CN109857146A - A kind of layering unmanned plane tracking and controlling method based on feedforward and weight distribution - Google Patents
A kind of layering unmanned plane tracking and controlling method based on feedforward and weight distribution Download PDFInfo
- Publication number
- CN109857146A CN109857146A CN201910013466.8A CN201910013466A CN109857146A CN 109857146 A CN109857146 A CN 109857146A CN 201910013466 A CN201910013466 A CN 201910013466A CN 109857146 A CN109857146 A CN 109857146A
- Authority
- CN
- China
- Prior art keywords
- unmanned plane
- control
- feedforward
- posture
- tracking
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The present invention discloses a kind of unmanned plane tracking and controlling method, belongs to motion control field.Comprise the steps of: that the structural parameters, the propeller that obtain unmanned plane itself can be provided lift tange and unmanned plane and the real-time movement state information of tracked target；Newton Euler's model is established, and adds saturated characteristic in the middle to model；It according to the model put up, is controlled in pitch angle, roll angle control, yaw angle control and height control and establish basic two close cycles cascade PID deviation controller in four subsystems；According to unmanned plane weight parameter and UAV Attitude information, gravity feedforward and posture feedforward are added；Addition saturation weight distributor, eliminates the invalid increment generated due to propeller saturated characteristic in pitch angle control subsystem and roll angle control subsystem；According to the posture angular limit of unmanned plane and the motion information of tracked target, settling time optimal track control device is connected on four subsystems of lower layer's posture and height control, finally realizes tenacious tracking.
Description
Technical field
The present invention relates to unmanned aerial vehicle (UAV) control, in particular to a kind of layering unmanned plane based on feedforward and weight distribution tracks control
Method processed.
Background technique
In recent years, with the development of unmanned air vehicle technique, the visual field of people is all progressed into all trades and professions, has incorporated people
Life；Unmanned plane is acted mainly as in military domain as reconnaissance plane, with small in size, low cost, it is easy to use, to operation
The advantages that environmental requirement is low, battlefield survival is stronger；And in civil field, it is broadly divided into and takes photo by plane and two aspect of loading, navigating
Bat field, video display industry carry out auxiliary shooting using unmanned plane and early have become normality；And other than video display industry, take photo by plane nobody
Machine also can be widely used to national ecological environmental protection, mineral resources exploration, marine environmental monitoring, land use survey, water money
It is source exploitation, crop growth monitoring and the yield by estimation, agricultural operation, Natural calamity monitoring and assessment, urban planning and municipal administration, gloomy
Woods pest and disease damage protection with monitoring, public safety, Defence business, etc. fields, have the wide market demand；And in loading field,
Agricultural irrigation unmanned plane, fire-fighting unmanned plane and express delivery launch unmanned plane and also enter the practical stage；
Although unmanned plane current unmanned plane and does not have autonomous flight ability using more and more extensive, in reality,
And be more that operation is remotely controlled by operator, with many limitations；So autonomous flight will be the following nothing
The Main way of man-machine research.
Summary of the invention
Stabilized flight in the height set in the sky the purpose of the present invention is control unmanned plane, can and speed unbiased with position
Unbiased carries out tenacious tracking to mobile surface targets；It is specific as follows:
A kind of layering unmanned plane tracking and controlling method based on feedforward and weight distribution, comprising the following steps:
(1) structural parameters of unmanned plane, including unmanned plane quality, acceleration of gravity, electrical axis and unmanned plane weight are obtained
Unmanned plane rotary inertia in the distance of the heart, tri- directions xyz；Obtaining unmanned plane propeller can be provided the range of lift；It is real
When obtain the velocity vector and displacement vector of the Eulerian angles posture information of unmanned plane itself, elevation information and tracked target；
(2) newton Euler's model is established according to the obtained unmanned plane parameter information of step (1), can be mentioned according to propeller
The lift tange of confession adds saturated characteristic to model in the middle；
(3) it according to the model put up, is controlled in pitch angle, roll angle control, yaw angle control and height control four
Basic two close cycles cascade PID deviation controller is established in subsystem, and is independently adjusted in the case where not considering coupling temporarily
Save parameter；
(4) according to unmanned plane weight parameter and UAV Attitude information, gravity feedforward and posture feedforward are added, by posture
Regard the disturbance of height control as with gravity coupling to offset, while by Taylor expansion come next discrete week to unmanned plane
Attitude value in phase is estimated；
(5) in pitch angle control subsystem and roll angle control subsystem addition saturation weight distributor, with eliminate by
