CN109084735B - A kind of tunnel monitoring abnormal state method based on unmanned plane device - Google Patents
A kind of tunnel monitoring abnormal state method based on unmanned plane device Download PDFInfo
- Publication number
- CN109084735B CN109084735B CN201810996777.6A CN201810996777A CN109084735B CN 109084735 B CN109084735 B CN 109084735B CN 201810996777 A CN201810996777 A CN 201810996777A CN 109084735 B CN109084735 B CN 109084735B
- Authority
- CN
- China
- Prior art keywords
- unmanned plane
- tunnel
- barrier
- degree
- unmanned
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 46
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 30
- 230000004888 barrier function Effects 0.000 claims abstract description 62
- 239000011159 matrix material Substances 0.000 claims description 35
- 230000005856 abnormality Effects 0.000 claims description 7
- 239000013598 vector Substances 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000007613 environmental effect Effects 0.000 claims description 5
- 230000006641 stabilisation Effects 0.000 claims description 5
- 238000011105 stabilization Methods 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 claims description 3
- 238000007667 floating Methods 0.000 claims description 3
- 238000009434 installation Methods 0.000 claims description 3
- 210000004556 brain Anatomy 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 4
- 230000002708 enhancing effect Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 2
- 241001669679 Eleotris Species 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004080 punching Methods 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
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
- 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)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Multimedia (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The invention discloses a kind of tunnel monitoring abnormal state method and unmanned plane devices, belong to air vehicle technique field.The device includes 550 wheelbase unmanned planes and the microcomputer that is connected with unmanned plane.Environment of the unmanned plane after entering in tunnel in whole shooting record tunnel, whether there are obstacles for real-time monitoring head diagonally above and frontward, horizontal direction and oblique front lower place, if having, unmanned plane is then measured respectively at a distance from barrier, when unmanned plane is less than or equal to its radius of safety at a distance from any one barrier, unmanned plane is in conflict situation, microcomputer recalculates and moves the position of unmanned plane, until unmanned plane flight forward will not clash, then unmanned plane continues flight forward and continues to shoot in tunnel environment up through tunnel.If barrier is too big or can not pass through too much, unmanned plane returns to entrance according to former route.The present invention effectively ensures that unmanned plane during flying safety, and precision is high, saves human cost, improves the efficiency of monitoring.
Description
Technical field
The invention belongs to air vehicle technique fields, are related to the method for environment in monitoring tunnel.Specifically one kind is based on
The tunnel monitoring abnormal state method of unmanned plane device.
Background technique
In recent years, unmanned air vehicle technique continues to develop, to widen the application field of unmanned plane, society is for unmanned plane
Using also constantly being promoted, unmanned plane gradually plays a significant role in military, commercial and civilian.
Environmental monitoring in tunnel is a very important job, is made including the monitoring of barrier in tunnel, rail
It whether there is the monitoring etc. in crack with status monitoring and tunnel.
Traditional mode is monitored by manpower, not only consumes human cost, but also in the environment of some danger
It may threaten to the life security of patrol officer down.If monitoring tunnel by way of installing fixed high definition camera
Interior environment, it is not only at high cost, and also later period scalability is poor.
One effective method is the environment monitored in tunnel using unmanned plane.It is high by being installed on unmanned plane
Clear camera realizes the shooting to environment in tunnel, and carries out emphasis shooting to abnormal environment, and then realize later period monitoring.
Summary of the invention
In view of the above-mentioned problems, the invention proposes a kind of tunnel monitoring abnormal state method based on unmanned plane device, real
The unmanned plane self monitor of environment in existing tunnel.Unmanned plane enters from tunnel one end, and autonomous positioning flight is passed through tunnel, while nothing
Man-machine energy whole process records the environment in tunnel and carries out emphasis shooting to abnormal environment.When in tunnel there are when barrier, nobody
Function detects barrier and bypasses, if barrier can not pass through greatly very much, unmanned plane auto-returned entrance.
The tunnel monitoring abnormal state unmanned plane device, comprising: 550 wheelbase unmanned planes and be connected with unmanned plane
Microcomputer；Microcomputer is for calculating adjustment unmanned plane during flying strategy, controlling unmanned plane during flying and analyzing environmental abnormality.
The 550 wheelbase unmanned planes are in top, the horizontal and direction of bottom three, while 360 degree of tops of corresponding installation are swashed
Optical sensor, 360 degree of horizontal laser light rangefinders and 360 degree of bottom laser sensors；Before monitoring unmanned plane and head are oblique respectively
Top barrier, horizontal obstacle object and at a distance from the oblique front lower place barrier of head, and each sensor the data obtained is transmitted
Into microcomputer.
