CN109283937A - A kind of plant protection based on unmanned plane sprays the method and system of operation - Google Patents
A kind of plant protection based on unmanned plane sprays the method and system of operation Download PDFInfo
- Publication number
- CN109283937A CN109283937A CN201811089345.3A CN201811089345A CN109283937A CN 109283937 A CN109283937 A CN 109283937A CN 201811089345 A CN201811089345 A CN 201811089345A CN 109283937 A CN109283937 A CN 109283937A
- Authority
- CN
- China
- Prior art keywords
- plant protection
- unmanned plane
- track
- remote sensing
- plant
- 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 64
- 239000007921 spray Substances 0.000 title claims abstract description 37
- 238000013507 mapping Methods 0.000 claims abstract description 30
- 238000013461 design Methods 0.000 claims abstract description 17
- 238000012544 monitoring process Methods 0.000 claims abstract description 10
- 235000013399 edible fruits Nutrition 0.000 claims description 25
- 238000000605 extraction Methods 0.000 claims description 16
- 230000004888 barrier function Effects 0.000 claims description 13
- 230000004927 fusion Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 238000013135 deep learning Methods 0.000 claims description 6
- 238000013136 deep learning model Methods 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 5
- 238000009826 distribution Methods 0.000 claims description 5
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000002360 preparation method Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000003702 image correction Methods 0.000 claims 1
- 238000005507 spraying Methods 0.000 abstract description 8
- 241000196324 Embryophyta Species 0.000 description 81
- 238000010586 diagram Methods 0.000 description 8
- 230000003044 adaptive effect Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 239000002420 orchard Substances 0.000 description 4
- 239000000284 extract Substances 0.000 description 3
- 238000013439 planning Methods 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 239000012530 fluid Substances 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 238000005304 joining Methods 0.000 description 2
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 244000038559 crop plants Species 0.000 description 1
- 238000013440 design planning Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 238000012546 transfer Methods 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)
- Image Processing (AREA)
Abstract
The present invention provides the method and system that a kind of plant protection based on unmanned plane sprays operation, method includes: the remote sensing surveying and mapping data that targeted plant protection region is obtained based on airborne agriculture feelings information monitoring platform；The plant protection drone operation track track under the plant protection region is calculated based on acquired remote sensing surveying and mapping data；Contrail tracker control unmanned plane execution carrying based on Backstepping sprays apparatus for work and completes plant protection task under designed plant protection drone operation track track.For obtained agriculture feelings geographic information data, as the subsequent basis for spraying route track design, design designs plant protection operation according to the topological relation of plant landform sequence and sprays route around the route track sprayed.
Description
Technical field
The present invention relates to air vehicle technique fields, and in particular to the side of operation is sprayed to a kind of plant protection based on unmanned plane
Method and system.
Background technique
Plant protection operation is required according to the regularity of distribution of crop plant, application, designs corresponding flight operation track, guarantees nothing
The man-machine effective spray good fortune width for spraying operation realizes agricultural precisely aviation operation.It is adjusted by effective flight parameter, it is ensured that agriculture
Industry plant protection drone selection preferably flight working path, reduce that aviation sprays operation resprays rate and drain spray rate, improves plant protection
Unmanned plane aviation sprays operation quality.
Plant protection operation measures farmland geography information by handhold GPS at present: by ground operation personnel handhold GPS measuring tool
Around operating area, the farmland geographic information map for establishing degree of precision is measured in walking.
Operation track planning based on geographic information map: according to the geodata design planning unmanned plane course line measured,
Course line is uploaded to UAV Flight Control System, completes unmanned plane along course line autonomous flight operation by flying control.
Existing route design technology concentrates on two parts, 1) geographic information map in Collecting operation region, navigates as operation
The premise of mark planning and designing；2) operation technique requirement is sprayed according to operating area geography information and correlation, the line of flight is set.
Disadvantage is that: 1) foundation to operating area walking is measured by operating personnel handhold GPS measuring device and made
The mode of the accurate map in industry region requires greatly labor intensity, and operating efficiency is impacted serious；2) AUTONOMOUS TASK at this stage
It is concentrated mainly on the plain topography of crop field, used flight path is generally with arc type route, reciprocal flight covering workspace
Domain, this route design are relatively easy, it is easy to accomplish, targeted plant is also that the almost the same wheat of growing way, rice field etc. are made
Object.But it is uneven for the hilly and mountainous land gradient, crop-planting field is small, it is dispersed it is big, irregular and fruit tree plant growing way is different
The case where when, it is difficult to meet application requirement.
Summary of the invention
In order to solve the existing problems, the present invention provides a kind of method that the plant protection based on unmanned plane sprays operation, needles
To obtained agriculture feelings geographic information data, as the subsequent basis for spraying route track design, design is around the course line rail sprayed
Mark, and plant protection operation is designed according to the topological relation of plant landform sequence and sprays route.
