CN108196573A - A kind of unmanned plane on-line identification and control method - Google Patents
A kind of unmanned plane on-line identification and control method Download PDFInfo
- Publication number
- CN108196573A CN108196573A CN201711498981.7A CN201711498981A CN108196573A CN 108196573 A CN108196573 A CN 108196573A CN 201711498981 A CN201711498981 A CN 201711498981A CN 108196573 A CN108196573 A CN 108196573A
- Authority
- CN
- China
- Prior art keywords
- unmanned plane
- control
- cloud server
- mathematical model
- swept
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000013178 mathematical model Methods 0.000 claims abstract description 50
- 230000004044 response Effects 0.000 claims abstract description 47
- 238000005086 pumping Methods 0.000 claims abstract description 10
- 238000001228 spectrum Methods 0.000 claims abstract description 5
- 238000004891 communication Methods 0.000 claims description 24
- 230000002068 genetic effect Effects 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000005096 rolling process Methods 0.000 claims description 7
- 230000001133 acceleration Effects 0.000 claims description 3
- 239000012141 concentrate Substances 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 2
- 238000012850 discrimination method Methods 0.000 claims description 2
- 230000006641 stabilisation Effects 0.000 claims description 2
- 238000011105 stabilization Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 5
- 238000005457 optimization Methods 0.000 abstract description 5
- 241000208340 Araliaceae Species 0.000 description 3
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 3
- 235000003140 Panax quinquefolius Nutrition 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000005284 excitation Effects 0.000 description 3
- 235000008434 ginseng Nutrition 0.000 description 3
- 238000011217 control strategy Methods 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a kind of unmanned plane on-line identification control methods, including 1), ground control station is according to the corresponding swept-frequency signal of state output of current unmanned plane, unmanned plane receives swept-frequency signal, and as the pumping signal of on-line identification, flight controller responds the swept-frequency signal received；2) it after, cloud server receives response data, is combined using FFT and window method is divided to carry out frequency-domain analysis to response data, obtained system frequency domain response and coherence spectrum, obtain unmanned plane mathematical model, calculate the control parameter of suitable unmanned plane current flight state；3), control parameter is returned to flight controller by cloud server again, and flight controller is precisely controlled unmanned plane according to control parameter.This on-line identification method can provide more accurate mathematical model for controlled device unmanned plane, according to the variation of external environment, and carry out real-time optimization to the control parameter of unmanned plane, reach better unmanned aerial vehicle (UAV) control effect and lower energy expenditure.
Description
Technical field
The present invention relates to a kind of unmanned plane on-line identification and control methods.
Background technology
As unmanned air vehicle technique is more and more ripe, the application field of unmanned plane is also more and more extensive, removes consumer level within nearly 2 years
Other than unmanned plane is hot, unmanned plane also obtains broad development in other many industries, and such as logistics unmanned plane, is detectd unmanned plane of taking photo by plane
Look into unmanned plane etc..
The key of unmanned plane during flying is accurately to control, and unmanned plane is precisely operated, and needs the standard of the standard and control measured.
If accurate unmanned plane model can be obtained, for current PID controller, on the basis for obtaining accurate unmanned plane model
On can calculate better control parameter, so as to achieve the purpose that precise control.
Existing unmanned aerial vehicle (UAV) control strategy is mostly based on pid control algorithm, and the acquisition of parameter needs not in default of model
It is disconnected to adjust ginseng, when large-scale in application, the model of each controlled device is different, and is controlled in the various process of work
The model of object can also change, and using same set of control parameter the control effect of all controlled devices can not reach
More satisfactory situation.
Invention content
The purpose of the present invention is to provide a kind of unmanned plane on-line identification and control methods, solve in the prior art due to nothing
The man-machine equipment computing capability of itself, the technical issues of on-line identification can not be carried out, cause unmanned plane that can not be precisely controlled.
