CN109002048A - A kind of scale centralization photovoltaic plant image data acquiring method based on multi-rotor unmanned aerial vehicle - Google Patents
A kind of scale centralization photovoltaic plant image data acquiring method based on multi-rotor unmanned aerial vehicle Download PDFInfo
- Publication number
- CN109002048A CN109002048A CN201810603936.1A CN201810603936A CN109002048A CN 109002048 A CN109002048 A CN 109002048A CN 201810603936 A CN201810603936 A CN 201810603936A CN 109002048 A CN109002048 A CN 109002048A
- Authority
- CN
- China
- Prior art keywords
- unmanned plane
- photovoltaic
- group string
- image data
- polygon
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 230000000007 visual effect Effects 0.000 claims abstract description 5
- 230000015572 biosynthetic process Effects 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000000354 decomposition reaction Methods 0.000 claims description 4
- 238000009826 distribution Methods 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000006073 displacement reaction Methods 0.000 abstract description 2
- 230000007613 environmental effect Effects 0.000 abstract description 2
- 238000001514 detection method Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 12
- 238000007689 inspection Methods 0.000 description 6
- 230000018109 developmental process Effects 0.000 description 5
- 230000000630 rising effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000002459 sustained effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/08—Control of attitude, i.e. control of roll, pitch, or yaw
- G05D1/0808—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft
-
- 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 Analysis (AREA)
Abstract
The scale centralization photovoltaic plant image data acquiring method based on multi-rotor unmanned aerial vehicle that the present invention relates to a kind of.Unmanned plane can be handled visible light video stream information by carrying processor;After unmanned plane lift-off, in conjunction with oneself state information, photovoltaic group displacement in the visual field can be calculated and set and orientation, according to calculated result, unmanned plane can realize the automatic tracking to photovoltaic group string;By calculating flying distance, unmanned plane can automatic collection visible images and infrared picture data;When unmanned plane reaches zone boundary, automatic execute turns to strategy;It repeats the above process, the image data acquiring task to entire centralized photovoltaic plant may be implemented in unmanned plane, and can be further used for fault detection.Compared to other collecting methods based on unmanned plane, this method implements simple, efficient, the location information independent of photovoltaic module, it is only necessary to obtain the area information of photovoltaic plant, have extremely strong environmental suitability.
Description
Technical field
The present invention relates to the inspections of the application and development of unmanned plane and scale centralization photovoltaic plant, more particularly, to one kind
Scale centralization photovoltaic plant image data acquiring method based on multi-rotor unmanned aerial vehicle.
Background technique
Unmanned plane industry has welcome the development of blowout in worldwide, in recent years Chinese market consumer level unmanned plane
Growth rate remains at 50% or more, occupies world consumption grade unmanned plane city by the Chinese unmanned plane enterprise of representative of big boundary
70% or more of field.What the Ministry of Industry and Information Technology printed and distributed " anticipates about promotion and the guidance of the civilian unmanned plane development of manufacturing of specification
See " it is clear, civilian unmanned plane industry is by sustained and rapid development, and the year two thousand twenty output value reaches 60,000,000,000 yuan, 40% or more average annual growth rate;
By 2025, the civilian unmanned plane output value reached 180,000,000,000 yuan.Unmanned plane industry is booming rising industry.
Due to can be simple with VTOL, control, multi-rotor unmanned aerial vehicle be able to fast-developing and universal, is very suitable to execute
The tasks such as near-earth investigates, monitors, taking photo by plane, early warning, agricultural plant protection;The nothings such as automatic obstacle avoiding, automatically track target, autonomous trajectory planning
The development of man-machine the relevant technologies, so that the robot that unmanned plane is gradually developed by the camera to fly as flight.
In photovoltaic art, the newly-increased installation 53.06GW of China's photovoltaic in 2017 creates the new highest record in history.Due to the photovoltaic module longevity
Life about 25 years, large-sized photovoltaic power station needed to stablize in the longer time period, efficiently generate electricity, and inspection is photovoltaic plant
The core content of operation.
However, this routine inspection mode efficiency is extremely low at present to the inspection of photovoltaic plant still based on manual inspection, and
Since centralized photovoltaic plant installation environment is complex, manual inspection is difficult to meet the requirements under most of scene.
Summary of the invention
For the existing deficiency towards large centralised photovoltaic plant image data acquiring method, the invention proposes one
Scale centralization photovoltaic plant image data acquiring method of the kind based on unmanned plane.