Meanwhile high definition camera is fixed on 550 wheelbase unmanned planes by three axis holders, unmanned plane is protected by light stream sensor
Hold the stabilization of itself and high definition camera.550 wheelbase unmanned planes carry forward sight high brightness supplementary lighting sources, bottom high brightness supplementary lighting sources
And air speed measuring apparatus.
The tunnel monitoring abnormal state unmanned plane device course of work is as follows:
After unmanned plane enters in tunnel, included forward sight high brightness supplementary lighting sources and bottom high brightness supplementary lighting sources are opened,
High definition camera starts the environment in whole shooting record tunnel simultaneously；Self-position, unmanned plane are determined using laser SLAM technology
360 degree of top lasers sensors, 360 degree of horizontal laser light rangefinders and 360 degree of bottom laser sensors distinguish real-time monitorings
Whether there are obstacles with head diagonally above and frontward for unmanned plane, unmanned plane horizontal direction whether there are obstacles and with unmanned plane with
Whether there are obstacles for the oblique front lower place of head.If unmanned plane does not encounter any obstacle in flight course, unmanned plane is pressed
Continue to fly according to when front direction and speed.
Otherwise, 360 degree of top lasers sensors, 360 degree of horizontal laser light rangefinders and 360 degree of bottom laser sensors point
Not Fan Hui unmanned plane and head diagonally above and frontward barrier, horizontal obstacle object and at a distance from the oblique front lower place barrier of head, when
When unmanned plane is less than or equal to its radius of safety at a distance from any one barrier, unmanned plane is in conflict situation.
After unmanned plane enters conflict situation, unmanned plane hovers at once, and microcomputer recalculates and moves unmanned plane
Position, until unmanned plane flight forward will not clash, then unmanned plane continues flight forward and continues to shoot in tunnel
Environment is up through tunnel.
If barrier is too big or can not pass through too much, unmanned plane returns to entrance according to former route.
Tunnel monitoring abnormal state method, the specific steps are as follows:
Step 1: 360 degree of top lasers sensors, 360 degree of horizontal laser light rangefinders and 360 degree will be installed on unmanned plane
Bottom laser sensor, high definition camera, supplementary lighting sources and air speed measuring apparatus；And it is connect with microcomputer and carries out debugging flight.
Step 2: unmanned plane after debugging is good is to flying in tunnel, high definition camera starts whole shooting record.
Step 3: each 360 degree of laser sensors of unmanned plane detect whether that there are barriers, and judge unmanned plane and each side
Whether the distance between upward barrier is less than or equal to radius of safety, if so, unmanned plane reaches conflict situation, enters step
Four, otherwise, enter step five；
Unmanned plane is regarded as a central point, in each of unmanned plane during flying process time step, is with the central point
Basis whether there is using three 360 degree of laser sensors to detect unmanned plane and tunnel top, bottom, both sides between the walls
Barrier, when not encountering any obstacle in flight course, then unmanned plane continues to fly according to when front direction and speed.
When encountering barrier, 360 degree of laser sensors in barrier direction calculate the central point and barrier
The distance between, if the distance is less than the radius of safety R of unmanned planecWhen, then determine that the unmanned plane enters conflict situation, otherwise
Do not enter conflict situation.
Step 4: microcomputer controls unmanned plane during flying, adjusts countermeasures and judge whether that conflict can be freed；If
It is to enter step five；Otherwise, unmanned plane shooting record obstacle information is returned by former route, and process terminates.
After unmanned plane enters conflict situation, speed is reduced to 0, carries out conflict Resolution in floating state:
Firstly, unmanned plane, centered on itself, x-axis is in the direction connected using tunnel both sides wall, with tunnel top bottom
The direction of connection is y-axis, establishes coordinate system as z-axis using the direction of unmanned plane during flying；
Then, it is L (L < R that unmanned plane, which measures the distance between barrier by 360 degree of laser sensors,c), with obstacle
The angle of line and x-y plane between object is θ, and barrier and unmanned plane are projected in x-y plane, at this time unmanned plane and barrier
Hindering the distance between object is Lcos θ, then it is R that unmanned plane, which frees conflict and needs the mobile shortest distance, in x-y planec-L·
Cos θ, direction are the direction along barrier and unmanned plane line.
Unmanned plane carries out conflict Resolution according to shortest moving distance, and unmanned plane continues to fly in tunnel after conflict Resolution
Row.