To solve the above-mentioned problems, the present invention adopts the following technical scheme that：
A kind of unmanned plane on-line identification control method, unmanned plane on-line identification control system include ground control station, high in the clouds
Server, unmanned plane ontology and the flight controller being arranged on unmanned plane ontology, flight controller control unmanned plane during flying, fly
Module communicates line control unit with cloud server, ground control station by radio communication；Ground control station can supervise in real time
Control the state of unmanned plane；
The method for being carried out on-line identification control to unmanned plane using ground control station, cloud server, is included the following steps：
1), according to the corresponding swept-frequency signal of state output of current unmanned plane, unmanned plane receives frequency sweep to be believed ground control station
Number, as the pumping signal of on-line identification, flight controller responds the swept-frequency signal received, to unmanned plane during flying into
Row adjusts posture, is provided with harvester on unmanned plane ontology, unmanned plane response data is acquired, and by radio communication
Module is transferred to cloud server；
2) after, cloud server receives response data, after obtaining unmanned plane mathematical model using corresponding algorithm, high in the clouds
Server is required according to flight, with reference to corresponding control method, calculates the control parameter of suitable unmanned plane current flight；
3), control parameter is returned to flight controller by cloud server again, and flight controller is according to control parameter to nothing
It is man-machine to be precisely controlled.
Mostly based on pid control algorithm, the acquisition of parameter needs constantly to adjust ginseng in default of model existing control strategy,
When large-scale in application, the model of each controlled device is different, and opposite direction is controlled in the various process of work
Model can also change, and using same set of control parameter the control effect of all controlled devices can not reach and compare reason
The situation thought.And on-line identification, more accurate unmanned plane mathematical model can be provided for controlled device unmanned plane, according to extraneous ring
The variation in border, and real-time optimization is carried out to the control parameter of unmanned plane, reach the energy that preferably control unmanned plane effect is lower
Consumption.
The computing system of cloud server, to analyzing unmanned plane response data, picks out nothing according to built-in algorithm
The man-machine mathematical model before unmanned plane, is then based on "current" model, is required with reference to control, choose suitable control method, right
The control parameter of unmanned plane optimizes, and calculates one group of optimal control parameter, and unmanned plane is returned to by cloud system, flight
Controller is precisely controlled unmanned plane according to the control parameter of optimization.
The model that on-line identification goes out can improve the precision of control, so as to expand the application range of unmanned machine, such as
Unmanned plane can be controlled to enter narrow operating area.Unmanned plane during flying stability is improved, improves unmanned plane continuation of the journey, it is suitable to control
Parameter processed can reduce oscillation, reduce energy expenditure.
It is further improved, the unmanned plane responds swept-frequency signal, i.e. pitch angle, roll angle and boat by changing unmanned plane
The posture of unmanned plane, position and speed are adjusted to angle；
The harvester includes three groups of 3-axis acceleration sensors, three groups of three-axis gyroscopes, three groups of magnetic sensors
With three groups of GNSS locating modules (if only one group, this group of sensor goes wrong, and aircraft is with regard to uncontrollable), harvester pair
The real-time response data of unmanned plane are acquired, and module is transferred to cloud server by radio communication；
Harvester acquires the posture response data information of current unmanned plane pitch angle, roll angle and course angle, and passes through
Wireless communication module is transferred to cloud server；Cloud server receives current unmanned plane pitch angle, roll angle and course angle
Posture response data after, obtain unmanned plane mathematical model using corresponding algorithm.
Pitch angle, roll angle and course angle are to influence the most important parameter of unmanned plane during flying posture, and unmanned plane
Position and speed are also by the way that these three attitude angles is controlled to be controlled, so nobody can be ensured by controlling three parameters
Machine carries out accurate, stabilized flight.
It is further improved, the unmanned plane is divided into three phases in flight course：Takeoff phase, steady state phase and
Landing phases；The swept-frequency signal of three phases may be the same or different, and the unmanned plane mathematical model of three phases is different,
It is required according to flight, the control method used of three phases is different；
After picking out unmanned plane mathematical model, cloud server can be according to the requirement of flight, such as stability of control, control
Rapidity, the robustness of control, to select different control methods, in the control of unmanned plane, roll angle and pitch angle control
System belongs to horizontal attitude control, and course angle control belongs to perpendicular attitude control；
Takeoff phase needs emphatically to control the posture of the horizontal direction of unmanned plane, and the posture of horizontal direction is main
It is roll angle and pitch angle.So swept-frequency signal can concentrate on rolling and pitch channel, ignore course channel.This when because
To be highly not high, wind is disturbed smaller, it may be considered that using PID control method, mainly using PI controllers.
Steady state phase, the posture weight of horizontal direction is higher, and the gesture stability weight of vertical direction is relatively low.So it sweeps
The output of frequency signal is concentrated mainly on rolling and pitch channel, and cannot ignore course channel.Because steady state phase height
More than 100m, some unmanned planes can even reach 10000m, and in this height section, wind is disturbed than stronger, in order to keep
Stablize it is contemplated that using LQG (Linear-Quadratic-Gauss control), so as to reach relatively good interference free performance.