Correspondingly, the present invention provides a kind of method that the plant protection based on unmanned plane sprays operation, the method includes with
Lower step:
The remote sensing surveying and mapping data in targeted plant protection region is obtained based on airborne agriculture feelings information monitoring platform；
The plant protection drone operation track track under the plant protection region is calculated based on acquired remote sensing surveying and mapping data；
Based on Backstepping contrail tracker control unmanned plane execute carry spray apparatus for work the plant protection without
Plant protection task is completed under man-machine operation's track track.
The remote sensing surveying and mapping data for obtaining targeted plant protection region based on airborne agriculture feelings information monitoring platform includes:
Remote sensing surveying and mapping data is acquired based on unmanned plane remote sensing survey unit mounted, the remote sensing surveying and mapping data includes:
Orthography, by the orthography acquired every time, collection point GPS position information, the posture information for acquiring the moment；
The remote sensing surveying and mapping data is based on ORB feature extraction algorithm and realizes image mosaic process, obtains the plant protection area
High-precision geographic information map under domain；
The operation key message in high-precision geographic information map is labeled based on YOLO V3 algorithm and extracts ginseng
Number.
The remote sensing survey unit set Visible Light Camera, infrared camera, multispectral module.
It is described that the remote sensing surveying and mapping data is based on ORB feature extraction algorithm realization image mosaic process, obtain the plant
Protect region under high-precision geographic information map include:
Original image and image to be spliced load are read and carry out grayscale image conversion pretreatment；
ORB characteristic point is extracted, the feature point description is calculated using image reform；
Splice new small figure on former big figure, rasterizing processing is carried out to big figure, and characteristic point is carried out to each grid and is mentioned
It takes, and thick matching is completed according to the method for arest neighbors characteristic matching；
Coordinate unification is carried out after obtaining overlapping region, and the characteristic point of grid and new figure carries out near big figure overlapping region
Image registration sorts to matching characteristic point, will match the characteristic point of robustness degree random distribution by after sorting consistence, by
It arranges out with the sequence for spending more higher more forward；
Corresponding intersection will be most matched by force and carries out meromixis processing, and fusion is weighted and averaged to two parts, it is real
The smooth connection of existing two images carries out two-dimensional histogram normalization by the image completed to fusion, quasi- by least square
It closes and color of image is corrected；
It obtains splicing new figure, continues to repeat the above process with next figure to be spliced, complete the splicing of all orthographies,
Obtain big map.
It is described that the operation key message in high-precision geographic information map is labeled and is extracted based on YOLO V3 algorithm
Parameter includes:
Agriculture feelings message sample data preparation of interest is come out for operating area actual conditions and obtains number of training
According to；
Detection layers of the Darknet-53 deep learning network structure as deep learning model are established, using full convolution and are drawn
Residual structure is entered；
Based on YOLO V3 algorithm sample data training deep learning network structure Darknet-53, after training model
The splicing big map established is imported, deep learning model will complete feature extraction and information labeling work, and operating personnel is by cloud
After the figure issued makees optimization amendment, final high-precision agriculture feelings geography information figure is obtained.
The plant protection drone operation track calculated based on acquired remote sensing surveying and mapping data under the plant protection region
Track includes:
According to the plant height of single plant fruit tree, tree crown range, distance parameter design is sprayed around the route track sprayed；
Motion track between operation fruit tree uses the motion profile of shortest path straight line fitting；
Using traditional bow font motion track avoiding barrier, the barrier region marked on map or when airborne
Sensor monitors after charging into barrier, first keeps hovering again by moving horizontally cut-through object.
The contrail tracker control unmanned plane executing agency based on Backstepping is in the plant protection drone operation
Plant protection task is completed under track track includes:
Modelling by mechanism is carried out to the unmanned plane that will carry out plant protection；
Based on the contrail tracker of Backstepping according to the expression formula and motor power of quadrotor drone power and torque
The revolving speed that model finally obtains quadrotor drone executing agency motor is learned, controls quadrotor drone in the plant protection drone
Plant protection task is completed under operation track track.
The unmanned plane for carrying out plant protection progress modelling by mechanism is included: by described pair
The unmanned plane uses nonlinear model, and the nonlinear model passes through Rigid Body Dynamics Model, power and torque table
It is stated up to formula, motor power model using traditional dynamics of rigid bodies.
Correspondingly, the present invention also provides the control system that a kind of unmanned plane realizes the plant protection of fruit tree plant, the control system
System executes the method as described in claim 1 to 8 any one.