In the invalid increment that propeller saturated characteristic generates；
(6) according to the posture angular limit of unmanned plane, it is input with the motion information of tracked target, establishes the time on upper layer
Optimal track control device is connected on four subsystems of lower layer's posture and height control, finally realizes tenacious tracking；
In the step (2), on the basis of basic newton Euler's mechanical model, it is added to the saturated characteristic of lift: Fmin
< F1、F2、F3、F4< Fmax, wherein F1、F2、F3、F4It is the lift that four propellers generate, FmaxIt is that single propeller can be provided
Maximum lift, FminIt is to maintain the minimum lift that propeller does not stall；
In the step (4):
1. the gravity feedforward added meets:
F′n=Fn+Fbase, n=1,2,3,4 (1)
Wherein Fn' it is the propeller lift being added after gravity feedforward, substitute original Fn, m is unmanned plane itself gross mass, and g is
Acceleration of gravity, FbaseIt is gravity feedforward；
2. the posture feedforward added meets:
Wherein FhIt is the inner loop control increment of height control subsystem, Fh' it is the height controller being added after posture feedforward
Increment substitutes original Fh；
3. since posture is continuous to the coupling of height, and actual controller and posture are anti-in height subsystem
Feedback is discrete, so the pose compensation directly obtained, there are certain hysteresis quality, delay degree increases with the increase of discrete periodic
By force；So being estimated using Taylor expansion the attitude disturbance in a next discrete periodic in control process:
Wherein θt、It is using time t as the pitch angle of parameter and roll angle；Discrete periodic is T, initial in a cycle
Moment is t0, the end cycle moment is t1, g (tε) be estimation after pose compensation value；
In the step (5), using the inner ring in pitch angle and roll angle serials control --- angular speed ring addition one
A saturation weight distributor is allocated controlling increment, is saturated weight distributor with pitch angle controlling increment, roll angle control
As input, pitch angle controlling increment and roll angle after distributing can control increment as exporting for increment and height gain processed；
If pitching angle increment and rolling angle increment will not make the lift of any one propeller reach saturation, initial value is kept to export, if
It goes beyond the scope, then makes pitching angle increment FθWith rolling angle incrementIn the case where keeping former symbol, make:
Wherein, FlimitIt is the maximum increment absolute value for making propeller reach saturation, FψlIt is that yaw angle is made to keep balance institute
There are adjusting allowance；F′θWithIt is then the output valve after supersaturated weight distributor；
In the step (6),
1. choosing the pitching angle theta and roll angle in attitude angleAs the control amount moved horizontally, the choosing of Eulerian angles attitude description
Take ZYX along rule；
2. being decomposed into the direction x and y using the relative position and relative velocity of tracked target and unmanned plane itself as input quantity
Direction:
A is the relative velocity on the direction x in parametric equation, and b is the relative position on the direction x, and c is opposite on the direction y
Speed, d are the relative position on the direction y；
3. the timing-optimal control of system is bang-bang control, as a result through deriving are as follows:
Max_x is the mould being most worth inputted on the direction tracking control unit x, and max_y is that tracking control unit is defeated in y-direction
The mould being most worth entered is the acceleration extreme value limited by attitude angle in unmanned plane tracking system；
4. since bang-bang control can generate concussion in steady-state value, so tracking control unit is close to stable situation
Under, i.e. the relative position of unmanned plane itself and tracked target and when smaller relative velocity is switched to deviation control mode, so that
System is restrained；
The utility model has the advantages that presently disclosed a kind of based on the layering unmanned plane tracing control side to feedover and weight distributes
System is specifically divided into upper layer tracking control unit and lower layer's posture and height controller by method；Lower layer's posture and height control
Device, by way of the feedforward of addition gravity and posture feedforward, counteract the gravity disturbance that is coupled in height control subsystem and
Attitude disturbance improves the dynamic property of height control subsystem；Attitude disturbance is estimated using Taylor expansion, is reduced
To the hysteresis quality of attitude disturbance estimation；Meanwhile while invalid increment is generated for because propeller lift has saturated characteristic
The problem of, saturation weight distributor is devised, the disturbance that invalid increment generates internal system is eliminated；Upper layer tracking control unit
Using position is unbiased and the unbiased final states as system of speed constrains, to reach the system operation time of stable state as performance indicator, if
It has counted time optimal tracking control unit, and has been switched to more smooth deviation control when system is close to final states, made system can be with
Moving target is tracked；Finally upper and lower level control system is concatenated, realizes quadrotor drone to mobile surface targets
Tracking, make system convergence.
