CN109606678B - A crawler-type unmanned aerial vehicle for automatic positioning of bridge supports - Google Patents
A crawler-type unmanned aerial vehicle for automatic positioning of bridge supportsDownload PDFInfo
- Publication number
- CN109606678B CN109606678BCN201811396449.9ACN201811396449ACN109606678BCN 109606678 BCN109606678 BCN 109606678BCN 201811396449 ACN201811396449 ACN 201811396449ACN 109606678 BCN109606678 BCN 109606678B
- Authority
- CN
- China
- Prior art keywords
- support
- crawler
- aerial vehicle
- unmanned aerial
- bridge
- 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
Images
Classifications
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64C—AEROPLANES; HELICOPTERS
- B64C39/00—Aircraft not otherwise provided for
- B64C39/02—Aircraft not otherwise provided for characterised by special use
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64D—EQUIPMENT FOR FITTING IN OR TO AIRCRAFT; FLIGHT SUITS; PARACHUTES; ARRANGEMENT OR MOUNTING OF POWER PLANTS OR PROPULSION TRANSMISSIONS IN AIRCRAFT
- B64D47/00—Equipment not otherwise provided for
- B—PERFORMING OPERATIONS; TRANSPORTING
- B64—AIRCRAFT; AVIATION; COSMONAUTICS
- B64D—EQUIPMENT FOR FITTING IN OR TO AIRCRAFT; FLIGHT SUITS; PARACHUTES; ARRANGEMENT OR MOUNTING OF POWER PLANTS OR PROPULSION TRANSMISSIONS IN AIRCRAFT
- B64D47/00—Equipment not otherwise provided for
- B64D47/08—Arrangements of cameras
- E—FIXED CONSTRUCTIONS
- E01—CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
- E01D—CONSTRUCTION OF BRIDGES, ELEVATED ROADWAYS OR VIADUCTS; ASSEMBLY OF BRIDGES
- E01D19/00—Structural or constructional details of bridges
- E01D19/02—Piers; Abutments ; Protecting same against drifting ice
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Architecture (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Manipulator (AREA)
- Image Analysis (AREA)
Abstract
Translated fromChinese
Description
Technical Field
The invention relates to civil engineering and artificial intelligence crossing technologies, in particular to a crawler-type unmanned aerial vehicle for automatically positioning a bridge bearing.
Background
With the rapid development of the infrastructure construction of China in recent years, the civil engineering industry develops rapidly, and after a large number of roads and bridges are constructed, the later-stage detection and maintenance work is carried out. The bridge support is an important component for connecting upper and lower structures of a bridge, can be the throat of the bridge, has a great relationship, and once a disease occurs, if the disease is not found and treated in time, the stress state and traffic safety of the structure are influenced. At present, the main detection way of the bridge support is manual detection, and the method is time-consuming, labor-consuming and can influence traffic. Some bridges built in mountains and on the sea are difficult to realize by a manual detection method, or the safety of bridge detection personnel is difficult to ensure. Therefore, an apparatus for automatically positioning a bridge support and collecting an image of the support is urgently needed.
Along with the rapid development of unmanned aerial vehicle technique, unmanned aerial vehicle's application also more permeates each industry, however, often because there is certain safe distance between unmanned aerial vehicle and the barrier when shooing the support with unmanned aerial vehicle among the prior art, the distance of unmanned aerial vehicle and bridge lower surface is far away when consequently shooing the support, this can influence the shooting angle and the shooting quality of support, can lead to the stability of unmanned aerial vehicle self to reduce because of the air current when unmanned aerial vehicle is close to the decking in addition, can't acquire high-quality support photo. And when using unmanned aerial vehicle to shoot bridge beam supports at present, the relative position between adjustment unmanned aerial vehicle and the bridge beam supports all relies on the manual work to go on, and this kind of method efficiency is lower.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide a crawler-type unmanned aerial vehicle capable of automatically positioning a bridge support, which can automatically adjust the relative position between the unmanned aerial vehicle and the bridge support and improve the stability of the unmanned aerial vehicle.
The technical scheme is as follows: the crawler-type unmanned aerial vehicle for automatically positioning the bridge support comprises a body, a crawler-type support and a connecting device, wherein the crawler-type support comprises a crawler and a support, the support comprises two frames which are arranged in parallel, a plurality of rotating rods are rotatably connected between the two frames, the crawler is sleeved on the rotating rods, and the body is connected with the frames through the connecting device; in addition, crawler-type unmanned aerial vehicle still includes image acquisition module and control module, and image acquisition module gathers bridge beam supports's image, and control module is according to the position of the automatic definite support of image and control unmanned aerial vehicle automatic adjustment and the relative position of support.