The technical solution adopted in the present invention is as follows:
Step (1) obtains scale collection according to design of photovoltaic power station drawing, high definition satellite image or unmanned plane high-altitude image
The zone boundary information of Chinese style photovoltaic plant;
Step (2) carries out the scale centralization photovoltaic plant zone boundary information obtained in step (1) with polygon
Fitting, formation zone polygon generate multiple regions polygon if photovoltaic group string is distributed in multiple regions;With side clockwise
To the apex coordinate of record area polygon;
Step (3) selects the geographically south or the vertex of the North rising as area polygonal to each area polygonal
Initial point;To the starting point of each area polygonal, if photovoltaic module is distributed in the east of starting point, prime direction is east, if
Photovoltaic module is distributed in the west of starting point, then prime direction is west;If starting point is located at area polygonal south, the side of cruise
Xiang Weibei, if starting point is located at the area polygonal north, south is in direction of cruising;
Step (4), the area polygonal as step (2) determine be not unique, it is determined that the data acquisition of area polygonal is suitable
Sequence successively executes step (5)~step (10) to each area polygonal;
Step (5) starts unmanned plane, in NED coordinate system, by the yaw angle of the yaw angle of unmanned plane and Visible Light Camera
It is set as 0, sets unmanned plane during flying rate v0, the initial flight speed of unmanned plane is determined according to the prime direction that step (3) determine
Degree, when inceptive direction is east, initial flight speed isIt is initial to fly when inceptive direction is western
Scanning frequency degree isSet dis=0;Wherein, vx、vyRespectively north, east to flying speed,
Dis is flying distance when unmanned plane carries out data acquisition to every a line photovoltaic module, and n_turn is unmanned plane in zone boundary
The number of turns at place and its initial value are 0;
Step (6), unmanned plane processor obtain visible light video stream, are detected to current video frame by image processing algorithm
Photovoltaic module, the boundary straight line for obtaining photovoltaic group string according to the slope of photovoltaic group string boundary straight line and position and combine flight high
Degree calculates the longitudinal pixel-shift amount of the orientation of photovoltaic group string and photovoltaic group string relative to video flowing central point;
Holder yaw angle is calculated in step (7), the photovoltaic group string orientation and offset obtained according to step (6)
Adjustment amount and speed value;
Step (8) repeats step (6), step (7) with fixed and not less than 5Hz frequency, until unmanned plane reaches
Zone boundary;During executing step (6), step (7), such as meet abs (dis-n*dis_photo)≤0.1, unmanned plane
Visible Light Camera and infrared camera acquire image data;Wherein, n is natural number, dis_photo be image data acquiring away from
It sows discord every abs is to ask signed magnitude arithmetic(al);
Step (9) executes when unmanned plane reaches zone boundary and turns to strategy, dis=0, n_turn=n_turn+1;
Step (10), unmanned plane continue to execute Image Acquisition task to the photovoltaic module of current line, are determined according to step (5)
Initial flight speed, execute step (6)~step (9) until unmanned plane is completed to the traversal of area polygonal.
Further, the following conditions are followed when the polygon of formation zone in the step (2):
R010: if two distributed areas of photovoltaic module are connections, then it is considered as a region;
R020: on the basis of R010, the fitted polygon of each distributed areas is generated according to area minimum principle;
R030: if fitted polygon is concave polygon, seeking the external convex polygon of this concave polygon, will according to connectivity
External convex polygon and concave polygon difference portion are divided into several non-interconnected regions;For the arbitrary region of difference portion, such as should
Region east and west sides intersect with concave polygon, then this concave polygon cannot function as area polygonal, otherwise concave polygon conduct
Area polygonal;
R040: all convex polygons all can serve as area polygonal;
R050: if concave polygon cannot function as area polygonal, needing to decompose concave polygon, until after decomposing
All polygons can act as area polygonal, and the principle of decomposition is all areas polygon North and South direction span and most after decomposing
It is small.
Further, the orientation of photovoltaic group string and photovoltaic group string are calculated in the step (6) relative in video flowing
Longitudinal pixel-shift amount of heart point specifically comprises the following steps:
S010: unmanned plane processor obtains visible light video frame, by color space conversion it will be seen that the conversion of light video frame
To HSV space, each channel threshold value of HSV space is set according to the color characteristic on photovoltaic module surface, HSV image is converted into two-value
Image;
S020: morphology closed operation is carried out to bianry image, the bianry image of close photovoltaic module is connected to, is expanded
Photovoltaic group string bianry image afterwards;
S030: the straight line information at photovoltaic group string edge after extension is extracted by Hough transformation, obtains photovoltaic group string edge line
Section, removal slope absolute value is greater than 0.8 or length is less than the line segment of longest line segment half, the remaining line segment G-bar k_ of calculating
Ave and extension after longitudinal pixel-shift amount pix_err of the photovoltaic group string up-and-down boundary center line relative to video flowing central point;
S040: in NED coordinate system, according to unmanned plane body yaw angle θ1, holder yaw angle θ2And photovoltaic module is in nothing
Inclination angle theta in man-machine video frame3Determine photovoltaic group string orientation, that is, being parallel to east-west direction angle is θ=θ1+θ2+θ3's
Straight line;Wherein, θ3=arctan (k_ave).
Further, the calculation method of holder yaw angle adjustment amount and speed value in the step (7) are as follows:
R110: according to the calculated result of step S040, the adjustment amount of unmanned machine head yaw angle is Δ θ2=θ3;
R120: according to the calculated result of step S030, photovoltaic group string up-and-down boundary center line is relative in image after extension
The real offset of heart point position is approximately d_err=pix_err/pix_height*H*tan (0.5*FOV);Wherein H
For the difference in height of unmanned plane Visible Light Camera and photovoltaic module, pix_height is video frame pixel line number, and FOV is vertical field of view
Angle;
R130: according to the calculated result in step R120, when unmanned plane is from western flight eastwards, speed command isWhen unmanned plane flight from east to west, speed command isIts
In,Can prevent excessive | d_err |, a, b are normal number.
Further, in the step (8) unmanned plane reach zone boundary decision condition are as follows: unmanned plane position arrival area
Outside the polygon of domain, and met according to the bianry image that step S010 is obtainedWherein, n=pix_
Width*pix_height is image pixel number,For the pixel quantity for belonging to photovoltaic module that step S010 is handled,
xiFor bianry image pixel value, ColorThreshold is the threshold value for judging to whether there is in the visual field photovoltaic module.
Further, the distance interval dis_photo of visible images and infrared image acquisition is not in the step (8)
Together, i.e., according to Visible Light Camera and infrared camera parameter, visible images and infrared image imaging characteristics, not for the two setting
Same distance interval.
Further, mainly comprising the processes of for strategy is turned in the step (9)
S110: unmanned plane flight S on cruise direction1Distance, S1About photovoltaic group string fore-and-aft distance interval;
S120: as UAV Video frame is unsatisfactory for the zone boundary decision condition in step (8), unmanned plane during flying side at this time
To identical as heading when previous row photovoltaic group string image data acquiring, until unmanned plane meets the region in step (8)
Edge determination condition, starting point of the position as new a line photovoltaic group string image data acquiring at this time;Such as unmanned aerial vehicle vision at this time
Frequency frame meets the zone boundary decision condition in step (8), and unmanned plane during flying direction is adopted with previous row photovoltaic group string image data
Heading when collection is on the contrary, until unmanned plane is just unsatisfactory for the zone boundary decision condition in step (8), position at this time
Starting point as new a line photovoltaic group string image data acquiring;
S130: after the starting point of the new a line photovoltaic group string image data acquiring of determination, unmanned plane during flying direction is set as
With heading when previous row photovoltaic group string image data acquiring on the contrary, steering procedure terminates.
