CN113534837B - Unmanned aerial vehicle flight state fusion control system and control method based on generation countermeasure network - Google Patents
Unmanned aerial vehicle flight state fusion control system and control method based on generation countermeasure network Download PDFInfo
- Publication number
- CN113534837B CN113534837B CN202110763783.9A CN202110763783A CN113534837B CN 113534837 B CN113534837 B CN 113534837B CN 202110763783 A CN202110763783 A CN 202110763783A CN 113534837 B CN113534837 B CN 113534837B
- Authority
- CN
- China
- Prior art keywords
- loss function
- input signal
- vgg
- unmanned aerial
- aerial vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 230000004927 fusion Effects 0.000 title claims abstract description 22
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 claims abstract description 36
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000001174 ascending effect Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 6
- 230000006872 improvement Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 230000003042 antagnostic effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000005405 multipole Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000007704 transition 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 or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
- G05D1/106—Change initiated in response to external conditions, e.g. avoidance of elevated terrain or of no-fly zones
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
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 an unmanned aerial vehicle flight state fusion control system and method based on a generation countermeasure network. The unmanned aerial vehicle comprises a plurality of rotors, fixed wings, a power system and a flight control system, wherein the flight control system controls the output power of the power system so as to integrally adjust the rotation speed of the plurality of rotors and the attack angle of the fixed wings, and an original flight state parameter output module reads rated working parameters of the plurality of rotors and the fixed wings in real time to generate a first input signal image; the actual flight state parameter acquisition module acquires actual working parameters of the multiple rotors and the fixed wings in real time to generate a second input signal image; the generator network module generates an enhanced picture according to the second input signal image, and the discriminator network module outputs a discrimination result according to the first input signal image and the enhanced picture. The unmanned aerial vehicle control system provided by the invention realizes the deep fusion of the multi-rotor wing and the fixed wing.
Description
[ field of technology ]
The invention relates to the technical field of unmanned aerial vehicles, in particular to a control system and a control method for combining flight states of a multi-rotor unmanned aerial vehicle and a fixed-wing unmanned aerial vehicle in a framework based on an countermeasure network model.
[ background Art ]
In recent years, with the rapid development of the unmanned aerial vehicle industry, the demand of unmanned aerial vehicles in the short-range transportation delivery field is more urgent. Particularly, the small carrying capacity of ten kilograms to fifty kilograms in a severe environment within a distance range of five kilometers to fifty kilometers, so that the unmanned aerial vehicle is widely applied in the production and living fields.
According to the different differentiation of the ways of providing power to the flight process, unmanned aerial vehicles are divided into multi-rotor unmanned aerial vehicles and fixed-wing unmanned aerial vehicles. The multi-rotor unmanned aerial vehicle is small and portable, convenient to take off and land, rapid to unfold, low in cost and good in environmental adaptability, and becomes a preferred type of short-distance transportation and throwing. However, there are two major technical stubs for existing multi-rotor unmanned aerial vehicles: short endurance time and small carrying capacity.
In order to solve the technical problems:
the prior art combines fixed wing and many rotors, forms "perpendicular fixed wing unmanned aerial vehicle" system from this, and this technique is mainly with fixed wing unmanned aerial vehicle, mainly utilizes many rotors to solve the problem in fixed wing unmanned aerial vehicle landing place, has also kept fixed wing bearing capacity strong, long-range performance simultaneously. The working states of the fixed wing and the multiple rotor wings are as follows:
in the take-off and landing stage, the fixed wing does not work, works through the multiple rotors and controls the take-off and landing of the unmanned aerial vehicle.
And in the cruising stage, the multi-rotor stops working, and the fixed wing works to control the unmanned aerial vehicle to navigate. However, the 'suspended fixed wing unmanned aerial vehicle' cannot realize hovering and accurate delivery, and is not suitable for transportation delivery in the existing market. The main reason is that in the whole flight process, the 'vertical fixed wing unmanned aerial vehicle' only implements simple fixed wing and multi-rotor mode switching at the take-off and landing stage, and the fixed wing and the multi-rotor mode switching work independently, and the fixed wing and the multi-rotor mode switching are not fused deeply, so that when accurate positioning and throwing are needed in the navigation process, the multi-rotor is in a stop working state, and accurate positioning and throwing cannot be realized.
