CN106873624A - Flight control method is hung based on the rotor wing unmanned aerial vehicle of Partial feedback linearization four - Google Patents

Abstract

The present invention relates to a kind of control method of four rotor wing unmanned aerial vehicle load flight, the hanging in-flight swing of load is preferably suppressed to propose a kind of gamma controller based on Partial feedback linearization method, while realizing four rotor wing unmanned aerial vehicle track followings.The technical solution adopted by the present invention is, flight control method is hung based on the rotor wing unmanned aerial vehicle of Partial feedback linearization four, realized being provided with the unmanned plane of the lifting rope for hanging loading product, step is, set up non-linear dynamic model, Partial Linear is carried out to the non-linear dynamic model of four rotor wing unmanned aerial vehicles hanging flight course with Partial feedback linearization method, and then designs controller to realize unmanned aerial vehicle (UAV) control.Present invention is mainly applied to the control occasion of four rotor wing unmanned aerial vehicles load flight.

Description

Flight control method is hung based on the rotor wing unmanned aerial vehicle of Partial feedback linearization four

Technical field

The present invention relates to a kind of four rotor wing unmanned aerial vehicle load flight control method, more particularly to four rotor wing unmanned aerial vehicles with The control method of suspending way load objects flight.

Background technology

Four rotor wing unmanned aerial vehicles are a kind of aircraft of many rotor structures.Due to four rotor wing unmanned aerial vehicles facility, cheap, mobility The strong advantage of energy, the flight system of four rotor wing unmanned aerial vehicles hanging in recent years is increasingly paid close attention to by people.

The correlative study work of current studies in China personnel is generally to be directed to full-scale helicopter hanger in-flight, towards flight The operational control performance issue of member, and for hanging rope and the air dynamic behaviour problem analysis (phase of hanging load Periodical：Nanjing Aero-Space University's journal；Author：Qi Wantao, Chen Renliang；Publish days：2011；Title of article：Helicopter hangs Hang flight stability and maneuverability analysis；The page number：406-412).And to the hanging flight problem of micro-, small-sized multi-rotor unmanned aerial vehicle It is related to relatively fewer.

Foreign study personnel propose some different control methods for the problem of unmanned plane helicopter hanger flight. G.Wu et al. (periodicals：IEEE Access；Author：G.Wu,K.Sreenath；Publish days：2015；Title of article： Variation-Based Linearization of Nonlinear Systems Evolving on SO(3)and S²；Page Code：1592-1604) a kind of time-varying linearization technique for SO (3) system is proposed, four rotor wing unmanned aerial vehicles are hung with the method Extension system has carried out analysis by Linearization, and devises the realization of LQR (linear quadratic regulator) controller with this To the stability contorting of the system.

Other research teams are studied for the track following problem of four rotor wing unmanned aerial vehicles hanging flight system. S.Notter et al. (periodicals：IFAC Papers On Line；Author：S.Notter,A.Heckmann,A.Mcfadyen, F.Gonzalez；Publish days：2016；Title of article：Modelling,Simulation and Flight Test of a Model Predictive Controlled Multirotor with Heavy Slung Load；The page number：182-187) transport Realized with Model Predictive Control (MPC, Model Predictive Control) method and hang the steady of flight for many rotors Control and be accurately positioned calmly.K.Sreenath et al. (meetings：In Proceedings of 2013IEEE International Conference on Robotics and Automation；Author：K.Sreenath,N.Michael,and V.Kumar； Publish days：2013；Title of article：Trajectory Generation and Control of a Quadrotor with a Cable-Suspended Load–A Differentially-Flat Hybrid System；The page number：4888- 4895) non-linear control strategy is devised for four rotors hanging flight system using differential flat method, realizes four rotors The flight of stabilization subtracts the control effect of pendulum with hanging.