Further, obstacle avoidance module and vertical range finder module are installed in the multi-rotor unmanned aerial vehicle that the present invention is based on,
Vertical range finder module ensures that the photovoltaic module of unmanned plane and underface keeps stable difference in height by the lifting of control unmanned plane.
Further, the Visible Light Camera and infrared camera that the present invention is based on multi-rotor unmanned aerial vehicle is carried vertically to
Down and have three axis from steady function.
Beneficial effects of the present invention are as follows: acquiring picture number using UAV flight's Visible Light Camera and infrared camera
According to, can greatly improve the image data acquiring efficiency of scale centralization photovoltaic plant, these image datas be accident analysis with
The important foundation of diagnosis, health degree assessment.Compared to traditional artificial acquisition method, this method can have very high automatic
Change level, the image data acquiring efficiency of scale centralization photovoltaic plant can be greatly improved;Unmanned plane is based on compared to other
Collecting method, this method implements simple, efficiently, the location information independent of photovoltaic module, it is only necessary to obtain photovoltaic electric
The area information stood has extremely strong environmental suitability.
Detailed description of the invention
Fig. 1 is the scale centralization photovoltaic plant image data acquiring plan of the present invention based on multi-rotor unmanned aerial vehicle
Slightly flow chart;
Fig. 2 is to generate fitted polygon schematic diagram according to photovoltaic module boundary information;
Fig. 3 .1 is the recessed fitted polygon schematic diagram without decomposition;
Fig. 3 .2 is the recessed fitted polygon schematic diagram for needing to decompose;
Fig. 4 is area polygonal starting point, prime direction and cruise direction schematic diagram;
Fig. 5 .1 is a width video frame instance graph;
Fig. 5 .2 is HSV Threshold segmentation result;
Fig. 5 .3 is schematic diagram calculation;
Fig. 5 .4 is the calculated result of photovoltaic group string boundary straight line G-bar and offset;
Fig. 6 is unmanned plane yaw angle in NED coordinate system, holder yaw angle, photovoltaic group string rotation angle relation schematic diagram;
Fig. 7 is that speed command calculates schematic diagram in NED coordinate system;
Fig. 8 .1 is that unmanned plane starts flight course schematic diagram of the initial position outside region after turning;
Fig. 8 .2 is that unmanned plane starts flight course schematic diagram of the initial position in region after turning.
Specific embodiment
The present invention is further described with reference to the accompanying drawing, the purpose is to technical solution of the present invention carry out it is clear,
Complete description, the following examples are a part of the embodiment rather than whole embodiments, based on the embodiment of the present invention, ability
Domain those of ordinary skill obtains other all embodiments without making creative work, belongs to guarantor of the present invention
The range of shield.
Directions vocabulary such as " east " " west " is used for multiple times in the present invention, is mainly based upon considered below: due to photovoltaic power generation
Feature, for all photovoltaic modulies all approximately towards geographic " south ", lateral photovoltaic group string arrangement is then approximate between east and west
To although photovoltaic module direction and orientation and geographic direction use these letters there are deviation under actual conditions
Breath is that unmanned plane specifies starting point and approximate heading is still feasible;On the basis for understanding photovoltaic module characteristic distributions
On, reduce deviation to make unmanned plane realization realize the accurate tracking of photovoltaic group string by visual servo process.
It is that the scale centralization photovoltaic plant image data of the present invention based on multi-rotor unmanned aerial vehicle is adopted shown in Fig. 1
Set method flow chart illustrates the main process of scale centralization photovoltaic plant image data acquiring, detailed and specific
Process can be expressed as follows:
Step (1) obtains scale collection according to design of photovoltaic power station drawing, high definition satellite image or unmanned plane high-altitude image
The zone boundary information of Chinese style photovoltaic plant;
Step (2) carries out the scale centralization photovoltaic plant zone boundary information obtained in step (1) with polygon
Fitting, formation zone polygon then generate multiple regions if photovoltaic group string is distributed between multiple regions and region relatively far apart
Polygon;The apex coordinate of record area polygon in a clockwise direction.It is deposited as an example, Fig. 2 illustrates the distribution of photovoltaic group string
The schematic diagram of multiple regions polygon is generated when apparent regional;
Further, the following conditions are followed when step (2) formation zone polygon:
R010: if two distributed areas of photovoltaic module are connections, then it is considered as a region;
R020: on the basis of condition R010, the fitted polygon of each distributed areas is generated according to area minimum principle,
Fitted polygon is either convex polygon, be also possible to concave polygon;
R030: if fitted polygon is concave polygon, seeking the external convex polygon of this concave polygon, will according to connectivity
External convex polygon and concave polygon difference portion are divided into several non-interconnected regions;For the arbitrary region of difference portion, such as should
Region east and west sides intersect with concave polygon, then this concave polygon cannot function as area polygonal, otherwise concave polygon conduct
Area polygonal;
R040: all convex polygons all can serve as area polygonal;
R050: if concave polygon cannot function as area polygonal, needing to decompose concave polygon, until after decomposing
All polygons all can serve as area polygonal, the principle of decomposition be after decomposing all areas polygon North and South direction span and
It is minimum that (Fig. 3 .1 intuitively illustrates the concave polygon that can be used as area polygonal and not can be used as the recessed polygon of area polygonal
Shape, and the concave polygon that not can be used as area polygonal is decomposed, the presence of " difference region 2 " makes recessed quasi- in Fig. 3 .2
It closes polygon to be decomposed);
