CN109211202A - A kind of method for optimizing route of the expressway slope inspection based on unmanned plane - Google Patents
A kind of method for optimizing route of the expressway slope inspection based on unmanned plane Download PDFInfo
- Publication number
- CN109211202A CN109211202A CN201811118398.3A CN201811118398A CN109211202A CN 109211202 A CN109211202 A CN 109211202A CN 201811118398 A CN201811118398 A CN 201811118398A CN 109211202 A CN109211202 A CN 109211202A
- Authority
- CN
- China
- Prior art keywords
- unmanned plane
- slope
- shooting
- point
- sunlight
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Multimedia (AREA)
- Astronomy & Astrophysics (AREA)
- Theoretical Computer Science (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to expressway slope inspection technical fields, and in particular to a kind of method for optimizing route of the expressway slope inspection based on unmanned plane: step 1, control unmanned plane shooting quality, to obtain best shooting angle；Step 2, classify to reference object, and style of shooting is determined according to classification situation；Step 3, it needs the object shot to be checked according to the picture of shooting by unmanned plane shooting, extracts disease suspicion point existing for side slope in picture；Step 4, the distance between each disease suspicion point is calculated, and constructs the adjacency matrix of disease suspicion point；Step 5, the shortest path of unmanned plane traversal disease suspicion point is calculated as the input of simulated annealing in adjacency matrix step 4 constructed.The present invention patrols to expressway slope by unmanned plane, and being capable of planning department the problem of reasonably patrolling path, being able to solve artificial inspection heavy workload existing in the prior art.
Description
Technical field
The invention belongs to expressway slope inspection technical fields, and in particular to a kind of highway side based on unmanned plane
The method for optimizing route of slope inspection.
Background technique
With the fast development of computer, communication etc., unmanned plane aerial photography system is also from initial military field research
Stage enters every profession and trade practical stage, and is widely used in forest fire protection, resource detection, search rescue, agriculture test, boat
Clap etc..The inspection application of unmanned plane has been covered to multiple fields such as electric power, petroleum, agricultural, communications both at home and abroad, but
It is also fewer in the research achievement of highway maintenance especially road slope inspection for unmanned plane, and side slope maintenance inspection is both
The emphasis of highway maintenance inspection is also difficult point.It is main at present to be patrolled using the operation mode side slope manually patrolled, but people
Work inspection heavy workload, side slope maintenance inspection work will spend a large amount of time and efforts of patrolman, but also there is inspection
The unreasonable situation in path, which increase the workloads of patrolman, reduce the whole working efficiency of road maintenance inspection.
Summary of the invention
In order to solve the problems existing in the prior art, the purpose of the present invention is to provide a kind of highways based on unmanned plane
The method for optimizing route of side slope inspection, this method patrol to expressway slope by unmanned plane, and being capable of planning department
Reasonable inspection path, the problem of being able to solve artificial inspection heavy workload existing in the prior art.
The present invention is to be achieved through the following technical solutions:
A kind of method for optimizing route of the expressway slope inspection based on unmanned plane, includes the following steps:
Step 1, unmanned plane shooting quality is controlled, to obtain best shooting angle；
Step 2, classify to reference object, and style of shooting is determined according to classification situation；
Step 3, it needs the object shot to be checked according to the picture of shooting by unmanned plane shooting, extracts picture
Disease suspicion point existing for middle side slope；
Step 4, the distance between each disease suspicion point is calculated, and constructs the adjacency matrix of disease suspicion point；
Step 5, unmanned plane traversal is calculated as the input of simulated annealing in adjacency matrix step 4 constructed
The shortest path of disease suspicion point.
In the step 1, carrying out control to unmanned plane shooting quality includes that pickup quality control and flight reappearance control；
Pickup quality control is to be carried out according to weather, season, region, time and projecting direction to unmanned plane, specifically:
Weather is divided into six classes, respectively strong sunlight, direct sunlight, thin cloud fine day, the bright cloudy day, rainy weather and
Dense black wet weather；The light exposure of weather of the weather of latter class relative to preceding one kind, unmanned plane increases by one grade；
Season is successively divided according to winter, summer and spring, or is successively divided according to winter, summer and autumn, after
One season increases by one grade compared to previous season, the light exposure of unmanned plane；
When carrying out pickup quality control to unmanned plane according to region, latitude is every to increase by 15 °, and unmanned plane aperture accordingly increases
Half grade or one grade；After having adjusted unmanned plane aperture according to latitude, in 1000m or less, unmanned plane aperture size is kept height above sea level
It is constant；For height above sea level in 1000m~2000m, unmanned plane aperture reduces one grade；Height above sea level is in 2000m~4000m, unmanned plane aperture
Reduce two grades；
When carrying out pickup quality control to unmanned plane according to the time, the period on daytime is chosen, with 1h before and after local high noon
Within period as reference；The unmanned plane aperture at local sunrise sunset moment is than period unmanned plane within 1h before and after high noon
The light exposure of aperture reduces third gear；After sunrise within 1h and before sunset within 1h after the light exposure calibration morning of unmanned plane aperture
Light exposure within 1h reduces two grades；The light exposure of unmanned plane is compared with high noon in 1h~2h before 1h~2h and sunset after sunrise
Light exposure within the 1h of front and back reduces one grade；
When carrying out pickup quality control to unmanned plane according to projecting direction, projecting direction is divided into five classes, respectively frontlighting,
Along sidelight, sidelight, side backlight and backlight；When sunlight is in the top of unmanned plane camera primary optical axis, sunlight incident direction with
It is frontlighting when the β of the angle of unmanned plane camera key light axis direction is 75 °~90 °；When sunlight is in unmanned plane camera primary optical axis
Top, when the β of the angle of sunlight incident direction and unmanned plane camera key light axis direction is 45 °~75 °, for along sidelight；When
Sunlight is in the top of unmanned plane camera primary optical axis, sunlight incident direction and the angle of unmanned plane camera key light axis direction
It is sidelight when β is 0 °~45 °；When sunlight is in the lower section of unmanned plane camera primary optical axis, sunlight incident direction and unmanned plane
It is side backlight when the β of the angle of camera key light axis direction is 0 °~45 °；When sunlight is under unmanned plane camera primary optical axis
It is inverse sidelight when the β of the angle of side, sunlight incident direction and unmanned plane camera key light axis direction is 45 °~75 °；Work as the sun
Light is in the lower section of unmanned plane camera primary optical axis, and sunlight incident direction and the β of the angle of unmanned plane camera key light axis direction are
It is backlight at 75 °~90 °；
Flight reappearance control includes control photograph overlap degree and ground sampling interval.