Further, the connecting device is telescopic.
Further, the connecting device comprises a holding part, a connecting rod and a fixing rod fixedly connected between the two frames, the holding part is connected with one end of the connecting wire, the connecting rod is provided with a boss, the fixing rod is fixedly connected with a track, the track is respectively connected with one end of the spring and one end of the adjusting rod, the other end of the spring is positioned below the boss, and the other end of the spring is respectively connected with the other end of the connecting wire and the other end of the adjusting rod.
Further, the shape of the portion of gripping is cyclic annular, and the connecting rod adopts elastic material to make, makes unmanned aerial vehicle have certain shock-absorbing function when leaning on the decking bottom like this.
Furthermore, the number of the tracks is two, and the two tracks are respectively positioned on two sides of the connecting rod.
Further, the boss is plural.
Furthermore, all the rotating rods are uniformly distributed on the peripheries of the two frames.
Further, the control module automatically determines the position of the support through a convolutional neural network, and the establishment of the convolutional neural network comprises the following steps:
s1: acquiring an image of a bridge support, adjusting the size of the image of the bridge support, calibrating position information of the support, and preprocessing the image;
s2: dividing the bridge support image into a training set and a test set, wherein the training set is used for training a convolutional neural network, and the test set is used for testing the convolutional neural network;
s3: establishing a convolutional neural network which comprises an input layer, an output layer, a convolutional layer and a pooling layer, wherein the output value of the convolutional neural network is the position information of the support in the bridge support image;
s4: and measuring the error between the actual output value of the convolutional neural network and the support position information which is calibrated in the step S1 through a loss function, and iteratively training the weight values of all layers of the convolutional neural network through a gradient descent method and a back propagation algorithm to obtain the convolutional neural network with the function of automatically positioning the bridge support.
Further, in step S1, the calibration of the support position information is performed by: and establishing a Cartesian coordinate system by taking the upper left corner of the picture as the origin of a coordinate axis, drawing a rectangular frame at the position of the support in the picture, and expressing the position of the support by using the corner coordinates of the rectangular frame.
Further, the loss function in step S4 is a mean square error MSE, which is obtained according to equation (1):
in the formula (1), n is convolution neural net for each inputThe number of images of the collaterals; y isiIs the output value of the convolutional neural network,
the information of the support position calibrated in step S1.
Has the advantages that: the invention discloses a crawler-type unmanned aerial vehicle capable of automatically positioning a bridge support, which has the following beneficial effects compared with the prior art:
1) according to the invention, through the matching of the control module and the image acquisition module, the relative position between the unmanned aerial vehicle and the bridge support can be automatically adjusted, manual adjustment is not needed, and the efficiency is improved;
2) according to the invention, through the design of the crawler-type support, the unmanned aerial vehicle can be positioned below the bridge deck by means of the crawler-type support; and there is static friction between track and the decking, when unmanned aerial vehicle rolled along the decking and shoots, only had the dwang to take place to rotate, can improve unmanned aerial vehicle's stability like this.
Drawings
Fig. 1 is a structural diagram of an unmanned aerial vehicle in the embodiment of the present invention;
FIG. 2 is a block diagram of a track frame according to an embodiment of the present invention;
fig. 3 is a structural diagram of the drone with the track removed in the particular embodiment of the invention;
FIG. 4 is a block diagram of a second embodiment of a coupling device in accordance with an embodiment of the present invention;
FIG. 5 is a block diagram of a second embodiment of a coupling device in accordance with the present invention in its natural state;
FIG. 6 is a view showing a construction of a second embodiment of a connecting device according to the present invention when a grip portion is pulled up.
Detailed Description
This embodiment discloses an automatic crawler-type unmanned aerial vehicle of location bridge beam supports, as shown in fig. 1, includingfuselage 1, crawler-type support 2, connectingdevice 3, image acquisition module and control module,fuselage 1 passes through connectingdevice 3 and connects crawler-type support 2, and image acquisition module gathers bridge beam supports's image, and control module is according to the position of the automatic definite support of image and control unmanned aerial vehicle automatic adjustment and the relative position of support.
As shown in fig. 1, thecrawler frame 2 includes acrawler 21 and aframe 22. As shown in fig. 2, thesupport 22 includes twoframes 221 arranged in parallel, a plurality of rotatingrods 222 are rotatably connected between the twoframes 221, and thecrawler belt 21 is sleeved on the rotatingrods 222. The rotatinglevers 222 are uniformly arranged in the circumferential direction of the twoframes 221.