Step (3) selects the geographically south or the vertex of the North rising as area polygonal to each area polygonal
Initial point;The east of starting point is distributed in the starting point of each area polygonal, such as photovoltaic module, then prime direction is east, such as
Photovoltaic module is distributed in the west of starting point, then prime direction is west;To the starting point of each area polygonal, point is such as originated
In area polygonal south, then north is in direction of cruising, and if starting point is located at the area polygonal north, then south is in direction of cruising;Fig. 4
The selection of area polygonal starting point, the determination of prime direction and direction of cruising are illustrated, what " south " " the North " herein referred to
It is the vertex in area polygonal southern side, north side boundary, in figureFor the coordinate of starting point 1, prime direction, patrol
The vector for direction composition of navigating, s2(x2,y2) be starting point 2 position coordinates,For the prime direction of starting point 3,For starting point
4 cruise direction
Step (4), the area polygonal as step (2) determine be not unique, it is determined that the data acquisition of area polygonal is suitable
Sequence successively executes step (5)~step (10) to each area polygonal;
Step (5) starts unmanned plane, in NED coordinate system (navigational coordinate system), by the yaw angle and visible light of unmanned plane
The yaw angle of camera is set as 0, sets unmanned plane during flying rate v0, unmanned plane is determined according to the prime direction that step (3) determine
Initial flight speed, when inceptive direction be east when, initial flight speed isWork as inceptive direction
When being western, initial flight speed isSet dis=0;Wherein, vx、vyRespectively north, east
To flying speed, dis be unmanned plane to every a line photovoltaic module carry out data acquisition when flying distance, n_turn is nobody
Number of turns and its initial value of the machine at zone boundary are 0;
Step (6), unmanned plane processor obtain visible light video stream, are detected to current video frame by image processing algorithm
Photovoltaic module, the boundary straight line for obtaining photovoltaic group string according to the slope of photovoltaic group string boundary straight line and position and combine flight high
Degree calculates the longitudinal pixel-shift amount of the orientation of photovoltaic group string and photovoltaic group string relative to video flowing central point;
Further, the orientation of photovoltaic group string and photovoltaic group string are calculated in step (6) relative to video flowing central point
Longitudinal pixel-shift amount specifically comprise the following steps:
S010: unmanned plane processor obtains visible light video frame, by color space conversion it will be seen that the conversion of light video frame
To HSV space, each channel threshold value of HSV space is set (such as to photovoltaic module surface according to the color characteristic on photovoltaic module surface
Blue-black tone it is more sensitive, to illumination-insensitive), HSV image is converted into bianry image;
S020: morphology closed operation is carried out to bianry image, the bianry image of close photovoltaic module is connected to, is expanded
Photovoltaic group string bianry image afterwards;
S030: the straight line information at photovoltaic group string edge after extension is extracted by Hough transformation, obtains photovoltaic group string edge line
Section, removal slope absolute value is greater than 0.8 or length is less than the line segment of longest line segment half, the remaining line segment G-bar k_ of calculating
Ave and extension after longitudinal pixel-shift amount pix_err of the photovoltaic group string up-and-down boundary center line relative to video flowing central point;
S040: in NED coordinate system (navigational coordinate system), according to unmanned plane body yaw angle θ1, holder yaw angle θ2And
Inclination angle theta of the photovoltaic module in UAV Video frame3It determines photovoltaic group string orientation, that is, is parallel to and is with east-west direction angle
θ=θ1+θ2+θ3Straight line;Wherein, θ3=arctan (k_ave);
Fig. 5 .1- Fig. 5 .4 is example using the unmanned plane figure of a secondary photovoltaic module, demonstrates calculating photovoltaic group string and is regarding
The process of angle and offset is rotated in frequency frame;Fig. 6 gives unmanned plane yaw angle in NED coordinate system, holder yaw angle, photovoltaic
Group string rotation angle relation schematic diagram;
Holder yaw angle is calculated in step (7), the photovoltaic group string orientation and offset obtained according to step (6)
Adjustment amount and speed value, it is therefore an objective to so that the displacement of photovoltaic group in video flow center, photovoltaic group string in video streaming
Transverse rotation angle is zero;
Further, the calculation method of holder yaw angle adjustment amount and speed value in step (7) are as follows:
R110: according to the calculated result of S040, the adjustment amount of unmanned machine head yaw angle is Δ θ2=θ3;
R120: according to the calculated result of S030, photovoltaic group string up-and-down boundary center line is relative to image center after extension
The real offset of position is approximately d_err=pix_err/pix_height*H*tan (0.5*FOV);Wherein H is nothing
The difference in height of man-machine Visible Light Camera and photovoltaic module, pix_height are video frame pixel line number, and FOV is vertical field of view angle;
R130: according to the calculated result in R120, when unmanned plane is from western flight eastwards, speed command isWhen unmanned plane flight from east to west, speed command is
Wherein,Can prevent excessive | d_err |, a, b are normal number;Fig. 7 illustrates NED coordinate system middling speed
Degree instruction calculates schematic diagram;
Step (8) repeats step (6), step (7) with fixed and not less than 5Hz frequency, until unmanned plane reaches
Zone boundary;During executing step (6), step (7), such as meeting abs (dis-n*dis_photo)≤0.1, (n is certainly
So number), unmanned plane Visible Light Camera and infrared camera acquire image data;Wherein, dis_photo is image data acquiring
Distance interval;Abs is to ask signed magnitude arithmetic(al);
Further, in step (8) unmanned plane reach zone boundary decision condition are as follows: unmanned plane position reach region it is more
Outside the shape of side, and met according to the bianry image that S010 is obtainedWherein, n=pix_width*