When control photograph overlap is spent, setting ship's control qx range is 60%~65%, and sidelapping degree qy is in 25%-
35%.
Ground sampling interval GSD is calculate by the following formula:
In formula, H indicates that unmanned plane distance boat takes the photograph the relative altitude in subregion mean baseline face；The lens focus of f expression camera；
μ indicates the physical size of pixel in unmanned plane photosensitive element.
In the step 2, reference object is divided into side slope and intercepting ditch, and side slope is divided into step-like side slope and non-step-like side
Slope, intercepting ditch are divided into trapezoidal intercepting ditch, rectangle intercepting ditch, L shape intercepting ditch and V-arrangement intercepting ditch, unmanned plane pair according to end pattern
Side slope and intercepting ditch are separately shot.
When side slope is shot, by the way of vertical photography, keep unmanned plane camera primary optical axis vertical with slope surface.
When shooting to step-like side slope, slope surface is split into several facets by step, when inclining for slope surfaces at different levels
When angle is identical, camera lens keep identical inclination angle to complete the shooting of entire slope surface when unmanned plane is shot.
When shooting to intercepting ditch, in such a way that unmanned plane camera primary optical axis is perpendicular to bottom of trench, camera primary optical axis exists
K1K1' in range, K1K1' it is maximum magnitude of the sun light direct beam into bottom of trench.
In the step 4, the adjacency matrix process for constructing disease suspicion point is as follows:
Step 4.1, it determines the landing point of unmanned plane, obtains the position letter of unmanned plane landing point and each disease suspicion point
Breath, the location information includes longitude, latitude and height；
Step 4.2, the distance between each point is calculated, constructs adjacency matrix, each point packet using the distance between each point
Include the landing point and each disease suspicion point of unmanned plane.
The step 5 calculates the shortest path of unmanned plane traversal disease suspicion point, simulated annealing using simulated annealing
Algorithm is provided that
Determine that solution space S, solution space S are the set in all paths that unmanned plane traverses a n point；
The objective function f (x) of solution space S all areas is set, objective function f (x) is the road that unmanned plane traverses all the points
Diameter total length, objective function f (x) such as following formula:
Wherein, x=1,2 ..., n, n are the number that unmanned plane traverses primary all the points；
Random number r when generating neighborhood solution is set, and random number r takes 0.5；
The cooling processing mode of simulated annealing is arranged to cool down for equal proportion, equal proportion cooling such as following formula:
tk+1=α tk
Wherein, k=0,1,2 ..., α take 0.95, tkIndicate Current Temperatures, tk+1Indicate the temperature at lower a moment；
The default outer circulation number as termination condition.
Compared with prior art, the invention has the following beneficial effects:
The present invention is based on the method for optimizing route of the expressway slope of unmanned plane inspection to carry out inspection energy using unmanned plane
During enough solving manual inspection, in the defect that precipitous position patrolman that is dangerous, being difficult to climb can not patrol, and utilize
Unmanned plane can accomplish all standing, not stay blind spot, have comprehensive；The present invention is by utilizing UAV Maneuver and flexibility
Feature carries out quick image data acquiring to the side slope position that needs are observed, and can take the position of precipitous danger,
Comprehensive discovery security risk in time, greatly reduces the workload of patrolman, improves the safety of staff；This hair
The bright environmental characteristic by analysis of slope, classifies to reference object, determines in conjunction with unmanned plane shooting and airmanship feature
Style of shooting, to guarantee the quality of inspection task, the image shot using manual analysis unmanned plane is checked, extracts figure
Disease suspicion point existing for side slope in piece；According to patrolling for the first time as a result, calculate the distance between each disease suspicion point, and construct
The adjacency matrix of disease suspicion point；It recycles the adjacency matrix of building and simulated annealing to solve unmanned plane and traverses collection point
Shortest path, in the hope of shortest path as secondary inspection path, secondary inspection is carried out to disease suspicion point, thus fastly
Speed, the evolution process for legibly grasping slide slope disease.
