CN106371453A - Visual servo method for use in the automatic line inspection by multi-rotor unmanned plane - Google Patents

Visual servo method for use in the automatic line inspection by multi-rotor unmanned plane Download PDF

Info

Publication number
CN106371453A
CN106371453A CN201610361571.7A CN201610361571A CN106371453A CN 106371453 A CN106371453 A CN 106371453A CN 201610361571 A CN201610361571 A CN 201610361571A CN 106371453 A CN106371453 A CN 106371453A
Authority
CN
China
Prior art keywords
straightway
angle
intersection point
line
threshold value
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.)
Pending
Application number
CN201610361571.7A
Other languages
Chinese (zh)
Inventor
白佶斌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Casting Intelligent Technology Co Ltd
Original Assignee
Shanghai Casting Intelligent Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Casting Intelligent Technology Co Ltd filed Critical Shanghai Casting Intelligent Technology Co Ltd
Priority to CN201610361571.7A priority Critical patent/CN106371453A/en
Publication of CN106371453A publication Critical patent/CN106371453A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/10Simultaneous control of position or course in three dimensions
    • G05D1/101Simultaneous control of position or course in three dimensions specially adapted for aircraft

Abstract

The line inspection by multi-rotor unmanned plane can greatly reduce cost, reduce the work intensity of people and resolve the safety problem posed by human line inspection. The navigating and positioning of a conventional multi-rotor unmanned plane mainly employs the combination of IMUs (inertia Measurement Unit) and satellites (GPS, Beidou, and etc.), which has some limitations. The precision of the prior non-differential positioning GPS navigation system is around 2.5 meters. And the inertial navigation system, due to its cumulative integration error, can only serve as a means of auxiliary navigation. As the transmission voltage across the transmission line is high, usually in the tens to hundreds of kilovolts. Such a high voltage level makes a very complex electromagnetic field around the transmission line, in which the existing navigation system, especially the geomagnetic sensor in the inertial navigation is likely to dysfunction due to interference, and ultimately air crash might occur. With the method of the invention, the probability of air crash by current unmanned aerial vehicles due to interference can be greatly reduced.