Pix_height is image pixel number,For the pixel quantity for belonging to photovoltaic module that S010 is handled, xiFor binary map
As pixel value, ColorThreshold is the threshold value for judging to whether there is in the visual field photovoltaic module;
Further, the distance interval dis_photo of visible images and infrared image acquisition can not in step (8)
Together, it can according to Visible Light Camera and infrared camera parameter, visible images and infrared image imaging characteristics, set for the two
Fixed different distance interval;
Step (9) executes when unmanned plane reaches zone boundary and turns to strategy, dis=0, n_turn=n_turn+1
Further, mainly comprising the processes of for strategy is turned in step (9)
S110: unmanned plane flight S on cruise direction1Distance, S1About photovoltaic group string fore-and-aft distance interval
S120: as UAV Video frame is unsatisfactory for the zone boundary decision condition in step (8), unmanned plane during flying side at this time
To identical as heading when previous row photovoltaic group string image data acquiring, until unmanned plane meets the region in step (8)
Edge determination condition, starting point of the position as new a line photovoltaic group string image data acquiring at this time;Such as unmanned aerial vehicle vision at this time
Frequency frame meets the zone boundary decision condition in step (8), and unmanned plane during flying direction is adopted with previous row photovoltaic group string image data
Heading when collection is on the contrary, until unmanned plane is just unsatisfactory for the zone boundary decision condition in step (8), position at this time
Starting point as new a line photovoltaic group string image data acquiring;
S130: after the starting point of the new a line photovoltaic group string image data acquiring of determination, unmanned plane during flying direction is set as
With heading when previous row photovoltaic group string image data acquiring on the contrary, steering procedure terminates;Fig. 8 features S110~S130
Steering strategy of the determining unmanned plane at area polygonal boundary;
Step (10), unmanned plane continue to execute Image Acquisition task to the photovoltaic module of current line, specifically i.e. basis
Method in step (5) determines initial flight speed, executes step (6)~step (9) until unmanned plane completion is polygon to region
The traversal of shape;
It is further noted that multi-rotor unmanned aerial vehicle stage+module described in this method has obstacle avoidance module and vertical survey
Away from module, vertical range finder module can ensure that the photovoltaic module of unmanned plane and underface keeps stablizing by controlling the lifting of unmanned plane
Difference in height;The Visible Light Camera and infrared camera that multi-rotor unmanned aerial vehicle platform described in this method carries vertically downward and have
Three axis are from steady function.
Claims (9)
1. a kind of scale centralization photovoltaic plant image data acquiring method based on multi-rotor unmanned aerial vehicle, which is characterized in that
Method includes the following steps:
Step (1) obtains scale centralization according to design of photovoltaic power station drawing, high definition satellite image or unmanned plane high-altitude image
The zone boundary information of photovoltaic plant;
Step (2) intends the scale centralization photovoltaic plant zone boundary information obtained in step (1) with polygon
It closes, formation zone polygon generates multiple regions polygon if photovoltaic group string is distributed in multiple regions;In a clockwise direction
The apex coordinate of record area polygon;
Step (3) selects the starting of the geographically south or the vertex of the North as area polygonal to each area polygonal
Point;To the starting point of each area polygonal, if photovoltaic module is distributed in the east of starting point, prime direction is east, if light
Components distribution is lied prostrate in the west of starting point, then prime direction is west;If starting point is located at area polygonal south, direction of cruising
For north, if starting point is located at the area polygonal north, south is in direction of cruising;
Step (4), the area polygonal as step (2) determine be not unique, it is determined that the data acquisition sequence of area polygonal, it is right
Each area polygonal successively executes step (5)~step (10);
Step (5) starts unmanned plane, and in NED coordinate system, the yaw angle of unmanned plane and the yaw angle of Visible Light Camera are arranged
It is 0, sets unmanned plane during flying rate v0, the initial flight speed of unmanned plane is determined according to the prime direction that step (3) determine, when
When inceptive direction is east, initial flight speed isWhen inceptive direction is western, initial flight speed
Degree isSet dis=0;Wherein, vx、vyRespectively north, east to flying speed, dis is
Unmanned plane carries out flying distance when data acquisition to every a line photovoltaic module, and n_turn is that unmanned plane turns at zone boundary
Curved number and its initial value are 0;
Step (6), unmanned plane processor obtain visible light video stream, detect photovoltaic by image processing algorithm to current video frame
Component, the boundary straight line for obtaining photovoltaic group string according to the slope of photovoltaic group string boundary straight line and position and combine flying height, meter
Calculate the longitudinal pixel-shift amount of the orientation of photovoltaic group string and photovoltaic group string relative to video flowing central point;
The tune of holder yaw angle is calculated in step (7), the photovoltaic group string orientation and offset that are obtained according to step (6)
Whole amount and speed value;
Step (8) repeats step (6), step (7) with fixed and not less than 5Hz frequency, until unmanned plane reaches region
Boundary;During executing step (6), step (7), such as meet abs (dis-n*dis_photo)≤0.1, unmanned plane is visible
Light camera and infrared camera acquire image data;Wherein, n is natural number, and dis_photo is between the distance of image data acquiring
Every abs is to ask signed magnitude arithmetic(al);
Step (9) executes when unmanned plane reaches zone boundary and turns to strategy, dis=0, n_turn=n_turn+1;
Step (10), unmanned plane continue to execute Image Acquisition task to the photovoltaic module of current line, are determined according to step (5) first
Beginning flying speed executes step (6)~step (9) until unmanned plane completes the traversal to area polygonal.