Method provided by the invention is based on airborne agriculture feelings information monitoring platform, carries a variety of camera models, can acquire such as
Lower information: orthography, landform, landforms, high-precision geographical location information, altitude data and multispectral image etc..Acquisition is believed
Breath is numbered according to geographical location information, and by high-precision graphic joining algorithm, obtains a wide range of, high-precision agriculture feelings geography letter
Data are ceased, as the subsequent basis for spraying route track design.Shop problem is sprayed for fruit tree plant, in obtained high-precision
On the topographic map of orchard, the long potential parameter of operation plant and serial number are marked out, according to the plant height of fruit tree plant, tree crown range, spray
The parameters such as distance, medical fluid flow velocity are applied, design is designed around the route track sprayed, and according to the topological relation of plant landform sequence
Route is sprayed, wherein unmanned plane is from plant x to plant x+1, the accessible straight path of optimizing, and height keeps the fixed height of profiling everywhere.
Method according to the present invention, the adaptive surveying and mapping data for acquiring plant protection region, and operation track rail is formed based on surveying and mapping data
Adaptive plant protection operation task is realized in road.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is that the plant protection based on unmanned plane of the embodiment of the present invention sprays the method flow diagram of operation；
Fig. 2 is that the remote sensing for obtaining targeted plant protection region based on airborne agriculture feelings information monitoring platform of the embodiment of the present invention is surveyed
Draw the method flow diagram of data；
Fig. 3 be the embodiment of the present invention by the remote sensing surveying and mapping data be based on ORB feature extraction algorithm realize image mosaic
The method flow diagram of process；
Fig. 4 is that the embodiment of the present invention based on YOLO V3 algorithm realizes that operation key message is labeled and extracting parameter
Method flow diagram；
Fig. 5 is that the embodiment of the present invention sprays operation flight path figure.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts all other
Embodiment shall fall within the protection scope of the present invention.
Fig. 1 shows the method flow diagram that the plant protection based on unmanned plane in the embodiment of the present invention sprays operation, specific to wrap
Include following steps:
S101, the remote sensing surveying and mapping data that targeted plant protection region is obtained based on airborne agriculture feelings information monitoring platform；
In specific implementation process, Fig. 2 shows obtain targeted plant protection region based on airborne agriculture feelings information monitoring platform
The method flow diagram of remote sensing surveying and mapping data, comprising:
S201, remote sensing surveying and mapping data is acquired based on unmanned plane remote sensing survey unit mounted；
It should be noted that remote sensing surveying and mapping data here includes: orthography, by the orthography acquired every time, adopt
Collection point GPS position information, the posture information for acquiring the moment.The remote sensing survey unit set Visible Light Camera, infrared camera, mostly light
Compose module.
In specific implementation process, used multi-rotor unmanned aerial vehicle carrier is modularized design, and mission payload can be according to making
The application demand of user is replaced, which carries low-altitude remote sensing measuring unit, gathers Visible Light Camera, infrared phase
Machine, multispectral module (can freely arrange in pairs or groups) keep certain height, by arc type rail by setting imitative ground offline mode over the ground
Mark flight, whole autonomous flight, without professional operate, do not limited by place, arbitrarily complicated landform can operation at any time,
And orthography is acquired in the collection point of setting, by the orthography acquired every time, collection point GPS position information, acquisition moment
Posture information be packaged as a data cell, after completing the data acquisition to operating area, fly back ground point, this is acquired
Data cell wrap and be transmitted to cloud server, carry out subsequent image real time transfer work.
S202, the remote sensing surveying and mapping data is based on to ORB feature extraction algorithm realization image mosaic process, obtains the plant
Protect the high-precision geographic information map under region；
It should be noted that ORB (Oriented FAST and Rotated BRIEF) is a kind of rapid characteristic points extraction
With the algorithm of description, Fig. 3 shows in the embodiment of the present invention and the remote sensing surveying and mapping data is based on to ORB feature extraction algorithm reality
The method flow diagram of existing image mosaic process, specifically includes:
S301, original image and image to be spliced load are read and carry out grayscale image conversion pretreatment；
Image mosaic process is that a width is merged by width splicing up to new figure splicing completion, first by original image and wait spell
The load of map interlinking picture reads and carries out grayscale image conversion pretreatment.
S302, ORB characteristic point is extracted, the feature point description is calculated using image reform；
In specific implementation process, ORB characteristic point is extracted, the description of this feature point is calculated using image reform, as follows:
Opq=∑ xpyqI (x, y)
θ=arctan (O01,O10)
Wherein, the principal direction for the characteristic point that θ value represents, solves the problems, such as rotational invariance.ORB feature point extraction algorithm
With obvious speed advantage, combines FAST detection feature spot speed fastly and BRIEF Feature Descriptor Project Realization is simple
Feature has obtained the Feature Points Extraction with engineering practical value.