On the other hand, when the multi-rotor working state is switched to the fixed-wing working state, or when the flying state of the unmanned aerial vehicle is changed, the flying attitude is severely changed or the power loss is serious at the moment of switching due to the inertial attribute of the self-motion state of the unmanned aerial vehicle.
[ invention ]
In order to solve the technical problems that in the prior art, the fusion of a plurality of rotary wings and fixed wings is insufficient, so that an unmanned aerial vehicle combined by the rotary wings and the fixed wings cannot realize accurate hovering and throwing for a long time, smooth and stable switching between different flight states cannot be realized, and power loss is serious in the switching process.
In view of the above, the invention provides an unmanned aerial vehicle flight state fusion control system and a control method based on generation of an countermeasure network.
An unmanned aerial vehicle flight state fusion control system based on a generation countermeasure network, wherein the unmanned aerial vehicle comprises ascending, cruising, fixed-point hovering and landing states, the unmanned aerial vehicle comprises a plurality of rotors, fixed wings, a power system and a flight control system, the flight control system controls the output power of the power system so as to fusion adjust the rotation speed of the plurality of rotors and the attack angle of the fixed wings, the flight control system comprises an original flight state parameter output module, an actual flight state parameter acquisition module and a generation countermeasure network model, and the original flight state parameter output module reads rated working parameters of the plurality of rotors and the fixed wings in real time to generate a first input signal image; the actual flight state parameter acquisition module acquires actual working parameters of the multiple rotors and the fixed wings in real time to generate a second input signal image; the generating countermeasure network model comprises a generator network module, a discriminator network module and a loss function, wherein the generator network module generates an enhanced picture according to the second input signal image, and the discriminator network module outputs a discrimination result according to the first input signal image and the enhanced picture.
As a further improvement of the present invention, the generator network module adopts an Attention network, wherein the Attention network includes downsampling, upsampling and Attention gate, the second input signal image learns deep features of the second input signal image through the downsampling, the deep features undergo deconvolution upsampling, and finally the enhanced image is output.
As a further improvement of the invention, the arbiter network module adopts PatchGAN, the enhanced image and the first input signal image output by the generator network module are input into the arbiter network module to obtain an n-n matrix, and finally a discrimination result is output, wherein the discrimination result takes the average value of the matrix as a true/false result.
As a further improvement of the present invention, the loss function includes an antagonistic loss function, a mean square error loss function, and a VGG loss function.
As a further improvement of the invention, the countermeasures loss function is:
wherein D represents the discriminator network module, G represents the generator network module, x represents the second input signal image, G (x) represents the enhanced image, y represents the first input signal image, k is the gain factor, and v is the airspeed of the aircraft.
As a further improvement of the present invention, the mean square error loss function is:
where W represents the width of the input image and H represents the height of the input image.
As a further improvement of the present invention, the VGG loss function is:
wherein phi is i The feature output of the ith layer convolution of the VGG model is the VGG loss function, and the definition of the generated image can be improved.
As a further development of the invention, the total loss function is:
Loss＝L G +λ MSE L MSE +λ VGG L VGG
wherein L is G Is the contrast loss function, L MSE Is a mean square error loss function, L VGG Is VGG loss function lambda MSE Penalty coefficient, lambda, of the mean square error loss function VGG Is the penalty coefficient for the VGG loss function. And setting a loss function, and then carrying out continuous alternate iterative training on the generated countermeasure network model by using the image data, and optimizing model parameters of the generated countermeasure network model.
An unmanned aerial vehicle flight state fusion control method based on a countermeasure network, wherein the unmanned aerial vehicle comprises a plurality of rotor wings, fixed wings, a power system and a flight control system, the flight control system comprises an original flight state parameter output module, an actual flight state parameter acquisition module and a generated countermeasure network model, the generated countermeasure network model comprises a generator network module, a discriminator network module and a loss function, and the method comprises the following steps:
step S01, providing an original flight state parameter output module, reading rated working parameters of the multiple rotors and the fixed wings, and generating a first input signal image;
step S02, providing an actual flight state parameter acquisition module, and acquiring actual working parameters of the multiple rotors and the fixed wings in real time to generate a second input signal image;
step S03, inputting the second input signal image to the generator network module to obtain an enhanced image;
step S04, inputting the enhanced image and the first input signal image into the discriminator network module, and outputting a discriminating result after discriminating;
and S05, the generated countermeasure network model correspondingly feeds back the fused enhanced images according to the output results of the step S03 and the step S04 to serve as flight control signals to drive the power system to adjust the working states of the multiple rotors and the fixed wings.