In landing phases, control focuses on horizontal attitude channel, i.e. roll angle and pitch angle.Swept-frequency signal is main
Concentrate on roll channel and pitch channel, it is impossible to ignore course channel.Because the requirement landed to horizontal attitude is harsher, institute
Hinf control method computing controllers can be considered, to reach accurate landing.
It is further improved, described in take-off process, swept-frequency signal when ground control station output unmanned plane takes off, nobody
Machine receives swept-frequency signal, and makes a response and start to take off, while the harvester on unmanned plane ontology is to the number of responses of unmanned plane
According to being acquired, then module is transferred to cloud server by radio communication；
After cloud server receives response data, the unmanned plane mathematical modulo of takeoff phase is obtained using corresponding algorithm
Then type is estimated the parameter of model using genetic algorithm, will be come out currently without man-machine digital's Model Distinguish, according to identification
The unmanned plane mathematical model gone out with reference to the control requirement of takeoff phase, selects suitable control method, calculates and suitably take off
Control parameter；Control parameter is returned to flight controller by cloud server again, and taking off for unmanned plane is precisely controlled.
It being further improved, the unmanned plane is in flight course, because external environment real-time change, such as height, wind speed
Variation, when particularly wind speed changes, be affected to the flight attitude of unmanned plane, so flight controller is every 1-
5min can detect whether mathematical model changes, and used detection method is specifically to detect excitation letter for one group to unmanned plane
Number, by response data with carrying out comparison comparison by the output that unmanned plane mathematical model is calculated：
1), if error is not above 15%, it is believed that unmanned plane mathematical model does not change, then with "current" model pair
The state modulator unmanned plane answered continues to fly；
2), if error is more than 15%, it is believed that unmanned plane mathematical model has occurred that variation, at this time ground control station root
According to the state output of current unmanned plane corresponding frequency sweep letter, unmanned plane receives swept-frequency signal, as the pumping signal of on-line identification,
Flight controller responds swept-frequency signal, and the response data of unmanned plane is acquired by harvester, then by wireless
Communication module is transferred to cloud server；
After cloud server receives response data, the unmanned plane mathematics of steady state phase is obtained using corresponding algorithm
Then model is estimated the parameter of model using genetic algorithm, current mathematics Model Distinguish is come out；According to what is picked out
Mathematical model with reference to the control requirement of steady state phase, selects suitable control method, calculates suitable stabilized flight rank
Section control parameter；Control parameter is returned to flight controller by cloud server again, the stabilized flight of unmanned plane is carried out accurate
Control.
It is further improved, the landing instruction for connecing unmanned plane and ground control station being received in flight course, and when confirmation
After information of landing, unmanned plane hovering, ground control station exports unmanned plane landing swept-frequency signal, and unmanned plane receives landing frequency sweep letter
Number, as the pumping signal of on-line identification, flight controller responds swept-frequency signal, and harvester is by the response of unmanned plane
Data are acquired, and then module is transferred to cloud server by radio communication；
After cloud server receives response data, the mathematical model of unmanned landing phases is obtained using corresponding algorithm,
Then the parameter of the mathematical model of landing phases is estimated using genetic algorithm, mathematical model is picked out to come, according to distinguishing
Know the mathematical model, with reference to the control requirement of landing phases, select suitable control method, recalculate suitable landing
Control parameter；Control parameter is returned to flight controller by cloud server again, and control unmanned plane is precisely landed.
It is further improved, the wireless communication module is 4G modules or WIFI module, realizes remote radio communication.
It is further improved, after the cloud server receives response data, unmanned plane number is obtained using corresponding algorithm
Model is learned, which includes following three kinds：
1), using the computing module built in cloud server, the number of unmanned plane is directly obtained using Least Square Recurrence method
Learn model；
2), being combined using FFT divides such identification algorithm of window method to carry out frequency-domain analysis to response data, obtains system frequency domain and rings
Should and coherence spectrum, obtain unmanned plane mathematical model, then the parameter of unmanned plane mathematical model estimated using genetic algorithm
Meter, picks out currently without man-machine digital's model；
3), according to the requirement of flight and type, different discrimination methods is selected to obtain unmanned plane mathematical model.
Compared with prior art, this programme has the advantages that：
1) on-line identification can provide more accurate mathematical model, according to external environment for controlled device unmanned plane
Variation, and real-time optimization is carried out to the control parameter of unmanned plane, reach the energy expenditure that preferably control unmanned plane effect is lower.