Detailed description of the invention
Fig. 1 is system entirety control structure block diagram；
Fig. 2 is the height control simulation comparison result figure before and after adding gravity feedforward；
Fig. 3 is that height before and after addition posture feedforward on the basis of having gravity feedforward controls simulation comparison result
Figure；
Fig. 4 is continuous signal, zero-order holder discrete signal and the Taylor series of the posture Product Perturbation item of height control
Estimate discrete signal figure；
Fig. 5 is that have weight distributor and the height control system rise time simulation result comparison without weight distributor；
Fig. 6 is response of the upper layer tracking control unit under ideal attitude motion model case, ideal pose movement here
Model refers to given UAV Attitude input, and unmanned plane can be non-delay without the response for shaking non-overshoot, to generate in the side x
To with the acceleration on the direction y；
Fig. 7 is the tracking simulation result of control system totality, attitude motion model use posture of the present invention and
Height controller.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawing and example, to control
Method carries out detailed demonstration and elaboration；
The present invention propose a kind of time optimal unmanned plane distributed based on feedforward compensation and propeller saturated characteristic weight with
Track control method, controller are broadly divided into two layers, as shown in Figure 1；Upper layer is tracking control unit, and lower layer is that posture and height control
Device；
Upper layer tracking control unit uses timing-optimal control method, and is decomposed into the direction x and the direction y, respectively with the direction x
Be input with the unmanned plane in the direction y and the relative velocity of tracked target and relative position, output then for desired pitch angle and
Roll angle, the input as pitch angle control subsystem and roll angle control subsystem；
Lower layer is divided into pitch angle control, roll angle control, and yaw angle control and height control four subsystems；All using string
Grade PID deviation controller corrects system input and output, has then done corresponding feedforward compensation and saturation weight distribution etc. and has improved
Method, specific structure are as shown in Figure 1；Wherein gravity feedforward is to support to the gravity being accumulated in height control subsystem
Disappear and adds；Posture feedforward is to be coupled to the posture triangle in height control to offset the attitude angle due to unmanned plane itself
Function product item, and since coupling influence of the posture to height control is continuous, and the measured value of posture is discrete, meeting
Certain hysteresis quality is generated, so estimating by Taylor expansion the attitude value in next discrete periodic, reduces it
Hysteresis quality；And weight distributor is saturated primarily to processing has saturated characteristic, three posture angle increments because of propeller lift
It is superimposed between height gain and the problem of invalid increment occurs；
The present invention proposes a kind of layering unmanned plane tracking and controlling method based on feedforward and weight distribution, real in the steps below
It applies:
(1) structural parameters of unmanned plane, including unmanned plane quality, acceleration of gravity, horn length, tri- sides xyz are obtained
Upward unmanned plane rotary inertia；Obtaining unmanned plane propeller can be provided the range of lift；Unmanned plane itself is obtained in real time
The velocity vector and displacement vector of Eulerian angles posture information, elevation information and tracked target；
By taking DJI M100 unmanned plane development platform as an example, gross mass m=3.2868kg, horn length l=0.325m turn
Dynamic inertia Jx=0.05503kg*m2, Jy=0.05503kg*m2, Jz=0.097175kg*m2, propeller maximum lift Fmax=
21N, the minimum lift that can be provided should be 0, but to avoid stalling, be set as Fmin=0.2N；The posture of unmanned plane itself
Information is obtained by inertial measurement component, and elevation information can be measured by barometer, ultrasonic wave, visual sensor etc.；By with
The velocity vector and displacement vector of track target can be measured by unmanned plane by visual sensor or be passed through by tracked target logical
The mode of letter sends information to unmanned plane；
Simulation result diagram of the invention, emulates by taking above-mentioned parameter as an example and is obtained；
(2) newton Euler's model is established, the lift tange that can be provided according to propeller is special to addition saturation in model
Property；Newton Euler's model of foundation is as follows:
Wherein, m is the quality of unmanned plane, θ,ψ is tri- Eulerian angles of F --- the pitching of UAV Attitude description respectively
Angle, roll angle and yaw angle；Sx、Sy、SzIt is the displacement on three change in coordinate axis direction of earth coordinates respectively；F1、F2、F3、F4Point
It is not the lift that four propellers generate, M1、M2、M3、M4It is the torque that four propellers generate, Jx、Jy、JzIt is unmanned plane around three
The rotary inertia of a reference axis, g are acceleration of gravity, and l is horn length of the propeller center to drone center of unmanned aerial vehicle；It can prove
It obtains: Mn=kFn, n=1,2,3,4, wherein k is related with propeller machine structure and horn length l, does not need to obtain quasi-
True occurrence can be adjusted in subsequent yaw angle control subsystem by pid parameter；So the 6th equation of formula (8) can
To become:
Then propeller saturated characteristic is added in model, due to there is 12 kinds of suitable rule when Eulerian angles describe posture,
But no matter using which kind of suitable rule, the first two equation of newton Euler model is all only influenced, four equations after not being related to；Therefore, originally
Invention be related to lower layer's posture and the model foundation of height controller is as follows:
