CN108491758A - A kind of track detection method and robot - Google Patents
A kind of track detection method and robot Download PDFInfo
- Publication number
- CN108491758A CN108491758A CN201810126117.2A CN201810126117A CN108491758A CN 108491758 A CN108491758 A CN 108491758A CN 201810126117 A CN201810126117 A CN 201810126117A CN 108491758 A CN108491758 A CN 108491758A
- Authority
- CN
- China
- Prior art keywords
- robot
- image
- mentioned
- track
- video stream
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 53
- 238000004458 analytical method Methods 0.000 claims abstract description 36
- 238000000034 method Methods 0.000 claims description 29
- 230000008569 process Effects 0.000 claims description 20
- 238000006243 chemical reaction Methods 0.000 claims description 18
- 230000004888 barrier function Effects 0.000 claims description 17
- 241001669679 Eleotris Species 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 230000005540 biological transmission Effects 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 3
- 238000007689 inspection Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 230000004927 fusion Effects 0.000 description 8
- 238000013523 data management Methods 0.000 description 7
- 230000001681 protective effect Effects 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 241001465754 Metazoa Species 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 239000004615 ingredient Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000008595 infiltration Effects 0.000 description 1
- 238000001764 infiltration Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000010079 rubber tapping Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000011426 transformation method Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a kind of track detection method and robots, wherein the track detection method includes：The live video stream of robot acquisition trajectory；Based on the live video stream, the position of the robot is determined by vision positioning；According to the traveling scheme of robot described in the location determination of the robot；The robot is controlled to be advanced according to the traveling scheme；In each frame image of the live video stream, detect whether that there are target images, wherein the target image is the image for the fault zone for showing the track；If in each frame image of the live video stream, there are the target images, the target image is then uploaded to server, so that the server carries out secondary analysis to the target image, and the target image is matched to by corresponding client based on the result of secondary analysis and is shown.The present invention program can realize the automatic detection to the infrastructure and infrastructure device of rail traffic, ensure Rail Transit System safe operation.
Description
Technical field
The invention belongs to robotic technology field more particularly to a kind of track detection methods and robot.
Background technology
The development of rail traffic is being concentrated mainly on a line big city and provincial capital nearly ten years, and has to two
The universal trend in three line cities.And safety is the first element of rail traffic.In order to ensure the safety of rail traffic, rail traffic
Infrastructure and infrastructure device follow inspection in 48 hours, monthly test, half annual test, annual test, 5 years standards examined greatly, and this its
In, some detections almost rely on artificial detection, such as inspection in 48 hours etc..Traditional artificial detection not only needs to spend big
Manpower and materials are measured, but also the problems such as there is false retrieval, missing inspection and the security risks of testing staff.
Invention content
In view of this, the present invention provides a kind of track detection method and robot, it can be achieved that basis to rail traffic
The automatic detection of facility and infrastructure device ensures Rail Transit System safe operation.
The first aspect of the present invention provides a kind of track detection method, and the track detection method includes：
The live video stream of robot acquisition trajectory；
Based on the live video stream, the position of the robot is determined by vision positioning；
According to the traveling scheme of robot described in the location determination of the robot；
The robot is controlled to be advanced according to the traveling scheme；
In each frame image of the live video stream, detect whether that there are target images, wherein the target image is
Show the image of the fault zone of the track；
If the target image is uploaded to there are the target image in each frame image of the live video stream
Server so that the server carries out secondary analysis to the target image, and based on the result of secondary analysis by the mesh
Logo image is matched to corresponding client and shows.
The second aspect of the present invention provides a kind of robot, and the robot application is in field of track traffic, the machine
Device people includes：
Video acquisition module is used for the live video stream of machine acquisition trajectory；
Vision positioning module determines the position of the robot by vision positioning for being based on the live video stream；
Scheme determining module, the traveling scheme for robot described in the location determination according to the robot；
Traveling control module is advanced for controlling the robot according to the traveling scheme；
Fault detection module, in each frame image of the live video stream, detecting whether there are target image,
In, the target image is the image for the fault zone for showing the track；
Information uploading module, for when in each frame image in the live video stream there are when the target image, will
The target image is uploaded to server so that the server carries out secondary analysis to the target image, and based on secondary
The target image is matched to corresponding client and shown by the result of analysis.
Therefore in the present invention program, first by the live video stream of robot acquisition trajectory, it is then based on described
Live video stream determines the position of the robot by vision positioning, described in the location determination then according to the robot
The traveling scheme of robot controls the robot and is advanced according to the traveling scheme, and in each frame of the live video stream
In image, detect whether that there are target images, wherein the target image is the figure for the fault zone for showing the track
The target image is uploaded to server by picture there are when the target image in each frame image of the live video stream,
So that the server carries out secondary analysis to the target image, and based on the result of secondary analysis by the target image
Corresponding client is assigned to show.The Daily Round Check operation of track is given robot to execute by the present invention program, is existed by robot
Quickly identification reduces false retrieval, the possibility that missing inspection occurs, realizes the base to rail traffic there are the region of failure during inspection
The automatic detection of Infrastructure and infrastructure device ensures Rail Transit System safe operation.
Description of the drawings
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description be only the present invention some
Embodiment for those of ordinary skill in the art without having to pay creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is the implementation process schematic diagram of track detection method provided in an embodiment of the present invention；
Fig. 2 (a) is the structure diagram of rail detection system provided in an embodiment of the present invention；
Fig. 2 (b) is the specific implementation flow schematic diagram of step 105 in embodiment illustrated in fig. 1；
Fig. 3 is the specific implementation flow schematic diagram of step 102 in embodiment illustrated in fig. 1；
Fig. 4 (a) is the schematic diagram of coordinate system in step 301 in embodiment illustrated in fig. 3；
Fig. 4 (b) is the schematic diagram of the center line of the center line of image and image middle orbit in step 303 in embodiment illustrated in fig. 3；
Fig. 5 is the specific implementation flow schematic diagram of step 103 in embodiment illustrated in fig. 1；
Fig. 6 is the structure diagram of robot provided in an embodiment of the present invention.
Specific implementation mode
In being described below, for illustration and not for limitation, it is proposed that such as tool of particular system structure, technology etc
Body details, to understand thoroughly the embodiment of the present invention.However, it will be clear to one skilled in the art that there is no these specific
The present invention can also be realized in the other embodiments of details.In other situations, it omits to well-known system, device, electricity
The detailed description of road and method, in case unnecessary details interferes description of the invention.
In order to illustrate the technical solution of the present invention, illustrated below by specific embodiment.
Embodiment one
Fig. 1 shows the implementation process for the track detection method that the embodiment of the present invention one provides, and details are as follows：
In a step 101, the live video stream of robot acquisition trajectory；
In embodiments of the present invention, service personnel is when starting the robot applied to field of track traffic, first by machine
People is placed in orbital position, for example, Gui Hang robots are installed on the rail of track, alternatively, flying robot is attached in track
Closely let fly away.Above-mentioned robot upon actuation, first sets the frame per second of image acquisition device (such as camera), so according to sampling thheorem
Start the live video stream of acquisition trajectory afterwards.Specifically, above-mentioned robot is equipped with multiple cameras, is mounted on based on camera
Different location in robot can collect the orbital image of different angle；Type based on camera can collect rail
The live video stream of the different video type in road, above-mentioned video type include but not limited to that infrared video, greyscale video, multichannel are super
HD video is not construed as limiting herein.Specifically, the multichannel ultra high-definition image in above-mentioned multichannel ultra high-definition video is for assessing bridge
The fault conditions such as crack, contact net foreign matter；Infrared image in above-mentioned infrared video is for assessing tunnel, bridge infiltration water and connecing
The fault conditions such as net-fault failure；That is, different types of video is for assessing different types of failure.Optionally, it is adopted in robot
After the various types of live video streams for collecting track, whether each frame image that can be detected in above-mentioned live video stream reaches pre-
If picture quality condition, if in above-mentioned live video stream exist not up to above-mentioned picture quality condition multichannel ultra high-definition image
And/or infrared image, then by the multichannel ultra high-definition image and/or infrared image of above-mentioned not up to above-mentioned picture quality condition from upper
It states in live video stream and rejects, to avoid the bad image contributions of quality to robot and server to the live video stream of track
Analyzing processing, reduce and the possibilities of error analysis results occur.