To realize that hanging subtracts the purpose of pendulum, the method that some research teams have used trajectory planning.S.Tang et al. (meetings View：In Proceedings of 2015IEEE International Conference on Robotics and Automation；Author：S.Tang,V.Kumar；Publish days：2013；Title of article：Mixed Integer Quadratic Program Trajectory Generation for a Quadrotor with a Cable- Suspended Payload；The page number：2216-2222.) using the track rule of Mixed Integer Quadratic Program The method of drawing, realizes four rotor flying Track Pick-ups, and hanging hunting of load gets a desired effect in making its flight course.

Other research team for multiple no-manned plane simultaneously hang same loading problem, hanging rope elastic problem and Four rotor wing unmanned aerial vehicle bands hanging load landing problem is hung to be studied.

The content of the invention

To overcome the deficiencies in the prior art, the present invention is directed to propose a kind of based on the non-linear of Partial feedback linearization method Controller, preferably suppresses hanging load in-flight swing while realizing four rotor wing unmanned aerial vehicle track followings.The present invention The technical scheme of use is, flight control method is hung based on the rotor wing unmanned aerial vehicle of Partial feedback linearization four, be provided with for Realized on the unmanned plane of the lifting rope for hanging loading product, step is to set up non-linear dynamic model, with Partial feedback linearization side Method carries out Partial Linear to the non-linear dynamic model of four rotor wing unmanned aerial vehicles hanging flight course, and then designs controller reality Existing unmanned aerial vehicle (UAV) control.

Further comprise the concrete steps that, first by the unmanned plane and hanging object in four rotor wing unmanned aerial vehicles hanging flight course Body carries out force analysis respectively, so as to obtain the non-linear dynamic model that four rotor wing unmanned aerial vehicles hang flight course：

Each variable-definition is as follows in formula (1)：m_QAnd m_LRespectively four rotor wing unmanned aerial vehicles and the quality of load, L is rope length, γ It is rope and the angle of vertical direction, g is acceleration of gravity, (x_Q,z_Q) it is that position of the quadrotor in two-dimensional space is sat Mark, F_nxAnd F_nyRespectively：

In formula (2), the total life that F is provided for four rotor wing unmanned aerial vehicles, θ is its angle of pitch；

The expression formula of the Tensity size T on hanging rope is as follows：

The controller design for the system as shown in formula (1) is proposed afterwards：

In formula (4), each symbol implication is as follows：

K in formula (5)_px,k_pz,k_dx,k_z,k_γControl gain is, and more than zero；e_xAnd e_zQuadrotor is represented respectively Site error e in x-axis and z-axis direction_x=x_Q-x_d, e_z=z_Q-z_d, (x_d,z_d) it is the target location of quadrotor；

K in formula (5) control gain_pxWith k_dxRepresent respectively in controller for x control directions " proportional " with it is " micro- Subitem "；Control gain k_pzWith k_dzRepresent respectively " proportional " and " differential term " for z control directions in controller；Control increases Beneficial k_γIt is " differential term " in controller for swinging angle control that pivot angle γ corresponding " proportional " is fixed valueSo k_γ's Value should be selected accordingly.

The step of entering line justification to controller asymptotic convergence characteristic be：

Controller (4) is substituted into mission nonlinear model (1) first, is obtained：

In formula (6),WhereinIt is the Position And Velocity of four rotor wing unmanned aerial vehicles Error.Then φ meets such as lower inequality：

|φ|<ρ(||p_Q||)||p_Q|| ⑺

Wherein | | | | 2 norms are represented, ρ () represents an increasing function, meets ρ (0)=0；

Design alternative liapunov function

WhereinIt is the total state amount of system, Then obvious P is positive definite matrix；

First derivative on the time is asked to formula (8), is obtained

By γ²<γ sin γ understand

Wherein

It is positive definite integral form,

Then obtained by formula (7)

Wherein ρ₂() represents an increasing function, meets ρ₂(0)=0；

So for arbitrary initial state, always there is one group of sufficiently large { k_px,k_pz,k_dx,k_z,k_γSo that

Wherein Q₂Positive definite integral form.It is hereby achieved that closed-loop system is half Globally asymptotic.Namely

The numerical simulation step of point stabilization and regulation control is carried out, test controller (4) hangs for four rotor wing unmanned aerial vehicles The control performance of flight system is hung, is emulated under MATLAB/SIMULINK environment, with the reasonability of access control device and can Row.

