CN109739254B - Unmanned aerial vehicle adopting visual image positioning in power inspection and positioning method thereof - Google Patents
Unmanned aerial vehicle adopting visual image positioning in power inspection and positioning method thereof Download PDFInfo
- Publication number
- CN109739254B CN109739254B CN201811380411.2A CN201811380411A CN109739254B CN 109739254 B CN109739254 B CN 109739254B CN 201811380411 A CN201811380411 A CN 201811380411A CN 109739254 B CN109739254 B CN 109739254B
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- positioning
- ground station
- uav
- 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
Landscapes
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
本发明公开了一种电力巡检中采用视觉图像定位的无人机及其定位方法，通过机载摄像头和地面站的算法程序实现对无人机位置的确定。在无人机云台装备一个定位单目摄像头获取无人机下方物体实时图像，图像回传给地面站，由地面站的视觉定位算法程序ORB‑SLAM2计算出无人机视觉定位摄像头的位姿，以此代表无人机的位置信息，从而对无人机实现视觉定位。并且在地面站系统中根据特征点信息构建3D数字地图，记录无人机运动轨迹以及分布的特征点，同时可以直观地了解电网输电线杆塔周围的地理环境信息。
The invention discloses an unmanned aerial vehicle using visual image positioning in electric power inspection and a positioning method thereof. The location of the unmanned aerial vehicle is determined through an airborne camera and an algorithm program of a ground station. A positioning monocular camera is equipped on the UAV PTZ to obtain real-time images of objects under the UAV, and the images are sent back to the ground station. The visual positioning algorithm program ORB‑SLAM2 of the ground station calculates the pose of the UAV visual positioning camera. , so as to represent the position information of the UAV, so as to realize the visual positioning of the UAV. And in the ground station system, a 3D digital map is constructed according to the feature point information to record the UAV's trajectory and distributed feature points, and at the same time, it can intuitively understand the geographical environment information around the power grid transmission line tower.
Description
Technical Field
The invention relates to the field of unmanned aerial vehicle power grid inspection, in particular to an unmanned aerial vehicle adopting visual image positioning in power inspection and a positioning method thereof.
Background
In recent years, the unmanned aerial vehicle trade is steadily developed, and the unmanned aerial vehicle performance obtains greatly promoting, and four rotor unmanned aerial vehicle especially have received the favor of trades such as fan and agricultural, express delivery owing to its stability is good, and is easy and simple to handle, characteristics such as can hover have received the hobbyist of taking photo by plane and agriculture.
Unmanned aerial vehicle electric wire netting is patrolled and examined has obtained high attention more recently. Because it has safe and reliable, high-efficient nimble, low cost's characteristic, must replace artifical the inspection step by step.
In addition, taking southern power grid as an example, the asset scale of 110kV and above power transmission lines is rapidly increased, and the annual average growth rate reaches 9.6%. The east-west span of the whole network overhead transmission line is 2000 kilometers, the altitude span is 4300 meters, wherein more than 80 percent of lines are located in mountain and green areas far away from towns, traffic main lines and rare people, the specific weight of the extra-dimensional lines needing extra patrol accounts for 20 percent of the total lines, and the transmission line has high operation and inspection difficulty and high quality requirement. And traditional manual patrol inspection work degree of difficulty is big, and the working cycle is long, and factor of safety is low, and the cost of labor is high, and it is serious to patrol inspection personnel's disappearance phenomenon in recent years simultaneously, and the annual average growth rate is less than 3%. The traditional manual mode can not meet the requirements of operation and maintenance of power grids in China more and more, and under the background, the exploration of a new mode of unmanned aerial vehicle power inspection is very necessary.
