrani A,Zinnecker A M,Fiorentini L,Bolender M A,Doman D B.Integrated adaptive guidance and control of constrained nonlinear air-breathing hypersonic vehicle models.In:Proceedings of the 2009 American ControlConference.St.Louis,USA:IEEE,2009.3172?3177

    [10]Wang Q,Stengel R F.Robustnonlinear controlof a hypersonic aircraft. Journal of Guidance,Control,and Dynamics,2000,23(4):577?585

    [11]Xu J H,Mirmirani M,loannou P A.Adaptive sliding mode control design for a hypersonic flightvehicle.Journal of Guidance,Control,and Dynamics,2004,27(5):829?838

    [12]Bowcutt K G.Multidisciplinary optimization of airbreathing hypersonic vehicles.Journal of Propulsion and Power,2001,17(6):1184?1190

    [13]Jiang C H,Zhang C Y,Zhu L.Research of robust adaptive trajectory linearization control based on T-S fuzzy system.Journal of Systems Engineering and Electronics,2008,19(3):537?545

    [14]Zhu L,Jiang C S,Zhang C Y.Adaptive trajectory linearization control for aerospace vehicle based on RBFNN disturbance observer.Acta Aeronautica etAstronauticaSinica,2007,28(3):673?677

    [15]Hu X X,Wu L G,Hu C H,Gao H J.Adaptive sliding mode tracking control for a flexible air-breathing hypersonic vehicle.Journal of the FranklinInstitute,2012,349(2):559?577

    [16]Yu S H,Yu XH,Shirinzadeh B,Man Z H.Continuous finite-time control for robotic manipulators with terminalsliding mode.Automatica,2005, 41(11):1957?1964

    [17]Slotine J J,Sastry S S.Tracking control of non-linear systems using sliding surfaces with application to robot manipulators.International Journal of Control,1983,38(2):465?492

    [18]Slotine J J,Li W P.Applied Nonlinear Control.New Jersey:Prentice Hall,1991.

    [19]Li S H,Sun H B,Sun C Y.Composite controldesign for an air-breathing hypersonic vehicle.Proceedings of the Institution of Mechanical Engineers,Part I:Journal of Systems and Control Engineering,2012,226(5): 651?664

    [20]Sun H B,Li S H,Sun C Y.Finite time integralsliding mode controlof hypersonic vehicles.Nonlinear Dynamics,2013,73(1?2):229?244

    [21]Zong Q,JiY H,Zeng F L,Liu H L.Outputfeedback back-stepping control for a generic hypersonic vehicle via small-gain theorem.Aerospace Science and Technology,2012,23(1):409?417

    [22]Zong,Q,Wang J,Tao Y.Adaptive high-order dynamic sliding mode control for a flexible air-breathing hypersonic vehicle.International Journal of RobustNonlinear Control,2013,23(15):1718?1736

    [23]Sigthorsson D O,Jankovsky P,Serrani A,Yurkovich S,Bolender M, Doman D B.Robust linear output feedback control of an air-breathing hypersonic vehicle.Journal of Guidance,Control and Dynamics,2008, 31(4):1052?1066

    [24]Bolender M A,Doman D B.Nonlinear longitudinaldynamicalmodelof an air-breathing hypersonic vehicle.Journal of Spacecraft and Rockets, 2007,44(2):374?387

    [25]Shaughnessy J D,Pinckney S Z.Hypersonic Vehicle Simulation Model: Winged-Cone Configuration,NASA Technical Memorandum,USA, NASA TM-102610,1991.

    [26]Li F H.Guidance and Control Technology of Hypersonic Aircrafts. Beijing:China Astronautic Publishing House,2012.

    [27]Levant A.High-order sliding modes,differentiation and outputfeedback control.International Journal of Control,2002,76(9?10):924?9412

    [28]Shtessel Y B,Shkolnikov A,Levant A.Smooth second-order sliding modes:missile guidance application.Automatica,2007,43(8): 1470?1476

    [29]Levant A.Robust exact differentiation via sliding mode technique. Automatica,1998,34(3):379?384

    Qun Zong Received the bachelor,master,and Ph.D. degrees all in automatic control from Tianjin University,in 1983,1995 and 2002,respectively.Since 1983,he has been with the School of Electrical Engineering and Automation,Tianjin University,where he is currently a professor.His research interests include complex system modeling and flightcontrol. Corresponding author of this paper.