The features of the present invention and beneficial effect are：

Present invention controller of the design based on Partial feedback linearization method, the hanging load that four rotor wing unmanned aerial vehicles are carried With preferably subtracting pendulum effect.Take into account to consider while four rotor wing unmanned aerial vehicles tracking intended trajectory is ensured and take load-carrying subtracting Pendulum problem, realize while four rotor wing unmanned aerial vehicle site error asymptotic convergences pivot angle also asymptotic convergence to zero.

Brief description of the drawings：

Fig. 1 is four rotor wing unmanned aerial vehicle hangar system structure diagrams.

System mode, control input and state error figure during Fig. 2 tracing control numerical simulations.

Four rotor wing unmanned aerial vehicle change in location curves when a is tracing control numerical simulation；

Four rotor wing unmanned aerial vehicle site error change curves when b is tracing control numerical simulation；

Control input speed change curve when c is tracing control numerical simulation；

Hanging load pivot angle change curve when d is tracing control numerical simulation.

Specific embodiment

To overcome the deficiencies in the prior art, the present invention is directed to propose a kind of based on the non-linear of Partial feedback linearization method Controller, preferably suppresses hanging load in-flight swing while realizing four rotor wing unmanned aerial vehicle track followings.The present invention The technical scheme of use is that the four rotor wing unmanned aerial vehicles hanging flight system control method based on Partial feedback linearization is being set Having realized on the unmanned plane of the lifting rope of loading product for hanging, and step is, with Partial feedback linearization method to four rotors nobody The non-linear dynamic model of machine hanging flight course carries out Partial Linear, and then design controller realizes unmanned aerial vehicle (UAV) control, Further comprise the concrete steps that, entered respectively by hanging unmanned plane and hanging object in flight course to four rotor wing unmanned aerial vehicles first Row force analysis, so as to obtain the non-linear dynamic model that four rotor wing unmanned aerial vehicles hang flight course：

Each variable-definition is as follows in formula (1)：m_QAnd m_LRespectively four rotor wing unmanned aerial vehicles and the quality of load, L is rope length, γ It is rope and the angle of vertical direction, g is acceleration of gravity, (x_Q,z_Q) it is that position of the quadrotor in two-dimensional space is sat Mark, F_nxAnd F_nyRespectively：

In formula (2), the total life that F is provided for four rotor wing unmanned aerial vehicles, θ is its angle of pitch；

The expression formula of the Tensity size T on hanging rope is as follows：

The controller design for the system as shown in formula (1) is proposed afterwards：

In formula (4), each symbol implication is as follows：

K in formula (5)_px,k_pz,k_dx,k_z,k_γControl gain is, and more than zero；e_xAnd e_zQuadrotor is represented respectively Site error e in x-axis and z-axis direction_x=x_Q-x_d, e_z=z_Q-z_d, (x_d,z_d) it is the target location of quadrotor.

K in formula (5) control gain_pxWith k_dxRepresent respectively in controller for x control directions " proportional " with it is " micro- Subitem ", the choosing method of its value is with general PD (proportional-plus-derivative) controller class seemingly；Control gain k_pzWith k_dzControl is represented respectively For " proportional " and " differential term " of z control directions in device, it chooses mode ibid.Control gain k_γTo be directed in controller " differential term " of swinging angle control, pivot angle γ corresponding " proportional " is fixed valueSo k_γValue should select accordingly.

The step of entering line justification to controller asymptotic convergence characteristic be：

Controller (4) is substituted into mission nonlinear model (1) first, is obtained：

In formula (6),WhereinIt is the Position And Velocity of four rotor wing unmanned aerial vehicles Error.Then φ meets such as lower inequality：

|φ|<ρ(||p_Q||)||p_Q|| ⑺

Wherein | | | | 2 norms are represented, ρ () represents an increasing function, meets ρ (0)=0.