Secondly, unmanned aerial vehicles used for power inspection currently rely on a GPS global positioning system. Because the coordinates of towers of general power transmission lines are mostly in regions with obvious topographic relief and remote positions, the positioning effect of the unmanned aerial vehicle is not ideal. Furthermore, the positioning effect is also affected by obstacles and electromagnetic interference. Once GPS satellite positioning is disturbed or signals are lost, operation of the drone in the event of a positioning information error must lead to catastrophic results.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to solve the problems that the existing unmanned aerial vehicle power inspection only adopts GPS positioning, because the tower coordinates of a common power transmission line are mostly in a region with obvious topographic relief and remote position, the positioning effect of the GPS on the unmanned aerial vehicle is not ideal, in addition, the positioning effect can be influenced by obstacles and electromagnetic interference, and once the GPS satellite positioning is interfered or signals are lost, the unmanned aerial vehicle can work under the condition of wrong positioning information to certainly cause disastrous results.
The technical scheme is as follows: in order to solve the above problems, the present invention provides the following technical solutions:
the utility model provides an adopt visual image to carry out unmanned aerial vehicle of location in electric power patrols and examines, has the cloud platform including the unmanned aerial vehicle base member that can realize flight function in unmanned aerial vehicle's bottom, is equipped with the camera lens on the cloud platform and can moves towards the monocular camera of bottom surface all the time.
Furthermore, a three-dimensional laser radar scanning device, an infrared thermal imager, an ultraviolet imager and a multispectral imager which can assist in imaging are further arranged in the base body of the unmanned aerial vehicle, and a high-resolution visible light camera/camera which can detect the electric power tower is further arranged on the holder.
Further, still be equipped with the wireless communication module that can carry out the communication with the ground satellite station in unmanned aerial vehicle's the base member.
An unmanned aerial vehicle positioning method for positioning by adopting visual images in power inspection comprises the following steps:
1) the unmanned aerial vehicle communicates with a ground station after taking off, and the ground station receives image information obtained by shooting of the unmanned aerial vehicle through a monocular camera without stopping;
2) the ground station processes the image frame returned by the unmanned aerial vehicle in real time, calculates R and t of the attitude change of the previous frame, and carries out real-time processing on the image frame
Storing the key frame;
3) eliminate the error, improve unmanned aerial vehicle positioning accuracy.
Further, in the step 2), the change of the pose between adjacent frames is described by a rotation matrix R and a translation vector t, and specifically includes the following steps:
a) the ground station extracts ORB feature points from the image frame and performs matching according to the feature points of the last frame stored in the system;
b) obtaining a plurality of pairs of feature point groups (x) after matchingi1,xi2) Using epipolar constraint equations
Where E is called the essential matrix and E ═ tΛR, and at least 8 pairs of the matched characteristic points construct an equation set; the essence matrix E obtained by adopting the eight-point method comprises the pose transformation information of the camera;
then, the intrinsic matrix E is subjected to SVD (singular value decomposition), i.e. E is decomposed into E ═ U ∑ VTIn the form of (a); there are two groups of solutions for each of R and t:
because the positions of the characteristic points are necessarily in front of the monocular camera, three groups of solutions which are not in conformity can be eliminated;
c) continuously inserting key frames, recording feature point information, performing local nonlinear optimization on the monocular camera pose information by adopting a BA algorithm, screening the recorded key frames, and removing redundant key frames.
Further, the step 3) is divided into a closed-loop detection link and a closed-loop correction link,
and a closed loop detection link calculates bag-of-word information of the key frame, and if similar bag-of-word descriptions exist in the system, the monocular camera returns to a scene which is obtained before.
When a closed loop is detected, calculating similarity transformation by a monocular SLAM through a Sim3 algorithm;
in the closed-loop correction link, firstly, repeated point clouds are fused, and a new edge is inserted into the Coovirility Graph to connect a closed loop; the pose information of the current frame and the key frames connected with the current frame are corrected. And then optimizing a pose Graph through the Essential Graph, and dispersing errors into the whole Graph.