Detailed description of the invention
Fig. 1 is shooting direction and projecting direction schematic diagram in method for optimizing route of the present invention；
Fig. 2 side slope schematic diagram, wherein Fig. 2 (a) is step-like side slope schematic diagram, and Fig. 2 (b) is non-step-like side slope signal
Figure；
Fig. 3 is intercepting ditch form of fracture, and wherein Fig. 3 (a) is trapezoidal intercepting ditch section, and Fig. 3 (b) is disconnected for rectangle intercepting ditch
Face, Fig. 3 (c) are L shape intercepting ditch section, and Fig. 3 (d) is V-arrangement intercepting ditch section；
Fig. 4 is slope surface style of shooting schematic diagram in method for optimizing route of the present invention；
Fig. 5 is intercepting ditch style of shooting schematic diagram in method for optimizing route of the present invention；
Fig. 6 is that step-like side slope boat takes the photograph schematic diagram in method for optimizing route of the present invention；
Fig. 7 is the secondary inspection collection point distribution map chosen in the embodiment of the present invention；
Fig. 8 is that simulated annealing solves process；
Fig. 9 is the optimization inspection path acquired in the embodiment of the present invention by simulated annealing.
Specific embodiment
It is next with reference to the accompanying drawings and examples that the present invention is further illustrated.
The present invention is based on the method for optimizing route of the expressway slope of unmanned plane inspection, including picture collection part and calculation
Method solves optimal path part, and each section is specific as follows:
1, picture collection part controls unmanned plane shooting quality, to obtain best shooting angle, claps unmanned plane
Taking the photograph quality and carrying out control includes pickup quality control and flight reappearance control；
1.1, pickup quality control is to be carried out according to weather, season, region, time and projecting direction to unmanned plane, specifically
Are as follows:
Weather: weather is divided into six classes, respectively strong sunlight, direct sunlight, thin cloud fine day, bright cloudy day, wet weather
The light exposure of weather and dense black overcast and rainy six class, weather of the weather of latter class relative to preceding one kind, unmanned plane increases by one grade.
Season: season is successively divided according to winter, summer and spring, or is successively drawn according to winter, summer and autumn
Point, the latter season increases by one grade compared to previous season, the light exposure of unmanned plane.
Region: when carrying out pickup quality control to unmanned plane according to region, latitude is every to increase by 15 °, and unmanned plane aperture is corresponding
Half grade or one grade of increase；After having adjusted unmanned plane aperture according to latitude, for height above sea level in 1000m or less, unmanned plane aperture is big
It is small to remain unchanged；For height above sea level in 1000m~2000m, unmanned plane aperture reduces one grade；Height above sea level in 2000m~4000m, nobody
Machine aperture reduces two grades.
Time: when carrying out pickup quality control to unmanned plane according to the time, choosing the period on daytime, with local positive morning
The period within 1h is as reference afterwards；The unmanned plane aperture at local sunrise sunset moment than the period within 1h before and after high noon without
The light exposure of man-machine aperture reduces third gear；After sunrise within 1h and before sunset within 1h the light exposure of unmanned plane aperture compared with high noon
Light exposure within the 1h of front and back reduces two grades；In 1h~2h before 1h~2h and sunset after sunrise the light exposure of unmanned plane compared with
Light exposure before and after high noon within 1h reduces one grade；
Projecting direction referring to Fig.1: when carrying out pickup quality control to unmanned plane according to projecting direction, projecting direction is divided into
Five classes, respectively frontlighting, along sidelight, sidelight, side backlight and backlight, according to this sequentially on the basis of frontlighting is shot f-number according to
It is secondary to increase half grade；When sunlight is in the top of unmanned plane camera primary optical axis, sunlight incident direction and unmanned plane camera key light
When the β of the angle of axis direction is 75 °~90 ° (i.e. ω≤15 ° in Fig. 1), shooting direction is almost the same with projecting direction, as suitable
Light；When sunlight is in the top of unmanned plane camera primary optical axis, sunlight incident direction and unmanned plane camera key light axis direction
When the β of angle is 45 °~75 ° (i.e. 15 ° of ω≤45 ° < in Fig. 1), as along sidelight；When sunlight is in unmanned plane phase owner
The top of optical axis, the β of the angle of sunlight incident direction and unmanned plane camera key light axis direction (i.e. 45 < in Fig. 1 that are 0 °~45 °
ω≤90 °) when, as sidelight；When sunlight is in the lower section of unmanned plane camera primary optical axis, sunlight incident direction and unmanned plane
That is, being side backlight when the β of the angle of camera key light axis direction is 0 °~45 ° (i.e. 90 ° of ω≤135 ° < in Fig. 1)；Work as sunlight
The β of the angle of lower section in unmanned plane camera primary optical axis, sunlight incident direction and unmanned plane camera key light axis direction is 45 °
When~75 ° (i.e. 135 ° of ω≤165 ° < in Fig. 1), as against sidelight；When sunlight is in the lower section of unmanned plane camera primary optical axis,
The β of the angle of sunlight incident direction and unmanned plane camera key light axis direction be 75 °~90 ° (i.e. 165 ° of < ω in Fig. 1≤
180 °) when, it is backlight.