Wherein FmaxIt is the maximum lift that single propeller can be provided, FminIt is to maintain the minimum liter that propeller does not stall
Power；
(3) it according to the model put up, is controlled in pitch angle, roll angle control, yaw angle control and height control four
Basic two close cycles cascade PID deviation controller is established in subsystem, and is independently adjusted in the case where not considering coupling temporarily
Save parameter；Specific two close cycles cascade PID deviation controller is built such as Fig. 1, and the tune ginseng problem of PID controller is no longer tired herein
It states；
(4) feedforward of addition gravity, posture feedover and by Taylor expansion come in next discrete periodic to unmanned plane
Attitude value estimated；
It can be seen from formula (11) in high kinetics model, the second-order differential amount of heightWith control amount F1、
F2、F3、F4In non-linear relation, will receive the cumulative item of a gravity and posture integer item influences, and forms coupling；
So the present invention compensates counteracting to two nonlinear disturbance items in such a way that gravity feedovers and posture feedovers,
So that system linearization, carries out feedback control after linearisation, as shown in Figure 1, concrete mode by the way of cascade PID again
Are as follows:
Firstly for lower layer's posture and four subsystems of height control, mechanical superposition model are as follows:
Fh,Fθ, FψIt is the increment of height controller and the output of three posture angle controllers respectively；
1. adding up a gravity feedforward amount F in input terminalbase, whereinMake Fn'=Fn+Fbase, n=1,2,3,
4；Fn' it is the propeller lift being added after gravity feedforward, substitute original Fn, and since the controlling increment that three attitude angles provide is
Increase by two lift simultaneously and reduce by two lift simultaneously, does not influence F1+F2+F3+F4Value, then:
2. the pitch angle and roll angle returned by attitude controller system feedback, the increment given by height controller
Signal does a feedforward product term to offset posture in formula (13) and couple to the product of height control；In order not to change Fbase's
Value, by FbaseThe product compensation needed is added to FhIn the middle, it obtains:
Fh' it is the height controller increment being added after posture feedforward, substitute original Fh, it finally obtains:
By formula (15) it is found that after overcompensation, the second-order differential of heightAnd former deviation controller increment FhIt is linear to close
System；
3. since the posture in formula (13) is continuous to the coupling of height, and actual controller and gesture feedback are
Discrete, so the pose compensation directly obtained, there are certain hysteresis quality, delay degree enhances with the increase of discrete periodic；Ifθt、It is using time t as the pitch angle of parameter and roll angle；Discrete periodic is T, a cycle
Interior initial time is t0, the end cycle moment is t1, from Lagrange mean value theorem: certainly existing a t0To t1It
Between certain moment tε, make:
So if g (t can be found outε), i.e., the discrete value g (t of alternative zero-order holder0), it is t when due to calculating0When
It carves, only g (t0) and cycle T be currently determining known quantity, so needing to g (t1) value and g (t) function estimated；
Again because the first differential angulation speed of attitude angle can measure to obtain, second-order differential amount can be counted by formula (11)
It acquires, it is possible to come using Taylor series to g (t1) estimated；
It is approximately again linear function by g (t), then is obtained by formula (16):
It can be obtained by formula (17) and formula (18):
Pose compensation value g (t has been acquired as a result,ε)；
(5) in pitch angle control subsystem and roll angle control subsystem addition saturation weight distributor, with eliminate by
In the invalid increment that propeller saturated characteristic generates；
It can be seen from formula (11) for three Euler's angle movement models, we can be by controlling its four
The lift difference of the pairs of youngster of propeller constructs controller, system can be made to become linear second-order system, but due to three attitude angle controls
The increment signal of device and height controller processed is to be superimposed upon power F1、F2、F3、F4On, it is contemplated that propeller can be provided lift tool
There is saturated characteristic, so if superposition without restriction, it is possible that following situation:
The increment signal of single attitude angle control, should make F1、F2、F3、F4Two power in the middle increase, and two power are reduced,
But due to the saturated characteristic of the effect of other signals and power itself, make:
1. increased amount makes propeller lift reach FmaxAnd can not continue growing, the amount of reduction can but make propeller
Lift is effectively reduced；
2. the amount of reduction makes propeller lift reach FminAnd can not continue to reduce, increased amount can but make propeller
Lift effectively increases；
This will lead to: posture increment signal has half or less than half is invalid, to reduce attitude regulation effect；
Meanwhile causing posture increment signal to F1+F2+F3+F4Value produce influence, be equivalent in height control and be added to one
Disturbing signal；So inner ring of the present invention in attitude controller, that is, angular speed ring, a saturation weight distributor is added,
Distribution is weighted to the increment for making propeller lift reach saturation, inhibits the generation of invalid increment；
Specifically: the angular speed increment signal of pitch angle and roll angle is input in saturation weight distributor, simultaneously
The increment signal of height control is also entered into saturation weight distributor, after the output distribution of saturation weight distributor
Pitch angle increment signal and roll angle increment signal；Its method of salary distribution are as follows: the preferential increment for guaranteeing height controller is unrestricted, but
Will there are certain attitude angle adjust allowance.Then in remaining incremental range, allowance is adjusted retaining certain yaw angle
In the case where, if pitching angle increment and rolling angle increment will not make the power of any one propeller reach FmaxOr Fmin, then keep
Initial value output, if going beyond the scope, makes pitching angle increment FθWith rolling angle incrementIn the case where keeping former symbol, make:
Wherein, FlimitIt is the maximum increment absolute value for making propeller reach saturation, FψlIt is that yaw angle is made to keep balance institute
There are adjusting allowance；F′θWithIt is then the output valve after supersaturated weight distributor；
So far, lower layer's posture and height controller design finish；Fig. 2 is to add gravity in the case that attitude disturbance is zero
Height before and after feedforward controls simulation comparison result figure, it can be seen that control effect after the feedforward of addition gravity in (2-b)
Be improved significantly；Fig. 3 be have gravity feedforward on the basis of addition posture feedforward before and after height control emulation
Comparing result figure, it can be seen that in (3-b) after addition posture feedforward, the drop of the output due to caused by coupling of the posture to height
It is low obviously to be offset and improved；Fig. 4 is the continuous signal of the posture Product Perturbation item of height control, the discrete letter of zero-order holder
Number and the discrete signal figure estimated of Taylor expansion, figure (4-a) be the continuous signal and zero-order holder of attitude disturbance
Discrete signal, figure (4-b) are the continuous signal of attitude disturbance and the discrete signal of Taylor series estimation, it can be clearly seen that, estimate
Evaluation can more represent former continuous signal in one cycle, and the discrete sluggishness on system for reducing signal influences；Fig. 5 is that have weight
Distributor and without weight distributor height control system rise time simulation result comparison；It is obvious to divide it can be seen that weight is added
After orchestration, height controls the disturbance due to having lacked posture validation increment, can reach stable state faster, chooses the rise time
It is because, according to output characteristic curve, overshoot is little in height control system as index, less, the rise time is more for concussion
Its dynamic property can be represented；
(6) be input with the motion information of tracked target according to the posture angular limit of unmanned plane, establish upper layer when
Between optimal track control device, be connected on four subsystems of lower layer's posture and height control, it is final to realize tenacious tracking
Control amount selected first, unmanned plane during flying are the horizontal acceleration provided by the inclination of three Eulerian angles, institute
With there are four types of the control modes that can choose: 1. controlling pitch angle+roll angle+yaw angle；2. controlling pitch angle+roll angle；③
Control pitch angle+yaw angle；4. controlling roll angle+yaw angle；It is each again because Eulerian angles attitude description has 12 kinds of suitable rule
The formula that kind can all cause unmanned plane to be displaced along rule changes, so selecting control amount appropriate and Eulerian angles appropriate suitable
Rule can simplify unmanned plane horizontal displacement model；The first two equation in formula (9) is unmanned plane horizontal position shifting formwork of the XYZ along rule
Type；
The present invention selects control mode 2. with ZYX along rule, then has:
Then, unmanned plane horizontal displacement model is simplified, our control amounts should select to bow it can be seen from formula (21)
Elevation angle theta and roll angleAnd can not image height degree control equally integer item is offset；Again because control target is solid
Surely it is arranged in height and mobile surface targets is tracked, so hasThat is:
Then formula (21) becomes:
The purpose of design of tracking control unit is to make unmanned plane in the sky finally can position is unbiased and the unbiased tracking of speed
Move up moving-target, that is to say, that, it is desirable that meet when control final states:
Due to not requiring energy consumption etc., so Con trolling index is system regulating time, optimal controller is designed accordingly, is obtained
It is optimal outWithOptimal attitude value is found out by formula (23) again, is concatenated on lower layer's tracking and attitude controller；In this way
The advantages of doing is will be to very intractable nonlinearity trigonometric function item in control system, by layering, in upper and lower level controller
Between carry out trigonometric function calculating, to make upper and lower level controller itself all without trigonometric function item, become linear system；
Initially set up state equation:
Wherein u1It is the acceleration on the direction x, i.e.,u2It is the acceleration on the direction y, i.e.,Then settling time is optimal
Performance indicator functional J:
tfIt is the system final states time, the parametric equation on the direction x and y is obtained according to the plane motion function of tracking target:
Wherein a is the relative speed difference on the direction x, and b is that the relative displacement on the direction x is poor；C is relatively fast on the direction y
It is poor to spend, and d is that the relative displacement on the direction y is poor；Since the direction x is identical with the direction y, so only proving the direction x, the direction y is similarly
It can；Construct the Hamilton's equation on the direction x:
H=1+ λ1x2+λ2u1 (28)
λ1And λ2It is Lagrange multiplier, is solved according to regular equation:
c1、c2All it is undetermined constant value, is solved further according to transversality condition:
c1=-c1*t+c2 (30)
It is available thus according to minimal principle:
Max_x is the mould being most worth inputted on the direction tracking control unit x, and max_y is that tracking control unit is defeated in y-direction
The mould being most worth entered is the acceleration extreme value limited by attitude angle in unmanned plane tracking system；It can be derived by formula (32)
Out, this optimum control meets bang-bang control law, and because the input that can be obtained when unmanned plane during flying is every when being per
Relative velocity and the relative position of itself and tracking target are carved, problem becomes a dynamic programming problems；The very corn of a subject becomes
Solving equations:
It solves:
Similarly, also there is solution on the direction y:
Max_y is the mould being most worth that tracking control unit inputs in y-direction, is by attitude angle in unmanned plane tracking system
The acceleration extreme value of limitation；Time optimal controller is constructed by formula (33), formula (34), the posture of perfect condition responds,
Non-delay no concussion, and it is any controllable；The simulation result of perfect condition posture response is as shown in fig. 6, be the tracking characteristics of x respectively
Curve, y tracking characteristics curve, and comprehensive tracking characteristics curve；Because reference axis all contains time shaft t, it is possible to find out,
System can achieve convergence, that is, realize that the position of tracking is unbiased and speed is unbiased；
In systems in practice, ideal attitude characteristic is not present, practical attitude characteristic has delay to have overshoot, bang-bang
Control can not make system reach stable, so when close to final states, it should and replace bang-bang to control with deviation controller,
Make system convergence, in the present invention, this attitude characteristic is the input-output characteristic of lower layer's posture and height control system；It will be upper
Layer tracking control unit and lower layer's posture and height controller are concatenated, then complete all controller designs of the invention, most
Whole simulated effect is as shown in Figure 7, it can be seen that unmanned plane is when to ground movable object tracking, after shaking several times most
System can be made to reach convergence eventually, realize that the position of tracking is unbiased and speed is unbiased.
Claims (5)
1. a kind of layering unmanned plane tracking and controlling method based on feedforward and weight distribution, it is characterised in that the following steps are included:
(1) structural parameters for obtaining unmanned plane, including unmanned plane quality, acceleration of gravity, electrical axis and drone center of unmanned aerial vehicle
Unmanned plane rotary inertia in distance, tri- directions xyz；Obtaining unmanned plane propeller can be provided the range of lift；It obtains in real time
Take the Eulerian angles posture information of unmanned plane itself, the velocity vector and displacement vector of elevation information and tracked target；
(2) newton Euler's model is established according to the obtained unmanned plane parameter information of step (1), can be provided according to propeller
Lift tange adds saturated characteristic to model in the middle；
(3) it according to the model put up, is controlled in pitch angle, roll angle control, yaw angle control and height control four subsystems
Basic two close cycles cascade PID deviation controller, and the separately adjustable ginseng in the case where not considering coupling temporarily are established in system
Number；
(4) according to unmanned plane weight parameter and UAV Attitude information, gravity feedforward and posture feedforward are added, by posture and gravity
Coupling regards the disturbance of height control as to offset, while by Taylor expansion come in next discrete periodic to unmanned plane
Attitude value is estimated；
(5) the addition saturation weight distributor in pitch angle control subsystem and roll angle control subsystem, to eliminate due to spiral shell
The invalid increment for revolving paddle saturated characteristic and generating；
(6) according to the posture angular limit of unmanned plane, it is input with the motion information of tracked target, establishes the time optimal on upper layer
Tracking control unit is connected on four subsystems of lower layer's posture and height control, finally realizes tenacious tracking.
2. a kind of layering unmanned plane tracking and controlling method based on feedforward and weight distribution, it is characterised in that: the step (2)
In, on the basis of basic newton Euler's mechanical model, it is added to the saturated characteristic of lift: Fmin< F1、F2、F3、F4< Fmax,
Wherein F1、F2、F3、F4It is the lift that four propellers generate, FmaxIt is the maximum lift that single propeller can be provided, FminIt is
The minimum lift for keeping propeller not stall.
3. a kind of layering unmanned plane tracking and controlling method based on feedforward and weight distribution, it is characterised in that: the step (4)
In:
1. the gravity feedforward added meets:
F′n=Fn+Fbgge, n=1,2,3,4 (1)
Wherein Fn' it is the propeller lift being added after gravity feedforward, substitute original Fn, m is unmanned plane itself gross mass, and g is gravity
Acceleration, FbggeIt is gravity feedforward；
2. the posture feedforward added meets:
Wherein FhIt is the inner loop control increment of height control subsystem, Fh' it is that the height controller being added after posture feedovers increases
Amount substitutes original Fh；
3. since posture is continuous to the coupling of height, and actual controller and gesture feedback are in height subsystem
Discrete, so the pose compensation directly obtained, there are certain hysteresis quality, delay degree enhances with the increase of discrete periodic；Institute
To be estimated using Taylor expansion the attitude disturbance in a next discrete periodic in control process:
Wherein θt、It is using time t as the pitch angle of parameter and roll angle；Discrete periodic is T, the initial time in a cycle
For t0, the end cycle moment is t1, g (tε) be estimation after pose compensation value.
4. it is a kind of based on feedforward and weight distribution layering unmanned plane tracking and controlling method, it is characterised in that: the step (5) when
In, using the inner ring in pitch angle and roll angle serials control --- angular speed ring adds a saturation weight distributor come to control
Increment processed is allocated, and saturation weight distributor is using pitch angle controlling increment, roll angle controlling increment and height gain as defeated
Enter, using distribute after pitch angle controlling increment and roll angle can control increment as export；If pitching angle increment and roll angle increase
Amount will not make the lift of any one propeller reach saturation, then keep initial value to export, if going beyond the scope, increase pitch angle
Measure FθWith rolling angle incrementIn the case where keeping former symbol, make:
Wherein, FlimitIt is the maximum increment absolute value for making propeller reach saturation, FψlBe make yaw angle keep balance institute there are
Adjusting allowance；F′θWithIt is then the output valve after supersaturated weight distributor.
5. it is a kind of based on feedforward and weight distribution layering unmanned plane tracking and controlling method, it is characterised in that: the step (6) when
In,
1. choosing the pitching angle theta and roll angle in attitude angleAs the control amount moved horizontally, Eulerian angles attitude description is chosen
ZYX is along rule；
2. being decomposed into the direction x and the side y using the relative position and relative velocity of tracked target and unmanned plane itself as input quantity
To:
A is the relative velocity on the direction x in parametric equation, and b is the relative position on the direction x, and c is the relative velocity on the direction y,
D is the relative position on the direction y；
3. the timing-optimal control of system is bang-bang control, as a result through deriving are as follows:
Max_x is the mould being most worth inputted on the direction tracking control unit x, and max_y is that tracking control unit inputs in y-direction
The mould being most worth is the acceleration extreme value limited by attitude angle in unmanned plane tracking system；
4. since bang-bang control can generate concussion in steady-state value, so this tracking control unit is close under stable situation,
That is the relative position of unmanned plane itself and tracked target and when smaller relative velocity, is switched to deviation control mode, so as to be
System is restrained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910013466.8A CN109857146B (en) | 2019-01-07 | 2019-01-07 | Layered unmanned aerial vehicle tracking control method based on feedforward and weight distribution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910013466.8A CN109857146B (en) | 2019-01-07 | 2019-01-07 | Layered unmanned aerial vehicle tracking control method based on feedforward and weight distribution |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109857146A true CN109857146A (en) | 2019-06-07 |
| CN109857146B CN109857146B (en) | 2021-10-15 |
Family
ID=66894080
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910013466.8A Active CN109857146B (en) | 2019-01-07 | 2019-01-07 | Layered unmanned aerial vehicle tracking control method based on feedforward and weight distribution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109857146B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110609568A (en) * | 2019-09-24 | 2019-12-24 | 广东名阳信息科技有限公司 | Strong self-coupling PI cooperative control method for large unmanned aerial vehicle UAV |
| CN112286224A (en) * | 2020-10-23 | 2021-01-29 | 中国有色金属长沙勘察设计研究院有限公司 | Method for realizing accurate autonomous take-off and landing of unmanned airport rotor aircraft |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080203230A1 (en) * | 2007-02-07 | 2008-08-28 | Keita Ogo | Attitude change control method, attitude change control system, attitude change control program and program recording medium |
| WO2015180171A1 (en) * | 2014-05-30 | 2015-12-03 | SZ DJI Technology Co., Ltd. | Aircraft attitude control methods |
| US20160210863A1 (en) * | 2015-01-19 | 2016-07-21 | The Aerospace Corporation | Autonomous nap-of-the-earth (anoe) flight path planning for manned and unmanned rotorcraft |
| CN106200665A (en) * | 2016-08-25 | 2016-12-07 | 东北大学 | Carry modeling and the self-adaptation control method of the four-axle aircraft of uncertain load |
| CN106681344A (en) * | 2016-12-26 | 2017-05-17 | 湖南纳雷科技有限公司 | Height control method and height control system for aerial vehicle |
| CN108845588A (en) * | 2018-06-22 | 2018-11-20 | 哈尔滨工业大学 | A kind of quadrotor Trajectory Tracking Control method based on Nonlinear Guidance |
| CN109062237A (en) * | 2018-09-17 | 2018-12-21 | 南京航空航天大学 | A kind of unmanned tiltrotor active disturbance rejection attitude control method |
-
2019
- 2019-01-07 CN CN201910013466.8A patent/CN109857146B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080203230A1 (en) * | 2007-02-07 | 2008-08-28 | Keita Ogo | Attitude change control method, attitude change control system, attitude change control program and program recording medium |
| WO2015180171A1 (en) * | 2014-05-30 | 2015-12-03 | SZ DJI Technology Co., Ltd. | Aircraft attitude control methods |
| US20160210863A1 (en) * | 2015-01-19 | 2016-07-21 | The Aerospace Corporation | Autonomous nap-of-the-earth (anoe) flight path planning for manned and unmanned rotorcraft |
| CN106200665A (en) * | 2016-08-25 | 2016-12-07 | 东北大学 | Carry modeling and the self-adaptation control method of the four-axle aircraft of uncertain load |
| CN106681344A (en) * | 2016-12-26 | 2017-05-17 | 湖南纳雷科技有限公司 | Height control method and height control system for aerial vehicle |
| CN108845588A (en) * | 2018-06-22 | 2018-11-20 | 哈尔滨工业大学 | A kind of quadrotor Trajectory Tracking Control method based on Nonlinear Guidance |
| CN109062237A (en) * | 2018-09-17 | 2018-12-21 | 南京航空航天大学 | A kind of unmanned tiltrotor active disturbance rejection attitude control method |
Non-Patent Citations (8)
| Title |
|---|
| LUCIO R. SALINAS 等: "Complete bilateral teleoperation system for a rotorcraft uav with time-varying delay", 《MATHEMATICAL PROBLEMS IN ENGINEERING》 * |
| 任丽莉 等: "基于图像处理的四旋翼自主跟踪智能车设计", 《长春师范大学学报》 * |
| 刘富 等: "基于视觉的四旋翼飞行器稳定飞行控制", 《吉林大学学报(信息科学版)》 * |
| 康冰 等: "四旋翼飞行器携带负载稳定飞行算法", 《吉林大学学报（工学版）》 * |
| 张纪敏: "面向空中-水面协作的自主起降系统设计及控制", 《万方数据知识服务平台》 * |
| 王向磊: "四旋翼无人机自抗扰飞行控制研究", 《万方数据知识服务平台》 * |
| 钱杰: "四旋翼飞行器整体设计与航迹跟踪地面站软件设计", 《万方数据知识服务平台》 * |
| 龙文 等: "变载荷下六旋翼无人机高度控制", 《传感技术学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110609568A (en) * | 2019-09-24 | 2019-12-24 | 广东名阳信息科技有限公司 | Strong self-coupling PI cooperative control method for large unmanned aerial vehicle UAV |
| CN110609568B (en) * | 2019-09-24 | 2021-01-05 | 广东名阳信息科技有限公司 | Strong self-coupling PI cooperative control method for large unmanned aerial vehicle UAV |
| CN112286224A (en) * | 2020-10-23 | 2021-01-29 | 中国有色金属长沙勘察设计研究院有限公司 | Method for realizing accurate autonomous take-off and landing of unmanned airport rotor aircraft |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109857146B (en) | 2021-10-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107844123B (en) | Nonlinear aircraft track control method | |
| CN107479567B (en) | The unknown quadrotor drone attitude controller of dynamic characteristic and method | |
| CN102073755B (en) | Motion control simulation method for near-space hypersonic aircraft | |
| Santos et al. | Intelligent fuzzy controller of a quadrotor | |
| Tang et al. | Nonlinear dynamic modeling and hybrid control design with dynamic compensator for a small-scale UAV quadrotor | |
| CN109465825A (en) | The adaptive dynamic surface control method of the RBF neural of mechanical arm flexible joint | |
| CN111399531B (en) | Hypersonic aircraft glide section guidance and attitude control integrated design method | |
| CN109725644A (en) | A kind of hypersonic aircraft linear optimization control method | |
| CN106843245A (en) | A kind of UAV Attitude control method, device and unmanned plane | |
| CN106527122B (en) | Stratospheric airship height-lock control nonlinear PID controller method | |
| CN106200665A (en) | Carry modeling and the self-adaptation control method of the four-axle aircraft of uncertain load | |
| CN112578805B (en) | Attitude control method of rotor craft | |
| CN103869817A (en) | Vertical take-off and landing control method for quad-tilt-rotor unmanned aerial vehicle | |
| CN108638068A (en) | A kind of flying robot's Control System Design method carrying redundancy mechanical arm | |
| CN111258216A (en) | Sliding mode repetitive controller suitable for four-rotor aircraft | |
| CN105425812A (en) | Unmanned aerial vehicle automatic landing locus control method based on double models | |
| Jiao et al. | Analysis and design the controller for quadrotors based on PID control method | |
| Kamanditya et al. | Elman recurrent neural networks based direct inverse control for quadrotor attitude and altitude control | |
| CN106896722A (en) | Adoption status feeds back the hypersonic vehicle composite control method with neutral net | |
| CN109857146A (en) | A kind of layering unmanned plane tracking and controlling method based on feedforward and weight distribution | |
| CN106774385B (en) | A kind of dirigible spot hover control method using adaptive variable structure | |
| CN106200383B (en) | A kind of three axis Inertially-stabilizeplatform platform control methods based on model reference adaptive neural network | |
| CN108445753B (en) | Method for balancing lateral force and restraining lateral reverse thrust of unmanned aerial vehicle | |
| CN108459611B (en) | Attitude tracking control method of near space vehicle | |
| CN110134135B (en) | Four-rotor aircraft control method based on improved MPC-PID |
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 |