S303, splice new small figure on former big figure, rasterizing processing is carried out to big figure, and feature is carried out to each grid
Point extracts, and completes thick matching according to the method for arest neighbors characteristic matching；
In specific implementation process, splice new small figure on former big figure, to reduce calculation amount, big figure is carried out at rasterizing
Reason, and feature point extraction is carried out to each grid, and thick matching is completed according to the method for arest neighbors characteristic matching, determining faster
Position goes out the overlapping region of big figure and new figure, reduces extra big figure feature extraction and calculation amount.
S304, carry out coordinate unification after obtaining overlapping region, and near big figure overlapping region grid and new figure feature
Point carries out image registration, sorts to matching characteristic point, and the characteristic point for matching robustness degree random distribution is passed through sorting consistence
Afterwards, by matching degree, more higher more forward sequence arranges out；
In specific implementation process, carry out coordinate unification after obtaining overlapping region, and near big figure overlapping region grid with
The characteristic point of new figure carries out image registration, sorts to matching characteristic point, and the characteristic point for matching robustness degree random distribution is led to
After crossing sorting consistence, by matching degree, more higher more forward sequence arranges out, promotes operation real-time to reduce operand, extracts
Preceding 20 matching characteristic point.Registration result is optimized using RANSAC robust method, removes error hiding.
S305, corresponding intersection progress meromixis processing will be most matched by force, two parts are weighted and averaged and are melted
It closes, realizes the smooth connection of two images, two-dimensional histogram normalization is carried out by the image completed to fusion, passes through minimum two
Multiply fitting to correct color of image；
In specific implementation process, corresponding intersection will be most matched by force and carries out meromixis processing, two parts will be carried out
Weighted average fusion, realizes the smooth connection of two images, carries out two-dimensional histogram normalization by the image completed to fusion,
Color of image is corrected by least square fitting.
S306, it obtains splicing new figure, continues to repeat the above process with next figure to be spliced, complete all orthographies
Splicing, obtains big map.
In specific implementation process, obtained splicing is newly schemed, and continues to repeat the above process with next figure to be spliced, completes institute
There is the splicing of orthography, obtains big map.
S203, the operation key message in high-precision geographic information map is labeled and is mentioned based on YOLO V3 algorithm
Take parameter.
YOLO (You Only Look Once) algorithm is object detection algorithm popular at present, speed
Fast and structure is simple.After having obtained the high-precision geographic information map of operating area, pass through cloud big data processing center, base
In YOLO V3 algorithm, simultaneously extracting parameter is labeled to the operation key message in geographic information map, if fruit tree plant kind
Parameters and some obstacle informations such as class, tree crown radius, tree height, farmland area.Fig. 4 shows real based on YOLO V3 algorithm
Existing operation key message is labeled and the method flow diagram of extracting parameter, includes the following:
S401, agriculture feelings message sample data preparation of interest is obtained out for operating area actual conditions to train sample
Notebook data；
It should be noted that here can be whole by agriculture feelings message sample data of interest for operating area actual conditions
Reason comes out (can with collection in worksite, database data with existing sample also can be used), obtains training sample data.
S402, detection layers of the Darknet-53 deep learning network structure as deep learning model are established, using full volume
It accumulates and introduces residual structure；
S403, deep learning network structure Darknet-53 is trained based on YOLO V3 algorithm sample data, trains mould
The splicing big map of foundation is imported after type, deep learning model will complete feature extraction and information labeling work, and operating personnel will
After the figure that cloud issues makees optimization amendment, final high-precision agriculture feelings geography information figure is obtained.
S102, plant protection drone operation track under the plant protection region is calculated based on acquired remote sensing surveying and mapping data
Track；
It should be noted that needing exist for the plant height according to single plant fruit tree, tree crown range, spraying distance parameter design
Around the route track sprayed；And the movement rail of shortest path straight line fitting is used according to the motion track between operation fruit tree
Mark；Using traditional bow font motion track avoiding barrier, the barrier region marked on map or work as airborne sensing
Device monitors after charging into barrier, first keeps hovering again by moving horizontally cut-through object.