Has the advantages that: compared with the prior art, the invention has the advantages that: according to the invention, a positioning monocular camera is arranged on the unmanned aerial vehicle holder to obtain a real-time image of an object below the unmanned aerial vehicle, the image is transmitted back to the ground station, and the position and attitude of the unmanned aerial vehicle visual positioning camera are calculated by a visual positioning algorithm program ORB-SLAM2 of the ground station, so that the position information of the unmanned aerial vehicle is represented, and the visual positioning of the unmanned aerial vehicle is realized. And a 3D digital map is constructed in the ground station system according to the characteristic point information, the movement track of the unmanned aerial vehicle and the distributed characteristic points are recorded, and meanwhile, the geographic environment information around the power grid transmission line tower can be intuitively known.
Drawings
FIG. 1 is a schematic structural diagram of a technical scheme of visual positioning adopted in the unmanned aerial vehicle power inspection of the invention;
FIG. 2 is a schematic diagram of feature points and monocular camera poses for visual positioning using the ORB algorithm according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
The utility model provides an adopt visual image to carry out unmanned aerial vehicle of location in electric power patrols and examines, has the cloud platform including the unmanned aerial vehicle base member that can realize flight function in unmanned aerial vehicle's bottom, is equipped with the camera lens on the cloud platform and can moves towards the monocular camera of bottom surface all the time.
And a high-resolution visible light camera/camera capable of assisting in imaging, three-dimensional laser radar scanning equipment, an infrared thermal imager, an ultraviolet imager and a multispectral imager are further arranged in the base body of the unmanned aerial vehicle.
Still be equipped with the wireless communication module that can carry out the communication with the ground satellite station in unmanned aerial vehicle's the base member.
An unmanned aerial vehicle positioning method for positioning by adopting visual images in power inspection comprises the following steps:
1) the unmanned aerial vehicle communicates with a ground station after taking off, and the ground station receives image information obtained by shooting of the unmanned aerial vehicle through a monocular camera without stopping;
2) the ground station processes the image frame returned by the unmanned aerial vehicle in real time, calculates R and t of the attitude change of the previous frame, and carries out real-time processing on the image frame
Storing the key frame;
3) eliminate the error, improve unmanned aerial vehicle positioning accuracy.
The change in pose between adjacent frames is described by the rotation matrix R and the translation vector t.
The ORB-SLAM2 program is divided into three threads, tracking, local mapping, and closed loop detection.
1) The tracking thread is responsible for extracting ORB feature points from the image frame and matching according to the last frame feature points stored in the system. Obtaining a plurality of pairs of feature point groups (x) after matchingi1,xi2) Using epipolar constraint equations
(where E is called the essential matrix, and E ═ tΛR) and at least 8 pairs of the above matched characteristic points construct an equation set. The essence matrix E obtained by adopting the eight-point method comprises the pose transformation information of the camera.
Then, the intrinsic matrix E is subjected to SVD decomposition (eigenvalue decomposition), i.e., E is decomposed into E ═ U Σ VTIn the form of (1). There are two groups of solutions for each of R and t:
since the positions of the feature points are necessarily in front of the monocular camera, three sets of solutions which are not in conformity can be eliminated.
2) Continuously inserting key frames into the local graph building process, recording the characteristic point information, and performing local nonlinear optimization on the monocular camera position and attitude information by adopting a BA (bundle adjustment) algorithm. And then screening the recorded key frames and eliminating redundant key frames.
3) Errors exist in the positions of the monocular cameras calculated by the first two threads and the point cloud positions of the feature points obtained by triangulation, and even if BA is adopted in the local mapping thread for local or global optimization, accumulated errors still exist. The closed-loop detection thread is used for eliminating accumulated errors and mainly comprises closed-loop detection and closed-loop correction.
And a closed loop detection link calculates bag of words (BOW) information of the key frame, and if similar bag of words description exists in the system, the monocular camera returns to a scene which is obtained before. The author of ORB-SLAM2 trained a large number of word bags based on ORB descriptions offline, requiring loading into the program.
When a closed loop is detected, the monocular SLAM computes a similarity transformation using the Sim3 algorithm.