The control of 1.2 flight reappearances, including control photograph overlap degree and ground sampling interval
It should be avoided using unmanned plane inspection and the case where acquisition zone is missed occur, guarantee that the image of acquisition is not in dynamic
It is fuzzy and being capable of the progress interpretation of side slope state.The present invention is controlled it from degree of overlapping and ground sampling interval to guarantee matter
Amount.
Photo degree of overlapping
Guarantee that aerophoto can be overlapped the entire air cover domain of covering, in order to avoid there is " aerial photographic gap ".Endlap is set
Spending qx range is 60%~65%, and sidelapping degree qy is in 25%-35%.
The ground sampling interval (GSD)
In digital airborne photography, exists between ground sampling interval, flying height, pixel dimension and the camera lens principal focal distance of camera and close
It is formula:
In formula, H unmanned plane distance boat takes the photograph the relative altitude in subregion mean baseline face；The lens focus of f expression camera；μ table
Show the physical size of pixel in photosensitive element.
The classification and selection of 1.3 styles of shooting
The present invention needs side slope to classify according to research.By side slope be divided into step-like side slope (such as Fig. 2 (a) shown in) and
Non- step-like side slope (shown in such as Fig. 2 (b)).
Intercepting ditch form of fracture is divided into trapezoidal intercepting ditch (shown in such as Fig. 3 (a)), rectangle intercepting ditch (such as according to end pattern
Shown in Fig. 3 (b)), L shape intercepting ditch (shown in such as Fig. 3 (c)) and V-arrangement intercepting ditch (shown in such as Fig. 3 (the d)) present invention utilizing nobody
When machine carries out Image Acquisition, slope surface and intercepting ditch are separately carried out.
Herein in slope images acquisition, by the way of vertical photography, keep unmanned plane camera primary optical axis vertical with slope surface,
The surface characteristics of side slope can be more intuitively shown, as shown in Figure 4.
As shown in figure 5, being shot in such a way that camera primary optical axis is perpendicular to bottom of trench for the intercepting ditch present invention, guarantee
Primary optical axis is in K1K1' range is interior, it is possible to prevente effectively from the appearance of acquisition blind area.K1K1' be sun light direct beam into bottom of trench most
On a large scale.
The slope surface inclination angle of non-step-like side slope is general constant or variation is little.For the side slope of not intercepting ditch, no
Need to divide air cover domain.Slope surface is split into facet one by one by " step " when step-like side slope boat is taken the photograph, when slopes at different levels
When the inclination angle in face is identical, i.e. β1=β2=β3=β, as shown in fig. 6, camera lens can keep identical and incline when unmanned plane is shot
Angle θpIt completes the shooting of entire slope surface, when shooting, makes camera lens primary optical axis and slope surface perpendicular as far as possible.According to certain overlapping
Degree keeps the direction in course line consistent with the trend of road.
2. solving optimal path using algorithm
According to the specific photographed scene of unmanned plane, after adjusting unmanned plane according to above-mentioned shooting quality control mode, recycle
Unmanned plane adjusted carries out the Image Acquisition of full coverage type to expressway slope, and the image transmitting of acquisition to inspection people
Member.Patrolman the distance between finds disease suspicion point, then calculates each disease suspicion point by the side slope image of observation acquisition,
And construct the adjacency matrix of disease suspicion point；Finally using the adjacency matrix of building as the input of simulated annealing, calculate
To the shortest path of unmanned plane traversal disease suspicion point.
Embodiment
It in the present embodiment, is tested by taking the peaceful high speed side slope K1308+450 of the Chinese as an example, first according to the situation of locality when experiment
Unmanned plane is adjusted, and is shot by unmanned plane, after shooting, patrolman is according to the image of shooting inspection people
Member finds disease suspicion point by the side slope image of observation acquisition, on the basis of first inspection, shares 14 places as two
The collection point of secondary inspection, each collection point side slope distribution situation as shown in fig. 7, collection point 14 be unmanned plane rise, drop point.It examines
The vehicle of road dealing and the safety of unmanned plane are considered, first by unmanned plane vertical ascent to 8 meters or so of the surface of landing point
Position, i.e. collection point 7 as inspection task start rise, stop, be also the beginning and end of path planning.
Since the location information of collection point is longitude, latitude and height, need first to convert thereof into three-dimensional rectangular coordinate, then
It is solved.
Unmanned plane generally mostly uses global position system GPS coordinate system, i.e. WGS-84 coordinate system at present.If WGS-84
The geodetic coordinates of collection point is (B, L, H) under coordinate system, and wherein B, L and H respectively represent geodetic latitude, longitude and altitude,
It is (X, Y, Z) corresponding to the coordinate in space three-dimensional rectangular coordinate system, there is corresponding relational expression therebetween.With Han Ninggao
It is tested for fast side slope K1308+450, by their corresponding relationship, is acquired between each collection point using MATLAB software
Distance, form 13 × 13 adjacency matrix, adjacency matrix is as follows:
When optimizing application algorithm solves shortest path, this adjacency matrix is input to algorithm routine.