Similarly, when unmanned plane and multiple barriers clash simultaneously to free, unmanned plane shooting record obstacle
Object information is returned by former route.
Step 5: unmanned plane enhances algorithm using the feature of built-in double velocity correlation information, to the figure of high definition camera shooting
As carrying out at-once monitor analysis, and the position in output abnormality region carries out emphasis shooting.
Specific step is as follows:
It is special to carry out image by neural network for step 501, the shooting picture for inputting the channel 1024dpi × 1024dpi × 3
Sign is extracted, and the picture in the channel 8dpi × 8dpi × 256 is generated；
Step 502, each channel are one 8 × 8 matrix Asi(i=1,2 ..., 256), for each matrix AiIf
Xit=(Xit_1,Xit_2,…,Xit_8) it is 8 n-dimensional random variable ns that 8 transversal vectors of the matrix are constituted；
Step 503 utilizes 8 n-dimensional random variable n XitFurther calculate covariance matrix Cit；
Wherein, cit_jk=Cov (Xit_j,Xit_k) j, k=1,2 ..., 8；Xit_jAnd Xit_kRespectively 8 n-dimensional random variable ns
Component.
Step 504, for each matrix Ai, calculate 8 n-dimensional random variable n X of 8 column vectors of matrix compositioniv=
(Xiv_1,Xiv_2,…,Xiv_8)；
Step 505 utilizes 8 n-dimensional random variable n XivFurther calculate covariance matrix Civ；
Step 506, finally to each matrix Ai, utilize covariance matrix CitWith covariance matrix CivAfter obtaining feature enhancing
Matrix
Step 507, by the enhanced matrix of featureBy average value pond become a length be 512 dimension features to
Amount, then by full articulamentum, judge existing abnormal area, to carry out emphasis shooting to abnormal area.
Step 6: determining whether unmanned plane passes through tunnel, shooting is fully completed, if so, process terminates；Otherwise, it skips to
Step 2.
The present invention has the advantages that
1. a kind of tunnel monitoring abnormal state method based on unmanned plane device, can be real by way of laser positioning
Existing unmanned plane autonomous flight can navigate or make a return voyage in the tunnel of no GPS positioning signal, and by being equipped on fuselage
High brightness supplementary lighting sources and high definition camera shoot the environment in tunnel, so as to carry out later period monitoring.
2. a kind of tunnel monitoring abnormal state method based on unmanned plane device, may be implemented unmanned plane and encounters in tunnel
Autonomous conflict Resolution and instant image procossing after barrier carry out emphasis shooting to the abnormal area of image.
3. a kind of tunnel monitoring abnormal state method based on unmanned plane device, the unmanned plane of environment is autonomous in tunnel
Monitoring carries out the monitoring of environment in tunnel from improving monitoring safety and two aspect of high efficiency.For ensuring that unmanned plane flies
Row safety, the efficiency for improving monitoring have great importance.
4. a kind of tunnel monitoring abnormal state method based on unmanned plane device, can be realized the self monitor of unmanned plane,
Precision is high, can effectively solve environment monitoring problem in the tunnel of complex condition.Meanwhile the present invention can replace traditional manpower
Human cost is saved in monitoring, reduces the risk of inspection member work.
Detailed description of the invention
Fig. 1 is a kind of hardware architecture diagram of tunnel monitoring abnormal state unmanned plane device of the present invention；
Fig. 2 is a kind of flow chart of the tunnel monitoring abnormal state method based on unmanned plane device of the present invention；
Fig. 3 is the schematic diagram that unmanned plane of the present invention clashes in tunnel；
Fig. 4 is conflict Resolution schematic diagram of the unmanned plane of the present invention during autonomous flight；
Fig. 5 is that unmanned plane of the present invention carries out at-once monitor analysis flow chart diagram to shooting image；
1-550 wheelbase unmanned plane in figure, 2- microcomputer, 3-360 degree horizontal laser light rangefinder, 4- light stream sensor, 5-
360 degree of top lasers sensors, 6-360 degree bottom laser sensor, tri- axis holder of 7-, 8- high definition camera, 9- forward sight high brightness
Supplementary lighting sources, the bottom 10- high brightness supplementary lighting sources, 11- air speed measuring apparatus.
Specific embodiment
Below in conjunction with attached drawing, the present invention is described in further detail.
The tunnel monitoring abnormal state method based on unmanned plane device that the present invention provides a kind of realizes environment in tunnel
Unmanned plane self monitor.