In specific implementation process, for the plant protection operation of hilly and orchard, be broadly divided into: 1. are directed to the ring of fruit tree plant
Around spraying operation；2. being moved to straight path and the part avoidance track of next fruit tree after the complete fruit tree of operation.By two
Partial track combines, and obtains the plant protection operation track in entire orchard by the Path Generation being respectively segmented.It is geographical in agriculture feelings
Fruit tree plant job task point relevant location information A (x, y), B (x, y), C (x, y) and other phases have been obtained on information map
After closing job parameter, key operation Task-decomposing are as follows:
(1) it surround and sprays track
Solve the problems, such as that single plant fruit tree sprays flight course planning first, according to plant height, tree crown range, spray distance etc. ginseng
Number, design is around the route track sprayed.Wherein, plant protection drone keeps fixed high operation, flight path ξ=(x, y, z)TMiddle height
Spending z is determined by plant height h, sprays operation flight path as shown in figure 5, single fruit tree plant sprays operation track expression
Formula:
Wherein, a, b are the long and short radius parameters obtained to the oval approximate fits of tree crown range；d0It is unmanned plane and fruit
Tree plant sprays distance；T0It is to spray Mission Operations time correlation coefficient；k0It is that drone flying height keeps parameter.
(2) motion profile between fruit tree setting
Motion track between operation fruit tree can use the motion profile of shortest path straight line fitting, consider arbitrary neighborhood
Spraying the start-stop coordinate between setting is P [xi,yi,zi]→P[xi+1,yi+1,zi+1], flight path are as follows:
Wherein, Tx,TyIt is time constant coefficient of the operation plant protection drone in x-y plane movement speed, determines and make from upper one
The industry fruit tree speed mobile to next operation fruit tree；k1It is that unmanned plane during flying highly keeps parameter everywhere, guarantees unmanned plane distance
Ground level maintains a safe distance.
(3) motion profile of avoiding barrier
Using traditional bow font motion track avoiding barrier, the barrier region marked on map or when airborne
Sensor monitors after charging into barrier, first keeps hovering again by moving horizontally cut-through object, compares in technology realization
Simply, it does not elaborate.
S103 executes to carry based on the contrail tracker control unmanned plane of Backstepping sprays apparatus for work in the plant
Protect completion plant protection task under unmanned machine operation track track.
Here the contrail tracker control unmanned plane executing agency based on Backstepping is in the plant protection drone operation
It includes: to carry out modelling by mechanism to the unmanned plane that will carry out plant protection that plant protection task is completed under track track；Track based on Backstepping
Tracking control unit according to the expression formula and motor dynamics model of quadrotor drone power and torque finally obtain quadrotor nobody
The revolving speed of machine executing agency motor, control quadrotor drone are completed plant protection under the plant protection drone operation track track and are appointed
Business.Here unmanned plane uses nonlinear model, and the nonlinear model passes through Rigid Body Dynamics Model, power and torque expression
Formula, motor power model use traditional dynamics of rigid bodies to state.
Unmanned plane according to the present invention realizes that the control system of fruit tree plant plant protection, the control system execute Fig. 1 to Fig. 5
Shown in method and application process.
Method provided by the invention is based on airborne agriculture feelings information monitoring platform, carries a variety of camera models, can acquire such as
Lower information: orthography, landform, landforms, high-precision geographical location information, altitude data and multispectral image etc..Acquisition is believed
Breath is numbered according to geographical location information, and by high-precision graphic joining algorithm, obtains a wide range of, high-precision agriculture feelings geography letter
Data are ceased, as the subsequent basis for spraying route track design.Shop problem is sprayed for fruit tree plant, in obtained high-precision
On the topographic map of orchard, the long potential parameter of operation plant and serial number are marked out, according to the plant height of fruit tree plant, tree crown range, spray
The parameters such as distance, medical fluid flow velocity are applied, design is designed around the route track sprayed, and according to the topological relation of plant landform sequence
Route is sprayed, wherein unmanned plane is from plant x to plant x+1, the accessible straight path of optimizing, and height keeps the fixed height of profiling everywhere.
Method according to the present invention, the adaptive surveying and mapping data for acquiring plant protection region, and operation track rail is formed based on surveying and mapping data
Adaptive plant protection operation task is realized in road.
Be provided for the embodiments of the invention above a kind of plant protection based on unmanned plane spray the method and system of operation into
It has gone and has been discussed in detail, used herein a specific example illustrates the principle and implementation of the invention, the above implementation
The explanation of example is merely used to help understand method and its core concept of the invention；Meanwhile for the general technology people of this field
Member, according to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion this explanation
Book content should not be construed as limiting the invention.
Claims (9)
1. a kind of method that the plant protection based on unmanned plane sprays operation, which is characterized in that the described method comprises the following steps:
The remote sensing surveying and mapping data in targeted plant protection region is obtained based on airborne agriculture feelings information monitoring platform；
The plant protection drone operation track track under the plant protection region is calculated based on acquired remote sensing surveying and mapping data；
Contrail tracker control unmanned plane based on Backstepping, which executes to carry, sprays apparatus for work in the plant protection drone
Plant protection task is completed under operation track track.