The present embodiment obtains the optimal path of secondary inspection using simulated annealing.The operation process of simulated annealing
As shown in figure 8, the step of simulated annealing is described as by initial solution i and control parameter t, to current in cycle period
Solution repeats the iteration of " generating new explanation --- deviation of calculating target function --- to give up/receive ", and the t value that gradually decays, directly
It terminates to obtain approximate optimal solution to algorithm.
When optimizing application algorithm solves shortest path, the setting of simulated annealing is first carried out:
1) solution space S is determined first, and solution space S is the collection in all paths that unmanned plane traverses a n collection point just
It closes, then S=[Sij] is the arrangement of (1,2 ..., n), and Sij indicates jth time access collection point i；
2) the objective function f (x) of solution space S all areas is set, the objective function f (x) of the present embodiment is unmanned plane time
The path total length for going through all collection points, sees below formula:
Wherein, x=1,2 ..., n, n are the number of the primary all collection points of unmanned plane traversal；
3) neighborhood solution is the new explanation generated according to certain rules, should make the new explanation generated empty throughout entire solution as far as possible
Between, it is main that new explanation is generated using SWAP exchange, backout, when generating neighborhood solution, arbitrary constant r is set, this reality
It applies random number r in example and takes 0.5: generating a random number r i.e. between 0 to 1, if r≤0.5, generated using SWAP transformation new
Solution；If r > 0.5, new explanation is generated using backout；
4) when carrying out cooling processing using simulated annealing, the cooling method that the present embodiment uses is equal proportion cooling, i.e.,
Each temperature all declines in the same scale, equal proportion cooling such as following formula:
tk+1=α tk
In formula, k=0,1,2 ...；α ∈ (0,1), tkIndicate Current Temperatures, tk+1Indicate the temperature at lower a moment.α is constant
And closer to 1, temperature decline is slower, and the present embodiment selects α=0.95；
5) Probability p that simulated annealing is obtained according to above-mentioned cooling processing and in conjunction with Metropolis criterion determines
Whether receive new explanation to replace old solution to become current solution, continue to generate next new explanation if not receiving, otherwise becomes current
Solution.The calculating of Probability p such as following formula:
In formula, Δ f is the deviation of new explanation and objective function, and T is the Current Temperatures of system；
6) stop criterion is set, what interior loop termination criterion, that is, Metropolis criterion was used to determine to generate under different temperatures
Candidate solution quantity, outer loop end, that is, algorithm termination terminate, and the present embodiment is using preset outer circulation number as termination condition.
After simulated annealing is provided with, the adjacency matrix for having acquired 13 × 13 is input to simulation as test data
In annealing algorithm program, shortest path is solved.
Emulation experiment is carried out using MATLAB, is tested by taking the peaceful high speed side slope K1308+450 of the Chinese as an example, obtained path
The results are shown in Table 1:
Table 1
| SA-solution=11 2 13 84 12 791 10 365 |
| SA-length=161.102 |
As shown in Table 1, the unmanned plane side slope with collection point 7 for task path, stop that simulated annealing acquires is used
The path of secondary inspection are as follows: 14- > 7- > 9- > 1- > 10- > 3- > 6- > 5- > 11- > 2- > 13- > 8- > 4- > 12- > 7- > 14, unmanned plane
Global path total length is needed plus the distance between unmanned plane landing point and task start-stop point, as 177m.It is moved back by simulation
Overlay of the path that pyrogenic process obtains in side slope is as shown in Figure 9.
To sum up, the present invention has understood the terrain issues of expressway slope in depth, checks such as intercepting ditch, gutter facility
Concrete condition, the inspection boat for having selected this unmanned plane to carry out expressway slope takes the photograph task, and flying quality is stable, aerial photographing effect
It is good, shrinkable easy to carry, and the barrier in 30 meters of front can be farthest perceived, it is suitble to work in locating for expressway slope
More complicated environment.
The present invention the side slope inspection of highway maintenance department actual demand and there are aiming at the problem that, have following four big characteristics
Advantage: (1) comprehensive: during side slope manual inspection, can not to be patrolled in precipitous position patrolman that is dangerous, being difficult to climb
It looks into, can accomplish all standing using unmanned plane inspection, not stay blind spot.
(2) side slope high efficiency: is considerably increased due to the accessibility of side slope low feature for the manual inspection of side slope
The difficulty of inspection work.A large amount of time and energy are taken, workload is increased, reduces the whole work of road maintenance inspection
Make efficiency.For this problem, the characteristics of present invention is by utilizing UAV Maneuver and flexibility, the side slope that needs are observed
Position carries out quick image data acquiring, and can take the position of precipitous danger, and comprehensive discovery safety is hidden in time
Suffer from.The workload for greatly reducing patrolman improves the safety of staff.
(3) intelligent: the present invention is automatically positioned disease position using image processing techniques intelligent recognition, avoids the people that should patrol
Member's subjective judgement finds security risk not in time；Maintenance department is instructed to carry out the processing to danger source in time, thus by accident wind
Danger is reduced in controlled range；Predict and optimize inspection management to provide reference for slide slope disease.
(4) convenience: traditional slope of highway inspection work can only be filled in recording table for check by patrolman and submit and deposit
Shelves, paper record are not easy to check mainly with verbal description.The present invention be then by side slope information observed by unmanned plane it is automatic on
Slope data library is passed to, corresponding information is generated into side slope inspection picture, does not need patrolman's hand filling.