The computing system of cloud server, to analyzing unmanned plane response data, picks out working as unmanned plane according to built-in algorithm
Preceding mathematical model, is then based on "current" model, and the control parameter of unmanned plane is optimized, and calculates one group of optimum control ginseng
Number, returns to unmanned plane, flight controller is precisely controlled unmanned plane according to the control parameter of optimization by cloud system.
2), the model that on-line identification goes out can improve the precision of control, so as to expand the application range of unmanned machine,
For example unmanned plane can be controlled to enter narrow operating area.Improve the stability of unmanned plane.
3) continuation of the journey, is improved, suitable parameter can reduce oscillation, reduce energy expenditure.
Description of the drawings
Fig. 1 is the block diagram of unmanned plane on-line identification control system of the present invention.
Fig. 2 is the flow chart that the on-line identification of unmanned plane takeoff phase controls.
Fig. 3 is the flow chart that the on-line identification of unmanned plane steady state phase controls.
Fig. 4 is the flow chart that the on-line identification of unmanned plane landing phases controls.
Specific embodiment
To make the purpose of the present invention and technical solution clearer, with reference to the embodiment of the present invention to the technology of the present invention
Scheme carries out clear, complete description.
Embodiment one：
As shown in Figure 1, unmanned plane on-line identification control system includes ground control station, cloud server, unmanned plane ontology
With the flight controller being arranged on unmanned plane ontology, flight controller control unmanned plane during flying, flight controller passes through wireless
Communication module communicates with cloud server, ground control station；Ground control station can monitor the state of unmanned plane in real time；
As shown in Fig. 2, described in take-off process, swept-frequency signal when ground control station output unmanned plane takes off, this is swept
Frequency signal is input to rolling and pitch channel, it is contemplated that takeoff phase interference it is smaller, swept-frequency signal can be frequency 0.5Hz, width
It is worth the sinusoidal signal for 50, for normal remote control control signal between -100~+100, unmanned plane receives this swept-frequency signal, and
It makes a response and starts to take off, while the harvester on unmanned plane ontology is acquired the response data of unmanned plane, Ran Houtong
It crosses wireless communication module and is transferred to cloud server；
The response data received is directly obtained the mathematical modulo of unmanned plane by cloud server using Least Square Recurrence method
Type, takeoff phase rolling and pitch channel may be used second mathematical model and matched.
Wherein：G (s) is transmission function, and s is plural, kt1,kt2,kt3,kt4To refer to coefficient.
Takeoff phase height is relatively low, and air speed influence is small, considers to control aircraft using PID control method, with reference to distinguishing
Know the mathematical model, calculate suitable control parameter；Control parameter is returned to flight controller by cloud server again, right
Unmanned plane, which takes off, to be precisely controlled.
Described three groups of 3-axis acceleration sensors of harvester, three groups of three-axis gyroscopes, three groups of magnetic sensors and three
Group GNSS locating modules, are acquired unmanned plane response data, and module is transferred to cloud server by radio communication；
In the present embodiment, the wireless communication module is 4G modules.In other embodiments, wireless communication module is can
Think WIFI module.
Embodiment two：
As shown in figure 3, unmanned plane, during stabilized flight, whether flight controller can detect mathematical model every 3min
It changes, it is windy to disturb, the characteristics of frequency is high, and span of control limit of control is small is controlled, a frequency is inputted respectively in rolling and pitch channel
For 20Hz, amplitude is 10 sinusoidal excitation signal, then to the output signal that model calculates and collected real response
Data are compared, if error is more than 15%, it is believed that model has occurred that variation：
1), when model does not change, then continue to fly with the corresponding state modulator unmanned plane of "current" model；
2), when model changes, ground control station is believed according to the corresponding frequency sweep of state output of current unmanned plane, surely
The characteristics of swept-frequency signal for determining mission phase is disturbed according to stabilized flight is windy, and control frequency is high, and span of control limit of control is small, use frequency for
100Hz, for the sinusoidal signal that amplitude is 10 as pumping signal, unmanned plane receives swept-frequency signal, and the excitation as on-line identification is believed
Number, unmanned plane responds this swept-frequency signal, and the response data of unmanned plane is acquired by harvester, then by wireless
Communication module is transferred to cloud server；
Cloud server, which combines the data received using FFT, divides window method, carries out frequency-domain analysis to data, obtains system
Frequency domain response and coherence spectrum obtain the new mathematical model of unmanned plane current generation, this mathematical model is on the basis of second order
On, a time delay process has been superimposed, the parameter of new mathematical model has been estimated using genetic algorithm, by current new mathematics
Model Distinguish comes out, and needs wind disturbance resistance and the requirement stablized according to stabilized flight, it may be considered that with LQG theories into line control unit
Design, with reference to the model picked out, calculates one group of optimal control parameter；Cloud server again returns to new control parameter
Flight controller is precisely controlled unmanned plane stabilized flight according to new control parameter.