The tunnel monitoring abnormal state method, specifically includes that through high definition camera the environment shot in tunnel, packet
Include rail state, barrier and tunnel slot etc. in tunnel；And image is carried out based on double velocity correlation information characteristics enhancing algorithm
Instant analysis shoots abnormal environment emphasis, and customizes the autonomous conflict Resolution strategy of the unmanned plane after encountering barrier.It is first
First pass through high brightness supplementary lighting sources solve the problems, such as it is very dark without light or light in tunnel, in combination with light stream sensor and three
Axis holder keeps the stabilization of high definition camera, to realize the normal photographing of environment in tunnel.Secondly in order to guarantee the peace of unmanned plane
Entirely, it is detected by laser sensor between unmanned plane and tunnel top, bottom, both sides wall and barrier that may be present
Distance, thus further determine unmanned plane countermeasures, finally realize the monitoring of environment in entire tunnel.
The tunnel monitoring abnormal state unmanned plane device, as shown in Figure 1, comprising: 550 wheelbase unmanned planes 1 and with
The connected microcomputer 2 of unmanned plane；Microcomputer 2 is for calculating adjustment unmanned plane during flying strategy, controlling unmanned plane during flying and dividing
Analyse environmental abnormality.
The 550 wheelbase unmanned planes 1 are in top, the horizontal and direction of bottom three, while 360 degree of tops of corresponding installation are swashed
5,360 degree of horizontal laser light rangefinders 3 of optical sensor and 360 degree of bottom laser sensors 6；It monitors unmanned plane respectively and head is oblique
Front upper place barrier, horizontal obstacle object and at a distance from the oblique front lower place barrier of head, and each sensor the data obtained is passed
It is defeated into microcomputer.
Meanwhile high definition camera 8 is fixed on 550 wheelbase unmanned planes 1 by three axis holders 7,550 wheelbase unmanned planes 1 are included
Forward sight high brightness supplementary lighting sources 9, bottom high brightness supplementary lighting sources 10 and air speed measuring apparatus 11, when detecting that wind speed is larger,
Unmanned aerial vehicle body generates vibration, opens the stabilization that light stream sensor 4 keeps itself fuselage and high definition camera 8.
The tunnel monitoring abnormal state unmanned plane device course of work is as follows:
After unmanned plane enters in tunnel, due to very dark without light or light in tunnel, included forward sight high brightness is opened
Supplementary lighting sources and bottom high brightness supplementary lighting sources, while high definition camera starts the environment in whole shooting record tunnel；Pass through bat
The picture taken the photograph carries out the artificial monitoring of environment in later period tunnel.
After unmanned plane enters tunnel, due to that may there is no GPS positioning signal in tunnel, determined certainly using laser SLAM technology
Body position, 360 degree of top lasers sensors of unmanned plane, 360 degree of horizontal laser light rangefinders and 360 degree of bottom laser sensors
Respectively real-time monitoring unmanned plane whether there are obstacles with head diagonally above and frontward, unmanned plane horizontal direction whether there are obstacles with
And whether there are obstacles with the oblique front lower place of head with unmanned plane.If unmanned plane does not encounter any barrier in flight course
Hinder, then unmanned plane continues to fly according to when front direction and speed.
Otherwise, 360 degree of top lasers sensors, 360 degree of horizontal laser light rangefinders and 360 degree of bottom laser sensors point
Not Fan Hui unmanned plane and head diagonally above and frontward barrier, horizontal obstacle object and at a distance from the oblique front lower place barrier of head, when
When unmanned plane is less than or equal to its radius of safety at a distance from any one barrier, unmanned plane is in conflict situation.
After unmanned plane enters conflict situation, unmanned plane hovers at once, and microcomputer recalculates and moves unmanned plane
Position, until unmanned plane flight forward will not clash, then unmanned plane continues flight forward and continues to shoot in tunnel
Environment is up through tunnel.
If barrier is too big or can not pass through too much, unmanned plane returns to entrance according to former route.
Unmanned plane can be handled immediately after the picture in shooting tunnel simultaneously, if it find that the environment in tunnel exists
It is abnormal, emphasis shooting can be carried out to abnormal scene.
Tunnel monitoring abnormal state method, as shown in Figure 2, the specific steps are as follows:
Step 1: 360 degree of top lasers sensors, 360 degree of horizontal laser light rangefinders and 360 degree will be installed on unmanned plane
Bottom laser sensor, high definition camera, supplementary lighting sources and air speed measuring apparatus；And it is connect with microcomputer and carries out debugging flight.