Design alternative liapunov function

WhereinIt is the total state amount of system, Then obvious P is positive definite matrix.

First derivative on the time is asked to formula (8), is obtained

By γ²<γ sin γ understand

Wherein

It is positive definite integral form,

Can then be obtained by formula (7)

Wherein ρ₂() represents an increasing function, meets ρ₂(0)=0.

So for arbitrary initial state, always there is one group of sufficiently large { k_px,k_pz,k_dx,k_z,k_γSo that

Wherein Q₂Positive definite integral form.It is hereby achieved that closed-loop system is half Globally asymptotic.Namely

The numerical simulation step of point stabilization and regulation control is carried out, test controller (4) hangs for four rotor wing unmanned aerial vehicles The control performance of flight system is hung, is emulated under MATLAB/SIMULINK environment, with the reasonability of access control device and can Row.

The technical problems to be solved by the invention are to propose a kind of nonlinear Control based on Partial feedback linearization method Device, preferably suppresses hanging load in-flight swing while realizing four rotor wing unmanned aerial vehicle track followings.

The technical solution adopted by the present invention is：Lyapunov Equation is designed based on Partial feedback linearization method, and then Design controller realizes control targe, comprises the following steps：

Stress point is carried out respectively by hanging unmanned plane and hanging object in flight course to four rotor wing unmanned aerial vehicles first Analysis, so as to obtain the non-linear dynamic model that four rotor wing unmanned aerial vehicles hang flight course：

Each variable-definition is as follows in formula (1)：m_QAnd m_LRespectively four rotor wing unmanned aerial vehicles and the quality of load, L is rope length, γ It is rope and the angle of vertical direction, g is acceleration of gravity, (x_Q,z_Q) it is that position of the quadrotor in two-dimensional space is sat Mark, F_nxAnd F_nyRespectively：

In formula (2), the total life that F is provided for four rotor wing unmanned aerial vehicles, θ is its angle of pitch；

The expression formula of the Tensity size T on hanging rope is as follows：

The controller design for the system as shown in formula (1) is proposed afterwards：

In formula (4), each symbol implication is as follows：

K in formula (5)_px,k_pz,k_dx,k_z,k_γControl gain is, and more than zero；e_xAnd e_zQuadrotor is represented respectively Site error e in x-axis and z-axis direction_x=x_Q-x_d, e_z=z_Q-z_d, (x_d,z_d) it is the target location of quadrotor.

K in formula (5) control gain_pxWith k_dxRepresent respectively in controller for x control directions " proportional " with it is " micro- Subitem ", the choosing method of its value is with general PD (proportional-plus-derivative) controller class seemingly；Control gain k_pzWith k_dzControl is represented respectively For " proportional " and " differential term " of z control directions in device, it chooses mode ibid.Control gain k_γTo be directed in controller " differential term " of swinging angle control, pivot angle γ corresponding " proportional " is fixed valueSo k_γValue should select accordingly.

The step of entering line justification to controller asymptotic convergence characteristic be：

Controller (4) is substituted into mission nonlinear model (1) first, is obtained：

In formula (6),WhereinIt is the Position And Velocity of four rotor wing unmanned aerial vehicles Error.Then φ meets such as lower inequality：

|φ|<ρ(||p_Q||)||p_Q|| ⑺

Wherein | | | | 2 norms are represented, ρ () represents an increasing function, meets ρ (0)=0.

Design alternative liapunov function

WhereinIt is the total state amount of system, Then obvious P is positive definite matrix.

First derivative on the time is asked to formula (8), is obtained

By γ²<γ sin γ understand

Wherein

It is positive definite integral form,

Can then be obtained by formula (7)

Wherein ρ₂() represents an increasing function, meets ρ₂(0)=0.