Claims (10)
1. a kind of method for optimizing route of the expressway slope inspection based on unmanned plane, which comprises the steps of:
Step 1, unmanned plane shooting quality is controlled, to obtain best shooting angle；
Step 2, classify to reference object, and style of shooting is determined according to classification situation；
Step 3, it needs the object shot to be checked according to the picture of shooting by unmanned plane shooting, extracts side in picture
Disease suspicion point existing for slope；
Step 4, the distance between each disease suspicion point is calculated, and constructs the adjacency matrix of disease suspicion point；
Step 5, unmanned plane traversal disease is calculated as the input of simulated annealing in adjacency matrix step 4 constructed
The shortest path of suspicion point.
2. a kind of method for optimizing route of expressway slope inspection based on unmanned plane according to claim 1, special
Sign is, in the step 1, carrying out control to unmanned plane shooting quality includes that pickup quality control and flight reappearance control；
Pickup quality control is to be carried out according to weather, season, region, time and projecting direction to unmanned plane, specifically:
Weather is divided into six classes, respectively strong sunlight, direct sunlight, thin cloud fine day, bright cloudy day, rainy weather and dense black
It is overcast and rainy；The light exposure of weather of the weather of latter class relative to preceding one kind, unmanned plane increases by one grade；
Season is successively divided according to winter, summer and spring, or is successively divided according to winter, summer and autumn, the latter
Season increases by one grade compared to previous season, the light exposure of unmanned plane；
When carrying out pickup quality control to unmanned plane according to region, latitude is every to increase by 15 °, and unmanned plane aperture accordingly increases half grade
Or one grade；After having adjusted unmanned plane aperture according to latitude, in 1000m or less, unmanned plane aperture size remains unchanged height above sea level；
For height above sea level in 1000m~2000m, unmanned plane aperture reduces one grade；For height above sea level in 2000m~4000m, unmanned plane aperture reduces two
Shelves；
When carrying out pickup quality control to unmanned plane according to the time, the period on daytime is chosen, within 1h before and after local high noon
Period as reference；The unmanned plane aperture at local sunrise sunset moment is than period unmanned plane aperture within 1h before and after high noon
Light exposure reduce third gear；After sunrise within 1h and before sunset within 1h after the light exposure calibration morning of unmanned plane aperture 1h with
Interior light exposure reduces two grades；In 1h~2h before 1h~2h and sunset after sunrise after the light exposure calibration morning of unmanned plane
Light exposure within 1h reduces one grade；
When carrying out pickup quality control to unmanned plane according to projecting direction, projecting direction is divided into five classes, respectively frontlighting, suitable side
Light, sidelight, side backlight and backlight；When sunlight is in the top of unmanned plane camera primary optical axis, sunlight incident direction and nobody
It is frontlighting when the β of the angle of machine camera key light axis direction is 75 °~90 °；When sunlight is in the upper of unmanned plane camera primary optical axis
When the β of the angle of side, sunlight incident direction and unmanned plane camera key light axis direction is 45 °~75 °, for along sidelight；Work as the sun
Light is in the top of unmanned plane camera primary optical axis, and sunlight incident direction and the β of the angle of unmanned plane camera key light axis direction are
It is sidelight at 0 °~45 °；When sunlight is in the lower section of unmanned plane camera primary optical axis, sunlight incident direction and unmanned plane phase
It is side backlight when the β of the angle of owner's optical axis direction is 0 °~45 °；When sunlight is in the lower section of unmanned plane camera primary optical axis,
It is inverse sidelight when the β of the angle of sunlight incident direction and unmanned plane camera key light axis direction is 45 °~75 °；At sunlight
In the lower section of unmanned plane camera primary optical axis, the β of the angle of sunlight incident direction and unmanned plane camera key light axis direction is 75 °~
It is backlight at 90 °；
Flight reappearance control includes control photograph overlap degree and ground sampling interval.
3. a kind of method for optimizing route of expressway slope inspection based on unmanned plane according to claim 2, special
Sign is, when control photograph overlap is spent, setting ship's control qx range is 60%~65%, and sidelapping degree qy is in 25%-
35%.
4. a kind of method for optimizing route of expressway slope inspection based on unmanned plane according to claim 2, special
Sign is that ground sampling interval GSD is calculate by the following formula:
In formula, H indicates that unmanned plane distance boat takes the photograph the relative altitude in subregion mean baseline face；The lens focus of f expression camera；μ table
Show the physical size of pixel in unmanned plane photosensitive element.
5. a kind of method for optimizing route of expressway slope inspection based on unmanned plane according to claim 1, special
Sign is, in the step 2, reference object is divided into side slope and intercepting ditch, and side slope is divided into step-like side slope and non-step-like side slope,
Intercepting ditch is divided into trapezoidal intercepting ditch, rectangle intercepting ditch, L shape intercepting ditch and V-arrangement intercepting ditch, unmanned plane side slope according to end pattern
It is separately shot with intercepting ditch.
6. a kind of method for optimizing route of expressway slope inspection based on unmanned plane according to claim 5, special
Sign is, when side slope is shot, by the way of vertical photography, keeps unmanned plane camera primary optical axis vertical with slope surface.