2. a kind of scale centralization photovoltaic plant image data based on multi-rotor unmanned aerial vehicle according to claim 1 is adopted
Set method, which is characterized in that follow the following conditions when the polygon of formation zone in the step (2):
R010: if two distributed areas of photovoltaic module are connections, then it is considered as a region;
R020: on the basis of R010, the fitted polygon of each distributed areas is generated according to area minimum principle;
R030: if fitted polygon is concave polygon, seeking the external convex polygon of this concave polygon, will be external according to connectivity
Convex polygon and concave polygon difference portion are divided into several non-interconnected regions;For the arbitrary region of difference portion, such as the region
East and west sides intersect with concave polygon, then this concave polygon cannot function as area polygonal, and otherwise concave polygon is as region
Polygon;
R040: all convex polygons all can serve as area polygonal;
R050: if concave polygon cannot function as area polygonal, needing to decompose concave polygon, until all after decomposing
Polygon can act as area polygonal, and the principle of decomposition is all areas polygon North and South direction span and minimum after decomposing.
3. a kind of scale centralization photovoltaic plant image data based on multi-rotor unmanned aerial vehicle according to claim 1 is adopted
Set method, which is characterized in that the orientation of photovoltaic group string and photovoltaic group string are calculated in the step (6) relative in video flowing
Longitudinal pixel-shift amount of heart point specifically comprises the following steps:
S010: unmanned plane processor obtains visible light video frame, by color space conversion it will be seen that light video frame is transformed into HSV
Space sets each channel threshold value of HSV space according to the color characteristic on photovoltaic module surface, HSV image is converted to bianry image;
S020: morphology closed operation is carried out to bianry image, is connected to the bianry image of close photovoltaic module, after being expanded
Photovoltaic group string bianry image;
S030: the straight line information at photovoltaic group string edge after extension is extracted by Hough transformation, photovoltaic group string edge line segment is obtained, goes
Except slope absolute value be greater than 0.8 or length be less than longest line segment half line segment, calculate residue line segment G-bar k_ave and
Longitudinal pixel-shift amount pix_err of the photovoltaic group string up-and-down boundary center line relative to video flowing central point after extension;
S040: in NED coordinate system, according to unmanned plane body yaw angle θ1, holder yaw angle θ2And photovoltaic module is in unmanned plane
Inclination angle theta in video frame3Determine photovoltaic group string orientation, that is, being parallel to east-west direction angle is θ=θ1+θ2+θ3It is straight
Line;Wherein, θ3=arctan (k_ave).
4. a kind of scale centralization photovoltaic plant image data based on multi-rotor unmanned aerial vehicle according to claim 3 is adopted
Set method, which is characterized in that the calculation method of holder yaw angle adjustment amount and speed value in the step (7) are as follows:
R110: according to the calculated result of step S040, the adjustment amount of unmanned machine head yaw angle is Δ θ2=θ3;
R120: according to the calculated result of step S030, photovoltaic group string up-and-down boundary center line is relative to image center after extension
The real offset of position is approximately d_err=pix_err/pix_height*H*tan (0.5*FOV);Wherein H is nothing
The difference in height of man-machine Visible Light Camera and photovoltaic module, pix_height are video frame pixel line number, and FOV is vertical field of view angle;
R130: according to the calculated result in step R120, when unmanned plane is from western flight eastwards, speed command isWhen unmanned plane flight from east to west, speed command is
Wherein,Can prevent excessive | d_err |, a, b are normal number.
5. a kind of scale centralization photovoltaic plant image data based on multi-rotor unmanned aerial vehicle according to claim 3 is adopted
Set method, which is characterized in that unmanned plane reaches the decision condition of zone boundary in the step (8) are as follows: unmanned plane position reaches
Outside area polygonal, and met according to the bianry image that step S010 is obtainedWherein, n=
Pix_width*pix_height is image pixel number,The pixel number for belonging to photovoltaic module handled for step S010
Amount, xiFor bianry image pixel value, ColorThreshold is the threshold value for judging to whether there is in the visual field photovoltaic module.
6. a kind of scale centralization photovoltaic plant image data based on multi-rotor unmanned aerial vehicle according to claim 1 is adopted
Set method, which is characterized in that the distance interval dis_photo of visible images and infrared image acquisition is not in the step (8)
Together, i.e., according to Visible Light Camera and infrared camera parameter, visible images and infrared image imaging characteristics, not for the two setting
Same distance interval.
7. a kind of scale centralization photovoltaic plant image data based on multi-rotor unmanned aerial vehicle according to claim 5 is adopted
Set method, which is characterized in that turn to mainly comprising the processes of for strategy in the step (9)
S110: unmanned plane flight S on cruise direction1Distance, S1About photovoltaic group string fore-and-aft distance interval;
S120: if UAV Video frame at this time is unsatisfactory for the zone boundary decision condition in step (8), unmanned plane during flying direction with
Heading when previous row photovoltaic group string image data acquiring is identical, until unmanned plane meets the zone boundary in step (8)
Decision condition, starting point of the position as new a line photovoltaic group string image data acquiring at this time;Such as UAV Video frame at this time
When meeting the zone boundary decision condition in step (8), unmanned plane during flying direction and previous row photovoltaic group string image data acquiring
Heading on the contrary, until unmanned plane be just unsatisfactory for the zone boundary decision condition in step (8), position conduct at this time
The starting point of new a line photovoltaic group string image data acquiring;
S130: after the starting point of the new a line photovoltaic group string image data acquiring of determination, unmanned plane during flying direction is set as with before
Heading when a line photovoltaic group string image data acquiring is on the contrary, steering procedure terminates.
8. a kind of scale centralization photovoltaic plant image data based on multi-rotor unmanned aerial vehicle according to claim 1 is adopted
Set method, which is characterized in that obstacle avoidance module and vertical range finder module are equipped in multi-rotor unmanned aerial vehicle, vertical range finder module is logical
The lifting for crossing control unmanned plane ensures that the photovoltaic module of unmanned plane and underface keeps stable difference in height.