2. the method that the plant protection based on unmanned plane sprays operation as described in claim 1, which is characterized in that described based on airborne
The remote sensing surveying and mapping data that agriculture feelings information monitoring platform obtains targeted plant protection region includes:
Remote sensing surveying and mapping data is acquired based on unmanned plane remote sensing survey unit mounted, the remote sensing surveying and mapping data includes: just to penetrate
Image, by the orthography acquired every time, collection point GPS position information, the posture information for acquiring the moment；
The remote sensing surveying and mapping data is based on ORB feature extraction algorithm and realizes image mosaic process, is obtained under the plant protection region
High-precision geographic information map；
Simultaneously extracting parameter is labeled to the operation key message in high-precision geographic information map based on YOLO V3 algorithm.
3. the method that the plant protection based on unmanned plane sprays operation as claimed in claim 2, which is characterized in that the remote sensing survey
Unit set Visible Light Camera, infrared camera, multispectral module.
4. the method that the plant protection based on unmanned plane sprays operation as claimed in claim 3, which is characterized in that it is described will be described distant
Feel surveying and mapping data and be based on ORB feature extraction algorithm realization image mosaic process, the high-precision obtained under the plant protection region is geographical
Information map includes:
Original image and image to be spliced load are read and carry out grayscale image conversion pretreatment；
ORB characteristic point is extracted, the feature point description is calculated using image reform；
Splice new small figure on former big figure, rasterizing processing is carried out to big figure, and feature point extraction is carried out to each grid, and
Thick matching is completed according to the method for arest neighbors characteristic matching；
Coordinate unification is carried out after obtaining overlapping region, and the characteristic point of grid and new figure carries out image near big figure overlapping region
Registration sorts to matching characteristic point, after the characteristic point for matching robustness degree random distribution is passed through sorting consistence, by matching degree
More higher, more forward sequence arranges out；
Corresponding intersection will be most matched by force and carries out meromixis processing, and fusion is weighted and averaged to two parts, realizes two
The smooth connection of width image carries out two-dimensional histogram normalization by the image completed to fusion, passes through least square fitting pair
Color of image correction；
It obtains splicing new figure, continues to repeat the above process with next figure to be spliced, complete the splicing of all orthographies, obtain
Big map.
5. the method that the plant protection based on unmanned plane sprays operation as claimed in claim 4, which is characterized in that described to be based on YOLO
V3 algorithm is labeled to the operation key message in high-precision geographic information map and extracting parameter includes:
Agriculture feelings message sample data preparation of interest is come out for operating area actual conditions and obtains training sample data；
Detection layers of the Darknet-53 deep learning network structure as deep learning model are established, using full convolution and are introduced
Residual structure；
Based on YOLO V3 algorithm sample data training deep learning network structure Darknet-53, imported after training model
The splicing big map of foundation, deep learning model will complete feature extraction and information labeling work, operating personnel issue cloud
Figure make optimization amendment after, obtain final high-precision agriculture feelings geography information figure.
6. the method that the plant protection based on unmanned plane sprays operation as claimed in claim 5, which is characterized in that described to be based on being obtained
The remote sensing surveying and mapping data taken calculates the plant protection drone operation track track under the plant protection region
According to the plant height of single plant fruit tree, tree crown range, distance parameter design is sprayed around the route track sprayed；
Motion track between operation fruit tree uses the motion profile of shortest path straight line fitting；
Using traditional bow font motion track avoiding barrier, the barrier region marked on map or work as airborne sensing
Device monitors after charging into barrier, first keeps hovering again by moving horizontally cut-through object.
7. the method that the plant protection based on unmanned plane sprays operation as claimed in claim 6, which is characterized in that described to be based on contragradience
The contrail tracker control unmanned plane executing agency of method completes plant protection under the plant protection drone operation track track and appoints
Business includes:
Modelling by mechanism is carried out to the unmanned plane that will carry out plant protection；
Based on the contrail tracker of Backstepping according to the expression formula and motor dynamics mould of quadrotor drone power and torque
Type finally obtains the revolving speed of quadrotor drone executing agency motor, controls quadrotor drone in the plant protection drone operation
Plant protection task is completed under track track.
8. the method that the plant protection based on unmanned plane sprays operation as claimed in claim 7, which is characterized in that described pair will carry out
The unmanned plane of plant protection carries out modelling by mechanism
The unmanned plane use nonlinear model, the nonlinear model by Rigid Body Dynamics Model, power and torque expression formula,
Motor power model uses traditional dynamics of rigid bodies to state.