The closed loop correction link first fuses the duplicate point clouds and inserts a new edge in the Covisibility Graph to connect the closed loops. The pose information of the current frame and the key frames connected with the current frame are corrected. And then optimizing a pose Graph through the Essential Graph, and dispersing errors into the whole Graph.
According to the attached drawing 1, the unmanned aerial vehicle adopts the visual positioning to firstly configure a monocular camera on an unmanned aerial vehicle cloud platform in the electric power inspection, the camera is different from a high-resolution camera used in the electric power facility inspection, the camera is not used for acquiring detection data, but is only used for shooting image data under the unmanned aerial vehicle, therefore, the monocular camera is required to be arranged on a movable cloud platform of the unmanned aerial vehicle, the camera is ensured to be always right on the ground, and the unmanned aerial vehicle is not influenced by the motion attitude of the unmanned aerial vehicle.
The real-time image that monocular camera acquireed is transmitted back to the ground station through wireless transmission module by unmanned aerial vehicle.
And after the ground station obtains the image returned by the unmanned aerial vehicle, the image is subjected to information extraction by using an ORB-SLAM2 program in the ground station system.
The monocular ORB-SLAM2 program requires an initialization process before positioning of the drone can be achieved. The purpose of initialization is to obtain the scale factor of the real physical world and the 3D digital world map by utilizing a plurality of frames of images after starting up, set the coordinates of the flying point of the unmanned aerial vehicle and record the coordinates in the 3D digital map data of the system.
The ground station utilizes an ORB-SLAM2 program to position the unmanned aerial vehicle in real time, records the flight track of the unmanned aerial vehicle, and presents the flight track on a ground station screen in a 3D digital map form. The recorded feature points are all points on an object below the unmanned aerial vehicle, so that the geographic environment condition below the unmanned aerial vehicle can be observed more intuitively by the feature point cloud pictures presented by the 3D digital map at the same time.
Unmanned aerial vehicle can acquire the 3D digital map information that the ground satellite station found through wireless transmission module to know the positional information of self, according to the orbit information of record, unmanned aerial vehicle can be full independently return voyage accurately.
Claims (1)
1. An unmanned aerial vehicle positioning method for positioning by adopting visual images in power inspection is characterized in that: the method comprises the following steps:
1) the unmanned aerial vehicle communicates with a ground station after taking off, and the ground station receives image information obtained by shooting of the unmanned aerial vehicle through a monocular camera without stopping;
2) the ground station processes the image frames transmitted back by the unmanned aerial vehicle in real time, calculates R and t of the attitude change of the previous frame, and stores key frames in the system;
3) errors are eliminated, and the positioning accuracy of the unmanned aerial vehicle is improved;
in the step 2), the change of the pose between the adjacent frames is described by a rotation matrix R and a translation vector t, and the method specifically comprises the following steps:
a) the ground station extracts ORB feature points from the image frame and performs matching according to the feature points of the last frame stored in the system;
b) obtaining a plurality of pairs of feature point groups (x) after matchingi1,xi2) Using epipolar constraint equations
Where E is called the essential matrix and E ═ tΛR, and at least 8 pairs of the matched characteristic points construct an equation set; the essence matrix E obtained by adopting the eight-point method comprises the pose transformation information of the camera;
then, the intrinsic matrix E is subjected to SVD (singular value decomposition), i.e. E is decomposed into E ═ U ∑ VTIn the form of (a); there are two groups of solutions for each of R and t:
because the positions of the characteristic points are necessarily in front of the monocular camera, three groups of solutions which are not in conformity can be eliminated;
c) continuously inserting key frames, recording feature point information, performing local nonlinear optimization on the monocular camera pose information by adopting a BA algorithm, screening the recorded key frames, and removing redundant key frames;
the step 3) is divided into a closed-loop detection link and a closed-loop correction link,
the closed loop detection link calculates bag-of-word information of the key frame, and if similar bag-of-word descriptions exist in the system, the monocular camera returns to a scene which is obtained before;
when a closed loop is detected, calculating similarity transformation by a monocular SLAM through a Sim3 algorithm;
in the closed-loop correction link, firstly, repeated point clouds are fused, and a new edge is inserted into the Coovirility Graph to connect a closed loop; the pose information of the current frame and the key frame connected with the current frame can be corrected, and then the pose Graph is optimized through the Essential Graph, and errors are dispersed into the whole Graph.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811380411.2A CN109739254B (en) | 2018-11-20 | 2018-11-20 | Unmanned aerial vehicle adopting visual image positioning in power inspection and positioning method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811380411.2A CN109739254B (en) | 2018-11-20 | 2018-11-20 | Unmanned aerial vehicle adopting visual image positioning in power inspection and positioning method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109739254A CN109739254A (en) | 2019-05-10 |
| CN109739254B true CN109739254B (en) | 2021-11-09 |
Family
ID=66355692
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811380411.2A Active CN109739254B (en) | 2018-11-20 | 2018-11-20 | Unmanned aerial vehicle adopting visual image positioning in power inspection and positioning method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109739254B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110135076A (en) * | 2019-05-17 | 2019-08-16 | 北京航空航天大学 | A multi-objective integrated optimization method for pan/tilt mechanical structure based on ISIGHT co-simulation |
| CN110702101A (en) * | 2019-08-29 | 2020-01-17 | 全球能源互联网研究院有限公司 | Positioning method and system for power inspection scene |
| CN110553734A (en) * | 2019-08-30 | 2019-12-10 | 国网福建省电力有限公司漳州供电公司 | A method for inspecting hidden dangers in power transmission channels by carrying multi-spectral equipment with drones |
| CN110631588B (en) * | 2019-09-23 | 2022-11-18 | 电子科技大学 | Unmanned aerial vehicle visual navigation positioning method based on RBF network |
| CN110609569B (en) * | 2019-09-26 | 2022-07-15 | 温岭市非普电气有限公司 | An autonomous control UAV precise inspection system and method applied to power towers |
| CN111272148B (en) * | 2020-01-20 | 2021-08-31 | 江苏方天电力技术有限公司 | Adaptive imaging quality optimization method for autonomous inspection of transmission lines by unmanned aerial vehicle |
| CN112102403B (en) * | 2020-08-11 | 2022-11-25 | 国网安徽省电力有限公司淮南供电公司 | High-precision positioning method and system for autonomous inspection unmanned aerial vehicle in power transmission tower scene |
| CN112381784B (en) * | 2020-11-12 | 2024-06-25 | 国网浙江省电力有限公司信息通信分公司 | A device detection system based on multispectral imaging |
| CN114020041B (en) * | 2021-12-14 | 2024-02-20 | 云南民族大学 | A multi-UAV multi-threaded two-dimensional exploration simulation method and system |
| CN117784120B (en) * | 2024-02-23 | 2024-05-28 | 南京新航线无人机科技有限公司 | Unmanned aerial vehicle flight state monitoring method and system |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2434256A2 (en) * | 2010-09-24 | 2012-03-28 | Honeywell International Inc. | Camera and inertial measurement unit integration with navigation data feedback for feature tracking |
| CN104298248A (en) * | 2014-10-08 | 2015-01-21 | 南京航空航天大学 | Accurate visual positioning and orienting method for rotor wing unmanned aerial vehicle |
| WO2017004799A1 (en) * | 2015-07-08 | 2017-01-12 | SZ DJI Technology Co., Ltd. | Camera configuration on movable objects |
| CN106529538A (en) * | 2016-11-24 | 2017-03-22 | 腾讯科技（深圳）有限公司 | Method and device for positioning aircraft |
| CN106803270A (en) * | 2017-01-13 | 2017-06-06 | 西北工业大学深圳研究院 | Unmanned aerial vehicle platform is based on many key frames collaboration ground target localization method of monocular SLAM |
| CN107193279A (en) * | 2017-05-09 | 2017-09-22 | 复旦大学 | Robot localization and map structuring system based on monocular vision and IMU information |
| CN107657640A (en) * | 2017-09-30 | 2018-02-02 | 南京大典科技有限公司 | Intelligent patrol inspection management method based on ORB SLAM |
| JP2018063512A (en) * | 2016-10-12 | 2018-04-19 | 本郷飛行機株式会社 | Mobile body attitude control system |
| CN108255189A (en) * | 2018-01-31 | 2018-07-06 | 佛山市神风航空科技有限公司 | A kind of power patrol unmanned machine system |
| CN108303099A (en) * | 2018-06-14 | 2018-07-20 | 江苏中科院智能科学技术应用研究院 | Autonomous navigation method in unmanned plane room based on 3D vision SLAM |
| CN108710824A (en) * | 2018-04-10 | 2018-10-26 | 国网浙江省电力有限公司信息通信分公司 | A kind of pedestrian recognition method divided based on regional area |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8340852B2 (en) * | 2009-04-29 | 2012-12-25 | Honeywell International Inc. | System and method for simultaneous localization and map building |
| KR101739996B1 (en) * | 2010-11-03 | 2017-05-25 | 삼성전자주식회사 | Moving robot and simultaneous localization and map-buliding method thereof |
| CN102707724B (en) * | 2012-06-05 | 2015-01-14 | 清华大学 | Visual localization and obstacle avoidance method and system for unmanned plane |
| CN102914294B (en) * | 2012-09-10 | 2015-07-01 | 中国南方电网有限责任公司超高压输电公司天生桥局 | System and method for measuring unmanned aerial vehicle electrical line patrol on basis of images |
| CN105388908A (en) * | 2015-12-11 | 2016-03-09 | 国网四川省电力公司电力应急中心 | Machine vision-based unmanned aerial vehicle positioned landing method and system |
| CN107357286A (en) * | 2016-05-09 | 2017-11-17 | 两只蚂蚁公司 | Vision positioning guider and its method |
| CN106054914A (en) * | 2016-08-17 | 2016-10-26 | 腾讯科技（深圳）有限公司 | Aircraft control method and aircraft control device |
| CN106168807B (en) * | 2016-09-09 | 2018-01-09 | 腾讯科技（深圳）有限公司 | The flight control method and flight control assemblies of a kind of aircraft |
| CN106527475A (en) * | 2016-10-28 | 2017-03-22 | 中国电力科学研究院 | Distribution network inspection unmanned aerial vehicle and inspection method thereof |
| CN107102647A (en) * | 2017-03-30 | 2017-08-29 | 中国人民解放军海军航空工程学院青岛校区 | Unmanned plane target tracking and controlling method based on image |
| US10593060B2 (en) * | 2017-04-14 | 2020-03-17 | TwoAntz, Inc. | Visual positioning and navigation device and method thereof |
| CN107117313B (en) * | 2017-05-24 | 2019-03-12 | 东南大学 | A kind of unmanned plane road detection system based on BIM |
| CN107479554B (en) * | 2017-09-07 | 2020-12-11 | 海南飞行者科技有限公司 | Robot system and outdoor map building navigation method thereof |
| CN107747941B (en) * | 2017-09-29 | 2020-05-15 | 歌尔股份有限公司 | Binocular vision positioning method, device and system |
-
2018
- 2018-11-20 CN CN201811380411.2A patent/CN109739254B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2434256A2 (en) * | 2010-09-24 | 2012-03-28 | Honeywell International Inc. | Camera and inertial measurement unit integration with navigation data feedback for feature tracking |
| CN104298248A (en) * | 2014-10-08 | 2015-01-21 | 南京航空航天大学 | Accurate visual positioning and orienting method for rotor wing unmanned aerial vehicle |