Step 2: unmanned plane after debugging is good is to flying in tunnel, high definition camera starts whole shooting record.
Since the environment in tunnel is very dark, forward sight high brightness supplementary lighting sources and bottom high brightness supplementary lighting sources are opened, thus
The environment in tunnel is shot using high definition camera.It might have irregular air-flow in tunnel simultaneously and cause unmanned plane machine
Body it is unstable, open light stream sensor to monitor and keep unmanned aerial vehicle body to stablize, while keeping using three axis holders high
The stabilization of clear camera further realizes instant and the later period tunnel so as to more clearly shoot the environment in record tunnel
Interior environmental monitoring.
Step 3: unmanned plane carries out conflict probe in flight course, each 360 degree of laser sensors detect whether there is barrier
Hinder object, and judge whether the distance between barrier is less than or equal to radius of safety in unmanned plane and all directions, if so, unmanned plane
Conflict situation is reached, four is entered step, otherwise, enters step five；
Unmanned plane is regarded as a central point, in each of unmanned plane during flying process time step, is with the central point
Basis whether there is using three 360 degree of laser sensors to detect unmanned plane and tunnel top, bottom, both sides between the walls
Barrier, when not encountering any obstacle in flight course, then unmanned plane continues to fly according to when front direction and speed.
As shown in figure 3,360 degree of laser sensors in barrier direction calculate the center when encountering barrier
The distance between point and barrier, if the distance is less than the radius of safety R of unmanned planecWhen, then determine that the unmanned plane enters punching
Prominent state, does not otherwise enter conflict situation.
Step 4: microcomputer controls unmanned plane during flying, adjusts countermeasures and judge whether that conflict can be freed；If
It is to enter step five；Otherwise, unmanned plane shooting record obstacle information is returned by former route, and process terminates.
After unmanned plane enters conflict situation, speed is reduced to 0, carries out conflict Resolution in floating state:
Firstly, unmanned plane, centered on itself, x-axis is in the direction connected using horizontal direction, that is, tunnel both sides wall, with perpendicular
The direction that histogram is connected to i.e. tunnel top bottom is y-axis, to build to direction, that is, unmanned plane during flying direction in tunnel as z-axis
Vertical coordinate system；
Then, it is assumed that conflict object is barrier, and unmanned plane is measured between barrier by 360 degree of laser sensors
Distance is L (L < Rc), the angle of line and x-y plane between barrier is θ, and barrier and unmanned plane are projected to x-y
In plane, as shown in figure 4, the distance between unmanned plane and barrier are Lcos θ, then the unmanned plane solution in x-y plane at this time
It is R that de- conflict, which needs the mobile shortest distance,c- Lcos θ, direction are the direction along barrier and unmanned plane line.
In order to reduce energy consumption to the greatest extent, increase the hours underway of unmanned plane, unmanned plane is carried out according to shortest moving distance
Conflict Resolution, unmanned plane continues to fly in tunnel after conflict Resolution.
Similarly, if unmanned plane clashes simultaneously with multiple barriers, unmanned plane still selects shortest path
Non conflicting of flying to region is further continued for flying in tunnel.If barrier is too many, it cannot free, unmanned plane shooting record obstacle
Object information is returned by former route.
Step 5: unmanned plane enhances algorithm using the feature of built-in double velocity correlation information, to the figure of high definition camera shooting
As carrying out at-once monitor analysis, and the position in output abnormality region carries out emphasis shooting.
As shown in Figure 5, the specific steps are as follows:
It is special to carry out image by neural network for step 501, the shooting picture for inputting the channel 1024dpi × 1024dpi × 3
Sign is extracted, and the picture in the channel 8dpi × 8dpi × 256 is generated；
Step 502, each channel are one 8 × 8 matrix Asi(i=1,2 ..., 256), for each matrix AiIf
Xit=(Xit_1,Xit_2,…,Xit_8) it is 8 n-dimensional random variable ns that 8 transversal vectors of the matrix are constituted；
Step 503 utilizes 8 n-dimensional random variable n XitFurther calculate covariance matrix Cit；
Wherein, cit_jk=Cov (Xit_j,Xit_k) j, k=1,2 ..., 8；Xit_jAnd Xit_kRespectively 8 n-dimensional random variable ns
Component.