7. a kind of method for optimizing route of expressway slope inspection based on unmanned plane according to claim 5, special
Sign is, when shooting to step-like side slope, slope surface is split into several facets by step, when the inclination angle of slope surfaces at different levels
When identical, camera lens keep identical inclination angle to complete the shooting of entire slope surface when unmanned plane is shot.
8. a kind of method for optimizing route of expressway slope inspection based on unmanned plane according to claim 5, special
Sign is, when shooting to intercepting ditch, in such a way that unmanned plane camera primary optical axis is perpendicular to bottom of trench, camera primary optical axis exists
K1K1' in range, K1K1' it is maximum magnitude of the sun light direct beam into bottom of trench.
9. a kind of method for optimizing route of expressway slope inspection based on unmanned plane according to claim 1, special
Sign is, in the step 4, the adjacency matrix process for constructing disease suspicion point is as follows:
Step 4.1, it determines the landing point of unmanned plane, obtains the location information of unmanned plane landing point and each disease suspicion point, institute
Stating location information includes longitude, latitude and height；
Step 4.2, the distance between each point is calculated, constructs adjacency matrix using the distance between each point, each point includes nothing
Man-machine landing point and each disease suspicion point.
10. a kind of method for optimizing route of expressway slope inspection based on unmanned plane according to claim 1, special
Sign is that the step 5 calculates the shortest path of unmanned plane traversal disease suspicion point using simulated annealing, and simulated annealing is calculated
Method is provided that
Determine that solution space S, solution space S are the set in all paths that unmanned plane traverses a n point；
The objective function f (x) of solution space S all areas is set, objective function f (x) is that the path of unmanned plane traversal all the points is total
Length, objective function f (x) such as following formula:
Wherein, x=1,2 ..., n, n are the number that unmanned plane traverses primary all the points；
Random number r when generating neighborhood solution is set, and random number r takes 0.5；
The cooling processing mode of simulated annealing is arranged to cool down for equal proportion, equal proportion cooling such as following formula:
tk+1=α tk
Wherein, k=0,1,2 ..., α take 0.95, tkIndicate Current Temperatures, tk+1Indicate the temperature at lower a moment；Default be used as terminates item
The outer circulation number of part.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811118398.3A CN109211202B (en) | 2018-09-21 | 2018-09-21 | Unmanned aerial vehicle-based highway slope patrol path optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811118398.3A CN109211202B (en) | 2018-09-21 | 2018-09-21 | Unmanned aerial vehicle-based highway slope patrol path optimization method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109211202A true CN109211202A (en) | 2019-01-15 |
| CN109211202B CN109211202B (en) | 2020-12-18 |
Family
ID=64981503
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811118398.3A Active CN109211202B (en) | 2018-09-21 | 2018-09-21 | Unmanned aerial vehicle-based highway slope patrol path optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109211202B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110243374A (en) * | 2019-06-26 | 2019-09-17 | 贵州省交通规划勘察设计研究院股份有限公司 | A kind of unmanned plane airline generation method convenient for side slope image collection |
| CN111595853A (en) * | 2020-05-29 | 2020-08-28 | 中国科学院、水利部成都山地灾害与环境研究所 | Slope disease inspection method and system |
| CN111751502A (en) * | 2020-07-08 | 2020-10-09 | 中国计量大学 | Multi-unmanned-aerial-vehicle cooperative pollutant tracing method based on improved simulated annealing |
| WO2020237478A1 (en) * | 2019-05-27 | 2020-12-03 | 深圳市大疆创新科技有限公司 | Flight planning method and related device |
| CN112146615A (en) * | 2020-09-23 | 2020-12-29 | 招商局重庆交通科研设计院有限公司 | Slope monitoring method based on multiple unmanned aerial vehicles |
| JP2021022846A (en) * | 2019-07-29 | 2021-02-18 | 砂防エンジニアリング株式会社 | Inspection method and inspection system |
| CN112461205A (en) * | 2020-11-12 | 2021-03-09 | 中国铁路设计集团有限公司 | Method for manufacturing cross section of existing railway line based on unmanned aerial vehicle oblique photogrammetry |
| CN112904894A (en) * | 2021-01-19 | 2021-06-04 | 招商局重庆交通科研设计院有限公司 | Slope live-action image acquisition method based on unmanned aerial vehicle oblique photography |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102147255A (en) * | 2011-01-12 | 2011-08-10 | 北京航空航天大学 | Real-time path planning method for unmanned aerial vehicle group under threat information sharing environment |
| CN105487554A (en) * | 2016-01-12 | 2016-04-13 | 武汉顶翔智控科技有限公司 | Multi-rotor unmanned aerial vehicle automatic homeward flight path planning algorithm |
| CN106969778A (en) * | 2017-02-28 | 2017-07-21 | 南京航空航天大学 | A kind of multiple no-manned plane cooperates with the paths planning method of dispenser |
| CN206990574U (en) * | 2017-04-19 | 2018-02-09 | 文翀 | Rural highway critical forecast system |