    Bailing Tian Received his Ph.D.degree in automatic controlfrom Tianjin University in 2011.Since 2011,he has been with the Schoolof Electrical Engineering and Automation,Tianjin University,where he is currently a lecturer.His research interests include trajectory optimization,guidance and control.

    Wei Xu Received the Ph.D.degree in electrical engineering from the Institute of Electrical Engineering,Chinese Academy of Sciences in 2008.He is currently a professor with the School of Electrical and Electronic Engineering,Huazhong University of Science and Technology.His research interests include controland electromagnetic design for electric machines.

    Received her Ph.D.degree in control theory and control engineering from School of Automation,Southeast University,in 2012.Since 2012,she is a lecturer at the School of Electricaland Automation Engineering,Tianjin University. Her research interests include sliding mode control, nonlinear system control,and intelligent control.

    Manuscriptreceived October 15,2013;accepted April11,2014.This work was supported by National Natural Science Foundation of China(61125306, 61273092,61301035,61304018,and 61411130160),National High-Technology Research and Development Program of China(2014AA051901), Tianjin Science and Technology Supporting Program(14JCQNJC05400), Research Innovation Program of Tianjin University(2013XQ0101),Hubei Science and Technology Supporting Program(XYJ2014000314),Aeronautical Science Foundation of China Supported by Science and Technology on Aircraft Control Laboratory(20125848004),and China Post-doctoral Science Foundation(2014M561559).Recommended by Associate Editor Bin Xian

    :Chaoxu Mu,Qun Zong,Bailing Tian,WeiXu.Continuous sliding mode controllerwith disturbance observerforhypersonic vehicles.IEEE/CAA Journalof Automatica Sinica,2015,2(1):45?55

    Chaoxu Mu,Qun Zong,and Bailing Tian are with the Departmentof Electrical Engineering and Automation,Tianjin University,Tianjin 300072,China (e-mail:cxmu@tju.edu.cn;zongqun@tju.edu.cn;bailing?tian@tju.edu.cn).

    Wei Xu is with the School of Electrical and Electronic Engineering, Huazhong University of Science and Technology,Wuhan 430074,China(email:weixu@hust.edu.cn).