Wherein：G (s) is transmission function, and s is plural, kf1,kf2,kf3,kf4To refer to coefficient, e natural constants, τ0Representative is prolonged
When coefficient.
In the present embodiment, other parts are in the same manner as in Example 1.
Embodiment three：
As shown in figure 4, the unmanned plane receives the landing instruction of ground control station in flight course, and land when confirming
After information, ground control station is according to the characteristics of landing, output unmanned plane landing swept-frequency signal, when this swept-frequency signal is according to landing,
Frequency is 200Hz by the characteristics of attitude stabilization requires and control frequency is higher, and amplitude is 5 sinusoidal signal, as distinguishing online
The pumping signal of knowledge, unmanned plane respond this swept-frequency signal, and the response data of unmanned plane is acquired by harvester, so
Module is transferred to cloud server by radio communication afterwards；
Cloud server, which combines the data received using FFT, divides window method, carries out frequency-domain analysis to data, obtains system
Frequency domain response and coherence spectrum obtain the mathematical model of unmanned landing phases, this mathematical model is second-order model, is calculated using heredity
Method estimates the parameter of the mathematical model of landing phases, and mathematical model is picked out to come.High appearance is needed according to landing phases
The requirement of state stability, it may be considered that be the design with Hinf theories into line control unit, according to anti-interference and control stability
Requirement, then the weight function of reasonable design calculates one group of optimal control parameter；Cloud server again returns control parameter
To flight controller, control unmanned plane is precisely landed.
Wherein：G (s) is transmission function, and s is plural, kl1,kl2,kl3,kl4To refer to coefficient.
In the present embodiment, other parts are identical with embodiment.
Do not done in the present invention illustrate be the prior art or can be realized by the prior art, and the present invention
Described in specific implementation case be only the present invention preferable case study on implementation, not be used for limit the present invention practical range.
The equivalent changes and modifications that i.e. all contents according to scope of the present invention patent are made all should be used as the technology scope of the present invention.
Claims (8)
1. a kind of unmanned plane on-line identification and control method, which is characterized in that unmanned plane on-line identification includes ground with control system
Flight controller on face control station, cloud server, unmanned plane ontology and unmanned plane ontology, flight controller control unmanned plane
Flight, module communicates flight controller with cloud server, ground control station by radio communication；Ground control station can
Monitor the state of unmanned plane in real time；
The method for being carried out on-line identification and control to unmanned plane using ground control station, cloud server, is included the following steps：
1), ground control station receives swept-frequency signal according to the corresponding swept-frequency signal of state output of current unmanned plane, unmanned plane, makees
For the pumping signal of on-line identification, unmanned plane responds the frequency sweep control signal received, and unmanned plane during flying posture is continuous
Changing, the harvester set on unmanned plane ontology can be acquired the response data i.e. attitude data of unmanned plane, and
Module is transferred to cloud server by radio communication；
2) after, cloud server receives response data, after obtaining unmanned plane mathematical model using corresponding algorithm, cloud service
Device is required according to flight, with reference to corresponding control method, calculates the control parameter of suitable unmanned plane current flight；
3), control parameter is returned to flight controller by cloud server again, and flight controller is according to control parameter to unmanned plane
It is precisely controlled.
2. unmanned plane on-line identification control method according to claim 1, which is characterized in that the unmanned plane is to specific
Swept-frequency signal is responded, and pitch angle, roll angle and the course angle of unmanned plane change；
The harvester includes three groups of 3-axis acceleration sensors, three groups of three-axis gyroscopes, three groups of magnetic sensors and three
Group GNSS locating modules, harvester are acquired the real-time response data of unmanned plane, and module is transmitted by radio communication
To cloud server；
Harvester acquires the posture response data information of current unmanned plane pitch angle, roll angle and course angle, and passes through wireless
Communication module is transferred to cloud server；Cloud server receives the appearance of current unmanned plane pitch angle, roll angle and course angle
After state response data, unmanned plane mathematical model is obtained using corresponding algorithm.