| JP6623184B2 (en) * | 2017-01-31 | 2019-12-18 | 日本電信電話株式会社 | Global structure extraction apparatus, method, and program for multilayer neural network |
-
2018
- 2018-09-21 CN CN201811118398.3A patent/CN109211202B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102147255A (en) * | 2011-01-12 | 2011-08-10 | 北京航空航天大学 | Real-time path planning method for unmanned aerial vehicle group under threat information sharing environment |
| CN105487554A (en) * | 2016-01-12 | 2016-04-13 | 武汉顶翔智控科技有限公司 | Multi-rotor unmanned aerial vehicle automatic homeward flight path planning algorithm |
| JP6623184B2 (en) * | 2017-01-31 | 2019-12-18 | 日本電信電話株式会社 | Global structure extraction apparatus, method, and program for multilayer neural network |
| CN106969778A (en) * | 2017-02-28 | 2017-07-21 | 南京航空航天大学 | A kind of multiple no-manned plane cooperates with the paths planning method of dispenser |
| CN206990574U (en) * | 2017-04-19 | 2018-02-09 | 文翀 | Rural highway critical forecast system |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020237478A1 (en) * | 2019-05-27 | 2020-12-03 | 深圳市大疆创新科技有限公司 | Flight planning method and related device |
| CN110243374A (en) * | 2019-06-26 | 2019-09-17 | 贵州省交通规划勘察设计研究院股份有限公司 | A kind of unmanned plane airline generation method convenient for side slope image collection |
| JP2021022846A (en) * | 2019-07-29 | 2021-02-18 | 砂防エンジニアリング株式会社 | Inspection method and inspection system |
| CN111595853A (en) * | 2020-05-29 | 2020-08-28 | 中国科学院、水利部成都山地灾害与环境研究所 | Slope disease inspection method and system |
| CN111751502A (en) * | 2020-07-08 | 2020-10-09 | 中国计量大学 | Multi-unmanned-aerial-vehicle cooperative pollutant tracing method based on improved simulated annealing |
| CN111751502B (en) * | 2020-07-08 | 2022-07-05 | 中国计量大学 | Multi-unmanned-aerial-vehicle cooperative pollutant tracing method based on improved simulated annealing |
| CN112146615A (en) * | 2020-09-23 | 2020-12-29 | 招商局重庆交通科研设计院有限公司 | Slope monitoring method based on multiple unmanned aerial vehicles |
| CN112461205A (en) * | 2020-11-12 | 2021-03-09 | 中国铁路设计集团有限公司 | Method for manufacturing cross section of existing railway line based on unmanned aerial vehicle oblique photogrammetry |
| CN112904894A (en) * | 2021-01-19 | 2021-06-04 | 招商局重庆交通科研设计院有限公司 | Slope live-action image acquisition method based on unmanned aerial vehicle oblique photography |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109211202B (en) | 2020-12-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109211202A (en) | A kind of method for optimizing route of the expressway slope inspection based on unmanned plane | |
| CN112422783B (en) | Unmanned aerial vehicle intelligent patrol system based on parking apron cluster | |
| US11378718B2 (en) | Unmanned aerial vehicle system and methods | |
| CN109416413A (en) | Solar energy forecast | |
| CN110276254B (en) | Unmanned aerial vehicle-based automatic recognition and early warning method for bootlegged area bootlegged | |
| US20210221506A1 (en) | Unmanned aerial vehicle system and methods | |
| CN109063532A (en) | A kind of field lost contact personnel's method for searching based on unmanned plane | |
| Adami et al. | Ultra light UAV systems for the metrical documentation of cultural heritage: Applications for architecture and archaeology | |
| CN112649900A (en) | Visibility monitoring method, device, equipment, system and medium | |
| CN112857356A (en) | Unmanned aerial vehicle water body environment investigation and air route generation method | |
| CN111460687A (en) | Simulation evaluation analysis-based camouflage scheme aided design system and method | |
| KR102262382B1 (en) | large area space information system for image acquistion base Unmanned Aerial Vehicle | |
| KR102548454B1 (en) | method for recommending installation location and installation angle of solar panel | |
| CN112904894A (en) | Slope live-action image acquisition method based on unmanned aerial vehicle oblique photography | |
| RU2716477C1 (en) | System for monitoring forest use and forest pathological changes | |
| CN115601471B (en) | Drawing method based on large unmanned aerial vehicle photoelectric reconnaissance swept area | |
| CN114659499B (en) | Smart city 3D map model photography establishment method based on unmanned aerial vehicle technology | |
| CN114627252B (en) | Unmanned plane for acquiring surface temperature distribution and surface temperature distribution map acquisition method | |
| CN115061497B (en) | Unmanned aerial vehicle photovoltaic panel inspection method and system | |
| CN114185365B (en) | Aviation camera controller supporting different-speed image motion compensation function | |
| CN113674343B (en) | Unmanned aerial vehicle inspection method, unmanned aerial vehicle inspection device, unmanned aerial vehicle inspection equipment and storage medium | |
| CN114440836A (en) | Unmanned aerial vehicle photogrammetry modeling method with glass curtain wall building | |
| CN115373416A (en) | Intelligent inspection method for railway power through line | |
| Yang et al. | Measuring the area of cultivated land reclaimed from rural settlements using an unmanned aerial vehicle | |
| Yuanting et al. | Research on image mosaic of low altitude UAV based on harris corner detection |
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 |