| WO2017004799A1 (en) * | 2015-07-08 | 2017-01-12 | SZ DJI Technology Co., Ltd. | Camera configuration on movable objects |
| JP2018063512A (en) * | 2016-10-12 | 2018-04-19 | 本郷飛行機株式会社 | Mobile body attitude control system |
| CN106529538A (en) * | 2016-11-24 | 2017-03-22 | 腾讯科技（深圳）有限公司 | Method and device for positioning aircraft |
| CN106803270A (en) * | 2017-01-13 | 2017-06-06 | 西北工业大学深圳研究院 | Unmanned aerial vehicle platform is based on many key frames collaboration ground target localization method of monocular SLAM |
| CN107193279A (en) * | 2017-05-09 | 2017-09-22 | 复旦大学 | Robot localization and map structuring system based on monocular vision and IMU information |
| CN107657640A (en) * | 2017-09-30 | 2018-02-02 | 南京大典科技有限公司 | Intelligent patrol inspection management method based on ORB SLAM |
| CN108255189A (en) * | 2018-01-31 | 2018-07-06 | 佛山市神风航空科技有限公司 | A kind of power patrol unmanned machine system |
| CN108710824A (en) * | 2018-04-10 | 2018-10-26 | 国网浙江省电力有限公司信息通信分公司 | A kind of pedestrian recognition method divided based on regional area |
| CN108303099A (en) * | 2018-06-14 | 2018-07-20 | 江苏中科院智能科学技术应用研究院 | Autonomous navigation method in unmanned plane room based on 3D vision SLAM |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109739254A (en) | 2019-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109739254B (en) | Unmanned aerial vehicle adopting visual image positioning in power inspection and positioning method thereof | |
| CN109034018B (en) | Low-altitude small unmanned aerial vehicle obstacle sensing method based on binocular vision | |
| CN107314762B (en) | Method for detecting ground object distance below power line based on monocular sequence images of unmanned aerial vehicle | |
| CN105197252B (en) | A kind of SUAV landing method and system | |
| CN105139002B (en) | Foreign matter identification system on power line | |
| CN109683629B (en) | Unmanned aerial vehicle electric power overhead line system based on combination navigation and computer vision | |
| CN103927731B (en) | Low-altitude remote sensing image rapid and automatic splicing method without POS assisting | |
| CN103822615B (en) | A kind of multi-control point extracts and the unmanned aerial vehicle target real-time location method be polymerized automatically | |
| CN110084785B (en) | A method and system for sag measurement of power lines based on aerial images | |
| CN107247458A (en) | UAV Video image object alignment system, localization method and cloud platform control method | |
| CN117437360A (en) | A fast and high-precision reconstruction method of power lines based on multi-source data fusion | |
| CN109792951B (en) | Unmanned aerial vehicle air route correction system for pollination of hybrid rice and correction method thereof | |
| CN106500669A (en) | A kind of Aerial Images antidote based on four rotor IMU parameters | |
| CN110850889B (en) | A UAV autonomous inspection system based on RTK navigation | |
| CN110706273B (en) | Real-time collapse area measurement method based on unmanned aerial vehicle | |
| KR20220163227A (en) | Drone used 3d mapping method | |
| CN111831010A (en) | A UAV Obstacle Avoidance Flight Method Based on Digital Space Slicing | |
| CN115950435A (en) | Real-time positioning method for unmanned aerial vehicle inspection image | |
| CN113920186B (en) | Low-altitude unmanned-machine multi-source fusion positioning method | |
| CN114779816B (en) | Searching and rescuing unmanned aerial vehicle for taking off and landing in ruins environment after earthquake and system thereof | |
| CN116297472B (en) | Unmanned aerial vehicle bridge crack detection method and system based on deep learning | |
| CN118882604A (en) | A geographic information surveying and collecting device | |
| CN114485549B (en) | Spatial ranging method and system for trees and overhead transmission line | |
| CN118642522A (en) | An adaptive power tower inspection method based on visual servoing | |
| CN116295340A (en) | A UAV Binocular Vision SLAM Method Based on Panoramic Camera |
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 |