3. unmanned plane on-line identification control method according to claim 1 or 2, which is characterized in that the unmanned plane is flying
It is divided into three phases during row：Takeoff phase, steady state phase and landing phases；The swept-frequency signal of three phases can phase
Together, it can also be different, the unmanned plane mathematical model of three phases is different, is required according to flight, the control used of three phases
Method is different；
In the control of unmanned plane, roll angle and pitch angle control belong to horizontal attitude control, and course angle control belongs to vertical appearance
State controls；
Takeoff phase needs emphatically to control the posture of the horizontal direction of unmanned plane, and the posture of horizontal direction is mainly rolled
Corner and pitch angle so swept-frequency signal can concentrate on rolling and pitch channel, ignore course channel, and control method uses
PID control method, mainly using PI controllers；
Steady state phase, the posture weight of horizontal direction is higher, and the gesture stability weight of vertical direction is relatively low, so frequency sweep is believed
Number output be concentrated mainly on rolling and pitch channel, but course channel cannot be ignored, use LQG (Linear-Quadratic-Gauss controls
System) control method；
In landing phases, control focuses on horizontal attitude channel, i.e. roll angle and pitch angle, and swept-frequency signal is mainly concentrated
In roll channel and pitch channel, it is impossible to ignore course channel, using Hinf control method computing controllers, to reach accurate drop
It falls.
4. unmanned plane on-line identification control method according to claim 4, which is characterized in that in the take-off process, ground
Swept-frequency signal when face control station output unmanned plane takes off, unmanned plane receives swept-frequency signal, and makes a response, while unmanned plane sheet
Harvester on body is acquired the response data of unmanned plane, and then response data is transferred to by module by radio communication
Cloud server；
After cloud server receives response data, the unmanned plane mathematical model of takeoff phase is obtained using corresponding algorithm, it will
It comes out currently without man-machine digital's Model Distinguish, according to the unmanned plane mathematical model picked out, is required with reference to the control of takeoff phase,
Suitable control method is selected, calculates suitable control parameter of taking off；Control parameter is returned to flight by cloud server again
Controller is precisely controlled taking off for unmanned plane.
5. unmanned plane on-line identification control method according to claim 4, which is characterized in that the unmanned plane flies stable
During row, flight controller can detect whether unmanned plane mathematical model changes every 1-5min, used detection side
Method is specifically to detect pumping signal for one group to unmanned plane, and response data is defeated with being calculated by unmanned plane mathematical model
Go out to be compared：
1), if error is not above 15%, it is believed that unmanned plane mathematical model does not change, then corresponding with "current" model
State modulator unmanned plane continues to fly；
If 2), error be more than 15%, it is believed that unmanned plane mathematical model has occurred that variation, at this time ground control station according to ought
The corresponding frequency sweep letter of state output of preceding unmanned plane, unmanned plane receives swept-frequency signal, as the pumping signal of on-line identification, flight
Controller responds swept-frequency signal, and the response data of unmanned plane is acquired by harvester, then by radio communication
Module is transferred to cloud server；
Cloud server obtains the unmanned plane mathematical modulo of stabilization sub stage by after the response data received using corresponding algorithm
Then type is estimated the parameter of model using genetic algorithm, current mathematics Model Distinguish is come out；According to the number picked out
Model is learned, with reference to the control requirement of steady state phase, suitable control method is selected, calculates suitable steady state phase
Control parameter；Control parameter is returned to flight controller by cloud server again, and the stabilized flight of unmanned plane is precisely controlled
System.
6. unmanned plane on-line identification control method according to claim 5, which is characterized in that the unmanned plane that connects is flying
The landing instruction of ground control station is received in the process, and after landing information is confirmed, ground control station output unmanned plane landing is swept
Frequency signal, unmanned plane receive landing swept-frequency signal, as the pumping signal of on-line identification, flight controller to swept-frequency signal into
The response data of unmanned plane is acquired by row response, harvester, and then module is transferred to cloud service by radio communication
Device；
After cloud server receives response data, the mathematical model of unmanned plane landing phases is obtained using corresponding algorithm, so
The parameter of the mathematical model of landing phases is estimated using genetic algorithm afterwards, mathematical model is picked out to come, according to identification
The mathematical model gone out with reference to the control requirement of landing phases, selects suitable control method, recalculates suitable landing control
Parameter processed；Control parameter is returned to flight controller by cloud server again, and control unmanned plane is precisely landed.
7. unmanned plane on-line identification control method according to claim 1, which is characterized in that the wireless communication module is
4G modules or WIFI module.
8. unmanned plane on-line identification control method according to claim 3, which is characterized in that the cloud server receives
To after response data, unmanned plane mathematical model is obtained using corresponding algorithm, which includes following three kinds：
1), using the computing module built in cloud server, the mathematical modulo of unmanned plane is directly obtained using Least Square Recurrence method
Type；
2), using FFT combine divide such identification algorithm of window method to response data carry out frequency-domain analysis, obtain system frequency domain response with
And coherence spectrum, unmanned plane mathematical model is obtained, then the parameter of unmanned plane mathematical model is estimated using genetic algorithm, is distinguished
Know and currently without man-machine digital's model；
3), according to the requirement of flight and type, different discrimination methods is selected to obtain unmanned plane mathematical model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711498981.7A CN108196573A (en) | 2017-12-29 | 2017-12-29 | A kind of unmanned plane on-line identification and control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711498981.7A CN108196573A (en) | 2017-12-29 | 2017-12-29 | A kind of unmanned plane on-line identification and control method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108196573A true CN108196573A (en) | 2018-06-22 |
Family
ID=62587839
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711498981.7A Pending CN108196573A (en) | 2017-12-29 | 2017-12-29 | A kind of unmanned plane on-line identification and control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108196573A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110427047A (en) * | 2019-07-26 | 2019-11-08 | 深圳市道通智能航空技术有限公司 | Wind speed measuring method, wind speed estimator and unmanned plane |
| CN110780603A (en) * | 2019-09-21 | 2020-02-11 | 苏州浪潮智能科技有限公司 | Server switching power supply control algorithm optimization method and system |
| CN111538237A (en) * | 2020-03-20 | 2020-08-14 | 北京航空航天大学 | Method for identifying and correcting non-linear light gray model of tilt rotor unmanned aerial vehicle |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101930494A (en) * | 2010-09-03 | 2010-12-29 | 清华大学 | Method for identifying aircraft model with undetermined order and parameters based on mode segmentation and genetic algorithm |
| CN103197545A (en) * | 2013-02-25 | 2013-07-10 | 西北工业大学 | Quick on-line comprehensive identification method of high-speed air vehicle |
| CN103777523A (en) * | 2014-02-28 | 2014-05-07 | 西安费斯达自动化工程有限公司 | Aircraft multi-loop model cluster composite PID (proportion integration differentiation) robust controller design method |
| CN103885450A (en) * | 2014-03-03 | 2014-06-25 | 天津大学 | Method for nonlinear control over unmanned helicopter posture and verification platform |
| CN103941742A (en) * | 2014-04-29 | 2014-07-23 | 中国科学院自动化研究所 | Unmanned aerial vehicle ground sliding deviation rectification control device and method |
| CN104182560A (en) * | 2014-01-08 | 2014-12-03 | 中国商用飞机有限责任公司北京民用飞机技术研究中心 | Aircraft flutter prediction and analysis method and device |
| WO2017161386A1 (en) * | 2016-03-17 | 2017-09-21 | Airspace Systems Inc. | System and method for aerial system discrimination and action |
-
2017
- 2017-12-29 CN CN201711498981.7A patent/CN108196573A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101930494A (en) * | 2010-09-03 | 2010-12-29 | 清华大学 | Method for identifying aircraft model with undetermined order and parameters based on mode segmentation and genetic algorithm |
| CN103197545A (en) * | 2013-02-25 | 2013-07-10 | 西北工业大学 | Quick on-line comprehensive identification method of high-speed air vehicle |
| CN104182560A (en) * | 2014-01-08 | 2014-12-03 | 中国商用飞机有限责任公司北京民用飞机技术研究中心 | Aircraft flutter prediction and analysis method and device |
| CN103777523A (en) * | 2014-02-28 | 2014-05-07 | 西安费斯达自动化工程有限公司 | Aircraft multi-loop model cluster composite PID (proportion integration differentiation) robust controller design method |
| CN103885450A (en) * | 2014-03-03 | 2014-06-25 | 天津大学 | Method for nonlinear control over unmanned helicopter posture and verification platform |
| CN103941742A (en) * | 2014-04-29 | 2014-07-23 | 中国科学院自动化研究所 | Unmanned aerial vehicle ground sliding deviation rectification control device and method |
| WO2017161386A1 (en) * | 2016-03-17 | 2017-09-21 | Airspace Systems Inc. | System and method for aerial system discrimination and action |
Non-Patent Citations (4)
| Title |
|---|
| SANGHYUK PARK: "Modeling with vortex lattice method and frequency sweep flight test for a fixed-wing UAV", 《CONTROL ENGINEERING PRACTICE》 * |
| 夏慧: "小型无人直升机的模型辨识与鲁棒控制技术研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑》 * |
| 王淼 等: "起飞阶段小型无人机航向模型的系统辨识", 《2012 2ND INTERNATIONAL CONFERENCE ON AEROSPACE ENGINEERING AND INFORMATION TECHNOLOGY》 * |
| 黄奔: "航拍用小型无人机的飞控系统算法的研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110427047A (en) * | 2019-07-26 | 2019-11-08 | 深圳市道通智能航空技术有限公司 | Wind speed measuring method, wind speed estimator and unmanned plane |
| CN110427047B (en) * | 2019-07-26 | 2022-06-03 | 深圳市道通智能航空技术股份有限公司 | Wind speed measuring and calculating method, wind speed estimator and unmanned aerial vehicle |
| CN110780603A (en) * | 2019-09-21 | 2020-02-11 | 苏州浪潮智能科技有限公司 | Server switching power supply control algorithm optimization method and system |
| CN110780603B (en) * | 2019-09-21 | 2023-01-10 | 苏州浪潮智能科技有限公司 | Server switching power supply control algorithm optimization method and system |
| CN111538237A (en) * | 2020-03-20 | 2020-08-14 | 北京航空航天大学 | Method for identifying and correcting non-linear light gray model of tilt rotor unmanned aerial vehicle |
| CN111538237B (en) * | 2020-03-20 | 2021-10-29 | 北京航空航天大学 | Method for identifying and correcting non-linear light gray model of tilt rotor unmanned aerial vehicle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107728642B (en) | Unmanned aerial vehicle flight control system and method thereof | |
| CN106774409B (en) | A kind of semi-autonomous imitative ground flight system and its control method of unmanned plane | |
| CN204178240U (en) | UAV flight control system | |
| CN102508493A (en) | Flight control method for small unmanned aerial vehicle | |
| CN106909169A (en) | A kind of full automatic power patrol UAV system | |
| CN105094138A (en) | Low-altitude autonomous navigation system for rotary-wing unmanned plane | |
| CN104881042B (en) | A kind of multiple dimensioned air remote sensing test platform | |
| CN108196573A (en) | A kind of unmanned plane on-line identification and control method | |
| CN105303899A (en) | Child-mother type robot cooperation system of combination of unmanned surface vessel and unmanned aerial vehicle | |
| CN109283937A (en) | A kind of plant protection based on unmanned plane sprays the method and system of operation | |
| CN105182992A (en) | Unmanned aerial vehicle control method and device | |
| CN104199462B (en) | Sea wave sensing based water unmanned plane autonomous taking-off and landing control system | |
| CN107608386A (en) | A kind of UAV Navigation System and method based on the distribution of high-voltage line electromagnetism fingerprint | |
| CN107783549B (en) | Single-rotor-wing plant protection unmanned aerial vehicle obstacle avoidance system based on multi-sensor information fusion technology | |
| CN202939489U (en) | Multi-rotor autobalance flight controller | |
| CN114217632B (en) | Self-adaptive fault-tolerant unmanned aerial vehicle tracking cruise system and method | |
| CN103728981A (en) | Non-linear navigation tracking control method for unmanned aerial vehicle | |
| CN110196454A (en) | Geological exploration integrated system based on unmanned plane | |
| CN103217981A (en) | Four-rotor aircraft speed control method based on integral variable structure control | |
| CN102289714A (en) | Method for controlling autonomous take-off and landing of small unmanned rotorcraft based on behavioral model | |
| CN108196563A (en) | A kind of multi-rotor unmanned aerial vehicle active disturbance rejection compensating control method and system | |
| CN106155075A (en) | A kind of detachable unmanned aerial vehicle control system | |
| CN105738909B (en) | One kind being suitable for low latitude plant protection unmanned helicopter operation boundary extraction method | |
| CN106904272A (en) | A kind of swingable flapping wing robot flight control assemblies of empennage and method | |
| CN202494924U (en) | Fixed-wing unmanned aerial vehicle (UAV) power patrol system on the basis of 3D laser scanning system |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180622 |