So for arbitrary initial state, always there is one group of sufficiently large { k_px,k_pz,k_dx,k_z,k_γSo that

Wherein Q₂Positive definite integral form.It is hereby achieved that closed-loop system is half Globally asymptotic.Namely

To verify the validity of the nonlinear control method for being directed to four rotor wing unmanned aerial vehicles hanging flight system of the invention, enter Numerical simulation checking is gone.With reference to numerical simulation and accompanying drawing to the present invention for four rotor wing unmanned aerial vehicles hanging flight system Nonlinear control method is described in detail.

The control performance of flight system is hung for four rotor wing unmanned aerial vehicles for test controller (4), in MATLAB/ Under SIMULINK environment, the numerical simulation of point stabilization and regulation control is carried out, with the reasonability of access control device and feasible Property.

First, numerical simulation brief introduction

The relevant parameter of unmanned plane hangar system is set as

L=0.9, m_Q=1.2, m_L=0.36, g=9.8

Choose target trajectory

Carry out numerical simulation.

2nd, tracing control numerical simulation.

Given above-mentioned parameter, and selecting system state initial value x_Q(0)=0, z_Q(0)=0, γ (0)=0 °, by controller (4) Be applied to four rotor wing unmanned aerial vehicle hangar systems (1), to position of aircraft control with subtract pendulum effect and consider, and in view of being The constraint of system input amplitude, chooses controller parameter as follows:

k_px=2.7, k_dx=1.1, k_pz=30, k_dz=4, k_γ=1.5

Then can now obtain shown in control effect such as Fig. 2 (a), Fig. 2 (b), Fig. 2 (c), Fig. 2 (d).Fig. 2 (a), Fig. 2 B (), Fig. 2 (c), Fig. 2 (d) respectively describes the curve that four rotor wing unmanned aerial vehicle positions change over time with target location, four rotors The curve that unmanned plane site error is changed over time, the curve that system control input is changed over time, with hanging load pivot angle with Time changing curve.Solid line represents the change curve of four rotor wing unmanned aerial vehicle physical locations in 2 (a), and dotted line is the change of target location Change curve.It can be seen that the controller designed by herein can well follow aim curve, pivot angle from Fig. 2 (a), Fig. 2 (b) The decay of concussion is very fast.

By above-mentioned analysis, it was demonstrated that the validity of carried algorithm of the invention.

Claims (4)

1. it is a kind of that flight control method is hung based on the rotor wing unmanned aerial vehicle of Partial feedback linearization four, it is characterized in that, it is being provided with use In realization on the unmanned plane of lifting rope for hanging loading product, step is to set up non-linear dynamic model, with Partial feedback linearization Method carries out Partial Linear to the non-linear dynamic model of four rotor wing unmanned aerial vehicles hanging flight course, and then designs controller Realize unmanned aerial vehicle (UAV) control.

2. flight control method, its feature are hung based on the rotor wing unmanned aerial vehicle of Partial feedback linearization four as claimed in claim 1 It is further to comprise the concrete steps that, is divided by hanging unmanned plane and hanging object in flight course to four rotor wing unmanned aerial vehicles first Force analysis is not carried out, so as to obtain the non-linear dynamic model that four rotor wing unmanned aerial vehicles hang flight course：

$\left\{\begin{array}{c}{m}_Q{\stackrel{\&CenterDot;\&CenterDot;}{x}}_Q=F_{nx}cos\mathrm{\&gamma;}-\frac{m_{Q}sin\mathrm{\&gamma;}}{m_Q+m_L}{F}_{ny}+\frac{m_Q{m}_{L}sin\mathrm{\&gamma;}}{m_Q+m_L}L{\stackrel{\&CenterDot;}{\mathrm{\&gamma;}}}^2,\\ m_Q{\stackrel{\&CenterDot;\&CenterDot;}{z}}_Q=F_{nx}sin\mathrm{\&gamma;}+\frac{m_Q\cos \mathrm{\&gamma;}}{m_Q+m_L}{F}_{ny}-\frac{m_Q{m}_{L}cos\mathrm{\&gamma;}}{m_Q+m_L}L{\stackrel{\&CenterDot;}{\mathrm{\&gamma;}}}^2-m_{Q}g,\\ m_{Q}L\stackrel{\&CenterDot;\&CenterDot;}{\mathrm{\&gamma;}}=-F_{nx}\end{array}\right.---\left(1\right)$