A first embodiment of theconnecting device 3, shown in fig. 1 and 3, comprises connecting rods and fixing rods perpendicular to each other and fixedly connected, the fixing rods being fixed between the twoframes 221. The connectingmeans 3 is not telescopic in this embodiment.
A second embodiment of the connectingdevice 3 is shown in fig. 4, and includes anannular grip portion 31, a connectingrod 33, and afixing rod 32 fixedly connected between twoframes 221. As shown in fig. 5, theholding portion 31 is connected to one end of the connectingline 341, the connectingrod 33 is provided with fourbosses 331, thefixing rod 32 is fixedly connected to tworails 34, the tworails 34 are respectively located at two sides of the connectingrod 33, therails 34 are respectively connected to one end of thespring 35 and one end of the adjustingrod 36, the other end of thespring 35 is located below thebosses 331, and the other end of thespring 35 is respectively connected to the other end of the connectingline 341 and the other end of the adjustingrod 36. Two groups ofsprings 35 and two groups of adjustingrods 36 are provided respectively. The connectingrod 33 is made of an elastic material, such as rubber. Thetrack 34 is also provided with ahousing 37 on the outside, as shown in fig. 4. In this embodiment, the connectingdevice 3 is retractable, and how to achieve the retractable is described as follows: when theholding portion 31 is pulled up, as shown in fig. 6, the connectingwire 341 rotates the adjustingrod 36, thespring 35 is compressed, the connectingrod 33 can be controlled to move upward or downward, the adjustingrod 36 can move in thetrack 34, and theholding portion 31 is released when the connectingrod 33 moves upward or downward to a proper position, that is, when thedevice 3 to be connected contracts or expands to a proper length. The length adjustment (expansion) of the connectingdevice 3 is discontinuous, and only integral multiple of the distance between thebosses 331 can be adjusted.
The image acquisition module may employ a device capable of taking high definition images, such as a Canon 5D3 camera.
The control module adopts an artificial intelligence chip, and the artificial intelligence chip can select the cambrian 1H8 which is issued by the Chinese academy and faces the low-power-consumption scene vision application. Different from the traditional chip, the new generation of cambrian artificial intelligence chip simulates neurons and synapses of the brain, one instruction can complete the processing of a group of neurons, and the efficiency of the calculation mode is hundreds of times higher than that of the traditional chip when the intelligent processing is carried out, and the performance power consumption ratio also realizes leap.
The control module automatically determines the position of the support through a convolutional neural network, and the establishment of the convolutional neural network comprises the following steps:
s1: acquiring an image of a bridge support, adjusting the size of the image of the bridge support, calibrating position information of the support, and preprocessing the image;
s2: dividing the bridge support image into a training set and a test set, wherein the training set is used for training a convolutional neural network, and the test set is used for testing the convolutional neural network;
s3: establishing a convolutional neural network which comprises an input layer, an output layer, a convolutional layer and a pooling layer, wherein the output value of the convolutional neural network is the position information of the support in the bridge support image;
s4: and measuring the error between the actual output value of the convolutional neural network and the support position information which is calibrated in the step S1 through a loss function, and iteratively training the weight values of all layers of the convolutional neural network through a gradient descent method and a back propagation algorithm to obtain the convolutional neural network with the function of automatically positioning the bridge support.
In step S1, the calibration of the support position information is performed by: and establishing a Cartesian coordinate system by taking the upper left corner of the picture as the origin of a coordinate axis, drawing a rectangular frame at the position of the support in the picture, and expressing the position of the support by using the corner coordinates of the rectangular frame. The image preprocessing method comprises the following steps: the sum of the pixel values of all the images is calculated and then divided by the number of images to obtain a mean image, and the pixel values of the mean image are subtracted in each image.
The loss function in step S4 is a mean square error MSE, which is obtained according to equation (1):
in the formula (1), n is the number of images input into the convolutional neural network each time; y isiIs the output value of the convolutional neural network,
the information of the support position calibrated in step S1.
In step S4, the gradient descent method specifically includes: and calculating the gradient of the loss function to each weight, starting from any point, moving for a certain distance along the opposite direction of the gradient of the point, and continuously moving for a certain distance along the opposite direction of the gradient at a new position, so that the weight of the network is continuously updated. The back propagation algorithm comprises the following specific steps: when the weights of all layers of the convolutional neural network are updated iteratively by using a gradient descent method, the gradients are propagated forwards from the last layer of the network in sequence according to a chain derivation method.