Description

A kind of visual servo method in the autonomous line walking of many rotor wing unmanned aerial vehicles
Technical field
The invention belongs to the technical field of UAV Flight Control and machine vision mixing together is and in particular to a kind of be based on straightway Many rotor wing unmanned aerial vehicles Visual servoing control method of intersection point azimuth determination.
Background technology
In recent years, many rotor wing unmanned aerial vehicles have obtained quick development, have flight stability, easily manipulation, low noise, safety hazard The features such as property is little, obtains increasingly wider application in multiple fields such as aeroplane photography, electric inspection process, the condition of a disaster investigation, military surveillances.
Using many rotor wing unmanned aerial vehicles line walking, can be good at reduces cost, the amount of lightening one's labor and solve what artificial line walking was brought Safety problem.The navigator fix of traditional many rotor wing unmanned aerial vehicles mainly adopts imu (inertia assembly) and satellite (such as gps, Big Dipper etc.) Compound mode, has certain limitation.The gps navigation system precision of front non-differential positioning about at 2.5 meters, inertial navigation System, due to there being accumulative integral error, is only used as the means of assisting navigation.Because the transmission voltage of transmission line of electricity is higher, typically All arrive hundreds of kilovolt tens, so high electric pressure causes the electromagnetic field environment of transmission line of electricity periphery more complicated, existing Geomagnetic sensor in navigation system, especially inertial navigation is easily interfered and failure, ultimately results in the generation of air crash accident
In order to improve the intelligent and reliability of the autonomous line walking of many rotor wing unmanned aerial vehicles, reduce the workload of manual intervention it is necessary to increase The ability of the autonomous Visual servoing control of many rotor wing unmanned aerial vehicles by force.
Content of the invention
It is an object of the invention to making up the deficiency of the autonomous line walking technology of existing many rotor wing unmanned aerial vehicles, a kind of straightway that is based on is provided to hand over The Visual servoing control method of point azimuth determination, the method is without manual intervention, and need not carry out edge during detection electric wire Detection, strong robustness and being easily achieved, computational efficiency is better than existing algorithm.
For reaching above-mentioned purpose, the present invention employs the following technical solutions and is achieved:
Visual servo method in a kind of autonomous line walking of many rotor wing unmanned aerial vehicles, comprises the following steps:
1) image acquisition: using the monocular camera being mounted in below many rotor wing unmanned aerial vehicles, the region comprising electric wire is acquired, obtains To image i.Wherein camera intrinsic parameter is demarcated, focal length f, x and y direction single pixel physical size dxAnd dyIt is known;
2) line segment detection: using the line Segment Detection growing based on gradient, line segment detection is carried out to image i, and will grow Degree is ignored less than the straightway of threshold value, and the straightway detecting is expressed as { l1, l2..., ln};
3) Straight-line segments mergence: end-point distances and angle are asked for any two in the straightway set detecting, to eligible Straightway merge, obtain straightway set { l1, l2..., lm};
4) straightway screening: to the straightway set { l after merging treatment1, l2..., lmIn each bar straight line judged, sieve Select qualified two straightway liAnd lj
5) straightway intersection point solves and azimuthal angle calculation: asks for straightway liAnd ljIntersection point p (x, y), and according to (x, y) Coordinate asks for azimuth angle alpha and the height angle beta that this intersection point is relative to unmanned plane;
6) visual servo scheme: azimuth angle alpha and height angle beta are input in intelligent controller, output control variable, control Rotor wing rotation direction and speed, so that unmanned plane is along electric wire autonomous flight, execute line walking task.