In a step 102, it is based on above-mentioned live video stream, the position of above-mentioned robot is determined by vision positioning；
In embodiments of the present invention, it is based on live video stream, realtime graphic can be got, specifically, it is believed that when
Before the image of a newest frame that gets be realtime graphic.Robot upon actuation, can be based on above-mentioned realtime graphic, pass through
Vision positioning determines the position of itself, realizes the positioning to itself.Optionally, GPS can be carried in above-mentioned robot interior
Module or other locating modules are realized the Primary Location to robot by above-mentioned GPS module or other locating modules, then are passed through
The realtime graphic of the collected track of vision positioning and robot realizes the precise positioning to robot, i.e., is drawn by Primary Location
Determine zonule residing for robot possibility, precise positioning is carried out in the zonule, lowers fortune when robot carries out precise positioning
Calculate pressure.Since track is after the completion of laying, position will not change, i.e. the position of track is fixed position, thus,
The environmental map that above-mentioned track can be obtained in advance is trained and is learnt to robot by the environmental map of above-mentioned track
Afterwards, restart above-mentioned robot；After above-mentioned robot starts and collects the realtime graphic of track, by the spy to orbital image
The image processing operations such as sign point extraction can determine the position of above-mentioned robot.Certainly, according to the difference of robot type, also may be used
To determine the position of robot otherwise, it is not construed as limiting herein.
In step 103, according to the traveling scheme of the above-mentioned robot of location determination of above-mentioned robot；
In embodiments of the present invention, it can be further determined that according to the current location of identified robot in step 102
The traveling scheme of above-mentioned robot specifically can pass through the current location planning robot's of the route and robot of track
Travelling route, when the current location of robot is Chong Die with the route of track, control robot follows the route of track to advance；When
When the position of the current location offset track of robot, control robot first marches to the position Chong Die with the route of track, then
The route of track is followed to advance.It is of course also possible to determine the traveling scheme of above-mentioned robot otherwise, do not limit herein
It is fixed.
At step 104, above-mentioned robot is controlled to be advanced according to above-mentioned traveling scheme；
In embodiments of the present invention, robot can advance according to traveling scheme for duration, and pass through step during advancing
Rapid 101 to 103 constantly update the traveling scheme of itself.
In step 105, in each frame image of above-mentioned live video stream, detect whether that there are target images；
In embodiments of the present invention, image procossing is carried out to live video stream by the embedded hardware that robot carries,
It whether there is target image in each frame image to detect above-mentioned live video stream, wherein above-mentioned target image is to show
The image of the fault zone of above-mentioned track.In fact, due to when being shot to track, since track is inherently by two
What parallel rail and the equidistant laying of several sleepers were constituted, thus the picture of the normal orbit of shooting gained is under normal conditions
It is very close, that is, if the live video stream of shooting acquisition normal orbit, adjacent two field pictures in the live video stream
Similarity should be higher.When there is the case where picture mutation, it is likely that mean that failure has occurred in track, can be examined based on this
It surveys and whether there is target image in each frame image.For example, if foreign matter occur in A points in track, other places are normal, then exist
During shooting live video stream, when shooting is to A points, include in live video stream A points image will with it is other not
Including the image of A points has significant difference, in other words, in live video stream, has the orbital image of foreign matter and does not have foreign matter
Orbital image necessarily have larger difference, therefore, it is possible to judge whether track is deposited according to the similarity of adjacent two field pictures
In foreign matter；Certainly, other obstacles, such as larger crack etc., can also according to the similarity of adjacent two field pictures judge obtain,
It is not construed as limiting herein.
In step 106, if there are above-mentioned target images in each frame image of above-mentioned live video stream, by above-mentioned mesh
Logo image is uploaded to server.
In embodiments of the present invention, robot, server, PC (Person Computer, PC) and/or movement
Terminal may make up a rail detection system, and above-mentioned server can also include data management platform and database, wherein above-mentioned number
It is used to store the various information accessed by robot according to library, above-mentioned data management platform provides an interactive interface, for showing
Show that the various information accessed by robot, i.e. service personnel can pass through data management platform realization and robot at server
Interaction, Fig. 2 (a) shows the structure diagram of the rail detection system.In the rail detection system, once it detects the presence of
Above-mentioned target image is then uploaded to server by target image by network so that above-mentioned server to above-mentioned target image into
Row secondary analysis, and the result notice service personnel based on secondary analysis.Specifically, PC and mobile terminal can be equipped with based on upper
State rail detection system carrying client, above-mentioned client login after, can the result based on above-mentioned secondary analysis will be upper
It states target image and is matched to corresponding client and show, for example, being shown in the client of PC and/or being shown in the client of mobile terminal
On end, it is not construed as limiting herein.By above-mentioned client, above-mentioned data management platform can also be accessed and database (accesses clothes
Business device).Above-mentioned network can be Wireless Fidelity (WiFi, Wireless Fidelity) network, can also be mobile data network
Network, such as general packet radio service technology (GPRS, General Packet Radio Service)) network, the third generation move
Dynamic communication network (i.e. 3G network), forth generation mobile communications network (i.e. 4G networks) etc., are not construed as limiting herein.Since robot exists
May also occur mistake during target image in detection live video stream, such as acquisition is made due to strong illumination
Occur fuzzy in image or do not know region, influence the judgement of robot, lead to machine human factor error will be not present faulty section
The image in domain is uploaded to server as target image, in this case, in order to improve the accuracy of fault detect, above-mentioned
In server secondary analysis and database purchase will be carried out to above-mentioned target image.If the result of secondary analysis determines above-mentioned target
Really there is fault zone in image, that is, determine that there are failures in the track shown by above-mentioned target image, then it can will be above-mentioned
The relevant information of target image is shown on the data management platform of server.Further, if the result of secondary analysis determines
Really there is fault zone in above-mentioned target image, the result that server is also based on secondary analysis judges whether to need manually
Go to and repair the fault zone, if so, the relevant information of above-mentioned target image is shown in the mobile terminal bound in advance and/
Or in the client of PC.It specifically, can be in the client of mobile terminal and/or PC if above-mentioned track is underground railway track
Show subway line, fault type indicated by above-mentioned target image by data management platform, fault degree, acquisition time and
Collecting location, naturally it is also possible to which other relevant informations of above-mentioned target image are shown in above-mentioned mobile terminal and/or above-mentioned number
According to management platform, it is not construed as limiting herein.The practical specific location for indicating above-mentioned track and breaking down of above-mentioned collecting location.Due to
There may be several robots to be run on the section route of each track, that is, may whithin a period of time, data management platform
Need to show the information of more than two target images, therefore, it is possible to based on acquisition time by sequence of the elder generation after to multiple targets
The relevant information of image is ranked up, and can also be believed the correlation of multiple target images by weight to light sequence based on fault degree
Breath is ranked up, be also based on collecting location apart from the machine (i.e. the mobile terminal of the relevant information of display target image and/or
PC position) shows the relevant information of multiple target images after sequence by being closely ranked up to remote sequence.
Optionally, in order to improve the accuracy of detection target image, Fig. 2 (b) shows the specific implementation stream of step 105
Journey, details are as follows：
In step 201, binary conversion treatment is carried out to each frame image of above-mentioned live video stream；
In embodiments of the present invention, the binary-state threshold used when first setting binary conversion treatment specifically may be used double
Peak method, maximum variance between clusters, maximum entropy threshold method or iterative method set above-mentioned binary-state threshold, and based on above-mentioned setting
Binary-state threshold carries out binary conversion treatment to each frame image of above-mentioned live video stream.
In step 202, the area-of-interest of above-mentioned each frame image after binary conversion treatment is extracted；
In embodiments of the present invention, above-mentioned area-of-interest (region of interest, ROI) is selected from image
The image-region selected, the area-of-interest are the emphasis for carrying out image analysis.Specifically, can by preset operator and
Function extracts the area-of-interest of above-mentioned each frame image after binary conversion treatment.
In step 203, denoising operation is carried out to the area-of-interest of above-mentioned each frame image based on Morphology Algorithm；
In embodiments of the present invention, carrying out denoising operation to the area-of-interest of above-mentioned each frame image will specifically open
Combine with closure operation and may make up morphology scratch filter, during this, needs to choose suitable structural element, ability
Realize preferable denoising effect.
In step 204, for each frame image after denoising, the area-of-interest and consecutive frame image of image are calculated
The characteristics of image difference of area-of-interest；
In embodiments of the present invention, above-mentioned consecutive frame image is figure adjacent with above-mentioned image in above-mentioned live video stream
Picture, for example, for the first frame image in live video stream, the image adjacent with the image is the rear frame image of the image；
For last frame image in live video stream, the image adjacent with the image is the prior image frame of the image.To regarding in real time
For intermediate frame image in frequency stream, the image adjacent with the image is the prior image frame of the image or rear frame image, for example, can
First to calculate the characteristics of image difference of the area-of-interest of intermediate frame image and the area-of-interest of prior image frame, if prior image frame
Picture quality it is bad, then can calculate the image of the area-of-interest of intermediate frame image and the area-of-interest of rear frame image again
Feature difference.Above-mentioned characteristics of image includes but not limited to：Color characteristic, textural characteristics, shape feature and spatial relation characteristics.
In step 205, if the image of the area-of-interest of above-mentioned image and the area-of-interest of above-mentioned consecutive frame image
Feature difference has exceeded preset characteristics of image disparity range, then above-mentioned image and/or above-mentioned consecutive frame image is determined as mesh
Logo image.
In embodiments of the present invention, ambient lighting feelings when above-mentioned preset characteristics of image disparity range can be according to shooting
Condition is adjusted, and is not construed as limiting herein.
Optionally, above-mentioned track detection method further includes：
In the traveling process of above-mentioned robot, based on the range sensor that above-mentioned robot carries, above-mentioned machine is obtained
People in above-mentioned robot local environment at a distance from barrier；
If above-mentioned distance is not more than preset distance threshold, above-mentioned traveling scheme is adjusted, so that robot
Far from above-mentioned barrier, advance alternatively, suspending above-mentioned robot.
Wherein, according to the type of above-mentioned robot, range sensor can be carried in the different location of robot.In one kind
Under application scenarios, the above-mentioned artificial flying robot of machine, then can robot all around orientation is respectively mounted distance up and down
Sensor knocks the barrier being likely to occur in the air on the way to avoid flight；Under another application scenarios, above-mentioned machine is artificial
Gui Hang robots then can knock rail in the front and back position mounting distance sensor of robot to avoid when moving forward or back
The barrier that road near zone is likely to occur.Further, when robot carry any range sensor measured by distance
When no more than preset distance threshold, above-mentioned traveling scheme can be adjusted, be advanced alternatively, suspending above-mentioned robot.
Under a kind of application scenarios, the above-mentioned artificial flying robot of machine, then in all around any orientation up and down
The distance that measures of range sensor when being not more than preset distance threshold, by the master of the above-mentioned Distance Transmission measured to robot
Dynamic security control plate, above-mentioned active safety control plate are advanced robot is controlled to the opposite direction of barrier (it is believed that above-mentioned
The direction of sensor is the direction of the barrier), until the distance that measures of above-mentioned range sensor beyond it is above-mentioned it is preset away from
After threshold value certain proportion (such as 20%), traveling scheme is adjusted and is continued on based on Obstacle Position.It needs to note
Meaning, during above-mentioned control robot advances to the opposite direction of barrier, it is also necessary to ensure other sensors institute
The value measured is also in normal range (NR).Further, above-mentioned active safety control plate can also connect accelerometer, barometer,
Light stream sensor and/or more than one ultrasonic sensor can not only measure the barrier in environment, can also obtain itself
Flight progress.As it can be seen that can be combined with flight control by above-mentioned active safety control plate, intelligent barrier avoiding can be realized；
And when above-mentioned active safety control plate detects data perturbation measured by a plurality of types of sensors, then need above-mentioned active to pacify
Full control panel control robot carries out emergency braking, and specially robot, which executes, declines instruction, declines instruction in above-mentioned execution
In the process, when the data measured by the range sensor being installed below robot are less than one meter, the flight electricity of closing machine people
Machine is to land；If during executing decline instruction, number measured by the range sensor being installed below robot is detected
According to still disorder, then the flight motor of closing machine people immediately, makes flying robot's free falling, by being set outside flying robot
The passive security protective cover set protects it.It is believed that above-mentioned passive security protective cover is mainly used for passive protection, keep away
When exempting from that flying robot is out of control because of unexpected factors in special circumstances, pedestrian or vehicular traffic are damaged.Specifically, above-mentioned
Passive security protective cover is designed using SOLIDWORKS design tools, and carries out finite element fraction with ANSYS analysis tools
Analysis under the premise of ensureing structural strength, reduces protective cover matter to greatest extent to realize the optimization to above-mentioned passive security protective cover
It measures (i.e. weight).
Above-mentioned robot can also provide remote monitoring function, and each biography of above-mentioned robot is specially sent to server
The collected data of sensor institute, at the same to server send above-mentioned robot collected track realtime graphic, above-mentioned
The collected data of each sensor institute and the data for carrying out fusion arrangement, and being obtained after fusion is arranged are sent on server
It is shown to the mobile terminal bound in advance.After above-mentioned fusion arrangement obtained data get muddled etc. abnormal phenomenon when,
It can be reminded on mobile terminals by modes such as vibrations, word and/or the tinkle of bells, it is abnormal existing to prompt service personnel to pay close attention to
As；And when above-mentioned robot carries out emergency braking, the side such as vibrations, word and/or the tinkle of bells can also be passed through on mobile terminals
Formula is reminded, to prompt service personnel that the place of robot emergency braking is gone to the behaviour such as robot to be recycled and overhauled
Make.Certainly, service personnel can also actively consult the data obtained after the above-mentioned fusion received by mobile terminal arranges, to sentence
Whether disconnected robot is during smooth flight, can be with when service personnel thinks the traveling of current robot there are when problem
Pause instruction is sent to robot by client or server, above-mentioned track machine people receives client or server is sent
Pause instruction after, above-mentioned pause instruction can be based on and suspend the flight of above-mentioned robot.The process of above-mentioned pause robot flight
It can refer to the process of above-mentioned emergency braking.
Under another application scenarios, the above-mentioned artificial Gui Hang robots of machine are then surveyed in the range sensor in front
It, will be upper when the distance obtained is not more than preset distance threshold (above-mentioned front is for the direction of travel of Gui Hang robots)
The Distance Transmission measured is stated to the active safety control plate of robot, above-mentioned active safety control plate control robot, which suspends, goes
Into；Also, since barrier may be animal, thus machine human hair can also be controlled and go out alarm audio, to disperse robot
Neighbouring animal.Further, above-mentioned robot can combine collected with installation manipulator and object frame of receiving, above-mentioned robot
The realtime graphic of track, can further determine whether the barrier falls into the range that manipulator can capture, if so, passing through
The manipulator captures barrier to receiving in object frame, with removing obstacles object.Optionally, the distance that range sensor measures is more than default
Distance threshold after, you can determine that the barrier has been not present, robot can be controlled and continued on.Further, above-mentioned master
Dynamic security control plate can also connect accelerometer, barometer, light stream sensor and/or more than one ultrasonic sensor, no
Barrier can be only measured, the detailed traveling situation of itself can also be obtained.As it can be seen that can be with by above-mentioned active safety control plate
Realize intelligent barrier avoiding.
Above-mentioned robot can also provide remote monitoring function, and each biography of above-mentioned robot is specially sent to server
The collected data of sensor institute, at the same to server send above-mentioned robot collected track realtime graphic, above-mentioned
The collected data of each sensor institute and the data for carrying out fusion arrangement, and being obtained after fusion is arranged are sent on server
It is shown to the mobile terminal bound in advance.After above-mentioned fusion arrangement obtained data get muddled etc. abnormal phenomenon when,
It can be reminded on mobile terminals by modes such as vibrations, word and/or the tinkle of bells, it is abnormal existing to prompt service personnel to pay close attention to
As；The data obtained after being arranged due to above-mentioned fusion include the realtime graphic of track, thus service personnel can also independently consult
Current realtime graphic, to judge robot whether in stable traveling process, when service personnel thinks current robot
Traveling can send pause instruction, above-mentioned track machine people receives by client or server there are when problem to robot
After the pause instruction sent to client or server, above-mentioned pause instruction can be based on and suspend above-mentioned robot traveling, so that
It obtains service personnel and rushes to robot position recycling machine people at the first time.
Optionally, Fig. 3 is shown as the artificial flying robot of machine, the specific implementation flow of step 102, is described in detail such as
Under：
In step 301, it is based on above-mentioned live video stream, obtains realtime graphic；
In embodiments of the present invention, it is believed that the image of the newest frame currently got is realtime graphic.
In step 302, coordinate system is initialized；
In embodiments of the present invention, the x-axis of above-mentioned coordinate system is the extending direction of track, and the y-axis of above-mentioned coordinate system is rail
Pillow is laid with direction, and the z-axis of above-mentioned coordinate system is perpendicular to the direction in track laying face.Fig. 4 (a) shows a kind of showing for coordinate system
It is intended to, can be seen that by Fig. 4 (a), x-axis coordinate represents the distance that Robot track has been advanced, and y-axis coordinate represents machine
The distance of people's offset track center line, z-axis coordinate represent the flying height of the relative orbit paved surface of robot.
In step 303, above-mentioned robot receives the RFID tag for carrying location information that radio-frequency unit is sent,
X-axis positioning is carried out according to above-mentioned RFID tag；
In embodiments of the present invention, above-mentioned radio-frequency unit is pre-set on the route of above-mentioned track, for example, can be every
A radio-frequency unit is just arranged in preset milimeter number (such as three kilometers), and robot, can at the set-point for passing by the radio-frequency unit
Distinguishing label with the radio frequency sent according to the radio-frequency unit positions the x-axis coordinate of itself, that is, knows itself along track
The distance of traveling.Alternatively, can also be according to being installed on the collected realtime graphic of camera institute in orientation under robot to track
In sleeper be identified, the quantity of sleeper passed through in robot traveling process is calculated, due to the laying distance of sleeper
It is fixed, thus the laying distance of sleeper can be multiplied by according to the quantity for the sleeper for calculating gained, knows itself along track
The distance of traveling.It is alternatively possible in conjunction with above two calculation, first the distance of traveling is counted according to the quantity of sleeper
It calculates, is then corrected further according to the distance for carrying traveling of the RFID tag of location information to being calculated so that x
The result of axis positioning is more accurate.
In step 304, above-mentioned robot is according in the center line of above-mentioned realtime graphic and above-mentioned realtime graphic middle orbit
The relative position of line carries out y-axis positioning；
In embodiments of the present invention, the realtime graphic in this step is acquired by the camera for being installed on orientation under robot
The realtime graphic arrived, theoretically considers, when robot towards track extending direction movement when, if positioned at track just on
Side, then the center line of track overlaps the center line with image, just shown as shown in Figure 4.Thus, it can calculate in above-mentioned realtime graphic,
The distance of the center line of realtime graphic to the center line of realtime graphic middle orbit is multiplied by certain engineer's scale, you can obtain according to the distance
Know the distance of robot offset track center line.
In step 305, above-mentioned robot acquires the flying height of above-mentioned robot, root by the range sensor of carrying
Z-axis positioning is carried out according to above-mentioned flying height；
In embodiments of the present invention, robot is acquired to track paving by being equipped on the range sensor in orientation under robot
If the distance in face, which is flying height, and z-axis positioning is carried out according to above-mentioned flying height；
Within step 306, according to the positioning of above-mentioned x-axis as a result, the result of above-mentioned y-axis positioning and the knot of above-mentioned z-axis positioning
Fruit determines above-mentioned robot coordinate residing in above-mentioned coordinate system.
In embodiments of the present invention, according to the positioning of above-mentioned x-axis as a result, the result and above-mentioned z-axis of above-mentioned y-axis positioning position
As a result, understanding that the mileage number of the track residing for robot, the distance of offset track center line and flying height namely robot exist
Residing coordinate in above-mentioned coordinate system.
Optionally, Fig. 5 is shown as the artificial flying robot of machine, the specific implementation flow of step 103, is described in detail such as
Under：
In step 501, it is based on above-mentioned live video stream, obtains realtime graphic；
In embodiments of the present invention, it is believed that the image of the newest frame currently got is realtime graphic.
In step 502, binary conversion treatment is carried out to above-mentioned realtime graphic；
In embodiments of the present invention, the binary-state threshold used when first setting binary conversion treatment specifically may be used double
Peak method, maximum variance between clusters, maximum entropy threshold method or iterative method set above-mentioned binary-state threshold, and based on above-mentioned setting
Binary-state threshold carries out binary conversion treatment to each frame image of above-mentioned live video stream.
In step 503, the area-of-interest of the above-mentioned realtime graphic after binary conversion treatment is extracted；
In embodiments of the present invention, above-mentioned each frame after binary conversion treatment can be extracted by preset operator and function
The area-of-interest of image.
In step 504, longitudinal rectangular area in above-mentioned area-of-interest is extracted based on Morphology Algorithm；
In embodiments of the present invention, which includes the image of two rail of track；
In step 505, by straight-line detection, two rail of above-mentioned track are determined in above-mentioned longitudinal rectangular area
Position；
In embodiments of the present invention, Hough transformation method may be used and carry out straight-line detection, LSD (Line can also be used
Segment Detector) straight-line detection is carried out, it is not construed as limiting herein.
In step 506, the position of above-mentioned two rail is fitted, obtains the center line of two rail；
In step 507, the position of center line and above-mentioned robot based on above-mentioned two rail determines above-mentioned robot
Traveling scheme.
It in embodiments of the present invention, can be based on the position of above-mentioned robot, in advance after obtaining the center line of two rail
The position of certain point on the center line of two rail is entered, then using the center line of two rail as the traveling scheme of robot；Alternatively,
Can also be directly from one route parallel with the above-mentioned center line of two rail of the location determination of robot as traveling scheme, this
Place is not construed as limiting.
Optionally, when the artificial Gui Hang robots of machine, since Gui Hang robots are advanced automatically on rail, y-axis
And z-axis is fixed, thus directly the quantity of sleeper can be identified by collected realtime graphic during traveling
And count, the quantity of the sleeper passed through in robot traveling process is obtained, multiplied by with the laying distance of sleeper, knows itself edge
The distance that track has been advanced.It is alternatively possible to first be calculated the distance of traveling according to the quantity of sleeper, then further according to taking
The distance of traveling of the RFID tag with location information to being calculated is corrected so that the result of x-axis positioning is more
It is accurate to add.
Optionally, when the artificial Gui Hang robots of machine, above-mentioned track detection method further includes：
In the traveling process of above-mentioned robot, the bolt of above-mentioned track is tapped by above-mentioned manipulator；
Acquire audio when above-mentioned robot taps the bolt of above-mentioned track；
Audio analysis is carried out to above-mentioned audio, whether the above-mentioned bolt of result judgement based on audio analysis loosens；
It has been loosened if it is determined that result is above-mentioned bolt, has then sent above-mentioned judgement knot to mobile terminal and/or above-mentioned server
Fruit, to notify service personnel to overhaul above-mentioned bolt.
Wherein, above-mentioned Gui Hang robots are also equipped with manipulator, and above-mentioned robot is during traveling, when by real-time
When the analysis of image finds the bolt installed in track, it can suspend and advance and bolt is positioned, based on determining bolt
Position output control manipulator taps above-mentioned bolt, and opens the sound-recording function of robot, when acquisition robot taps above-mentioned bolt
Audio.When bolt looseness, tapping the tone color of bolt can change, i.e., the ingredient of the middle higher hamonic wave of collected audio
With the amplitude proportional of each ingredient also can different from, by comparing collected audio and preset audio volume control, judgement is above-mentioned
Whether bolt loosens, if collected audio and preset audio volume control similarity are higher, it is determined that judges result for bolt not
It loosens, robot can continue on；It is preset less than one if collected audio and preset audio volume control similarity are relatively low
Similarity threshold, it is determined that judgement result is that above-mentioned bolt has loosened, and can be sent to mobile terminal and/or above-mentioned server
Judgement is stated as a result, to notify service personnel to overhaul above-mentioned bolt.
Therefore through the embodiment of the present invention so that robot realizes base according to the realtime graphic of collected track
It advances in the intelligence of track and intelligent measurement operates, the doubtful orbital region to break down can be quickly found out, and will be in the failure
It reports to server so that server can carry out secondary analysis, and the result based on secondary analysis notifies service personnel in time.
During this, it is no longer necessary to manually carry out inspection to track, but give robot to execute the operation of the Daily Round Check of track, lead to
Crossing robot, quickly identification reduces false retrieval, the possibility that missing inspection occurs, realizes to rail there are the region of failure during inspection
The automatic detection of the infrastructure and infrastructure device of road traffic ensures Rail Transit System safe operation.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
Embodiment two
Fig. 6 shows that the concrete structure block diagram of robot provided by Embodiment 2 of the present invention is only shown for convenience of description
With relevant part of the embodiment of the present invention.The robot 6 includes：Image capture module 61, vision positioning module 62, scheme is true
Cover half block 63, traveling control module 64, fault detection module 65, information uploading module 66.
Wherein, video acquisition module 61 are used for the live video stream of machine acquisition trajectory；
Vision positioning module 62 determines the position of above-mentioned robot by vision positioning for being based on above-mentioned live video stream
It sets；
Scheme determining module 63 is used for the traveling scheme of the above-mentioned robot of location determination according to above-mentioned robot；
Traveling control module 64 is advanced for controlling above-mentioned robot according to above-mentioned traveling scheme；
Fault detection module 65, in each frame image of above-mentioned live video stream, detecting whether there are target image,
Wherein, above-mentioned target image is the image for the fault zone for showing above-mentioned track；
Information uploading module 66, for when in each frame image in above-mentioned live video stream there are when above-mentioned target image,
Above-mentioned target image is uploaded to server so that above-mentioned server carries out secondary analysis to above-mentioned target image, and is based on two
Above-mentioned target image is matched to corresponding client and shown by the result of secondary analysis.
Optionally, above-mentioned fault detection module 65 includes：
First binarization unit carries out binary conversion treatment for each frame image to above-mentioned live video stream；
First extraction unit, the area-of-interest for extracting above-mentioned each frame image after binary conversion treatment；
Denoising unit, for carrying out denoising operation to the area-of-interest of above-mentioned each frame image based on Morphology Algorithm；
Difference calculation units, for for each frame image after denoising, calculating the area-of-interest and consecutive frame of image
The characteristics of image difference of the area-of-interest of image, wherein above-mentioned consecutive frame image be in above-mentioned live video stream with it is above-mentioned
The adjacent image of image；
Target image determination unit, if the region of interest of the area-of-interest and above-mentioned consecutive frame image for above-mentioned image
The characteristics of image difference in domain has exceeded preset characteristics of image disparity range, then by above-mentioned image and/or above-mentioned consecutive frame image
It is determined as target image.
Optionally, above-mentioned robot 6 further includes：
Detection of obstacles module, the distance in the traveling process of above-mentioned robot, being carried based on above-mentioned robot
Sensor, obtain above-mentioned robot in above-mentioned robot local environment at a distance from barrier；
Project setting module carries out above-mentioned traveling scheme if being not more than preset distance threshold for above-mentioned distance
Adjustment is advanced alternatively, suspending above-mentioned robot.
Optionally, above-mentioned robot 6 further includes：
Command reception module is used in the traveling process of above-mentioned robot, if above-mentioned track machine people receives movement
The pause instruction that terminal is sent then is based on above-mentioned pause instruction and suspends above-mentioned robot traveling.
Optionally, above-mentioned image capture module 61 includes：
Multichannel ultra high-definition image acquisition units are used for the multichannel ultra high-definition image of the above-mentioned track of captured in real-time；
Infrared image acquisition unit is used for the infrared image of the above-mentioned track of captured in real-time；
Picture quality detection unit, for detecting whether above-mentioned multichannel ultra high-definition image and above-mentioned infrared image reach default
Picture quality condition；
Image culling unit, for when there is the not up to multichannel ultra high-definition image of above-mentioned picture quality condition and/or red
When outer image, the multichannel ultra high-definition image and/or infrared image of above-mentioned not up to above-mentioned picture quality condition are rejected.
Optionally, above- mentioned information uploading module 66 includes：
Acquisition time acquiring unit, the acquisition time for obtaining above-mentioned target image；
Collecting location acquiring unit, the collecting location for obtaining above-mentioned target image；
Image information uploading unit is used for above-mentioned target image, the acquisition time of above-mentioned target image and above-mentioned target
The collecting location of image is uploaded to server.
Optionally, as the artificial flying robot of above-mentioned machine, above-mentioned vision positioning module 62 includes：
Realtime graphic acquiring unit obtains realtime graphic for being based on above-mentioned live video stream；
Initialization unit, for initializing coordinate system, wherein the x-axis of above-mentioned coordinate system is the extending direction of track, above-mentioned
The y-axis of coordinate system is sleeper laying direction, and the z-axis of above-mentioned coordinate system is perpendicular to the direction in track laying face；
X-axis positioning unit, the RFID tag for carrying location information for receiving radio-frequency unit transmission, according to upper
It states RFID tag and carries out x-axis positioning, wherein above-mentioned radio-frequency unit is pre-set on the route of above-mentioned track；
Y-axis positioning unit is used for the phase of the center line and the center line of above-mentioned realtime graphic middle orbit according to above-mentioned realtime graphic
Y-axis positioning is carried out to position；
Z-axis positioning unit, the flying height for acquiring above-mentioned robot by the range sensor of carrying, according to above-mentioned
Flying height carries out z-axis positioning；
Coordinate determination unit, for according to above-mentioned x-axis position as a result, result and the positioning of above-mentioned z-axis of above-mentioned y-axis positioning
As a result, determining above-mentioned robot coordinate residing in above-mentioned coordinate system.
Optionally, as the artificial flying robot of above-mentioned machine, said program determination unit 63 includes：
Realtime graphic acquiring unit obtains realtime graphic for being based on above-mentioned live video stream；
Second binarization unit, for carrying out binary conversion treatment to above-mentioned realtime graphic；
Second extraction unit, the area-of-interest for extracting the above-mentioned realtime graphic after binary conversion treatment；
Third extraction unit, for extracting longitudinal rectangular area in above-mentioned area-of-interest based on Morphology Algorithm；
Straight-line detection unit, for by straight-line detection, the two of above-mentioned track to be determined in above-mentioned longitudinal rectangular area
The position of bar rail；
Fitting unit is fitted for the position to above-mentioned two rail, obtains the center line of two rail；
Traveling scheme determination unit is used for the position of center line and above-mentioned robot based on above-mentioned two rail, in determination
State the traveling scheme of robot.
Optionally, when the artificial Gui Hang robots of above-mentioned machine, above-mentioned robot is equipped with manipulator, and above-mentioned robot is also
Including：
Module is tapped, for during traveling, the bolt of above-mentioned track to be tapped by above-mentioned manipulator；
Audio collection module, for acquiring audio when above-mentioned robot taps the bolt of above-mentioned track；
Audio analysis module, for carrying out audio analysis, the above-mentioned spiral shell of result judgement based on audio analysis to above-mentioned audio
Whether bolt loosens；
Bolt looseness determination module, for when it is that above-mentioned bolt has loosened to judge result, to mobile terminal and/or above-mentioned
Server sends above-mentioned judgement as a result, to notify service personnel to overhaul above-mentioned bolt.
Therefore through the embodiment of the present invention, robot can be realized according to the live video stream of collected track
Intelligence based on track is advanced and intelligent measurement operation, is quickly found out the doubtful orbital region to break down, and will be in the failure
It reports to server so that server can carry out secondary analysis, and be matched above-mentioned target image based on the result of secondary analysis
It is shown to corresponding client, to notify service personnel in time.In the process, it is no longer necessary to inspection manually is carried out to track, and
It is to give robot to execute the Daily Round Check operation of track, by robot, quickly there are failures for identification during inspection
False retrieval is reduced, the possibility that missing inspection occurs in region, realizes that the automation to the infrastructure and infrastructure device of rail traffic is examined
It surveys, ensures Rail Transit System safe operation.
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each work(
Can unit, module division progress for example, in practical application, can be as needed and by above-mentioned function distribution by different
Functional unit, module complete, i.e., the internal structure of above-mentioned robot is divided into different functional units or module, with complete with
The all or part of function of upper description.Each functional unit, module in embodiment can be integrated in a processing unit,
Can be that each unit, module physically exist alone, can also two or more modules, unit be integrated in one unit
In, the form that hardware had both may be used in above-mentioned integrated module, unit is realized, software function module, unit can also be used
Form is realized.In addition, the specific name of each functional unit, module is also only to facilitate mutually differentiation, is not limited to this
The protection domain of application.The specific work process of unit in above system, module can refer to pair in preceding method embodiment
Process is answered, details are not described herein.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, is not described in detail or remembers in some embodiment
The part of load may refer to the associated description of other embodiments.
Those of ordinary skill in the art may realize that lists described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, depends on the specific application and design constraint of technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
In embodiment provided by the present invention, it should be understood that disclosed robot and track detection method, it can be with
It realizes by another way.For example, robot described above is only schematical, for example, above-mentioned module or unit
Division, only a kind of division of logic function, formula that in actual implementation, there may be another division manner, such as multiple units or group
Part can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown
Or the mutual coupling or direct-coupling or communication connection discussed can be by some interfaces, device or unit it is indirect
Coupling or communication connection can be electrical, machinery or other forms.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although with reference to aforementioned reality
Applying example, invention is explained in detail, it will be understood by those of ordinary skill in the art that：It still can be to aforementioned each
Technical solution recorded in embodiment is modified or equivalent replacement of some of the technical features；And these are changed
Or replace, the spirit and scope for various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of track detection method, which is characterized in that the track detection method includes：
The live video stream of robot acquisition trajectory；
Based on the live video stream, the position of the robot is determined by vision positioning；
According to the traveling scheme of robot described in the location determination of the robot；
The robot is controlled to be advanced according to the traveling scheme；
In each frame image of the live video stream, detect whether that there are target images, wherein the target image is display
There is the image of the fault zone of the track；
If in each frame image of the live video stream, there are the target images, and the target image is uploaded to service
Device so that the server carries out secondary analysis to the target image, and based on the result of secondary analysis by the target figure
It is shown as being matched to corresponding client.
2. track detection method as described in claim 1, which is characterized in that each frame image in the live video stream
In, detect whether that there are target images, including：
Binary conversion treatment is carried out to each frame image of the live video stream；
Extract the area-of-interest of each frame image after binary conversion treatment；
Denoising operation is carried out to the area-of-interest of each frame image based on Morphology Algorithm；
For each frame image after denoising, the image of the area-of-interest of image and the area-of-interest of consecutive frame image is calculated
Feature difference, wherein the consecutive frame image is image adjacent with described image in the live video stream；
If the characteristics of image difference of the area-of-interest of described image and the area-of-interest of the consecutive frame image has exceeded pre-
If characteristics of image disparity range, then described image and/or the consecutive frame image are determined as target image.
3. track detection method as described in claim 1, which is characterized in that the track detection method further includes：
In the traveling process of the robot, based on the robot carry range sensor, obtain the robot with
The distance of barrier in the robot local environment；
If the distance is not more than preset distance threshold, the traveling scheme is adjusted, alternatively, suspending the machine
People advances.
4. track detection method as described in claim 1, which is characterized in that the track detection method further includes：
In the traveling process of the robot, if the track machine people receives the pause instruction of mobile terminal transmission,
Suspend the robot based on the pause instruction to advance.
5. track detection method as described in claim 1, which is characterized in that the real-time video of the robot acquisition trajectory
Stream, including：
The infrared image of the multichannel ultra high-definition image of track and the track described in the robot captured in real-time；
It detects the multichannel ultra high-definition image and whether the infrared image reaches preset picture quality condition；
If there is the multichannel ultra high-definition image and/or infrared image for being not up to described image quality requirements, described will be not up to
The multichannel ultra high-definition image and/or infrared image of described image quality requirements are rejected.
6. track detection method as described in claim 1, which is characterized in that described that the target image is uploaded to service
Device, including：
Obtain the acquisition time and collecting location of the target image；
The collecting location of the target image, the acquisition time of the target image and the target image is uploaded to service
Device.
7. such as claim 1 to 6 any one of them track detection method, which is characterized in that the artificial flying machine of machine
People, it is described to be based on the live video stream, the position of the robot is determined by vision positioning, including：
Based on the live video stream, realtime graphic is obtained；
Initialize coordinate system, wherein the x-axis of the coordinate system is the extending direction of track, and the y-axis of the coordinate system spreads for sleeper
The z-axis of set direction, the coordinate system is perpendicular to the direction in track laying face；
The robot receives the RFID tag for carrying location information that radio-frequency unit is sent, according to the radio frequency identification
Label carries out x-axis positioning, wherein the radio-frequency unit is pre-set on the route of the track；
The robot carries out y according to the relative position of the center line of the realtime graphic and the center line of the realtime graphic middle orbit
Axis positions；
The robot acquires the flying height of the robot by the range sensor of carrying, according to the flying height into
Row z-axis positions；
According to x-axis positioning as a result, that the result and the z-axis that the y-axis positions position as a result, determining the robot
The residing coordinate in the coordinate system.
8. such as claim 1 to 6 any one of them track detection method, which is characterized in that the artificial flying machine of machine
People, the traveling scheme of robot described in the location determination according to the robot, including：
Based on the live video stream, realtime graphic is obtained；
Binary conversion treatment is carried out to the realtime graphic；
Extract the area-of-interest of the realtime graphic after binary conversion treatment；
Longitudinal rectangular area in the area-of-interest is extracted based on Morphology Algorithm；
By straight-line detection, the position of two rail of the track is determined in longitudinal rectangular area；
The position of two rail is fitted, the center line of two rail is obtained；
The position of center line and the robot based on two rail, determines the traveling scheme of the robot.
9. such as claim 1 to 6 any one of them track detection method, which is characterized in that the artificial rail row machine of machine
People, the robot are equipped with manipulator, and the track detection method further includes：
In the traveling process of the robot, the bolt of the track is tapped by the manipulator；
Acquire the audio when robot taps the bolt of the track；
Audio analysis is carried out to the audio, whether bolt loosens described in the result judgement based on audio analysis；
Loosened if it is determined that result is the bolt, then send the judgement to mobile terminal and/or the server as a result, with
Service personnel is notified to overhaul the bolt.
10. a kind of robot, which is characterized in that the robot application includes in field of track traffic, the robot：
Video acquisition module is used for the live video stream of machine acquisition trajectory；
Vision positioning module determines the position of the robot by vision positioning for being based on the live video stream；
Scheme determining module, the traveling scheme for robot described in the location determination according to the robot；
Traveling control module is advanced for controlling the robot according to the traveling scheme；
Fault detection module, in each frame image of the live video stream, detecting whether that there are target images, wherein
The target image is the image for the fault zone for showing the track；
Information uploading module, for when in each frame image in the live video stream there are when the target image, will be described
Target image is uploaded to server so that the server carries out secondary analysis to the target image, and is based on secondary analysis
Result the target image be matched to corresponding client show.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810126117.2A CN108491758B (en) | 2018-02-08 | 2018-02-08 | Track detection method and robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810126117.2A CN108491758B (en) | 2018-02-08 | 2018-02-08 | Track detection method and robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108491758A true CN108491758A (en) | 2018-09-04 |
| CN108491758B CN108491758B (en) | 2020-11-20 |
Family
ID=63339994
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810126117.2A Expired - Fee Related CN108491758B (en) | 2018-02-08 | 2018-02-08 | Track detection method and robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108491758B (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109580137A (en) * | 2018-11-29 | 2019-04-05 | 东南大学 | A kind of bridge structure displacement influence line measurement method based on computer vision technique |
| CN111024431A (en) * | 2019-12-26 | 2020-04-17 | 江西交通职业技术学院 | Bridge rapid detection vehicle based on multi-sensor unmanned driving |
| CN111077159A (en) * | 2019-12-31 | 2020-04-28 | 北京京天威科技发展有限公司 | Fault detection method, system, equipment and readable medium for track circuit box |
| CN111380537A (en) * | 2018-12-27 | 2020-07-07 | 奥迪股份公司 | Method and device for positioning target position in navigation map |
| CN111668925A (en) * | 2019-03-05 | 2020-09-15 | 特变电工智能电气有限责任公司 | Transformer inspection tour inspection device based on intelligent vision |
| CN111672045A (en) * | 2020-05-21 | 2020-09-18 | 国网湖南省电力有限公司 | Fire-fighting robot, fire-fighting system and fire-fighting control method |
| CN112014848A (en) * | 2020-02-11 | 2020-12-01 | 深圳技术大学 | Sleeper positioning method, sleeper positioning device and electronic equipment |
| CN112444519A (en) * | 2019-08-30 | 2021-03-05 | 比亚迪股份有限公司 | Vehicle fault detection device and method |
| CN112508911A (en) * | 2020-12-03 | 2021-03-16 | 合肥科大智能机器人技术有限公司 | Rail joint touch net suspension support component crack detection system based on inspection robot and detection method thereof |
| CN113085923A (en) * | 2021-04-15 | 2021-07-09 | 北京智川科技发展有限公司 | Track detection method and device, automatic track detection vehicle and storage medium |
| CN113111704A (en) * | 2021-03-02 | 2021-07-13 | 郑州大学 | Airport pavement disease and foreign matter detection method and system based on deep learning |
| CN114821165A (en) * | 2022-04-19 | 2022-07-29 | 北京运达华开科技有限公司 | Track detection image acquisition and analysis method |
| CN114973694A (en) * | 2022-05-19 | 2022-08-30 | 杭州中威电子股份有限公司 | Tunnel traffic flow monitoring system and method based on inspection robot |
| CN115601719A (en) * | 2022-12-13 | 2023-01-13 | 中铁十二局集团有限公司（Cn） | Climbing robot and method for detecting invasion of foreign objects in subway tunnel |
| CN115760989A (en) * | 2023-01-10 | 2023-03-07 | 西安华创马科智能控制***有限公司 | Hydraulic support robot track alignment method and device |
| CN116596731A (en) * | 2023-05-25 | 2023-08-15 | 北京贝能达信息技术股份有限公司 | Rail transit intelligent operation and maintenance big data management method and system |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101519981A (en) * | 2009-03-19 | 2009-09-02 | 重庆大学 | Mine locomotive anti-collision early warning system based on monocular vision and early warning method thereof |
| CN202178515U (en) * | 2011-07-30 | 2012-03-28 | 山东鲁能智能技术有限公司 | Transformer station intelligent robot inspection system |
| CN102490764A (en) * | 2011-12-13 | 2012-06-13 | 天津卓朗科技发展有限公司 | Automatic detection method of track turnout notch |
| CN104331910A (en) * | 2014-11-24 | 2015-02-04 | 沈阳建筑大学 | Track obstacle detection system based on machine vision |
| CN104796664A (en) * | 2015-03-26 | 2015-07-22 | 成都市斯达鑫辉视讯科技有限公司 | Video monitoring device |
| CN105700532A (en) * | 2016-04-19 | 2016-06-22 | 长沙理工大学 | Vision-based transformer substation inspection robot navigation positioning control method |
| CN106341661A (en) * | 2016-09-13 | 2017-01-18 | 深圳市大道智创科技有限公司 | Patrol robot |
| CN106444588A (en) * | 2016-11-30 | 2017-02-22 | 国家电网公司 | Inspection system and inspection method of valve hall robot based on video monitoring linkage system |
| CN106428558A (en) * | 2016-11-28 | 2017-02-22 | 北京交通大学 | Rail comprehensive inspection method based on air-rail double-purpose unmanned aerial vehicle |
| CN107071344A (en) * | 2017-01-22 | 2017-08-18 | 深圳英飞拓科技股份有限公司 | A kind of large-scale distributed monitor video data processing method and device |
| CN107084754A (en) * | 2017-04-27 | 2017-08-22 | 深圳万发创新进出口贸易有限公司 | A kind of transformer fault detection device |
| CN206653397U (en) * | 2017-03-29 | 2017-11-21 | 张胜雷 | The towed electricity of hanger rail leads to integrated piping lane environmental monitoring intelligent miniature robot |
| CN107433952A (en) * | 2017-05-12 | 2017-12-05 | 北京瑞途科技有限公司 | A kind of intelligent inspection robot |
-
2018
- 2018-02-08 CN CN201810126117.2A patent/CN108491758B/en not_active Expired - Fee Related
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101519981A (en) * | 2009-03-19 | 2009-09-02 | 重庆大学 | Mine locomotive anti-collision early warning system based on monocular vision and early warning method thereof |
| CN202178515U (en) * | 2011-07-30 | 2012-03-28 | 山东鲁能智能技术有限公司 | Transformer station intelligent robot inspection system |
| CN102490764A (en) * | 2011-12-13 | 2012-06-13 | 天津卓朗科技发展有限公司 | Automatic detection method of track turnout notch |
| CN104331910A (en) * | 2014-11-24 | 2015-02-04 | 沈阳建筑大学 | Track obstacle detection system based on machine vision |
| CN104796664A (en) * | 2015-03-26 | 2015-07-22 | 成都市斯达鑫辉视讯科技有限公司 | Video monitoring device |
| CN105700532A (en) * | 2016-04-19 | 2016-06-22 | 长沙理工大学 | Vision-based transformer substation inspection robot navigation positioning control method |
| CN106341661A (en) * | 2016-09-13 | 2017-01-18 | 深圳市大道智创科技有限公司 | Patrol robot |
| CN106428558A (en) * | 2016-11-28 | 2017-02-22 | 北京交通大学 | Rail comprehensive inspection method based on air-rail double-purpose unmanned aerial vehicle |
| CN106444588A (en) * | 2016-11-30 | 2017-02-22 | 国家电网公司 | Inspection system and inspection method of valve hall robot based on video monitoring linkage system |
| CN107071344A (en) * | 2017-01-22 | 2017-08-18 | 深圳英飞拓科技股份有限公司 | A kind of large-scale distributed monitor video data processing method and device |
| CN206653397U (en) * | 2017-03-29 | 2017-11-21 | 张胜雷 | The towed electricity of hanger rail leads to integrated piping lane environmental monitoring intelligent miniature robot |
| CN107084754A (en) * | 2017-04-27 | 2017-08-22 | 深圳万发创新进出口贸易有限公司 | A kind of transformer fault detection device |
| CN107433952A (en) * | 2017-05-12 | 2017-12-05 | 北京瑞途科技有限公司 | A kind of intelligent inspection robot |
Non-Patent Citations (1)
| Title |
|---|
| 柳斐: "变电站定轨自主巡视机器人***研究", 《中国优秀硕士学位论文全文数据库（电子期刊）工程科技Ⅱ辑》 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109580137A (en) * | 2018-11-29 | 2019-04-05 | 东南大学 | A kind of bridge structure displacement influence line measurement method based on computer vision technique |
| CN111380537A (en) * | 2018-12-27 | 2020-07-07 | 奥迪股份公司 | Method and device for positioning target position in navigation map |
| CN111668925A (en) * | 2019-03-05 | 2020-09-15 | 特变电工智能电气有限责任公司 | Transformer inspection tour inspection device based on intelligent vision |
| CN112444519B (en) * | 2019-08-30 | 2022-07-15 | 比亚迪股份有限公司 | Vehicle fault detection device and method |
| CN112444519A (en) * | 2019-08-30 | 2021-03-05 | 比亚迪股份有限公司 | Vehicle fault detection device and method |
| CN111024431A (en) * | 2019-12-26 | 2020-04-17 | 江西交通职业技术学院 | Bridge rapid detection vehicle based on multi-sensor unmanned driving |
| CN111024431B (en) * | 2019-12-26 | 2022-03-11 | 江西交通职业技术学院 | Bridge rapid detection vehicle based on multi-sensor unmanned driving |
| CN111077159A (en) * | 2019-12-31 | 2020-04-28 | 北京京天威科技发展有限公司 | Fault detection method, system, equipment and readable medium for track circuit box |
| CN112014848A (en) * | 2020-02-11 | 2020-12-01 | 深圳技术大学 | Sleeper positioning method, sleeper positioning device and electronic equipment |
| CN112014848B (en) * | 2020-02-11 | 2023-06-23 | 深圳技术大学 | Sleeper positioning method, sleeper positioning device and electronic equipment |
| CN111672045B (en) * | 2020-05-21 | 2021-11-30 | 国网湖南省电力有限公司 | Fire-fighting robot, fire-fighting system and fire-fighting control method |
| CN111672045A (en) * | 2020-05-21 | 2020-09-18 | 国网湖南省电力有限公司 | Fire-fighting robot, fire-fighting system and fire-fighting control method |
| CN112508911A (en) * | 2020-12-03 | 2021-03-16 | 合肥科大智能机器人技术有限公司 | Rail joint touch net suspension support component crack detection system based on inspection robot and detection method thereof |
| CN113111704A (en) * | 2021-03-02 | 2021-07-13 | 郑州大学 | Airport pavement disease and foreign matter detection method and system based on deep learning |
| CN113085923B (en) * | 2021-04-15 | 2022-01-25 | 北京智川科技发展有限公司 | Track detection method and device, automatic track detection vehicle and storage medium |
| CN113085923A (en) * | 2021-04-15 | 2021-07-09 | 北京智川科技发展有限公司 | Track detection method and device, automatic track detection vehicle and storage medium |
| CN114821165A (en) * | 2022-04-19 | 2022-07-29 | 北京运达华开科技有限公司 | Track detection image acquisition and analysis method |
| CN114973694A (en) * | 2022-05-19 | 2022-08-30 | 杭州中威电子股份有限公司 | Tunnel traffic flow monitoring system and method based on inspection robot |
| CN114973694B (en) * | 2022-05-19 | 2024-05-24 | 杭州中威电子股份有限公司 | Tunnel traffic flow monitoring system and method based on inspection robot |
| CN115601719A (en) * | 2022-12-13 | 2023-01-13 | 中铁十二局集团有限公司（Cn） | Climbing robot and method for detecting invasion of foreign objects in subway tunnel |
| CN115760989A (en) * | 2023-01-10 | 2023-03-07 | 西安华创马科智能控制***有限公司 | Hydraulic support robot track alignment method and device |
| CN116596731A (en) * | 2023-05-25 | 2023-08-15 | 北京贝能达信息技术股份有限公司 | Rail transit intelligent operation and maintenance big data management method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108491758B (en) | 2020-11-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108491758A (en) | A kind of track detection method and robot | |
| WO2022037278A1 (en) | Substation inspection robot system based on artificial intelligence | |
| KR101949525B1 (en) | Safety management system using unmanned detector | |
| CN108957240A (en) | Electric network fault is remotely located method and system | |
| CN110492607A (en) | A kind of intelligent substation condition monitoring system based on ubiquitous electric power Internet of Things | |
| CN108494097A (en) | Robot inspection system for substation | |
| CN108956640A (en) | Vehicle-mounted detection apparatus and detection method suitable for distribution line inspection | |
| CN104951775A (en) | Video technology based secure and smart recognition method for railway crossing protection zone | |
| CN102915638A (en) | Surveillance video-based intelligent parking lot management system | |
| CN108710827B (en) | A kind of micro- police service inspection in community and information automatic analysis system and method | |
| CN103729908A (en) | Intelligent inspection device of railway tunnel and application method thereof | |
| CN109208468B (en) | Automatic bridge inspection system based on video analysis | |
| CN110874866B (en) | Video-based three-dimensional monitoring method and system for transformer substation | |
| CN108008666A (en) | One kind building neutral net and its method of work | |
| CN106444684A (en) | Electrical equipment operating environment data acquisition and telecontrol comprehensive linkage device and use method | |
| CN109905667A (en) | A kind of video monitoring system suitable for large span special construction bridge | |
| CN115661964A (en) | Intelligent inspection system for comprehensive pipe gallery | |
| CN205827660U (en) | A kind of open type parking ground parking management equipment | |
| KR101989376B1 (en) | Integrated track circuit total monitoring system | |
| CN113014870B (en) | Subway gate passage ticket evasion identification method based on passenger posture rapid estimation | |
| CN113691778A (en) | Panoramic station patrol system for urban rail transit station | |
| CN213518003U (en) | A patrol and examine robot and system of patrolling and examining for airport pavement | |
| CN109934161A (en) | Vehicle identification and detection method and system based on convolutional neural network | |
| CN112532927A (en) | Intelligent safety management and control system for construction site | |
| CN107991953A (en) | One kind building cental system |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20201120 |