9. a kind of scale centralization photovoltaic plant image data based on multi-rotor unmanned aerial vehicle according to claim 1 is adopted
Set method, which is characterized in that the Visible Light Camera and infrared camera that multi-rotor unmanned aerial vehicle is carried are vertically downward and with three axis
From steady function.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810603936.1A CN109002048B (en) | 2018-06-12 | 2018-06-12 | Multi-rotor unmanned aerial vehicle large-scale centralized photovoltaic power station image data acquisition method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810603936.1A CN109002048B (en) | 2018-06-12 | 2018-06-12 | Multi-rotor unmanned aerial vehicle large-scale centralized photovoltaic power station image data acquisition method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109002048A true CN109002048A (en) | 2018-12-14 |
CN109002048B CN109002048B (en) | 2020-06-09 |
Family
ID=64601247
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810603936.1A Active CN109002048B (en) | 2018-06-12 | 2018-06-12 | Multi-rotor unmanned aerial vehicle large-scale centralized photovoltaic power station image data acquisition method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109002048B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109765930A (en) * | 2019-01-29 | 2019-05-17 | 理光软件研究所(北京)有限公司 | A kind of unmanned plane vision navigation system |
CN110221621A (en) * | 2019-05-30 | 2019-09-10 | 深圳市道通智能航空技术有限公司 | A kind of unmanned plane and its planing method of flight range, device, storage medium |
CN112132898A (en) * | 2020-09-21 | 2020-12-25 | 重庆中电自能科技有限公司 | Photovoltaic module positioning method |
CN112365673A (en) * | 2020-11-12 | 2021-02-12 | 光谷技术股份公司 | Forest fire monitoring system and method |
CN113326752A (en) * | 2021-05-20 | 2021-08-31 | 淮阴工学院 | Unmanned aerial vehicle-based photovoltaic power station identification method and system |
CN113379703A (en) * | 2021-06-08 | 2021-09-10 | 西安理工大学 | Photovoltaic panel dark spot defect detection method based on Yolo-v4 network structure |
CN113433958A (en) * | 2021-06-16 | 2021-09-24 | 广州发展新能源股份有限公司 | Unmanned aerial vehicle inspection method and device |
WO2022225449A1 (en) * | 2021-04-19 | 2022-10-27 | Quantified Energy Labs Pte. Ltd. | Automated imaging of photovoltaic devices using an aerial vehicle and automated flight of the aerial vehicle for performing the same |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2649702A1 (en) * | 2010-12-09 | 2013-10-16 | The Boeing Company | Unmanned vehicle and system |
KR101541032B1 (en) * | 2015-02-13 | 2015-08-03 | (주)하이레벤 | System for collecting and analyzing data regarding module surface temperature of solar photovoltaic power facilities using unmanned air vehicle |
CN105700544A (en) * | 2016-04-08 | 2016-06-22 | 暨南大学 | UAV tour inspection system and implementation method for electrical equipment of photovoltaic power station |
CN105786017A (en) * | 2016-04-01 | 2016-07-20 | 常州天合光能有限公司 | Photovoltaic power station survey, operation and maintenance system based on unmanned aerial vehicle |
CN106527475A (en) * | 2016-10-28 | 2017-03-22 | 中国电力科学研究院 | Distribution network inspection unmanned aerial vehicle and inspection method thereof |
CN106873627A (en) * | 2017-03-31 | 2017-06-20 | 湘潭大学 | A kind of multi-rotor unmanned aerial vehicle and method of automatic detecting transmission line of electricity |
CN107015560A (en) * | 2017-03-02 | 2017-08-04 | 浙江大学 | A kind of method for inspecting of the photovoltaic array based on unmanned plane |
CN107026612A (en) * | 2017-05-17 | 2017-08-08 | 许昌学院 | A kind of solar power station automatic tour inspection system |
EP3214758A1 (en) * | 2016-03-03 | 2017-09-06 | Solaredge Technologies Ltd. | Methods for mapping power generation installations |
CN107219852A (en) * | 2017-06-27 | 2017-09-29 | 成都阿普奇科技股份有限公司 | A kind of large-sized photovoltaic power station Intelligent line patrolling device and its implementation |
CN107330957A (en) * | 2017-05-24 | 2017-11-07 | 浙江大学 | A kind of image processing method with mapping interaction relation between figure layer |
CN108037767A (en) * | 2017-12-13 | 2018-05-15 | 王俊梅 | A kind of unmanned plane visual dead angles turn to escape flight system |
CN108132675A (en) * | 2017-11-23 | 2018-06-08 | 东南大学 | Unmanned plane is maked an inspection tour from main path cruise and intelligent barrier avoiding method by a kind of factory |
-
2018
- 2018-06-12 CN CN201810603936.1A patent/CN109002048B/en active Active
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2649702A1 (en) * | 2010-12-09 | 2013-10-16 | The Boeing Company | Unmanned vehicle and system |
KR101541032B1 (en) * | 2015-02-13 | 2015-08-03 | (주)하이레벤 | System for collecting and analyzing data regarding module surface temperature of solar photovoltaic power facilities using unmanned air vehicle |
EP3214758A1 (en) * | 2016-03-03 | 2017-09-06 | Solaredge Technologies Ltd. | Methods for mapping power generation installations |
CN105786017A (en) * | 2016-04-01 | 2016-07-20 | 常州天合光能有限公司 | Photovoltaic power station survey, operation and maintenance system based on unmanned aerial vehicle |
CN105700544A (en) * | 2016-04-08 | 2016-06-22 | 暨南大学 | UAV tour inspection system and implementation method for electrical equipment of photovoltaic power station |
CN106527475A (en) * | 2016-10-28 | 2017-03-22 | 中国电力科学研究院 | Distribution network inspection unmanned aerial vehicle and inspection method thereof |
CN107015560A (en) * | 2017-03-02 | 2017-08-04 | 浙江大学 | A kind of method for inspecting of the photovoltaic array based on unmanned plane |
CN106873627A (en) * | 2017-03-31 | 2017-06-20 | 湘潭大学 | A kind of multi-rotor unmanned aerial vehicle and method of automatic detecting transmission line of electricity |