9. a kind of system that the plant protection based on unmanned plane sprays operation, which is characterized in that the control system is executed as right is wanted
Seek method described in 1 to 8 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811089345.3A CN109283937A (en) | 2018-09-18 | 2018-09-18 | A kind of plant protection based on unmanned plane sprays the method and system of operation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811089345.3A CN109283937A (en) | 2018-09-18 | 2018-09-18 | A kind of plant protection based on unmanned plane sprays the method and system of operation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109283937A true CN109283937A (en) | 2019-01-29 |
Family
ID=65181391
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811089345.3A Pending CN109283937A (en) | 2018-09-18 | 2018-09-18 | A kind of plant protection based on unmanned plane sprays the method and system of operation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109283937A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109792951A (en) * | 2019-02-21 | 2019-05-24 | 华南农业大学 | For the unmanned plane course line correction system of hybrid rice pollination and its bearing calibration |
| CN109828578A (en) * | 2019-02-22 | 2019-05-31 | 南京天创电子技术有限公司 | A kind of instrument crusing robot optimal route planing method based on YOLOv3 |
| CN109871029A (en) * | 2019-02-21 | 2019-06-11 | 华南农业大学 | A kind of plant protection drone flight course planning optimization method based on image processing techniques |
| CN109885051A (en) * | 2019-02-21 | 2019-06-14 | 彭劲松 | A kind of ecological environment health quality appraisal procedure |
| CN110371294A (en) * | 2019-07-12 | 2019-10-25 | 仲恺农业工程学院 | Plant protection drone system and plant protection drone operational method |
| CN110597279A (en) * | 2019-08-30 | 2019-12-20 | 南京精微迅智能科技有限公司 | Operation method of agricultural unmanned aerial vehicle and control method thereof |
| CN111145235A (en) * | 2019-12-26 | 2020-05-12 | 长光禹辰信息技术与装备(青岛)有限公司 | Crop medicine spraying method, device, equipment and computer readable storage medium |
| CN112565726A (en) * | 2020-12-10 | 2021-03-26 | 广州极飞科技有限公司 | Method for determining job prescription chart, job control method and related device |
| CN113009884A (en) * | 2019-12-19 | 2021-06-22 | 广州极飞科技股份有限公司 | Method, device, equipment and storage medium for controlling operation of unmanned equipment |
| CN113807309A (en) * | 2021-09-28 | 2021-12-17 | 北京石油化工学院 | Orchard machine walking route planning method based on deep learning |
| CN114237301A (en) * | 2022-02-24 | 2022-03-25 | 南京南机智农农机科技研究院有限公司 | Air-ground integrated plant protection operation system and method |
| CN117274844A (en) * | 2023-11-16 | 2023-12-22 | 山东科技大学 | Rapid extraction method for field peanut seedling emergence condition by using unmanned aerial vehicle remote sensing image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120095621A1 (en) * | 2009-03-26 | 2012-04-19 | Ohio University | Trajectory tracking flight controller |
| CN106485655A (en) * | 2015-09-01 | 2017-03-08 | 张长隆 | A kind of taken photo by plane map generation system and method based on quadrotor |
| CN108320510A (en) * | 2018-04-03 | 2018-07-24 | 深圳市智绘科技有限公司 | One kind being based on unmanned plane video traffic information statistical method and system |
-
2018
- 2018-09-18 CN CN201811089345.3A patent/CN109283937A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120095621A1 (en) * | 2009-03-26 | 2012-04-19 | Ohio University | Trajectory tracking flight controller |
| CN106485655A (en) * | 2015-09-01 | 2017-03-08 | 张长隆 | A kind of taken photo by plane map generation system and method based on quadrotor |
| CN108320510A (en) * | 2018-04-03 | 2018-07-24 | 深圳市智绘科技有限公司 | One kind being based on unmanned plane video traffic information statistical method and system |
Non-Patent Citations (2)
| Title |
|---|
| 李鲜: "无人机三维立体重构与物体识别系统研究与实现", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 陈再励 等: "面向丘陵山地果树植株的植保无人机轨迹跟踪控制器设计", 《自动化与信息工程》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109871029A (en) * | 2019-02-21 | 2019-06-11 | 华南农业大学 | A kind of plant protection drone flight course planning optimization method based on image processing techniques |
| CN109885051A (en) * | 2019-02-21 | 2019-06-14 | 彭劲松 | A kind of ecological environment health quality appraisal procedure |
| CN109792951A (en) * | 2019-02-21 | 2019-05-24 | 华南农业大学 | For the unmanned plane course line correction system of hybrid rice pollination and its bearing calibration |
| CN109792951B (en) * | 2019-02-21 | 2020-07-31 | 华南农业大学 | Unmanned aerial vehicle air route correction system for pollination of hybrid rice and correction method thereof |
| CN109828578B (en) * | 2019-02-22 | 2020-06-16 | 南京天创电子技术有限公司 | Instrument inspection robot optimal route planning method based on YOLOv3 |