Step 504, for each matrix Ai, calculate 8 n-dimensional random variable n X of 8 column vectors of matrix compositioniv=
(Xiv_1,Xiv_2,…,Xiv_8)；
Step 505 utilizes 8 n-dimensional random variable n XivFurther calculate covariance matrix Civ；
Step 506, finally to each matrix Ai, utilize covariance matrix CitWith covariance matrix CivAfter obtaining feature enhancing
Matrix
Step 507, by the enhanced matrix of featureBy average value pond become a length be 512 dimension features to
Amount, then by full articulamentum, abnormal area is judged whether there is, to carry out emphasis shooting, and output abnormality to abnormal area
The diagonal line coordinates in region.
It is if there is abnormal area (may be lack sleeper by rail, there are the reasons such as crack to cause for tunnel wall), then defeated
The diagonal line coordinates of abnormal area out, and emphasis shooting is carried out to abnormal area, to further realize the artificial observation in later period.
Step 6: determining whether unmanned plane passes through tunnel, shooting is fully completed, if it is, unmanned plane is to tunnel inner ring
The monitoring in border is completed；Otherwise, unmanned plane is to flying in tunnel, and continues to shoot.
The present invention completes the self monitor of unmanned plane under tunnel environment, can save human cost, improves monitoring efficiency,
Convenience is provided for subsequent artificial monitoring, there is positive meaning.
Claims (1)
1. a kind of tunnel monitoring abnormal state method based on unmanned plane device, the unmanned plane device include 550 wheelbases without
Microcomputer that is man-machine and being connected with unmanned plane；Microcomputer controls unmanned plane for calculating adjustment unmanned plane during flying strategy
It flies and analyzes environmental abnormality；
550 wheelbase unmanned planes correspond to 360 degree of top lasers sensors of installation in top, the horizontal and direction of bottom three,
360 degree of horizontal laser light rangefinders and 360 degree of bottom laser sensors；Monitoring unmanned plane and head diagonally above and frontward obstacle respectively
Object, horizontal obstacle object and is transferred to micro electric at a distance from the oblique front lower place barrier of head, and by each sensor the data obtained
In brain；
High definition camera is fixed on 550 wheelbase unmanned planes by three axis holders, unmanned plane by light stream sensor keep itself and
The stabilization of high definition camera；
The 550 wheelbase unmanned planes carry forward sight high brightness supplementary lighting sources, bottom high brightness supplementary lighting sources and measuring wind speed
Instrument；
It is characterized in that, specific step is as follows:
Step 1: 360 degree of top lasers sensors, 360 degree of horizontal laser light rangefinders and 360 degree of bottoms will be installed on unmanned plane
Laser sensor, high definition camera, supplementary lighting sources and air speed measuring apparatus；And it is connect with microcomputer and carries out debugging flight；
Step 2: unmanned plane after debugging is good is to flying in tunnel, high definition camera starts whole shooting record；
After unmanned plane enters in tunnel, included forward sight high brightness supplementary lighting sources and bottom high brightness supplementary lighting sources are opened, simultaneously
High definition camera starts the environment in whole shooting record tunnel；
Step 3: each 360 degree of laser sensors of unmanned plane detect whether that there are barriers, and judge in unmanned plane and all directions
Whether the distance between barrier is less than or equal to radius of safety, if so, unmanned plane reaches conflict situation, enters step four, no
Then, five are entered step；
Specifically: unmanned plane is regarded as a central point, self-position is determined using laser SLAM technology, in unmanned plane during flying mistake
Each of journey time step detects unmanned plane and head using three 360 degree of laser sensors based on the central point
Whether there are obstacles for diagonally above and frontward, unmanned plane horizontal direction whether there are obstacles and with unmanned plane and the oblique front lower place of head
Whether there are obstacles；If unmanned plane does not encounter any obstacle in flight course, unmanned plane according to when front direction and
Speed continues to fly；
When encountering barrier, 360 degree of top lasers sensors, 360 degree of horizontal laser light rangefinders and 360 degree of bottom lasers
Sensor returns to unmanned plane and head diagonally above and frontward barrier respectively, horizontal obstacle object and with the oblique front lower place barrier of head away from
From when unmanned plane is less than or equal to the radius of safety R of unmanned plane at a distance from any one barriercWhen, then determine this nobody
Machine enters conflict situation, does not otherwise enter conflict situation；
Step 4: microcomputer controls unmanned plane during flying, adjusts countermeasures and judge whether that conflict can be freed；If so, into
Enter step 5；Otherwise, unmanned plane shooting record obstacle information is returned by former route, and process terminates；
After unmanned plane enters conflict situation, speed is reduced to 0, carries out conflict Resolution in floating state, microcomputer recalculates
And the position of mobile unmanned plane, until unmanned plane flight forward will not clash, then unmanned plane continue flight forward and
Continue to shoot environment in tunnel；
Specifically:
Firstly, unmanned plane centered on itself, using the direction of tunnel both sides wall connection as x-axis, is connected with tunnel top bottom
Direction be y-axis, coordinate system is established as z-axis using the direction of unmanned plane during flying；
Then, it is L that unmanned plane, which measures the distance between barrier by 360 degree of laser sensors, and L is less than the peace of unmanned plane
Full radius Rc, the angle of line and x-y plane between barrier is θ, and barrier and unmanned plane are projected to x-y plane
Interior, the distance between unmanned plane and barrier are Lcos θ at this time, then unmanned plane frees conflict and needs to move in x-y plane
The shortest distance be Rc- Lcos θ, direction are the direction along barrier and unmanned plane line；
Unmanned plane carries out conflict Resolution according to shortest moving distance, and unmanned plane continues to fly in tunnel after conflict Resolution；
Similarly, when unmanned plane and multiple barriers clash cannot free simultaneously or barrier is too big or too many nothing
Method passes through, and unmanned plane shooting record obstacle information returns to entrance by former route；
Step 5: unmanned plane using built-in double velocity correlation information feature enhance algorithm, to high definition camera shooting image into
The analysis of row at-once monitor, and the position in output abnormality region carries out emphasis shooting；
Specific step is as follows:
Step 501, the shooting picture for inputting the channel 1024dpi × 1024dpi × 3 carry out characteristics of image by neural network and mention
It takes, generates the picture in the channel 8dpi × 8dpi × 256；
Step 502, each channel are one 8 × 8 matrix Asi(i=1,2 ..., 256), for each matrix AiIf Xit=
(Xit_1,Xit_2,…,Xit_8) it is 8 n-dimensional random variable ns that 8 transversal vectors of the matrix are constituted；
Step 503 utilizes 8 n-dimensional random variable n XitFurther calculate covariance matrix Cit；
Wherein, cit_jk=Cov (Xit_j,Xit_k) j, k=1,2 ..., 8；Xit_jAnd Xit_kThe respectively component of 8 n-dimensional random variable ns；
Step 504, for each matrix Ai, calculate 8 n-dimensional random variable n X of 8 column vectors of matrix compositioniv=(Xiv_1,
Xiv_2,…,Xiv_8)；
Step 505 utilizes 8 n-dimensional random variable n XivFurther calculate covariance matrix Civ；
Step 506, finally to each matrix Ai, utilize covariance matrix CitWith covariance matrix CivObtain the enhanced square of feature
Battle array
Step 507, by the enhanced matrix of featureBecome the feature vector that a length is 512 dimensions by average value pond,
Again by full articulamentum, existing abnormal area is judged, to carry out emphasis shooting to abnormal area；
Step 6: determining whether unmanned plane passes through tunnel, shooting is fully completed, if so, process terminates；Otherwise, step is skipped to
Two.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810996777.6A CN109084735B (en) | 2018-08-29 | 2018-08-29 | A kind of tunnel monitoring abnormal state method based on unmanned plane device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810996777.6A CN109084735B (en) | 2018-08-29 | 2018-08-29 | A kind of tunnel monitoring abnormal state method based on unmanned plane device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109084735A CN109084735A (en) | 2018-12-25 |
| CN109084735B true CN109084735B (en) | 2019-07-26 |
Family
ID=64795090
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810996777.6A Active CN109084735B (en) | 2018-08-29 | 2018-08-29 | A kind of tunnel monitoring abnormal state method based on unmanned plane device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109084735B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109991977B (en) * | 2019-03-04 | 2022-04-29 | 斯坦德机器人（深圳）有限公司 | Path planning method and device for robot |
| CN109883407B (en) * | 2019-03-20 | 2021-05-18 | 武汉建工集团股份有限公司 | Wall surface monitoring method and system based on infrared distance measurement |
| CN110086973B (en) * | 2019-05-10 | 2020-11-27 | 中国计量大学 | Cloud platform image stabilization system based on light stream camera |
| CN111598846B (en) * | 2020-04-26 | 2024-01-05 | 鲲鹏通讯(昆山)有限公司 | Method for detecting rail defects in tunnel based on YOLO |
| CN111999522B (en) * | 2020-07-17 | 2022-09-06 | 山西省交通建设工程质量检测中心（有限公司） | High-precision tunnel wind speed detection system and method thereof |
| CN112180943B (en) * | 2020-10-19 | 2022-07-01 | 山东交通学院 | Underwater robot navigation obstacle avoidance method based on visual image and laser radar |
| CN113220013A (en) * | 2021-04-07 | 2021-08-06 | 同济大学 | Multi-rotor unmanned aerial vehicle tunnel hovering method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105739519A (en) * | 2014-12-11 | 2016-07-06 | 中孚航空科技（天津）有限公司 | Unmanned plane laser obstacle avoidance system |
| CN107885227A (en) * | 2017-11-30 | 2018-04-06 | 广州市华科尔科技股份有限公司 | A kind of unmanned plane automatic obstacle-avoiding method |
| CN107990876A (en) * | 2017-11-20 | 2018-05-04 | 北京科技大学 | The quick scanning means in underground mine goaf and method based on unmanned vehicle |
| CN108229587A (en) * | 2018-02-06 | 2018-06-29 | 贵州电网有限责任公司 | A kind of autonomous scan method of transmission tower based on aircraft floating state |
| CN108447075A (en) * | 2018-02-08 | 2018-08-24 | 烟台欣飞智能系统有限公司 | A kind of unmanned plane monitoring system and its monitoring method |
-
2018
- 2018-08-29 CN CN201810996777.6A patent/CN109084735B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105739519A (en) * | 2014-12-11 | 2016-07-06 | 中孚航空科技（天津）有限公司 | Unmanned plane laser obstacle avoidance system |
| CN107990876A (en) * | 2017-11-20 | 2018-05-04 | 北京科技大学 | The quick scanning means in underground mine goaf and method based on unmanned vehicle |
| CN107885227A (en) * | 2017-11-30 | 2018-04-06 | 广州市华科尔科技股份有限公司 | A kind of unmanned plane automatic obstacle-avoiding method |
| CN108229587A (en) * | 2018-02-06 | 2018-06-29 | 贵州电网有限责任公司 | A kind of autonomous scan method of transmission tower based on aircraft floating state |
| CN108447075A (en) * | 2018-02-08 | 2018-08-24 | 烟台欣飞智能系统有限公司 | A kind of unmanned plane monitoring system and its monitoring method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109084735A (en) | 2018-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109084735B (en) | A kind of tunnel monitoring abnormal state method based on unmanned plane device | |
| CN108496129B (en) | Aircraft-based facility detection method and control equipment | |
| CN206691107U (en) | Pilotless automobile system and automobile | |
| US8554395B2 (en) | Method and system for facilitating autonomous landing of aerial vehicles on a surface | |
| CN108447075B (en) | Unmanned aerial vehicle monitoring system and monitoring method thereof | |
| CN106965946B (en) | A kind of method and apparatus improving landing safety based on detection obstacle | |
| CN111123983B (en) | Interception net capture control system and control method for unmanned aerial vehicle | |
| CN106527426A (en) | Indoor multi-target track planning system and method | |
| CN108258613B (en) | Intelligent line inspection photoelectric pod and line inspection realization method | |
| CN107783103B (en) | A kind of flying bird tracking intelligent method of the lightweight with self-learning function | |
| CN108897342B (en) | Positioning and tracking method and system for fast-moving civil multi-rotor unmanned aerial vehicle | |
| JP2006027448A (en) | Aerial photographing method and device using unmanned flying body | |
| CN112068539A (en) | Unmanned aerial vehicle automatic driving inspection method for blades of wind turbine generator | |
| CN109573088A (en) | A kind of Shipborne UAV photoelectricity guidance carrier landing system and warship method | |
| CN108829136A (en) | The a wide range of synergic monitoring method and apparatus of unmanned aerial vehicle group | |
| KR20120036684A (en) | An intelligent aviation robot using gps | |
| CN115291219A (en) | Method and device for realizing dynamic obstacle avoidance of unmanned aerial vehicle by using monocular camera and unmanned aerial vehicle | |
| CN108170139A (en) | A kind of photoelectricity multitask system for unmanned boat and perform method | |
| CN107449786B (en) | Device for finely observing surface of object | |
| CN111692916B (en) | Aiming device and aiming method | |
| González et al. | Vision-based UAV detection for air-to-air neutralization | |
| CN116753778A (en) | Unmanned aerial vehicle countering system and method based on information fusion and task distribution | |
| JPH0737908B2 (en) | Mobile noise monitoring system | |
| CN114353667A (en) | Ground target measurement method based on AR and unmanned aerial vehicle monocular vision and application thereof | |
| KR20110136225A (en) | An intelligent aviation robot using gps |
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 |