    在线永久观看黄色视频| 久久精品国产综合久久久| 美女 人体艺术 gogo| 国产精品精品国产色婷婷| 国产又爽黄色视频| 国产1区2区3区精品| 欧美日韩亚洲国产一区二区在线观看| 在线观看日韩欧美| 国产成人影院久久av| 亚洲国产欧洲综合997久久, | 国产免费av片在线观看野外av| 日本五十路高清| 久久久久亚洲av毛片大全| 亚洲人成网站在线播放欧美日韩| netflix在线观看网站| 国产精品电影一区二区三区| 免费观看精品视频网站| 91老司机精品| 丰满人妻熟妇乱又伦精品不卡| 老司机午夜福利在线观看视频| 欧美日韩福利视频一区二区| 精品欧美国产一区二区三| 午夜福利免费观看在线| 美女高潮喷水抽搐中文字幕| 国产一区在线观看成人免费| 国产精品久久久久久亚洲av鲁大| 午夜两性在线视频| 美女国产高潮福利片在线看| 午夜免费鲁丝| svipshipincom国产片| 波多野结衣巨乳人妻| 美国免费a级毛片| 韩国av一区二区三区四区| 亚洲专区字幕在线| 岛国在线观看网站| 在线观看免费视频日本深夜| 国产亚洲精品久久久久久毛片| 在线观看免费视频日本深夜| 亚洲一区高清亚洲精品| x7x7x7水蜜桃| 一区二区三区精品91| 久久久久国内视频| 男人的好看免费观看在线视频 | 日本精品一区二区三区蜜桃| 2021天堂中文幕一二区在线观 | 亚洲人成电影免费在线| 国产一卡二卡三卡精品| 丰满的人妻完整版| av免费在线观看网站| 欧美性长视频在线观看| 岛国在线观看网站| 国产免费男女视频| 久久青草综合色| 91老司机精品| 正在播放国产对白刺激| 丰满人妻熟妇乱又伦精品不卡| 亚洲国产精品久久男人天堂| 欧洲精品卡2卡3卡4卡5卡区| svipshipincom国产片| 午夜久久久在线观看| 国产区一区二久久| 中文字幕精品亚洲无线码一区 | 九色国产91popny在线| www.精华液| 久久久久精品国产欧美久久久| 欧美另类亚洲清纯唯美| www.精华液| 久久伊人香网站| 国产精品 欧美亚洲| 法律面前人人平等表现在哪些方面| 亚洲免费av在线视频| 国内精品久久久久久久电影| 黄色片一级片一级黄色片| 中文字幕精品亚洲无线码一区 | 正在播放国产对白刺激| 国产aⅴ精品一区二区三区波| 国产久久久一区二区三区| svipshipincom国产片| 老司机福利观看| 久久久久国产一级毛片高清牌| 婷婷丁香在线五月| 欧美一级毛片孕妇| 国产精品亚洲av一区麻豆| 亚洲第一电影网av| 伦理电影免费视频| 亚洲 欧美一区二区三区| 亚洲精品国产一区二区精华液| 国产成人啪精品午夜网站| 校园春色视频在线观看| 国产高清videossex| 亚洲欧洲精品一区二区精品久久久| 伦理电影免费视频| 久久人妻av系列| 人人妻人人看人人澡| 国产免费男女视频| 成年版毛片免费区| 变态另类成人亚洲欧美熟女| 午夜两性在线视频| 国产精品久久久久久亚洲av鲁大| 精华霜和精华液先用哪个| 国产精品久久久久久精品电影 | 一本一本综合久久| 亚洲色图 男人天堂 中文字幕| 免费看日本二区| 国产aⅴ精品一区二区三区波| 中文字幕久久专区| 国产精品亚洲美女久久久| 欧美绝顶高潮抽搐喷水| 麻豆成人av在线观看| 伊人久久大香线蕉亚洲五| 午夜a级毛片| 国产精品精品国产色婷婷| 亚洲国产精品成人综合色| 久久久精品国产亚洲av高清涩受| 校园春色视频在线观看| 可以在线观看毛片的网站| 午夜日韩欧美国产| 中文字幕精品免费在线观看视频| 精品国产亚洲在线| 国产野战对白在线观看| 88av欧美| 听说在线观看完整版免费高清| 