Each variable-definition is as follows in formula (1)：m_QAnd m_LRespectively four rotor wing unmanned aerial vehicles and the quality of load, L are rope length, and γ is rope The angle of rope and vertical direction, g is acceleration of gravity, (x_Q,z_Q) it is position coordinates of the quadrotor in two-dimensional space, F_nxAnd F_nyRespectively：

$\left\{\begin{array}{c}{F}_{nx}=Fsin(\mathrm{\&gamma;}-\mathrm{\&theta;}),\\ F_{ny}=Fcos(\mathrm{\&gamma;}-\mathrm{\&theta;})\end{array}\right.---\left(2\right)$

In formula (2), the total life that F is provided for four rotor wing unmanned aerial vehicles, θ is its angle of pitch；

The expression formula of the Tensity size T on hanging rope is as follows：

$T=\frac{m_L(F_{ny}+{\mathrm{LM}}_Q{\stackrel{\&CenterDot;}{\mathrm{\&gamma;}}}^2)}{m_Q+m_L}---\left(3\right)$

The controller design for the system as shown in formula (1) is proposed afterwards：

$\left\{\begin{array}{c}{F}_{nx}=m_Q\left(\cos \mathrm{\&gamma;}\right({\stackrel{\&CenterDot;\&CenterDot;}{x}}_d+u)+\sin \mathrm{\&gamma;}({\stackrel{\&CenterDot;\&CenterDot;}{z}}_d+g+v\left)\right),\\ F_{ny}=(m_Q+m_L)(-\sin \mathrm{\&gamma;}({\stackrel{\&CenterDot;\&CenterDot;}{x}}_d+u)+\cos \mathrm{\&gamma;}({\stackrel{\&CenterDot;\&CenterDot;}{z}}_d+g+v\left)\right)+m_{L}L{\stackrel{\&CenterDot;}{\mathrm{\&gamma;}}}^2\end{array}\right.---\left(4\right)$

In formula (4), each symbol implication is as follows：

$\left\{\begin{array}{c}u=-k_{dx}{\stackrel{\&CenterDot;}{e}}_x-k_{px}{e}_x-\frac{{\mathrm{Lk}}_{\mathrm{\&gamma;}}}{\cos \mathrm{\&gamma;}}\stackrel{\&CenterDot;}{\mathrm{\&gamma;}},\\ v=-k_{dz}{\stackrel{\&CenterDot;}{e}}_z-k_{pz}{e}_z\end{array}\right.---\left(5\right)$

K in formula (5)_px,k_pz,k_dx,k_z,k_γControl gain is, and more than zero；e_xAnd e_zRepresent quadrotor in x respectively Site error e on axle and z-axis direction_x=x_Q-x_d, e_z=z_Q-z_d, (x_d,z_d) it is the target location of quadrotor；

K in formula (5) control gain_pxWith k_dxRepresent respectively " proportional " and " differential term " for x control directions in controller； Control gain k_pzWith k_dzRepresent respectively " proportional " and " differential term " for z control directions in controller；Control gain k_γFor For " differential term " of swinging angle control in controller, pivot angle γ corresponding " proportional " is fixed valueSo k_γValue should evidence This selection.