CN107026612A (en) * | 2017-05-17 | 2017-08-08 | 许昌学院 | A kind of solar power station automatic tour inspection system |
CN107330957A (en) * | 2017-05-24 | 2017-11-07 | 浙江大学 | A kind of image processing method with mapping interaction relation between figure layer |
CN107219852A (en) * | 2017-06-27 | 2017-09-29 | 成都阿普奇科技股份有限公司 | A kind of large-sized photovoltaic power station Intelligent line patrolling device and its implementation |
CN108132675A (en) * | 2017-11-23 | 2018-06-08 | 东南大学 | Unmanned plane is maked an inspection tour from main path cruise and intelligent barrier avoiding method by a kind of factory |
CN108037767A (en) * | 2017-12-13 | 2018-05-15 | 王俊梅 | A kind of unmanned plane visual dead angles turn to escape flight system |
Non-Patent Citations (3)
Title |
---|
TANG WEN YANG等: "Overhead Power Line Detection from UAV Video", 《2012 19TH INTERNATIONAL CONFERENCE ON MECHATRONICS AND MACHINE VISION IN PRACTICE 》 * |
王斌 等: "红外热像仪搭载无人机在光伏电站中的应用", 《新能源科技》 * |
邓荣军: "基于遗传算法的输电线路无人机巡检路径规划", 《算法、设计与应用》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109765930A (en) * | 2019-01-29 | 2019-05-17 | 理光软件研究所(北京)有限公司 | A kind of unmanned plane vision navigation system |
CN110221621A (en) * | 2019-05-30 | 2019-09-10 | 深圳市道通智能航空技术有限公司 | A kind of unmanned plane and its planing method of flight range, device, storage medium |
CN110221621B (en) * | 2019-05-30 | 2021-11-09 | 深圳市道通智能航空技术股份有限公司 | Unmanned aerial vehicle, and method, device and storage medium for planning flight area of unmanned aerial vehicle |
CN112132898A (en) * | 2020-09-21 | 2020-12-25 | 重庆中电自能科技有限公司 | Photovoltaic module positioning method |
CN112365673A (en) * | 2020-11-12 | 2021-02-12 | 光谷技术股份公司 | Forest fire monitoring system and method |
CN112365673B (en) * | 2020-11-12 | 2022-08-02 | 光谷技术有限公司 | Forest fire monitoring system and method |
WO2022225449A1 (en) * | 2021-04-19 | 2022-10-27 | Quantified Energy Labs Pte. Ltd. | Automated imaging of photovoltaic devices using an aerial vehicle and automated flight of the aerial vehicle for performing the same |
CN113326752A (en) * | 2021-05-20 | 2021-08-31 | 淮阴工学院 | Unmanned aerial vehicle-based photovoltaic power station identification method and system |
CN113326752B (en) * | 2021-05-20 | 2024-04-30 | 淮阴工学院 | Unmanned aerial vehicle-based photovoltaic power station identification method and system |
CN113379703A (en) * | 2021-06-08 | 2021-09-10 | 西安理工大学 | Photovoltaic panel dark spot defect detection method based on Yolo-v4 network structure |
CN113433958A (en) * | 2021-06-16 | 2021-09-24 | 广州发展新能源股份有限公司 | Unmanned aerial vehicle inspection method and device |
Also Published As
Publication number | Publication date |
---|---|
CN109002048B (en) | 2020-06-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109002048A (en) | A kind of scale centralization photovoltaic plant image data acquiring method based on multi-rotor unmanned aerial vehicle | |
CN110873565B (en) | Unmanned aerial vehicle real-time path planning method for urban scene reconstruction | |
CN109060281B (en) | Integrated bridge detection system based on unmanned aerial vehicle | |
CN110244766B (en) | Planning method and system for unmanned aerial vehicle routing inspection route of photovoltaic power station | |
CN105318888B (en) | Automatic driving vehicle paths planning method based on unmanned plane perception | |
CN106780524A (en) | A kind of three-dimensional point cloud road boundary extraction method | |
CN109992006A (en) | A kind of accurate recovery method and system of power patrol unmanned machine | |
CN110554704A (en) | unmanned aerial vehicle-based fan blade autonomous inspection method | |
CN106584451B (en) | automatic transformer substation composition robot and method based on visual navigation | |
CN112033414A (en) | Unmanned aerial vehicle inspection route generation method, device, equipment and medium | |
CN103984355B (en) | Routing inspection flying robot and overhead power line distance prediction and maintaining method | |
CN105045274B (en) | A kind of intelligent shaft tower connected graph construction method for unmanned plane inspection trajectory planning | |
CN112068539A (en) | Unmanned aerial vehicle automatic driving inspection method for blades of wind turbine generator | |
CN114114314A (en) | Power transmission line inspection detection system and detection method based on laser point cloud | |
CN108196538B (en) | Three-dimensional point cloud model-based field agricultural robot autonomous navigation system and method | |
Xi et al. | A vision-based inspection strategy for large-scale photovoltaic farms using an autonomous UAV | |
CN110850889B (en) | Unmanned aerial vehicle autonomous inspection system based on RTK navigation | |
CN114895711A (en) | Automatic unmanned aerial vehicle flight path line planning method for fan blade inspection | |
CN113345094A (en) | Electric power corridor safety distance analysis method and system based on three-dimensional point cloud | |
CN114332647B (en) | River channel boundary detection and tracking method and system for unmanned ship | |
CN114898061A (en) | Power transmission corridor fast reconstruction method based on double-mode visual information of flying away | |
Guo et al. | Curb detection and compensation method for autonomous driving via a 3-D-LiDAR sensor | |
CN111198576A (en) | Control method, medium and unit for particle unmanned aerial vehicle under artificial intelligence big data | |
CN111323026A (en) | Ground filtering method based on high-precision point cloud map | |
CN115329932A (en) | Airplane landing attitude monitoring method based on digital twins |
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 |