| CN109828578A (en) * | 2019-02-22 | 2019-05-31 | 南京天创电子技术有限公司 | A kind of instrument crusing robot optimal route planing method based on YOLOv3 |
| CN110371294A (en) * | 2019-07-12 | 2019-10-25 | 仲恺农业工程学院 | Plant protection drone system and plant protection drone operational method |
| CN110597279A (en) * | 2019-08-30 | 2019-12-20 | 南京精微迅智能科技有限公司 | Operation method of agricultural unmanned aerial vehicle and control method thereof |
| CN113009884A (en) * | 2019-12-19 | 2021-06-22 | 广州极飞科技股份有限公司 | Method, device, equipment and storage medium for controlling operation of unmanned equipment |
| CN111145235A (en) * | 2019-12-26 | 2020-05-12 | 长光禹辰信息技术与装备(青岛)有限公司 | Crop medicine spraying method, device, equipment and computer readable storage medium |
| CN111145235B (en) * | 2019-12-26 | 2023-10-31 | 长光禹辰信息技术与装备(青岛)有限公司 | Crop drug spraying method, device, equipment and computer readable storage medium |
| CN112565726A (en) * | 2020-12-10 | 2021-03-26 | 广州极飞科技有限公司 | Method for determining job prescription chart, job control method and related device |
| CN113807309A (en) * | 2021-09-28 | 2021-12-17 | 北京石油化工学院 | Orchard machine walking route planning method based on deep learning |
| CN114237301A (en) * | 2022-02-24 | 2022-03-25 | 南京南机智农农机科技研究院有限公司 | Air-ground integrated plant protection operation system and method |
| CN117274844A (en) * | 2023-11-16 | 2023-12-22 | 山东科技大学 | Rapid extraction method for field peanut seedling emergence condition by using unmanned aerial vehicle remote sensing image |
| CN117274844B (en) * | 2023-11-16 | 2024-02-06 | 山东科技大学 | Rapid extraction method for field peanut seedling emergence condition by using unmanned aerial vehicle remote sensing image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109283937A (en) | A kind of plant protection based on unmanned plane sprays the method and system of operation | |
| Gao et al. | Review of wheeled mobile robots’ navigation problems and application prospects in agriculture | |
| CN107514993B (en) | The collecting method and system towards single building modeling based on unmanned plane | |
| Bacik et al. | Autonomous flying with quadrocopter using fuzzy control and ArUco markers | |
| Rathinam et al. | Vision-based monitoring of locally linear structures using an unmanned aerial vehicle | |
| Barrientos et al. | Aerial remote sensing in agriculture: A practical approach to area coverage and path planning for fleets of mini aerial robots | |
| CN108519775A (en) | A kind of UAV system and its control method precisely sprayed | |
| CN103389699B (en) | Based on the supervisory control of robot of distributed intelligence Monitoring and Controlling node and the operation method of autonomous system | |
| CN107356230A (en) | A kind of digital mapping method and system based on outdoor scene threedimensional model | |
| CN107608371A (en) | Four rotor automatic obstacle avoiding unmanned plane under the environment of community in urban areas | |
| CN106500669A (en) | A kind of Aerial Images antidote based on four rotor IMU parameters | |
| CN106840148A (en) | Wearable positioning and path guide method based on binocular camera under outdoor work environment | |
| CN109341706A (en) | A kind of production method of the multiple features fusion map towards pilotless automobile | |
| CN107589752A (en) | Unmanned plane cooperates with formation realization method and system with ground robot | |
| CN108415032A (en) | A kind of point cloud semanteme map constructing method based on deep learning and laser radar | |
| CN109959377A (en) | A kind of robot navigation's positioning system and method | |
| CN107085422A (en) | A kind of tele-control system of the multi-functional Hexapod Robot based on Xtion equipment | |
| CN109240284A (en) | A kind of autonomous paths planning method and device of unmanned agricultural machinery | |
| CN110333738A (en) | A kind of unmanned plane cluster verification method based on analogue simulation software | |
| Chen et al. | DroneTalk: An Internet-of-Things-based drone system for last-mile drone delivery | |
| CN109683629A (en) | Unmanned plane electric stringing system based on integrated navigation and computer vision | |
| CN104850120B (en) | Wheeled mobile robot air navigation aid based on IHDR autonomous learning frameworks | |
| CN102506867B (en) | SINS (strap-down inertia navigation system)/SMANS (scene matching auxiliary navigation system) combined navigation method based on Harris comer matching and combined navigation system | |
| CN106767791A (en) | A kind of inertia/visual combination air navigation aid using the CKF based on particle group optimizing | |
| Xing et al. | Multi-UAV cooperative system for search and rescue based on YOLOv5 |
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 |
Application publication date: 20190129 |
|
| RJ01 | Rejection of invention patent application after publication |