3. flight control method, its feature are hung based on the rotor wing unmanned aerial vehicle of Partial feedback linearization four as claimed in claim 1 It is that the step of entering line justification to controller asymptotic convergence characteristic is：

Controller (4) is substituted into mission nonlinear model (1) first, is obtained：

$\left\{\begin{array}{c}{\stackrel{\&CenterDot;\&CenterDot;}{e}}_x=-k_{dx}{\stackrel{\&CenterDot;}{e}}_x-k_{px}{e}_x-\frac{{\mathrm{Lk}}_{\mathrm{\&gamma;}}}{\cos \mathrm{\&gamma;}}\stackrel{\&CenterDot;}{\mathrm{\&gamma;}},\\ {\stackrel{\&CenterDot;\&CenterDot;}{e}}_z=-k_{dz}{\stackrel{\&CenterDot;}{e}}_z-k_{pz}{e}_z,\\ \stackrel{\&CenterDot;\&CenterDot;}{\mathrm{\&gamma;}}=-k_{\mathrm{\&gamma;}}\stackrel{\&CenterDot;}{\mathrm{\&gamma;}}-\frac{g}{L}\sin \mathrm{\&gamma;}+\mathrm{\&phi;}+\frac{1}{L}(-{\stackrel{\&CenterDot;\&CenterDot;}{x}}_d\cos \mathrm{\&gamma;}+{\stackrel{\&CenterDot;\&CenterDot;}{z}}_d\sin \mathrm{\&gamma;})\end{array}\right.---\left(6\right)$

In formula (6),WhereinIt is the Position And Velocity error of four rotor wing unmanned aerial vehicles. Then φ meets such as lower inequality：

|φ|<ρ(||p_Q||)||p_Q|| ⑺

Wherein | | | | 2 norms are represented, ρ () represents an increasing function, meets ρ (0)=0；

Design alternative liapunov function

WhereinIt is the total state amount of system, Then obvious P is positive definite matrix；

First derivative on the time is asked to formula (8), is obtained

By γ²<γ sin γ understand

Wherein

$Q=diag(\&lsqb;k_{px}{k}_{dx},k_{px}{k}_{dx},k_{pz}{k}_{dz},k_{pz}{k}_{dz},\frac{g}{L}{k}_{\mathrm{\&gamma;}},\frac{g}{L}{k}_{\mathrm{\&gamma;}}\&rsqb;),---\left(11\right)$

It is positive definite integral form,

$\begin{array}{c}N=-\left(\right(1+k_{px}){\stackrel{\&CenterDot;}{e}}_x+k_{dx}{e}_x)\frac{{\mathrm{Lk}}_{\mathrm{\&gamma;}}}{\cos \mathrm{\&gamma;}}\stackrel{\&CenterDot;}{\mathrm{\&gamma;}}+\left(\frac{g}{L}+\frac{g^2}{L^2}\right)\stackrel{\&CenterDot;}{\mathrm{\&gamma;}}(\mathrm{\&gamma;}-\sin \mathrm{\&gamma;})\\ +\left(\left(\frac{g}{L}+\frac{g^2}{L^2}\right)\stackrel{\&CenterDot;}{\mathrm{\&gamma;}}+k_{\mathrm{\&gamma;}}\mathrm{\&gamma;}\right)\mathrm{\&phi;}\end{array}---\left(12\right)$

Then obtained by formula (7)

Wherein ρ₂() represents an increasing function, meets ρ₂(0)=0；

So for arbitrary initial state, always there is one group of sufficiently large { k_px,k_pz,k_dx,k_z,k_γSo that

Wherein Q₂Positive definite integral form.It is hereby achieved that closed-loop system is half Globally asymptotic.Namely

$\underset{t\&RightArrow;\mathrm{\&infin;}}{\lim }\&lsqb;{\stackrel{\&CenterDot;}{e}}_x,e_x,{\stackrel{\&CenterDot;}{e}}_z,e_z,\stackrel{\&CenterDot;}{\mathrm{\&gamma;}},\mathrm{\&gamma;}\&rsqb;=0---\left(15\right).$

4. flight control method, its feature are hung based on the rotor wing unmanned aerial vehicle of Partial feedback linearization four as claimed in claim 3 It is the numerical simulation step for carrying out point stabilization and regulation control, test controller (4) is hung for four rotor wing unmanned aerial vehicles and flown The control performance of system, is emulated under MATLAB/SIMULINK environment, with the reasonability and feasibility of access control device.