Claims (10)
Translated fromChinese
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811396449.9ACN109606678B (en) | 2018-11-22 | 2018-11-22 | A crawler-type unmanned aerial vehicle for automatic positioning of bridge supports |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811396449.9ACN109606678B (en) | 2018-11-22 | 2018-11-22 | A crawler-type unmanned aerial vehicle for automatic positioning of bridge supports |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109606678A CN109606678A (en) | 2019-04-12 |
| CN109606678Btrue CN109606678B (en) | 2021-09-07 |
Family
ID=66003455
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811396449.9AActiveCN109606678B (en) | 2018-11-22 | 2018-11-22 | A crawler-type unmanned aerial vehicle for automatic positioning of bridge supports |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109606678B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112376419A (en)* | 2020-12-07 | 2021-02-19 | 湖南桥康智能科技有限公司 | Bridge jib intellectual detection system mechanism |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003099656A1 (en)* | 2002-05-23 | 2003-12-04 | Gerhard Mellmann | Aircraft comprising a rescue device |
| CN103459250A (en)* | 2011-03-29 | 2013-12-18 | 法国高等航空和航天学院 | Remotely controlled micro/nanoscale aerial vehicle comprising a system for traveling on the ground, vertical takeoff, and landing |
| CN103712035A (en)* | 2014-01-08 | 2014-04-09 | 北京理工大学 | Cage type pipeline aircraft |
| CN204161620U (en)* | 2014-10-16 | 2015-02-18 | 云南电网公司红河供电局 | The full landform retractable landing gear of depopulated helicopter |
| CN105354568A (en)* | 2015-08-24 | 2016-02-24 | 西安电子科技大学 | Convolutional neural network based vehicle logo identification method |
| CN106467168A (en)* | 2015-08-17 | 2017-03-01 | 富士通株式会社 | Aircraft frame structure, aircraft, aircraft using method |
| CN106956768A (en)* | 2017-05-02 | 2017-07-18 | 锐合防务技术（北京）有限公司 | Aircraft |
| CN206367593U (en)* | 2016-12-28 | 2017-08-01 | 芜湖元一航空科技有限公司 | A kind of inspection multi-purpose unmanned aerial vehicle |
| CN107352022A (en)* | 2017-06-08 | 2017-11-17 | 国蓉科技有限公司 | A kind of spherical UAS of rotor of impact resistant four |
| CN207550496U (en)* | 2017-12-15 | 2018-06-29 | 江苏航丰智控无人机有限公司 | Collapsible damping device and the unmanned plane of rising and falling |
| CN108288269A (en)* | 2018-01-24 | 2018-07-17 | 东南大学 | Bridge pad disease automatic identifying method based on unmanned plane and convolutional neural networks |
| CN207670666U (en)* | 2017-12-20 | 2018-07-31 | 深圳市中联讯科技有限公司 | Unmanned plane with shock-damping structure |
| CN207956042U (en)* | 2018-02-07 | 2018-10-12 | 北京交通大学 | A kind of novel unmanned plane undercarriage |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7946531B2 (en)* | 2008-02-14 | 2011-05-24 | The Boeing Company | Debris deflecting device, system, and method |
| CN102541064A (en)* | 2012-03-27 | 2012-07-04 | 沈阳中兴电力通信有限公司 | Magnetic navigation routing inspection robot |
| US9725161B2 (en)* | 2013-12-10 | 2017-08-08 | Borealis Technical Limited | Method for maximizing powered aircraft drive wheel traction |
| US9708934B2 (en)* | 2014-07-11 | 2017-07-18 | General Electric Company | System and method for inspecting turbomachines |
| US9418319B2 (en)* | 2014-11-21 | 2016-08-16 | Adobe Systems Incorporated | Object detection using cascaded convolutional neural networks |
| CN206756706U (en)* | 2017-05-26 | 2017-12-15 | 中铁十八局集团有限公司 | Bridge bottom disease follow shot system |
| CN107737755B (en)* | 2017-11-07 | 2021-04-09 | 雅砻江流域水电开发有限公司 | Intelligent mobile trash removal system based on hydroelectric power generation and control method thereof |
- 2018
- 2018-11-22CNCN201811396449.9Apatent/CN109606678B/enactiveActive
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003099656A1 (en)* | 2002-05-23 | 2003-12-04 | Gerhard Mellmann | Aircraft comprising a rescue device |
| CN103459250A (en)* | 2011-03-29 | 2013-12-18 | 法国高等航空和航天学院 | Remotely controlled micro/nanoscale aerial vehicle comprising a system for traveling on the ground, vertical takeoff, and landing |
| CN103712035A (en)* | 2014-01-08 | 2014-04-09 | 北京理工大学 | Cage type pipeline aircraft |
| CN204161620U (en)* | 2014-10-16 | 2015-02-18 | 云南电网公司红河供电局 | The full landform retractable landing gear of depopulated helicopter |
| CN106467168A (en)* | 2015-08-17 | 2017-03-01 | 富士通株式会社 | Aircraft frame structure, aircraft, aircraft using method |
| CN105354568A (en)* | 2015-08-24 | 2016-02-24 | 西安电子科技大学 | Convolutional neural network based vehicle logo identification method |
| CN206367593U (en)* | 2016-12-28 | 2017-08-01 | 芜湖元一航空科技有限公司 | A kind of inspection multi-purpose unmanned aerial vehicle |
| CN106956768A (en)* | 2017-05-02 | 2017-07-18 | 锐合防务技术（北京）有限公司 | Aircraft |
| CN107352022A (en)* | 2017-06-08 | 2017-11-17 | 国蓉科技有限公司 | A kind of spherical UAS of rotor of impact resistant four |
| CN207550496U (en)* | 2017-12-15 | 2018-06-29 | 江苏航丰智控无人机有限公司 | Collapsible damping device and the unmanned plane of rising and falling |
| CN207670666U (en)* | 2017-12-20 | 2018-07-31 | 深圳市中联讯科技有限公司 | Unmanned plane with shock-damping structure |
| CN108288269A (en)* | 2018-01-24 | 2018-07-17 | 东南大学 | Bridge pad disease automatic identifying method based on unmanned plane and convolutional neural networks |
| CN207956042U (en)* | 2018-02-07 | 2018-10-12 | 北京交通大学 | A kind of novel unmanned plane undercarriage |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109606678A (en) | 2019-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108288269A (en) | Bridge pad disease automatic identifying method based on unmanned plane and convolutional neural networks | |
| CN104978580B (en) | A kind of insulator recognition methods for unmanned plane inspection transmission line of electricity | |
| CN108363961A (en) | Bridge pad disease recognition method based on transfer learning between convolutional neural networks | |
| CN112164015A (en) | Monocular vision autonomous inspection image acquisition method and device and power inspection unmanned aerial vehicle | |
| JPWO2016059877A1 (en) | Control apparatus, control method, and air vehicle device | |
| CN113344953B (en) | Machine vision tidal bore flow velocity measurement method based on unmanned aerial vehicle | |
| CN109631854B (en) | A low-altitude unmanned aerial vehicle photography method for measuring coal mining subsidence cracks in mining areas | |
| CN112859905A (en) | Unmanned aerial vehicle inspection route generation method and device for overhead power line and unmanned aerial vehicle | |
| CN111928825B (en) | Unmanned aerial vehicle continuous circle surrounding oblique photography aerial photography method | |
| Sulistijono et al. | Implementation of victims detection framework on post disaster scenario | |
| CN109606678B (en) | A crawler-type unmanned aerial vehicle for automatic positioning of bridge supports | |
| CN103954314A (en) | Unmanned aerial vehicle low-altitude remote sensing simulating device | |
| CN110104167A (en) | A kind of automation search and rescue UAV system and control method using infrared thermal imaging sensor | |
| CN111814720A (en) | A method for detection and classification of foreign objects in airport runways based on UAV vision | |
| AU2018309905A1 (en) | Image processing device, image processing method, and program | |
| JP2017026577A (en) | Creating method for three-dimensional survey photographs | |
| CN111860175A (en) | A method and device for detecting vehicles in unmanned aerial vehicle images based on lightweight network | |
| CN117115252B (en) | A vision-based spatial pose estimation method for bionic flapping-wing UAVs | |
| JP2017201261A (en) | Shape information generation system | |
| CN110009696A (en) | Optimization of BP Neural Network Tri-objective Vision Calibration Based on Bee Colony Algorithm | |
| CN105784346A (en) | Aircraft holder stability testing device and method based on digital camera | |
| CN107741702A (en) | Attitude disturbance generating device for image acquisition of complex opto-mechanical components on ground simulation platform | |
| CN109342439A (en) | Cable Structure appearance detecting method based on unmanned plane | |
| JP2020076674A (en) | Current situation survey drawing creation system, method, and program | |
| Jiang et al. | Thrust vectoring control for infrastructure inspection multirotor vehicle |
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 |