久久天躁狠狠躁夜夜2o2o| 很黄的视频免费| 欧美大码av| 国产伦在线观看视频一区| 白带黄色成豆腐渣| 色哟哟哟哟哟哟| 中文字幕人成人乱码亚洲影| 国产蜜桃级精品一区二区三区| 日韩成人在线观看一区二区三区| 一级a爱视频在线免费观看| 热re99久久国产66热| 男女那种视频在线观看| av超薄肉色丝袜交足视频| 99在线人妻在线中文字幕| 美女午夜性视频免费| 精品久久久久久,| 12—13女人毛片做爰片一| 日韩大尺度精品在线看网址| 欧美日韩一级在线毛片| av视频在线观看入口| 欧美日韩亚洲综合一区二区三区_| 欧美成人一区二区免费高清观看 | 亚洲国产欧美日韩在线播放| 天天躁夜夜躁狠狠躁躁| 欧美又色又爽又黄视频| 午夜福利一区二区在线看| 久久精品影院6| 老鸭窝网址在线观看| 热99re8久久精品国产| 国产精品久久电影中文字幕| 免费一级毛片在线播放高清视频| 757午夜福利合集在线观看| 国产爱豆传媒在线观看 | 午夜福利欧美成人| 变态另类丝袜制服| 少妇裸体淫交视频免费看高清 | 天堂√8在线中文| 久久中文字幕一级| 久久精品91无色码中文字幕| av福利片在线| 丰满人妻熟妇乱又伦精品不卡| 韩国av一区二区三区四区| 日韩中文字幕欧美一区二区| 国产亚洲精品第一综合不卡| 91麻豆精品激情在线观看国产| 国产精品自产拍在线观看55亚洲| 一本久久中文字幕| 少妇的丰满在线观看| 日韩大尺度精品在线看网址| www日本在线高清视频| 欧美+亚洲+日韩+国产| 草草在线视频免费看| 精品久久久久久久毛片微露脸| 亚洲一码二码三码区别大吗| 午夜精品在线福利| 亚洲欧美精品综合久久99| 啦啦啦韩国在线观看视频| 欧美精品亚洲一区二区| 久久国产乱子伦精品免费另类| 一二三四社区在线视频社区8| av超薄肉色丝袜交足视频| 日本撒尿小便嘘嘘汇集6| 正在播放国产对白刺激| 热99re8久久精品国产| 午夜激情福利司机影院| 欧美中文综合在线视频| 波多野结衣巨乳人妻| 成年版毛片免费区| 此物有八面人人有两片| avwww免费| 视频区欧美日本亚洲| 这个男人来自地球电影免费观看| e午夜精品久久久久久久| 国内精品久久久久精免费| 午夜激情av网站| av天堂在线播放| 精品乱码久久久久久99久播| 欧美日韩亚洲综合一区二区三区_| 黄片大片在线免费观看| 欧美色欧美亚洲另类二区| 一本精品99久久精品77| 美国免费a级毛片| 好男人电影高清在线观看| 国产精品二区激情视频| 欧美成人一区二区免费高清观看 | 日韩成人在线观看一区二区三区| 亚洲精品美女久久久久99蜜臀| 99国产精品一区二区蜜桃av| 国产真实乱freesex| 国产精品一区二区免费欧美| 超碰成人久久| 久久精品人妻少妇| 亚洲成av片中文字幕在线观看| 免费搜索国产男女视频| 久久久久久亚洲精品国产蜜桃av| 中文字幕av电影在线播放| 欧美精品啪啪一区二区三区| 波多野结衣高清无吗| 亚洲av日韩精品久久久久久密| 久久久国产成人免费| 两个人看的免费小视频| 国产亚洲精品综合一区在线观看 | 亚洲 欧美 日韩 在线 免费| www日本黄色视频网| 成人亚洲精品一区在线观看| 操出白浆在线播放| 精品第一国产精品| 亚洲最大成人中文| 女性被躁到高潮视频| 国产精品综合久久久久久久免费| 色综合亚洲欧美另类图片| 一夜夜www| xxxwww97欧美| 国产1区2区3区精品| 变态另类成人亚洲欧美熟女| 久久久国产欧美日韩av| АⅤ资源中文在线天堂| 俺也久久电影网| 亚洲国产中文字幕在线视频| 国产伦在线观看视频一区| 精品少妇一区二区三区视频日本电影| 美女大奶头视频| 久久中文字幕一级| 午夜老司机福利片| 99久久综合精品五月天人人| 成人精品一区二区免费| 悠悠久久av| 18禁国产床啪视频网站| 母亲3免费完整高清在线观看| 久久草成人影院| 国产激情偷乱视频一区二区| 色哟哟哟哟哟哟| 日韩欧美三级三区| 久久精品成人免费网站| 老司机福利观看| 免费无遮挡裸体视频| 久久人人精品亚洲av| 亚洲无线在线观看| 国产精华一区二区三区| 久久久久精品国产欧美久久久| 欧美激情久久久久久爽电影| 村上凉子中文字幕在线| 精品久久久久久久毛片微露脸| 国产精品美女特级片免费视频播放器 | 中国美女看黄片| 校园春色视频在线观看| 欧美激情久久久久久爽电影| 国产蜜桃级精品一区二区三区| 亚洲欧美一区二区三区黑人| 老司机福利观看| 久热爱精品视频在线9| 国产不卡一卡二| 黄频高清免费视频| 午夜久久久在线观看| 国产乱人伦免费视频| 一区福利在线观看| 亚洲av日韩精品久久久久久密| 午夜老司机福利片| 成人精品一区二区免费| www.自偷自拍.com| 中文字幕另类日韩欧美亚洲嫩草| 亚洲av电影不卡..在线观看| av免费在线观看网站| 国产主播在线观看一区二区| 夜夜爽天天搞| 观看免费一级毛片| 久久婷婷成人综合色麻豆| 十八禁网站免费在线| 99久久国产精品久久久| 国产精品乱码一区二三区的特点| 午夜福利高清视频| 精品乱码久久久久久99久播| 日韩大码丰满熟妇| 国产成人av激情在线播放| 欧美一区二区精品小视频在线| 女性生殖器流出的白浆| 久久久久久亚洲精品国产蜜桃av| 少妇熟女aⅴ在线视频| 国产成人系列免费观看| 国产视频一区二区在线看| 男人舔女人下体高潮全视频| 老鸭窝网址在线观看| 国产av一区在线观看免费| 国产一级毛片七仙女欲春2 | 亚洲在线自拍视频| 国产99白浆流出| 天堂√8在线中文| 午夜老司机福利片| 久久 成人 亚洲| 亚洲国产精品999在线| 1024视频免费在线观看| 久久久久久久久免费视频了| 日本免费一区二区三区高清不卡| 黑丝袜美女国产一区| 女人被狂操c到高潮| 免费无遮挡裸体视频| 欧美激情 高清一区二区三区| 国产精华一区二区三区| 日日摸夜夜添夜夜添小说| 女人高潮潮喷娇喘18禁视频| 99国产精品一区二区蜜桃av| 正在播放国产对白刺激| 国产99白浆流出| 听说在线观看完整版免费高清| 亚洲第一电影网av| 国产亚洲av高清不卡| 久久久国产欧美日韩av| 欧美av亚洲av综合av国产av| 国产精品久久久av美女十八| 妹子高潮喷水视频| 一本大道久久a久久精品| 久久国产亚洲av麻豆专区| 很黄的视频免费| 日韩欧美国产在线观看| 国产视频一区二区在线看| 久久天堂一区二区三区四区| 欧美乱妇无乱码| 一级毛片高清免费大全| 日本五十路高清| 黑丝袜美女国产一区| 欧美另类亚洲清纯唯美| 99国产精品一区二区三区| 婷婷六月久久综合丁香| 一区二区三区精品91| 日日摸夜夜添夜夜添小说| 日韩欧美国产一区二区入口| 在线天堂中文资源库| 精品一区二区三区四区五区乱码| 在线观看舔阴道视频| 国产真人三级小视频在线观看| 欧美在线一区亚洲| 中文字幕最新亚洲高清| 中文字幕精品免费在线观看视频| 精品一区二区三区视频在线观看免费| 美女高潮喷水抽搐中文字幕| 中亚洲国语对白在线视频| 91麻豆av在线| 正在播放国产对白刺激| 一区二区三区激情视频| www日本黄色视频网| 丝袜美腿诱惑在线| www.