The present invention is further improved by:
Described step 3) in, the concrete grammar of Straight-line segments mergence is as follows:
To { l1, l2..., lnIn any two straightway { li, lj, two end points are respectively (xi1, yi1) and (xi2, yi2)、 (xj1, yj1) and (xj2, yj2), it is respectively θ with the angle of x-axis1And θ2;Calculate (xi1, yi1) and (xj1, yj1)、(xj2, yj2) Between Euclidean distance d11And d12, (xi2, yi2) and (xj1, yj1)、(xj2, yj2) between Euclidean distance be d21And d22;dmin For d11、d12、d21And d22In minima, obtain l simultaneouslyiAnd ljBetween angle theta12=abs (θ12);Calculate straight Line liTwo end points (xi1, yi1) and (xi2, yi2) arrive straight line ljTwo vertical line sections, note intersection point be respectively t1, t2, vertical line segment length Degree is respectively d11, d12;Calculate straight line ljTwo end points (xj1, yj1) and (xj2, yj2) arrive straight line liTwo vertical line sections, remember intersection point It is respectively t3, t4, vertical line segment length is respectively d21, d22.Defining criterion one is: dminLess than 10 pixels;Define criterion two For: θ12Less than 5 degree;Defining criterion three is: b1||b2||b3||b4, wherein b1=t1∈lj&&d11< threshold value tp, b2=t2 ∈lj&&d12< threshold value tp, b3=t3∈lj&&d21< threshold value tp, b4=t4∈li&&d22< threshold value tp.
For any two straightway liAnd lj, when there being a satisfaction in criterion one, two, three.Then merging two straightways is one Bar.Straightway set { l is obtained after all straightways are all processed1, l2..., lm}.
Described step 4) in, straightway screening method particularly includes:
Calculate straightway { l1, l2..., lmLength and slope, choose the maximum straightway of length, and obtain and its length difference Value is less than other straightways of threshold value with angle.Choose two of angle maximum as l in the straightway solvingiAnd lj.
Described step 5) in, straightway intersection point solves and azimuthal angle calculation method particularly includes:
liAnd ljTwo end points be respectively (ix1, iy1)、(ix2, iy2) and (jx1, jy1)、(jx2, jy2), liAnd ljIntersection point be p (x, y).By (ix1, iy1)、(ix2, iy2), (jx1, jy1)、(jx2, jy2) (ij can be obtained according to the following formulax, ijy):
X=
(ix1*jx1*iy2-ix2*jx1*iy1-ix1*jx2*iy2+ix2*jx2*iy1-ix1*jx1*jy2+ix1*jx2*jy1+ix2*jx1*jy2-ix2*jx2*jy1)/(ix1*jy1-jx1*iy1-ix1*jy2-ix2*jy1+jx1*iy2+jx2*iy1+ix2*jy2-jx2*iy2)
Y=
(ix1*iy2*jy1-ix2*iy1*jy1-ix1*iy2*jy2+ix2*iy1*jy2-jx1*iy1*jy2+jx2*jy1*jy1+jx1*iy2*jy2-jx2*iy2*jy1)/(ix1*jy1-jx1*iy1-ix1*jy2-ix2*jy1+jx1*iy2+jx2*iy1+ix2*jy2-jx2*iy2)
Then azimuth angle alpha and height angle beta are obtained by p (x, y).
Described step 6) in, visual servo scheme method particularly includes:
The controling power of azimuth and the angle of site is fα、fβ.Controller is controlled using the control of the pd with dead band, controling power meter Calculation such as following formula:
Wherein, α0、β0It is azimuth controlling dead error respectively.It is controlled by azimuthal, the angle of site, can reach inclined to position The control of difference.
Compared with prior art, the method have the advantages that
The present invention using the line Segment Detection based on gradient growth, thus rim detection need not be carried out to image, it is to avoid binaryzation The impact that threshold value is chosen.Method using Straight-line segments mergence, it is to avoid the interference of short and small straightway in small, broken bits, reduces calculating simultaneously Amount.Using the property that the straightway of less parallel line is intersecting at a distance, ask for the azimuth of intersection point, the angle of site as control deviation, Realize the line walking function under the autonomous characteristic condition of circuit.Finally, the present invention can be used for many rotor wing unmanned aerial vehicles to power circuit from In main tour task.
Brief description
Fig. 1 is the application scenarios schematic diagram of the present invention;
Fig. 2 is the breakpoint detection schematic diagram to power circuit for the present invention.
Fig. 3 is the algorithm overall block flow diagram of the present invention;
Specific embodiment
Below in conjunction with the accompanying drawings the present invention is described in further detail:
Fig. 1 is typical application scenarios of the present invention, and many rotor wing unmanned aerial vehicles are autonomous by a point along electric lines of force 1,2,3 rectilinear direction Move to b point.Fig. 2 illustrates the present invention and to the detection of electric lines of force 1,2,3 and solves intersection point schematic diagram.
Referring to Fig. 3, the present invention comprises the following steps:
1st, image acquisition
I is obtained after target image is acquired.Picture size width is m, highly for n.
2nd, line segment detection
Using the line Segment Detection growing based on gradient, image i is carried out with line segment detection, and length is less than the straight line of 0.5 × n Section is ignored, and the straightway detecting is expressed as { l1, l2..., ln}.
Wherein with regard to above-mentioned steps how concrete operations, it is known in the art technology, refers to document 1 " Qin Xunhui, Ma Rong, Fu Wei Flat, etc. a kind of line segment detection algorithm [j] based on gradient. photon journal, 2012,41 (2), 205-209. ", herein no longer Repeat.
3rd, Straight-line segments mergence
To { the l obtaining in previous step1, l2..., lnIn any two straightway { li, ljIt is known that two end points are respectively (xi1, yi1) and (xi2, yi2)、(xj1, yj1) and (xj2, yj2), it is respectively θ with the angle of x-axis1And θ2;Calculate (xi1, yi1) and (xj1, yj1)、(xj2, yj2) between Euclidean distance d11And d12, (xi2, yi2) and (xj1, yj1)、(xj2, yj2) between Euclidean distance be d21And d22;dminFor d11、d12、d21And d22In minima, obtain l simultaneouslyiAnd ljBetween angle theta12=abs (θ12); Calculate straight line liTwo end points (xi1, yi1) and (xi2, yi2) arrive straight line ljTwo vertical line sections, note intersection point be respectively t1, t2, hang down Line segment length is respectively d11, d12;Calculate straight line ljTwo end points (xj1, yj1) and (xj2, yj2) arrive straight line liTwo vertical line sections, Note intersection point is respectively t3, t4, vertical line segment length is respectively d21, d22.Defining criterion one is: dminLess than 10 pixels;Definition Criterion two is: θ12Less than 5 degree;Defining criterion three is: b1||b2||b3||b4, wherein b1=t1∈lj&&d11< threshold value tp, b2=t2∈lj&&d12< threshold value tp, b3=t3∈li&&d21< threshold value tp, b4=t4∈li&&d22< threshold value tp.
For any two straightway liAnd lj, when there being a satisfaction in criterion one, two, three.Then merge two straightways and be one. Straightway set { l is obtained after all straightways are all processed1, l2..., lm}.
4th, straightway screening:
Calculate straightway { l1, l2..., lmLength and slope, choose the maximum straightway of length, and obtain with its length difference and Angle is less than other straightways of threshold value.Choose two of angle maximum as l in the straightway solvingiAnd lj.
5th, straightway intersection point solves and azimuthal angle calculation
liAnd ljTwo end points be respectively (ix1, iy1)、(ix2, iy2) and (jx1, jy1)、(jx2, jy2), liAnd ljIntersection point is p (x, y). By (ix1, iy1)、(ix2, iy2), (jx1, jy1)、(jx2, jy2) (ij can be obtained according to the following formulax, ijy):
X=
(ix1*jx1*iy2-ix2*jx1*iy1-ix1*jx2*iy2+ix2*jx2*iy1-ix1*jx1*jy2+ix1*jx2*jy1+ix2*jx1*jy2-ix2*jx2*jy1)/(ix1*jy1 -jx1*iy1-ix1*jy2-ix2*jy1+jx1*iy2+jx2*iy1+ix2*jy2-jx2*iy2)
Y=
(ix1*iy2*jy1-ix2*iy1*jy1-ix1*iy2*jy2+ix2*iy1*jy2-jx1*iy1*jy2+jx2*jy1*jy1+jx1*iy2*jy2-jx2*iy2*jy1)/(ix1*jy1 -jx1*iy1-ix1*jy2-ix2*jy1+jx1*iy2+jx2*iy1+ix2*jy2-jx2*iy2)
Then azimuth angle alpha and height angle beta are obtained by p (x, y), acquiring method is known in the art technology, referred to document 2 " once Moral is virtuous, Dong Xurong, Li Rui. spacecraft relative angle measuring method [c]. Chinese intelligent automation meeting, and Lanzhou, 2007 August 24-27 day in year. "
6th, visual servo scheme
Controller is controlled using the pd with dead band.The controling power of azimuth and the angle of site is fα、fβ.Controling power is calculated as follows formula:
Wherein, α0、β0It is azimuth controlling dead error respectively.It is controlled by azimuthal, the angle of site, can reach inclined to position The control of difference.
Above content technological thought only to illustrate the invention, it is impossible to limit protection scope of the present invention with this, every carries according to the present invention The technological thought going out, any change done on the basis of technical scheme, each fall within claims of the present invention protection domain it Interior.