As a further refinement of the invention, the loss function comprises an antagonistic loss function, a mean square error loss function and a VGG loss function, wherein:
the countermeasures loss function is:
wherein D represents the discriminator network module, G represents the generator network module, x represents the second input signal image, G (x) represents the enhanced image, y represents the first input signal image, k is the gain factor, and v is the airspeed of the aircraft;
the mean square error loss function is:
where W represents the width of the input image and H represents the height of the input image;
the VGG loss function is:
wherein phi is i The feature output of the ith layer convolution of the VGG model is that the VGG loss function can improve the definition of the generated image;
the total loss function is:
Loss＝L G +λ MSE L MSE +λ VGG L VGG
wherein L is G Is the contrast loss function, L MSE Is a mean square error loss function, L VGG Is VGG loss function lambda MSE Penalty coefficient, lambda, of the mean square error loss function VGG Is the penalty coefficient for the VGG loss function. The loss function is set, and then the model parameters of the generated countermeasure network model are optimized by continuously and alternately training the generated countermeasure network model in an iterative manner by using the image data.
Compared with the prior art, the unmanned aerial vehicle flight state fusion control method based on the countermeasure network is provided. Through take off, land, cruise, fixed point hover and coordinate the turn in-process to unmanned aerial vehicle, flight control system passes through simultaneously many rotors with fixed wing cooperation collaborative work, control unmanned aerial vehicle's horizontal position, vertical altitude, forward speed, lateral velocity and aircraft course angle for unmanned aerial vehicle is taking off, landing, cruising, fixed point hover and coordinate the turn in-process, many rotors with fixed wing simultaneous working. According to the flight state, the attack angle of the fixed wing is adjusted to control the unmanned aerial vehicle to fly in the state that the ratio of lift force to resistance is maximum. The aerodynamic lift of the unmanned aerial vehicle in a high-speed flight state is improved, and the system power consumption is reduced.
Secondly, the loss function in the generated countermeasure network model can better restore the actual input parameter information, the defects of excessive sharp texture characteristics and lack of high-frequency information of the generated picture are overcome, and VGG loss enables the enhanced image regenerated by the countermeasure network model to be more similar to the first input signal image effect in terms of bottom-layer characteristic pixel values and high-layer abstract characteristics.
More importantly, in the flight control process of the unmanned aerial vehicle, a feedback control mechanism is built between the actual flight state of the unmanned aerial vehicle and an expected flight control plan under the control of a flight control system, so that the flight accurate control of the unmanned aerial vehicle is enhanced, and the flight parameter errors and power losses caused by loss and unsmooth switching between different flight states are reduced.
[ description of the drawings ]
For a clearer description of the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly introduced below, it being obvious that the drawings in the description below are only some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art, wherein:
FIG. 1 is a schematic perspective view of a unmanned aerial vehicle according to the present invention;
FIG. 2 is a block diagram of the configuration of the drone shown in FIG. 1;
FIG. 3 is a block diagram of the flight control system of FIG. 2;
FIG. 4 is a block diagram of the architecture of the generator network module shown in FIG. 3;
fig. 5 is a flowchart of a method for controlling unmanned aerial vehicle flight state fusion based on generating an countermeasure network model according to the present invention.
[ detailed description ] of the invention
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1 and fig. 2 in combination, fig. 1 is a schematic perspective view of a drone according to the present invention, and fig. 2 is a block diagram of the drone shown in fig. 1. The unmanned aerial vehicle 10 includes a fuselage 11, a plurality of rotors 13, fixed wings 15, a power system 17, and a flight control system 19. The plurality of rotor wings 13 and the stationary wing 15 are fixed to the fuselage 11. The power system 17 outputs power to adjust the working states of the plurality of rotor wings 13 and the fixed wing 15. The flight control system 19 controls the output power of the power system 17 to adjust the rotation speed of the multi-rotor 13 and the attack angle of the fixed wing 15 in a fusion manner.
The fuselage 11 supports and fixes the plurality of rotor wings 13, the fixed wing 15, and the power system 17.
The plurality of rotor wings 13 are fixed to the body 11 and uniformly distributed around the body 11. Each rotor 13 itself drives the operating state of the propeller by means of a rotating electric machine, i.e. the speed and direction of rotation of the propeller of each rotor 13 are driven by said rotating electric machine. The rotating motors are respectively and electrically connected with the flight control system 19, and receive the control signals of the power system 17, that is, the power system 17 receives the control signals of the flight control system 19 to drive the rotating motors to rotate, so as to adjust the working state of the rotor wing 13. The rotating motor drives the propeller to work in a set rotating direction and rotating speed range according to a control signal from the flight control system 19, namely the power system 17 adjusts the working state of the multi-rotor 13 corresponding to the output power.
The fixed wing 15 is fixed to the top end of the fuselage 11 and is electrically connected to the flight control system 19 through a support motor. The supporting motor drives the fixed wing 15 to incline relative to the horizontal direction in the flying process, namely: angle of attack. The change of the attack angle value of the fixed wing 15 correspondingly adjusts the flight resistance of the unmanned aerial vehicle 10 in the flight process.
The power system 17 includes a multipole output driving device with multiple output branches, which includes a rotating motor for driving the multiple rotors 13 to work and a supporting motor for adjusting the attack angle of the fixed wing 15. Each rotating motor and the supporting motor correspond to one output branch of the power system, and the rotating speed and the rotating direction of the multiple output branches are coordinated; the supporting motor corresponds to another output branch, and dynamically adjusts the attack angle of the fixed wing, so that the unmanned aerial vehicle 10 can smoothly switch between different states in real time in a space flight state.
Specifically, the unmanned aerial vehicle 10 includes a rising state, a cruising state and a landing state in the actual flight process, and the dynamic flight of the unmanned aerial vehicle 10 is switched between the three working states. Meanwhile, the unmanned aerial vehicle 10 is in different working states, and the rotation speed and the rotation direction of the rotor wing and the attack angle of the fixed wing are combined together.
Referring to fig. 2 and 3 in combination, fig. 3 is a block diagram of the flight control system shown in fig. 2. The flight control system 19 is electrically connected to the input shaft of the power system 17. The flight control system 19 acquires the flight resistance value, the speed value, the altitude value, etc. of the unmanned aerial vehicle 10, that is, the actual flight state parameter value of the unmanned aerial vehicle 10 in real time. The flight control system 19 includes an original flight status parameter output module 191, an actual flight status parameter acquisition module 193, and a generated countermeasure network model 195.
The original flight state parameter output module 191 reads the rated working parameters of the multiple rotor wings 13 and the rated working parameters of the fixed wings 15 from the unmanned aerial vehicle 10 in real time, and generates an original flight state curve of the unmanned aerial vehicle 10 as a first input signal image 1910.
The actual flight state parameter collection module 193 collects the actual working parameters of the multiple rotor wings 13 and the actual working parameters of the fixed wings 15 of the unmanned aerial vehicle 10 in real time, and generates an actual working state flight curve of the unmanned aerial vehicle 10 as a second input signal image 1930. Specifically, the actual flight status parameter collection module 193 includes a rotation speed sensor of the rotation motor, a detector for the included angle between the fixed wing and the horizontal direction, a height sensor, a flight speed sensor, a resistance sensor, etc., and senses the rotation speed, the attack angle, the flight height parameter, the flight speed parameter, and the flight resistance value of the rotation motor of the unmanned aerial vehicle 10 in real time, and draws a flight curve as the second input signal image 1930.
The generating the countermeasure network model 195 includes a generator network module 1951, a arbiter network module 1953, and a loss function 1955.
Please refer to fig. 4, which is a block diagram of the generator network module shown in fig. 3. The generator network module 1951 uses an Attention Unet. Wherein the Attention Unet network comprises downsampling, upsampling and Attention gate. The second input signal image 1930 is downsampled, deep features of the second input signal image 1930 are learned, the deep features are deconvoluted and upsampled, and finally an enhanced image 19510 is output.
The attribute Unet network has the advantage that features of different layers can be grasped, and integrated in a feature superposition manner, so that more information of the second input signal image 1930 can be carried, and an enhanced image 19510 can be generated better.
The discriminator network module 1953 receives the first input signal image 1910 output from the original flight status parameter output module 191 and the enhanced image 19510 generated by the generator network module 1951, and gives a discriminating result 19530, that is, the authenticity between the enhanced image 19510 and the first input signal image 1910, and obtains the enhanced image 19510 more similar to the first input signal image 1910 through continuous iterative training.
The arbiter network module 1953 uses PatchGAN to input the enhanced image 19510 and the first input signal image 1910 output from the generator network module 1951 into the arbiter network module 1953 to obtain an n matrix, and finally outputs the discrimination result 19530, where the discrimination result uses the average value of the matrix as a true/false result.
The loss function 1955 calculates image loss, optimizes model parameters that generate the countermeasure network model 195, the loss function 1955 including a countermeasure loss function, a mean square error loss function, and a VGG loss function, wherein:
the countermeasures loss function is:
wherein D represents the discriminator network module, G represents the generator network module, x represents the second input signal image, G (x) represents the enhanced image, y represents the first input signal image, k is the gain factor, and v is the airspeed of the aircraft.
The mean square error loss function is:
where W represents the width of the input image and H represents the height of the input image.
The VGG loss function is:
wherein phi is i The feature output of the ith layer convolution of the VGG model is the VGG loss function, and the definition of the generated image can be improved.
The total loss function is:
Loss＝L G +λ MSE L MSE +λ VGG L VGG
wherein L is G Is the function of the contrast loss,L MSE is a mean square error loss function, L VGG Is VGG loss function lambda MSE Penalty coefficient, lambda, of the mean square error loss function VGG Is the penalty coefficient for the VGG loss function. Setting the loss function, then carrying out continuous alternate iterative training on the generated countermeasure network model by using image data, optimizing model parameters of the generated countermeasure network model, and enabling the trained generated countermeasure network model to have better enhancement effect on the second input signal image.
Compared with the related art, in the unmanned aerial vehicle flight state fusion control system 10 based on the generation of the countermeasure network, the arbiter network module 1953 adopts the PatchGAN to ensure that all the enhanced images 19510 smoothly transition, and the enhanced images are closer to the rated first input signal image.
Meanwhile, the loss function in the generated countermeasure network model 195 can better restore the actual input parameter information, overcomes the defect of excessive tip texture characteristics of the generated picture and lack of high-frequency information, and the VGG loss enables the enhanced image regenerated by the countermeasure network model to be more similar to the first input signal image effect in terms of the pixel value of the bottom layer characteristic and the high-layer abstract characteristic.
The loss function senses the flight loss according to the second input signal image, adjusts the enhanced image, and further feeds back the discrimination result 19530, so that the enhanced image 19510 is closer to the first input signal image 1910, and improves the flight enhancement effect.
Referring to fig. 5, a flowchart of an unmanned aerial vehicle flight state fusion control method based on generating an countermeasure network model is disclosed in the present invention. The unmanned aerial vehicle flight state fusion control system based on the generation of the countermeasure network model is applied to the flight state control of the unmanned aerial vehicle 10, and comprises the following steps:
step S01, providing an original flight state parameter output module 191, reading rated working parameters of the multi-rotor 13 and the fixed wing 15, and generating a first input signal image 1910;
step S02, providing an actual flight status parameter acquisition module 193, acquiring actual working parameters of the multi-rotor 13 and the fixed wing 15 in real time, and generating a second input signal image 1930;
step S03, inputting the second input signal image 1930 to the generator network module 195 to obtain an enhanced image 19510;
step S04, inputting the enhanced image 19510 and the first input signal image 1910 to the discriminator network module 1953, and outputting the discriminating result 19530 after discriminating;
in step S05, the generated countermeasure network model 195 correspondingly feeds back the fused enhanced image 19510 as a flight control signal according to the output results of step S03 and step S04 to drive the power system 17 to adjust the working states of the multiple rotor wings 13 and the fixed wing 15.
When the unmanned aerial vehicle 10 works, the multi-rotor wings 13 and the fixed wings 15 work synchronously. The actual flight state parameter acquisition module 193 of the flight control system 19 acquires actual working parameters of the multi-rotor 13 and the fixed wing 15, and transmits the actual working parameters as a second input signal image 1930 to the pair of generated anti-network models 195; meanwhile, the original flight state parameter acquisition module 191 reads rated working parameters of the multi-rotor 13 and the fixed wing 15 to generate a first input signal image 1910. In the actual flight process, the first input signal image 1910 and the second input signal image 1930 have a difference, and when the difference is large, the phenomenon of jamming and drifting of the unmanned aerial vehicle 10 is caused, so that the flight state of the controller cannot be accurately controlled.
The arbiter network module 1953 combines the first input signal image 1910 and the second input signal image 1930 to output and synthesize the flight control signal, and outputs enhanced flight parameters, so as to dynamically modify the dynamic control of the flight control system 19 on the output power of the power system 17, thereby meeting the requirements of actual flight; the loss function senses the flight loss according to the actual working parameters, adjusts the enhancement of the enhancement flight parameters, and feedback-controls the output power of the power system 17 to improve the flight enhancement effect.
The unmanned aerial vehicle 10 includes a flying working state, a cruising working state and a landing flying state in the actual flight process. According to the actual flight requirement, the actual flight state of the unmanned aerial vehicle 10 is continuously adjusted, so that the unmanned aerial vehicle 10 is smoothly switched between different flight states, and in the switching moment, the generation countermeasure network model 19 correspondingly feeds back the flight control signals according to the first input signal image 1910 and the second input signal image 1930 to enhance and control the smooth switching of the unmanned aerial vehicle 10, reduce loss, improve cruising ability and precisely control the flight state of the unmanned aerial vehicle 10.
More importantly, in the process, a feedback control mechanism is built between the actual flight state of the unmanned aerial vehicle 10 and the expected flight control plan under the control of a flight control system, so that the flight precision control of the unmanned aerial vehicle 10 is enhanced, and the flight parameter errors and power losses caused by loss and unsmooth switching between different flight states are reduced.
Compared with the related art, the invention provides a flight control method for fusing multiple rotor wings 13 and fixed wings 15. Through to unmanned aerial vehicle 10 in the cruising process, control system passes through simultaneously many rotors with the fixed wing cooperation collaborative work, control unmanned aerial vehicle 10's position for unmanned aerial vehicle is in the cruising process, many rotors with the fixed wing simultaneous working. And adjusting the attack angle of the fixed wing according to the flight state so as to control the unmanned aerial vehicle to fly in the state that the ratio of the lifting force to the resistance is maximum. The aerodynamic lift of the unmanned aerial vehicle in a high-speed flight state is improved, and the system power consumption is reduced.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes or direct or indirect application in other related technical fields are included in the scope of the present invention.
Claims (5)
1. An unmanned aerial vehicle flight state fusion control system based on a generated countermeasure network, wherein the unmanned aerial vehicle comprises ascending, cruising and landing states, the unmanned aerial vehicle comprises a plurality of rotary wings, fixed wings, a power system and a flight control system, and the flight control system controls the output power of the power system so as to fusion adjust the rotation speed of the plurality of rotary wings and the attack angle of the fixed wings, and the unmanned aerial vehicle flight state fusion control system is characterized by comprising an original flight state parameter output module, an actual flight state parameter acquisition module and a generated countermeasure network model, wherein the original flight state parameter output module reads rated working parameters of the plurality of rotary wings and the fixed wings in real time and generates a first input signal image; the actual flight state parameter acquisition module acquires actual working parameters of the multiple rotors and the fixed wings in real time to generate a second input signal image; the generating an countermeasure network model comprises a generator network module, a discriminator network module and a loss function, wherein the generator network module generates an enhanced picture according to the second input signal image, the discriminator network module outputs a discrimination result according to the first input signal image and the enhanced picture, the loss function comprises a countermeasure loss function, a mean square error loss function and a VGG loss function,
the countermeasures loss function is:
wherein D represents the discriminator network module, G represents the generator network module, x represents the second input signal image, G (x) represents the enhanced image, y represents the first input signal image, k is the gain coefficient, v is the aircraft airspeed, and the mean square error loss function is:
where W represents the width of the input image and H represents the height of the input image,
the VGG loss function is:
wherein phi is i Is the characteristic output of the ith layer convolution of the VGG model, the VGG loss function can improve the definition of the generated image,
the total loss function is:
Loss＝L G +λ MSE L MSE +λ VGG L VGG
wherein L is G Is the contrast loss function, L MSE Is a mean square error loss function, L VGG Is VGG loss function lambda MSE Penalty coefficient, lambda, of the mean square error loss function VGG Is the penalty coefficient for the VGG loss function.
2. The unmanned aerial vehicle flight state fusion control system based on a generation countermeasure network of claim 1, wherein the generator network module employs an Attention network, wherein the Attention network comprises downsampling, upsampling, and Attention gate, the second input signal image is subjected to the downsampling, deep features of the second input signal image are learned, the deep features are subjected to deconvolution upsampling, and the enhanced image is finally output.
3. The unmanned aerial vehicle flight state fusion control system based on the generation countermeasure network according to claim 2, wherein the discriminator network module adopts patch gan, the enhanced image and the first input signal image output from the generator network module are input into the discriminator network module to obtain an n-by-n matrix, and finally a discrimination result is output, and the discrimination result takes the average value of the matrix as a true/false result.
4. The unmanned aerial vehicle flight state fusion control system based on generating an countermeasure network of claim 1, wherein the loss function is set and then the model parameters of the generating countermeasure network model are optimized by continuously and alternately iterative training of the generating countermeasure network model with image data.
5. The unmanned aerial vehicle flight state fusion control method based on the generation countermeasure network, wherein the unmanned aerial vehicle comprises a plurality of rotor wings, fixed wings, a power system and a flight control system, the flight control system comprises an original flight state parameter output module, an actual flight state parameter acquisition module and a generation countermeasure network model, and the generation countermeasure network model comprises a generator network module, a discriminator network module and a loss function, and is characterized by comprising the following steps:
step S01, providing an original flight state parameter output module, reading rated working parameters of the multiple rotors and the fixed wings, and generating a first input signal image;
step S02, providing an actual flight state parameter acquisition module, and acquiring actual working parameters of the multiple rotors and the fixed wings in real time to generate a second input signal image;
step S03, inputting the second input signal image to the generator network module to obtain an enhanced image; step S04, inputting the enhanced image and the first input signal image into the discriminator network module, and outputting a discriminating result after discriminating;
step S05, the generated countermeasure network model correspondingly feeds back the fused enhanced images according to the output results of the steps S03 and S04 as flight control signals to drive the power system to adjust the working states of the multiple rotors and the fixed wings,
the loss functions include a counterloss function, a mean square error loss function, and a VGG loss function, wherein:
the countermeasures loss function is:
wherein D represents the discriminator network module, G represents the generator network module, x represents the second input signal image, G (x) represents the enhanced image, y represents the first input signal image, k is the gain factor, and v is the airspeed of the aircraft;
the mean square error loss function is:
where W represents the width of the input image and H represents the height of the input image;
the VGG loss function is:
wherein phi is i The feature output of the ith layer convolution of the VGG model is that the VGG loss function can improve the definition of the generated image;
the total loss function is:
Loss＝L G +λ MSE L MSE +λ VGG L VGG
wherein L is G Is the contrast loss function, L MSE Is a mean square error loss function, L VGG Is VGG loss function lambda MSE Penalty coefficient, lambda, of the mean square error loss function VGG Is the penalty coefficient for the VGG loss function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110763783.9A CN113534837B (en) | 2021-07-06 | 2021-07-06 | Unmanned aerial vehicle flight state fusion control system and control method based on generation countermeasure network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110763783.9A CN113534837B (en) | 2021-07-06 | 2021-07-06 | Unmanned aerial vehicle flight state fusion control system and control method based on generation countermeasure network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113534837A CN113534837A (en) | 2021-10-22 |
| CN113534837B true CN113534837B (en) | 2024-03-22 |
Family
ID=78126938