精华液| 一区二区三区激情视频| 国产精品日韩av在线免费观看| 国产成人一区二区三区免费视频网站| 9191精品国产免费久久| 男女午夜视频在线观看| 色播亚洲综合网| 人人澡人人妻人| 欧美亚洲日本最大视频资源| 亚洲天堂国产精品一区在线| 女同久久另类99精品国产91| 日韩三级视频一区二区三区| 国产亚洲欧美98| 久久国产精品影院| 天堂√8在线中文| 两人在一起打扑克的视频| 女性被躁到高潮视频| 日韩欧美三级三区| bbb黄色大片| 超碰成人久久| 日韩欧美一区二区三区在线观看| 欧美色视频一区免费| 韩国av一区二区三区四区| 国产蜜桃级精品一区二区三区| 在线观看免费午夜福利视频| 美国免费a级毛片| 黄色女人牲交| 成人手机av| 91在线观看av| 丰满的人妻完整版| 三级毛片av免费| 日韩精品中文字幕看吧| 国产一区二区在线av高清观看| 精品日产1卡2卡| 午夜视频精品福利| 亚洲七黄色美女视频| 久久伊人香网站| 波多野结衣高清无吗| 这个男人来自地球电影免费观看| 免费在线观看日本一区| 一区福利在线观看| 麻豆国产av国片精品| 91九色精品人成在线观看| 精品国产美女av久久久久小说| 久久久精品国产亚洲av高清涩受| 国产黄片美女视频| 美女高潮到喷水免费观看| 成人国语在线视频| 国产亚洲精品久久久久久毛片| a级毛片a级免费在线| 正在播放国产对白刺激| 日韩欧美国产一区二区入口| 国产亚洲精品久久久久5区| 国产成人av激情在线播放| 国产精品久久久久久亚洲av鲁大| 一边摸一边抽搐一进一小说| 一本综合久久免费| 美女高潮喷水抽搐中文字幕| 国产精品亚洲av一区麻豆| 久热爱精品视频在线9| 叶爱在线成人免费视频播放| 两个人看的免费小视频| www日本黄色视频网| 给我免费播放毛片高清在线观看| 欧美性长视频在线观看| 亚洲熟妇中文字幕五十中出| 亚洲精品国产一区二区精华液| 精品电影一区二区在线| 精品免费久久久久久久清纯| 狠狠狠狠99中文字幕| 给我免费播放毛片高清在线观看| 好看av亚洲va欧美ⅴa在| 亚洲精品色激情综合| 首页视频小说图片口味搜索| 999久久久精品免费观看国产| 99精品欧美一区二区三区四区| 成年免费大片在线观看| 精品不卡国产一区二区三区| 丝袜美腿诱惑在线| 可以在线观看毛片的网站| 亚洲成人国产一区在线观看| 成年女人毛片免费观看观看9| 999久久久国产精品视频| 高清毛片免费观看视频网站| 亚洲专区字幕在线| 亚洲精品粉嫩美女一区| 日本在线视频免费播放| 亚洲欧美日韩无卡精品| 操出白浆在线播放| 国产精品一区二区精品视频观看| 亚洲av电影在线进入| 久久人妻av系列| 国产av一区二区精品久久| 国产视频内射| 老司机靠b影院| 久久久久久久午夜电影| 亚洲avbb在线观看| 亚洲一区二区三区色噜噜| 久久亚洲真实| 不卡av一区二区三区| 午夜福利免费观看在线| 精品久久久久久久久久免费视频| 亚洲片人在线观看| 久久久国产成人精品二区| 久久精品人妻少妇| 18美女黄网站色大片免费观看| 无限看片的www在线观看| 亚洲成人国产一区在线观看| 亚洲精品国产一区二区精华液| 妹子高潮喷水视频| 国产精品久久视频播放| 岛国在线观看网站| 18禁美女被吸乳视频| 免费在线观看视频国产中文字幕亚洲| 99国产综合亚洲精品| 波多野结衣巨乳人妻| 99精品欧美一区二区三区四区| 欧美另类亚洲清纯唯美| 亚洲第一av免费看| 十分钟在线观看高清视频www| 久久欧美精品欧美久久欧美| 国产欧美日韩一区二区精品| 亚洲中文字幕日韩| 欧美黑人巨大hd| www日本在线高清视频| 叶爱在线成人免费视频播放| 真人做人爱边吃奶动态| 日本精品一区二区三区蜜桃| 亚洲男人的天堂狠狠| 国产成人系列免费观看| 丝袜人妻中文字幕| 欧美 亚洲 国产 日韩一| 免费高清在线观看日韩| 丁香六月欧美| 一二三四社区在线视频社区8| 一本精品99久久精品77| 亚洲精品久久成人aⅴ小说| 欧美色欧美亚洲另类二区| 男女午夜视频在线观看| 国产成人av教育| 亚洲电影在线观看av| 日韩大码丰满熟妇| 久久国产精品影院| a级毛片在线看网站| 在线观看66精品国产| 亚洲一码二码三码区别大吗| 国产精品乱码一区二三区的特点| 免费高清在线观看日韩| 波多野结衣巨乳人妻| 国产一区二区在线av高清观看| 日韩大尺度精品在线看网址| 中文字幕人妻丝袜一区二区| 我的亚洲天堂| 99热只有精品国产| 黄色丝袜av网址大全| 亚洲片人在线观看| 成人三级做爰电影| 美女午夜性视频免费| 亚洲免费av在线视频| 天天躁狠狠躁夜夜躁狠狠躁| 久久久久久亚洲精品国产蜜桃av| 99国产极品粉嫩在线观看| 在线观看免费午夜福利视频| 国产精品综合久久久久久久免费| 国产真实乱freesex| 一进一出抽搐gif免费好疼| 两性夫妻黄色片| 久久久久久免费高清国产稀缺| 久久九九热精品免费| 精品少妇一区二区三区视频日本电影| 亚洲aⅴ乱码一区二区在线播放 | 9191精品国产免费久久| 成人免费观看视频高清| 看黄色毛片网站| 黑丝袜美女国产一区| 色综合婷婷激情| 极品教师在线免费播放| 国产亚洲精品av在线| 搡老妇女老女人老熟妇| 无限看片的www在线观看| 最近最新中文字幕大全免费视频| 女性被躁到高潮视频| 一个人观看的视频www高清免费观看 | 日韩欧美三级三区| 精品一区二区三区四区五区乱码| 大型黄色视频在线免费观看| 一进一出抽搐动态| 日韩精品青青久久久久久| 三级毛片av免费| 亚洲美女黄片视频| 中出人妻视频一区二区| av欧美777| 日本在线视频免费播放| 99热只有精品国产| 窝窝影院91人妻| 91成年电影在线观看| 国产成+人综合+亚洲专区| 日日干狠狠操夜夜爽| 一二三四在线观看免费中文在| 亚洲国产精品成人综合色| 国产精品亚洲美女久久久| 欧美成人性av电影在线观看| 国产乱人伦免费视频| 这个男人来自地球电影免费观看| 久久香蕉精品热| 午夜精品久久久久久毛片777| 国产精品久久久久久亚洲av鲁大| 成人国语在线视频| 久久亚洲精品不卡| 欧美久久黑人一区二区| 亚洲精品美女久久av网站| 91在线观看av| 最近最新中文字幕大全免费视频| 国产精品98久久久久久宅男小说| 一边摸一边抽搐一进一小说| 妹子高潮喷水视频| 欧美日韩福利视频一区二区| 亚洲精品一区av在线观看| 久久久久国产一级毛片高清牌| 国产主播在线观看一区二区| 九色国产91popny在线| 亚洲精品粉嫩美女一区| 在线观看舔阴道视频| 两个人视频免费观看高清| 欧美日本视频| 一夜夜www| 精品不卡国产一区二区三区| 国产精华一区二区三区| 国产成人欧美| 热99re8久久精品国产| 一级a爱片免费观看的视频| 校园春色视频在线观看| 亚洲午夜精品一区,二区,三区| 亚洲av第一区精品v没综合| 亚洲精品一卡2卡三卡4卡5卡| 久久九九热精品免费| 午夜福利高清视频| 桃色一区二区三区在线观看| 美女午夜性视频免费| 九色国产91popny在线| 99在线人妻在线中文字幕| 国产亚洲精品av在线| 满18在线观看网站| 国产亚洲精品久久久久久毛片| 麻豆成人午夜福利视频| 久久久久亚洲av毛片大全| 色综合欧美亚洲国产小说| 两个人免费观看高清视频| 日韩大尺度精品在线看网址| 国产精品自产拍在线观看55亚洲| 亚洲午夜精品一区,二区,三区| 国产成人影院久久av|