Claims (5)

1. the visual servo method in a kind of autonomous line walking of many rotor wing unmanned aerial vehicles is it is characterised in that comprise the following steps:
1) image acquisition: using the monocular camera being mounted in below many rotor wing unmanned aerial vehicles, the region comprising electric wire is acquired, obtains To image i, wherein camera intrinsic parameter is demarcated, focal length f, x and y direction single pixel physical size dxAnd dyIt is known;
2) line segment detection: using the line Segment Detection growing based on gradient, line segment detection is carried out to image i, and will grow Degree is ignored less than the straightway of threshold value, and the straightway detecting is expressed as { l1, l2..., ln};
3) Straight-line segments mergence: end-point distances and angle are asked for any two in the straightway set detecting, to eligible Straightway merge, obtain straightway set { l1, l2..., lm};
4) straightway screening: to the straightway set { l after merging treatment1, l2..., lmIn each bar straight line judged, filter out Qualified two straightway liAnd lj
5) straightway intersection point solves and azimuthal angle calculation: asks for straightway liAnd ljIntersection point p (x, y), and asked according to (x, y) coordinate Take azimuth angle alpha and the height angle beta that this intersection point is relative to unmanned plane;
6) visual servo scheme: azimuth angle alpha and height angle beta are input in intelligent controller, output control variable, control rotation Wing direction of rotation and speed are so that unmanned plane is along electric wire autonomous flight.
2. according to claim 1 based on merge lbp and gray feature description Algorism of Matching Line Segments method it is characterised in that Described step 3) in, the concrete grammar of Straight-line segments mergence is as follows:
To { l1, l2..., lnIn any two straightway { li, lj, two end points are respectively (xi1, yi1) and (xi2, yi2)、(xj1, yj1) (xj2, yj2), it is respectively θ with the angle of x-axis1And θ2;Calculate (xi1, yi1) and (xj1, yj1)、(xj2, yj2) between Euclidean distance d11 And d12, (xi2, yi2) and (xj1, yj1)、(xj2, yj2) between Euclidean distance be d21And d22;dminFor d11、d12、d21And d22 In minima, obtain l simultaneouslyiAnd ljBetween angle theta12=abs (θ12);Calculate straight line liTwo end points (xi1, yi1) and (xi2, yi2) arrive straight line ljTwo vertical line sections, note intersection point be respectively t1, t2, vertical line segment length is respectively d11, d12;Calculate straight line lj Two end points (xj1, yj1) and (xj2, yj2) arrive straight line liTwo vertical line sections, note intersection point be respectively t3, t4, vertical line segment length is respectively d21, d22.Defining criterion one is: dminLess than 10 pixels;Defining criterion two is: θ12Less than 5 degree;Defining criterion three is: b1||b2||b3||b4, wherein b1=t1∈lj&&d11< threshold value tp, b2=t2∈lj&&d12< threshold value tp, b3=t3∈li&&d21< Threshold value tp, b4=t4∈li&&d22< threshold value tp;For any two straightway liAnd lj, when having one in criterion one, two, three Individual satisfaction, then merge two straightways and be one, obtain straightway set { l after all straightways are all processed1, l2..., lm}.
3. according to claim 1 based on merge lbp and gray feature description Algorism of Matching Line Segments method it is characterised in that Described step 4) in, straightway screening method particularly includes:
Calculate straightway { l1, l2..., lmLength and slope, choose the maximum straightway of length, and obtain with its length difference and Angle is less than other straightways of threshold value, chooses two of angle maximum as l in the straightway solvingiAnd lj.
4. according to claim 1 based on merge lbp and gray feature description Algorism of Matching Line Segments method it is characterised in that Described step 5) in, straightway intersection point solves and azimuthal angle calculation method particularly includes:
liAnd ljTwo end points be respectively (ix1, iy1)、(ix2, iy2) and (jx1, jy1)、(jx2, jy2), liAnd ljIntersection point is p (x, y), by (ix1, iy1)、(ix2, iy2), (jx1, jy1)、(jx2, jy2) (ij can be obtained according to the following formulax, ijy):
X=
(ix1*jx1*iy2-ix2*jx1*iy1-ix1*jx2*iy2+ix2*jx2*iy1-ix1*jx1*jy2+ix1*jx2*jy1+ix2*jx1*jy2-ix2*jx2*jy1)/(ix1*jy1-jx1 *iy1-ix1*jy2-ix2*jy1+jx1*iy2+jx2*iy1+ix2*jy2-jx2*iy2)
Y=
(ix1*iy2*jy1-ix2*iy1*jy1-ix1*iy2*jy2+ix2*iy1*jy2-jx1*iy1*jy2+jx2*iy1*jy1+jx1*iy2*jy2-jx2*iy2*jy1)/(ix1*jy1-jx1 *iy1-ix1*jy2-ix2*jy1+jx1*iy2+jx2*iy1+ix2*jy2-jx2*iy2)
Then azimuth angle alpha and height angle beta are obtained by p (x, y).
5. according to claim 1 based on merge lbp and gray feature description Algorism of Matching Line Segments method it is characterised in that Described step 6) in, visual servo scheme method particularly includes:
The controling power of azimuth and the angle of site is fα、fβ.Controller is controlled using the control of the pd with dead band, controling power meter Calculation such as following formula:
Wherein, α0、β0It is azimuth controlling dead error respectively, is controlled by azimuthal, the angle of site, can reach inclined to position The control of difference.
CN201610361571.7A 2016-05-27 2016-05-27 Visual servo method for use in the automatic line inspection by multi-rotor unmanned plane Pending CN106371453A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610361571.7A CN106371453A (en) 2016-05-27 2016-05-27 Visual servo method for use in the automatic line inspection by multi-rotor unmanned plane