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110763783.9A Active CN113534837B (en) | 2021-07-06 | 2021-07-06 | Unmanned aerial vehicle flight state fusion control system and control method based on generation countermeasure network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113534837B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10373026B1 (en) * | 2019-01-28 | 2019-08-06 | StradVision, Inc. | Learning method and learning device for generation of virtual feature maps whose characteristics are same as or similar to those of real feature maps by using GAN capable of being applied to domain adaptation to be used in virtual driving environments |
| CN110866472A (en) * | 2019-11-04 | 2020-03-06 | 西北工业大学 | Unmanned aerial vehicle ground moving target identification and image enhancement system and method |
| CN111783545A (en) * | 2020-06-02 | 2020-10-16 | 山西潞安环保能源开发股份有限公司五阳煤矿 | Coal mine unmanned aerial vehicle image enhancement method based on generation of countermeasure network |
| CN112508929A (en) * | 2020-12-16 | 2021-03-16 | 奥比中光科技集团股份有限公司 | Method and device for training generation of confrontation network |
| WO2021088101A1 (en) * | 2019-11-04 | 2021-05-14 | 中国科学院自动化研究所 | Insulator segmentation method based on improved conditional generative adversarial network |
-
2021
- 2021-07-06 CN CN202110763783.9A patent/CN113534837B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10373026B1 (en) * | 2019-01-28 | 2019-08-06 | StradVision, Inc. | Learning method and learning device for generation of virtual feature maps whose characteristics are same as or similar to those of real feature maps by using GAN capable of being applied to domain adaptation to be used in virtual driving environments |
| CN110866472A (en) * | 2019-11-04 | 2020-03-06 | 西北工业大学 | Unmanned aerial vehicle ground moving target identification and image enhancement system and method |
| WO2021088101A1 (en) * | 2019-11-04 | 2021-05-14 | 中国科学院自动化研究所 | Insulator segmentation method based on improved conditional generative adversarial network |
| CN111783545A (en) * | 2020-06-02 | 2020-10-16 | 山西潞安环保能源开发股份有限公司五阳煤矿 | Coal mine unmanned aerial vehicle image enhancement method based on generation of countermeasure network |
| CN112508929A (en) * | 2020-12-16 | 2021-03-16 | 奥比中光科技集团股份有限公司 | Method and device for training generation of confrontation network |
Non-Patent Citations (1)
| Title |
|---|
| 基于生成对抗网络的机载遥感图像超分辨率重建;毕晓君;潘梦迪;;智能系统学报(01);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113534837A (en) | 2021-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP2021176757A (en) | Vertical takeoff and landing (vtol) air vehicle | |
| CN100503366C (en) | Method and apparatus for flight control of tiltrotor aircraft | |
| US8777152B2 (en) | Method and an aircraft provided with a swiveling tail rotor | |
| EP3445652A1 (en) | Combined pitch and forward thrust control for unmanned aircraft systems | |
| CN104044734A (en) | Multi-rotor unmanned airplane with inclined wings and rotors and control system and method | |
| CN106970531B (en) | Method for determining mode conversion control strategy of tilt wing vertical take-off and landing unmanned aerial vehicle | |
| CN109606674A (en) | Tail sitting posture vertical take-off and landing drone and its control system and control method | |
| CN108639332A (en) | The compound multi-modal flight control method of three rotor wing unmanned aerial vehicles | |
| CN107329484A (en) | The dynamic displacement multi-rotor aerocraft control system of oil and control method | |
| CN110329497A (en) | The multi-rotor unmanned aerial vehicle and its control method of a kind of paddle face variable-angle | |
| JP2009143268A (en) | Flight control system for aircraft and aircraft with the flight control system | |
| CN113534837B (en) | Unmanned aerial vehicle flight state fusion control system and control method based on generation countermeasure network | |
| US20220350348A1 (en) | Aircraft with a multi-fan propulsion system for controlling flight orientation transitions | |
| JP6952389B1 (en) | Aircraft | |
| CN109606680A (en) | The multi-modal aircraft of a kind of pair of hair full vector and flight system | |
| WO2019241617A2 (en) | Systems and methods for controlling a vehicle | |
| WO2022145045A1 (en) | Flying object control method | |
| CN108545182A (en) | A kind of VTOL fixed-wing unmanned plane | |
| He et al. | Simulation verification of Flight Control of a tilt tri-rotor UAV Using X-plane | |
| CN114035601A (en) | Tilt rotor unmanned aerial vehicle carrier landing method based on H infinite control | |
| CN113879051A (en) | Vertical take-off and landing and fixed wing aerocar | |
| CN111674545A (en) | Tilting type vertical take-off and landing aircraft and control method thereof | |
| CN112896485B (en) | Two-axis inclined wing aircraft with streamline fuselage and control method | |
| CN115230963A (en) | Unmanned aerial vehicle, unmanned aerial photography system and unmanned aerial photography method | |
| CN215043671U (en) | Vertical take-off and landing unmanned aerial vehicle |
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 |