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610361571.7A CN106371453A (en) 2016-05-27 2016-05-27 Visual servo method for use in the automatic line inspection by multi-rotor unmanned plane

Publications (1)

Publication Number Publication Date
CN106371453A true CN106371453A (en) 2017-02-01

Family

ID=57881010

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610361571.7A Pending CN106371453A (en) 2016-05-27 2016-05-27 Visual servo method for use in the automatic line inspection by multi-rotor unmanned plane

Country Status (1)

Country Link
CN (1) CN106371453A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106934832A (en) * 2017-03-23 2017-07-07 电子科技大学 A kind of simple straight line automatic positioning method towards vision line walking
CN107358570A (en) * 2017-06-30 2017-11-17 深圳市强华科技发展有限公司 A kind of IMAQ control method and its system
CN110298845A (en) * 2019-06-17 2019-10-01 中国计量大学 It transmits electricity under a kind of complex background based on image procossing line detecting method

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106934832A (en) * 2017-03-23 2017-07-07 电子科技大学 A kind of simple straight line automatic positioning method towards vision line walking
CN106934832B (en) * 2017-03-23 2019-07-09 电子科技大学 A kind of simple straight line automatic positioning method towards vision line walking
CN107358570A (en) * 2017-06-30 2017-11-17 深圳市强华科技发展有限公司 A kind of IMAQ control method and its system
CN110298845A (en) * 2019-06-17 2019-10-01 中国计量大学 It transmits electricity under a kind of complex background based on image procossing line detecting method

Similar Documents

Publication Publication Date Title
US11573544B2 (en) Performing 3D reconstruction via an unmanned aerial vehicle
Stokkeland et al. Autonomous visual navigation of unmanned aerial vehicle for wind turbine inspection
CN105644785B (en) A kind of UAV Landing method detected based on optical flow method and horizon
CN106708073B (en) A kind of quadrotor system of independent navigation power-line patrolling fault detection
CN105352495A (en) Unmanned-plane horizontal-speed control method based on fusion of data of acceleration sensor and optical-flow sensor
Xu et al. Use of land’s cooperative object to estimate UAV’s pose for autonomous landing
CN106371453A (en) Visual servo method for use in the automatic line inspection by multi-rotor unmanned plane
CN106774374B (en) Automatic unmanned aerial vehicle inspection method and system
CN104360688A (en) Guide device of line-cruising unmanned aerial vehicle and control method of guide device
CN109747824A (en) Device and barrier-avoiding method for unmanned plane avoidance inside chimney
Bian et al. A point-line-based SLAM framework for UAV close proximity transmission tower inspection
CN103278159B (en) Airborne 2D range finder using laser obtains the method for 3D point cloud
CN114419109A (en) Aircraft positioning method based on visual and barometric information fusion
CN103453875A (en) Real-time calculating method for pitch angle and roll angle of unmanned aerial vehicle
CN102175227A (en) Quick positioning method for probe car in satellite image
CN116429098A (en) Visual navigation positioning method and system for low-speed unmanned aerial vehicle
CN107883917B (en) High-voltage line tower height extraction method based on multi-temporal shadow difference
Qiu et al. Design and implementation of an autonomous landing control system of unmanned aerial vehicle for power line inspection
RU2466355C1 (en) Method of obtaining navigation information for automatic landing of unmanned aerial vehicle
Tehrani et al. Low-altitude horizon-based aircraft attitude estimation using UV-filtered panoramic images and optic flow
CN111610800B (en) Loosely-coupled unmanned aerial vehicle control system
CN108873924B (en) Airborne video ground sweep area calculation method
CN113433958A (en) Unmanned aerial vehicle inspection method and device
CN112797982A (en) Unmanned aerial vehicle autonomous landing measurement method based on machine vision
Xiong et al. Scale-aware monocular visual-inertial pose estimation for aerial robots

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination