CN109857133A - Multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision - Google Patents
Multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision Download PDFInfo
- Publication number
- CN109857133A CN109857133A CN201910232714.8A CN201910232714A CN109857133A CN 109857133 A CN109857133 A CN 109857133A CN 201910232714 A CN201910232714 A CN 201910232714A CN 109857133 A CN109857133 A CN 109857133A
- Authority
- CN
- China
- Prior art keywords
- unmanned plane
- barrier
- image
- pixel
- avoidance
- 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.)
- Withdrawn
Links
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision that the invention discloses a kind of, including binocular range finder module, UAV Attitude identification module, flight control modules and transmission module.Binocular camera is vertical in front of unmanned plane, two camera left and right settings of binocular camera and is parallel to each other, for shooting obstructions chart picture.Go out the true bearing of barrier and the actual distance of barrier and unmanned plane according to the pitch angle comprehensive analysis that the image of barrier and UAV Attitude identification module obtain;According to the true bearing of barrier, the actual distance and pitch angle of barrier and unmanned plane construct unmanned plane in the rectangle frame size of pixel coordinate system;Unmanned plane plans flight path according to whether there are obstacles in the rectangle frame;User is allowed to manipulate remote control control unmanned plane simultaneously.The present invention reduces unmanned planes to tilt the influence to binocular ranging, so that unmanned plane avoidance is more acurrate.
Description
Technical field
The present invention relates to unmanned plane avoidance technical field more particularly to a kind of multi-rotor unmanned aerial vehicle choosings based on binocular vision
Selecting property avoidance obstacle method.
Background technique
With the development of unmanned plane the relevant technologies and its apply complex scene, to surrounding perception environment ability have more
High requirement, wherein the power of barrier avoiding function plays conclusive influence to unmanned plane Self-adaptive flight.
Currently, unmanned plane obstacle avoidance system mostly uses greatly ultrasonic wave, laser, radar and binocular vision etc..Ultrasonic distance measurement
It is shorter and more sensitive to external condition;Laser is more sensitive to external conditions such as illumination;Radar is less sensitive to external condition,
But technology is not overripened, there are problems that erroneous judgement;And binocular ranging has measurement range wide, contains much information, the advantages such as at low cost, especially
It is to capture ambient enviroment rapidly.Therefore this patent chooses binocular vision and carries out avoidance.Specifically, binocular camera acquires
The first image and the second image of barrier, can directly restore barrier according to disparity map and the Stereo Vision of generation
Three dimensional space coordinate makes unmanned plane avoiding barrier to obtain the depth value of barrier.
When multi-rotor unmanned aerial vehicle flight forward, unmanned plane is heeling condition, and distance measured by binocular vision is not
Horizontal actual range between unmanned plane and ground or barrier, it may appear that unnecessary avoidance or hovering, so that unmanned plane is not
Can according to it is anticipated that rule carry out avoidance, will be greatly promoted a possibility that so that unmanned plane is collided, to reduce double
The practicability and reliability of mesh obstacle avoidance system.
Summary of the invention
In view of the drawbacks described above of the prior art, the present invention provides a kind of multi-rotor unmanned aerial vehicle selectivity based on binocular vision
Avoidance obstacle method overcomes unmanned plane itself to tilt bring error, solves the problems, such as the avoidance under unmanned plane complex environment.
The invention patent solves technical solution used by its technical problem:
Multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision, which is characterized in that the obstacle avoidance system
Including binocular range finder module, UAV Attitude identification module, flight control modules and transmission module.
The binocular range finder module includes the binocular camera with USB interface, a tree plum group.Binocular camera hangs down
It is directly set in front of unmanned plane, two camera left and right settings of binocular camera and is parallel to each other;Setting plum group is binocular camera
The processing system for acquiring data connects binocular camera by USB interface, for handling disparity map, and passes through pitch angle
Compensation rebuilds three-dimensional, calculates depth, while flying control by I/O port connection, controls the pitch angle of unmanned plane, roll angle and inclined
Navigate the flight attitudes such as angle, to realize selective avoidance.
The gesture recognition module includes accelerometer, gyroscope, magnetometer, barometer, for obtaining UAV Attitude
Angle and height.
The flight control modules are mainly embedded scm and earth station, for controlling the flight attitude of unmanned plane.
The transmission module is mainly digital transmission module, figure transmission module, remote controler.Digital transmission module is used for transmission with figure transmission module
Data and video information, remote controler is for artificially controlling unmanned plane.
Further, the multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method and step based on binocular vision is as follows:
1. obtaining the ambient image in front of unmanned plane by binocular camera;The ambient image is the two of binocular camera
A camera first image and second image captured in synchronization difference;
2. identifying in front of unmanned plane whether there is barrier according to the ambient image of acquisition;
3. barrier if it exists, then based on unmanned plane current pose information and the barrier depth obtained according to image
Value;The UAV Attitude information is unmanned plane pitch angle acquired in the attitude transducer set on unmanned plane, attitude transducer
For accelerometer, gyroscope and magnetometer;The depth value of the barrier is obtained by following steps:
It includes the first image and the second image that 3.1 binocular cameras, which obtain the ambient image in front of unmanned plane,;
Two cameras of 3.2 binocular cameras obtain the inside and outside parameter of camera, then lead to by Zhang Shi standardization
The external parameter and distortion factor for crossing camera correct the first image and the second image;By after correction the first image and
Second image carries out Stereo matching using SGAM algorithm, obtains the disparity map of two images;
3.3 obtain depth image based on the disparity map, extract barrier depth value according to the depth image;
3.4 setting avoidance threshold value lmax, identification is less than lmaxBarrier and reservation, remove lmaxBarrier.
4. if image, which exists, is less than lmaxBarrier, tree plum group carries out judging whether avoidance, and otherwise unmanned plane is kept original
Route continues to fly;
4.1 are existed based on image less than lmaxBarrier, calculate unmanned plane and nearest barrier horizontal distance d:
If v0When-v ≠ 0,
In formula, z is the vertical pivot coordinate of nearest barrier, and y is the ordinate of orthogonal axes of nearest barrier, and θ is pitch angle, v0For figure
The pixel ordinate of inconocenter, v are the pixel ordinate of nearest barrier.
If v0When-v=0,
D=zcos | θ | (2)
4.2 calculate the practical rectangle width of frame w of unmanned planefactWith length Ifact,
wfact=h cos | θ |+bsin | θ | (3)
In formula, h is unmanned plane height, and b is unmanned plane width, and θ is pitch angle, and θ ∈ (- 15 °, 15 °) is positive downwards, to
On be negative;Ifact=l, wherein l is unmanned plane length.
4.3 calculating pixel flight frames length Δ u and width Delta v: the parallax of image center when depth value d be
D=Bf/d (4)
In formula, B is the distance between two cameras, and f is the focal length of binocular camera;
Δ x=lD/B (5)
In formula, Δ x is the correspondingly-sized under image coordinate system of the l in depth value d.
In formula, Δ u1For the corresponding size under pixel coordinate system of the l in depth value d, dx is unit pixel in horizontal axis
Physical size;
In formula, Δ v1For wfactThe corresponding size under pixel coordinate system in depth value d, dy are unit pixel vertical
The physical size of axis;
By Δ u1With Δ v1Original 1.2 times are all expanded as, can reduce extraneous factor influences and handles in time emergency case,
Then have
Δ u=1.2 Δ u1, Δ v=1.2 Δ v1 (8)
4.4, when unmanned plane tilts, obtain the seat of pixel flight safety frame.If pixel flight safety frame center and pixel
The separation delta c of the flight frame line of centres:
Δ e=d tan | θ | (9)
In formula, Δ e is unmanned plane actual frames center offset;
Δ y_c=D Δ e/B (10)
In formula, Δ y_c is the offset of rectangle frame central point of the unmanned plane under image coordinate system;
Δ c=Δ y_c/dy=ftan | θ |/dy (11)
4.5 under pixel coordinate system, if θ >=0 °, pixel flight frame is obtained pixel flight safety to surface translation Δ c
Frame;If 0 ° of θ <, pixel flight frame is obtained pixel flight safety frame to underface translation Δ c;
4.6 traversal pixel flight safety frames: if barrier (1) is not present in pixel flight safety frame, unmanned plane is to pixel
The flight of flight safety frame direction;(2) if there are barrier in pixel flight safety frame, unmanned plane carries out corresponding selective avoidance,
Specific steps are as follows:
4.6.1 unmanned plane, which reduces speed now, goes slowly, and records the depth value deep of the nearest barrier of pixel flight safety frame, into
Row selects corresponding avoidance mode 1:
Unmanned plane flies upwards, if barrier is not present in pixel flight safety frame in upward flight course, nobody
Machine is by the airline operation originally planned;If there are barrier in pixel flight safety frame after upward 5 meters of flight, unmanned plane is vertical
Slow down and hover, and information is sent by digital transmission module and figure transmission module, prompts artificial be remotely controlled;
4.6.2 compared with the threshold value set_d of setting, unmanned plane carries out selecting corresponding avoidance mode 2 deep:
1. deep enters jogging area when being greater than set_d, unmanned plane adjustment pitch angle suitably reduces speed;
2. deep enters alert zone when being less than or equal to set_d, unmanned plane slows down hovering immediately, and by digital transmission module with
Figure transmission module sends information, prompts artificial remote control.
The invention adopts the above technical scheme compared with prior art, has following technical effect that
(1) the binocular range finder module small volume and less weight based on tree plum group, installation is simple, while reducing embeded processor
Operating quantity, convenient for UAV Attitude control adjustment in time;
(2) interference to binocular ranging, and the model that will be detected in rectangle frame are tilted by angle modification unmanned plane
Enclose and be divided into three parts according to distance and unmanned plane speed and be respectively processed, make the avoidance route planning of unmanned plane more rationally,
Accurately.
Detailed description of the invention
Fig. 1 is the structure chart of more rotation unmanned plane selectivity avoidance obstacle methods based on binocular vision;
Fig. 2 is the flow chart of the multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision;
Fig. 3 is the horizontal distance of unmanned plane Yu nearest barrier;
Fig. 4 is the practical rectangle frame size of unmanned plane;
Fig. 5 is that the practical rectangle of unmanned plane is converted into pixel flight frame;
Fig. 6 is that pixel flight frame is converted into pixel flight safety frame;
Fig. 7 is the schematic diagram between unmanned plane and barrier apart from processing mode.
Specific embodiment
In order to keep the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing 1-7, to the present invention into
One step detailed description.
Referring to Fig.1, the multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method provided by the present invention based on binocular vision,
It is characterized in that, the obstacle avoidance system includes binocular range finder module, UAV Attitude identification module, flight control modules and transmission
Module.
The binocular range finder module includes the binocular camera with USB interface, a tree plum group.Binocular camera hangs down
It is directly set in front of unmanned plane, two camera left and right settings of binocular camera and is parallel to each other;Setting plum group is binocular camera
The processing system for acquiring data connects binocular camera by USB interface, carries out weight for handling disparity map, and to three-dimensional
New building calculates depth, while being connected by I/O port and flying control, controls the flight such as pitch angle, roll angle and yaw angle of unmanned plane
Posture, to realize selective avoidance.
The gesture recognition module includes accelerometer, gyroscope, magnetometer, barometer, for obtaining UAV Attitude
Angle and height.
The flight control modules are mainly embedded scm and earth station, for controlling the flight attitude of unmanned plane.
The transmission module is mainly digital transmission module, figure transmission module, remote controler.Digital transmission module is used for transmission with figure transmission module
Data and video information, remote controler is for artificially controlling unmanned plane.
As shown in Fig. 2, the multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method and step based on binocular vision is as follows:
1. obtaining the ambient image in front of unmanned plane by binocular camera;The ambient image is the two of binocular camera
A camera first image and second image captured in synchronization difference;
2. identifying in front of unmanned plane whether there is barrier according to the ambient image of acquisition;
3. barrier if it exists, then based on unmanned plane current pose information and the barrier depth obtained according to image
Value;The UAV Attitude information is unmanned plane pitch angle acquired in the attitude transducer set on unmanned plane, attitude transducer
For accelerometer, gyroscope and magnetometer;The depth value of the barrier is obtained by following steps:
It includes the first image and the second image that 3.1 binocular cameras, which obtain the ambient image in front of unmanned plane,;
Two cameras of 3.2 binocular cameras obtain the inside and outside parameter of camera, then lead to by Zhang Shi standardization
The external parameter and distortion factor for crossing camera correct the first image and the second image;By after correction the first image and
Second image carries out Stereo matching using SGAM algorithm, obtains the disparity map of two images;
3.3 obtain depth image based on the disparity map, extract barrier depth value according to the depth image;
3.4 setting avoidance threshold value lmax, identification is less than lmaxBarrier and reservation, remove lmaxBarrier.
4. if image, which exists, is less than lmaxBarrier, tree plum group carries out judging whether avoidance, and otherwise unmanned plane is kept original
Route continues to fly;
4.1 are existed based on image less than lmaxBarrier, calculate the horizontal distance d of unmanned plane and nearest barrier (see figure
3):
If v0When-v ≠ 0,
In formula, z is the vertical pivot coordinate of nearest barrier, and y is the ordinate of orthogonal axes of nearest barrier, and θ is pitch angle, v0For figure
The pixel ordinate of inconocenter, v are the pixel ordinate of nearest barrier.
If v0When-v=0,
D=zcos | θ | (2)
4.2 as shown in figure 4, calculate the practical rectangle width of frame w of unmanned planefactWith length Ifact,
wfact=h cos | θ |+b sin | θ | (3)
In formula, h is unmanned plane height, and b is unmanned plane width, and θ is pitch angle, and θ ∈ (- 15 °, 15 °) is positive downwards, to
On be negative;Ifact=l, wherein l is unmanned plane length.
4.3 as shown in figure 5, calculate the length Δ u's and width Delta v: image center when depth value d of pixel flight frame
Parallax is
D=Bf/d (4)
In formula, B is the distance between two cameras, and f is the focal length of binocular camera;
Δ x=lD/B (5)
In formula, Δ x is the correspondingly-sized under image coordinate system of the l in depth value d.
In formula, Δ u1For the corresponding size under pixel coordinate system of the l in depth value d, dx is unit pixel in horizontal axis
Physical size;
In formula, Δ v1For wfactThe corresponding size under pixel coordinate system in depth value d, dy are unit pixel vertical
The physical size of axis;
By Δ u1With Δ v1Original 1.2 times are all expanded as, can reduce extraneous factor influences and handles in time emergency case,
Then have
Δ u=1.2 Δ u1, Δ v=1.2 Δ v1 (8)
4.4 as shown in fig. 6, obtain the seat of pixel flight safety frame when unmanned plane inclination.If pixel flight safety frame
The separation delta c at center and the pixel flight frame line of centres:
Δ e=dtan | θ | (9)
In formula, Δ e is unmanned plane actual frames center offset;
Δ y_c=D Δ e/B (10)
In formula, Δ y_c is the offset of rectangle frame central point of the unmanned plane under image coordinate system;
Δ c=Δ y_c/dy=f tan | θ |/dy (11)
4.5 under pixel coordinate system, if θ >=0 °, pixel flight frame is obtained pixel flight safety to surface translation Δ c
Frame;If 0 ° of θ <, pixel flight frame is obtained pixel flight safety frame to underface translation Δ c;
4.6 traversal pixel flight safety frames: if barrier (1) is not present in pixel flight safety frame, unmanned plane is to pixel
The flight of flight safety frame direction;(2) if there are barrier in pixel flight safety frame, unmanned plane carries out corresponding selective avoidance,
Specific steps are as follows:
4.6.1 unmanned plane, which reduces speed now, goes slowly, and records the depth value deep of the nearest barrier of pixel flight safety frame, into
Row selects corresponding avoidance mode 1:
Unmanned plane flies upwards, if barrier is not present in pixel flight safety frame in upward flight course, nobody
Machine is by the airline operation originally planned;If there are barrier in pixel flight safety frame after upward 5 meters of flight, unmanned plane is vertical
Slow down and hover, and information is sent by digital transmission module and figure transmission module, prompts artificial be remotely controlled;
4.6.2deep compared with the threshold value set_d of setting, unmanned plane carries out selecting corresponding avoidance mode 2:
1. deep enters jogging area when being greater than set_d, unmanned plane adjustment pitch angle suitably reduces speed;
2. deep enters alert zone when being less than or equal to set_d, unmanned plane slows down hovering immediately, and by digital transmission module with
Figure transmission module sends information, prompts artificial remote control.
The purpose of the present invention, technical scheme and beneficial effects are described in detail in above-mentioned specific embodiment, institute
It should be understood that the foregoing is merely a specific embodiment of the invention, it is not intended to restrict the invention, it is all in the present invention
Spirit and principle within, any modification, equivalent substitution, improvement and etc. done, should be included in protection scope of the present invention it
It is interior.
Claims (4)
1. the multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision, which comprises the following steps:
Step 1: the ambient image in front of unmanned plane is obtained by binocular camera;The ambient image is the two of binocular camera
A camera first image and second image captured in synchronization difference;
Step 2: identifying in front of unmanned plane whether there is barrier according to the ambient image of acquisition;
Step 3: barrier if it exists, then based on unmanned plane current pose information and the barrier depth obtained according to image
Value;The UAV Attitude information is unmanned plane pitch angle acquired in the attitude transducer set on unmanned plane, attitude transducer
For accelerometer, gyroscope and magnetometer;
Step 4: setting avoidance threshold value lmaxIf image, which exists, is less than lmaxBarrier, tree plum group carries out judging whether avoidance, it is no
Then unmanned plane keeps original route to continue to fly.
2. the multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision as described in claim 1, feature exist
In the depth value of the barrier is obtained by following steps:
1) it includes the first image and the second image that the binocular camera, which obtains the ambient image in front of unmanned plane,;
2) two cameras of the binocular camera obtain the inside and outside parameter of camera by Zhang Shi standardization, then by taking the photograph
As the external parameter and distortion factor of head correct the first image and the second image;By the first image and second after correction
Image carries out Stereo matching using SGAM algorithm, obtains the disparity map of two images;
3) depth image is obtained based on the disparity map, barrier depth value is extracted according to the depth image;
4) avoidance threshold value l is setmax, identification is less than lmaxBarrier and reservation, removal be greater than or equal to lmaxBarrier.
3. the multi-rotor unmanned aerial vehicle selectivity avoidance obstacle side based on binocular vision as described in claim 1 and claim 2
Method, which is characterized in that if image, which exists, is less than lmaxBarrier, tree plum group judges to be kept away by the following steps
Barrier, is also to maintain original route and continues to fly:
1) existed based on image and be less than lmaxBarrier, calculate unmanned plane and nearest barrier horizontal distance d:
If v0When-v ≠ 0,
In formula, z is the vertical pivot coordinate of nearest barrier, and y is the ordinate of orthogonal axes of nearest barrier, and θ is pitch angle, v0For in image
The pixel ordinate of the heart, v are the pixel ordinate of nearest barrier;
If v0When-v=0,
D=zcos | θ | (2)
2) the practical rectangle width of frame w of unmanned plane is calculatedfactWith length Ifact:
wfact=hcos | θ |+bsin | θ | (3)
In formula, h is unmanned plane height, and b is unmanned plane width, and θ is pitch angle, and θ ∈ (- 15 °, 15 °) is positive downwards, is upwards
It is negative;Ifact=l, wherein l is unmanned plane length;
3) the length Δ u and width Delta v: the parallax of image center when depth value d of calculating pixel flight frame are
D=Bf/d (4)
In formula, B is the distance between two cameras, and f is the focal length of binocular camera;
Δ x=lD/B (5)
In formula, Δ x is the correspondingly-sized under image coordinate system of the l in depth value d;
In formula, Δ u1For the corresponding size under pixel coordinate system of the l in depth value d, dx is physics of the unit pixel in horizontal axis
Size;
In formula, Δ v1For wfactThe corresponding size under pixel coordinate system in depth value d, dy are unit pixel in the longitudinal axis
Physical size;
By Δ u1With Δ v1Original 1.2 times are all expanded as, can reduce extraneous factor influences and handle emergency case in time, then has
Δ u=1.2 Δ u1, Δ v=1.2 Δ v1 (8)
4) when unmanned plane tilts, the seat of pixel flight safety frame is obtained;If pixel flight safety frame center and pixel are flown
The separation delta c of the frame line of centres:
Δ e=dtan | θ | (9)
In formula, Δ e is unmanned plane actual frames center offset;
Δ y_c=D Δ e/B (10)
In formula, Δ y_c is the offset of rectangle frame central point of the unmanned plane under image coordinate system;
Δ c=Δ y_c/dy=ftan | θ |/dy (11)
5) under pixel coordinate system, if θ >=0 °, pixel flight frame is obtained into pixel flight safety frame to surface translation Δ c;If
0 ° of θ <, pixel flight frame is obtained into pixel flight safety frame to underface translation Δ c;
6) traverse pixel flight safety frame: if barrier (1) is not present in pixel flight safety frame, unmanned plane flies to pixel pacifies
Full frame direction flight;(2) if there are barrier in pixel flight safety frame, unmanned plane carries out corresponding selective avoidance.
4. the multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision as described in claim 3, feature exist
In if unmanned plane carries out 2 kinds of selective avoidance modes there are barrier in pixel flight safety frame;
1) unmanned plane, which reduces speed now, goes slowly, and records the depth value deep of the nearest barrier of pixel flight safety frame, is selected
Following avoidance mode 1:
Unmanned plane flies upwards, if barrier is not present in pixel flight safety frame in upward flight course, unmanned plane is pressed
The airline operation originally planned;If unmanned plane subtracts immediately there are barrier in pixel flight safety frame after upward 5 meters of flight
Speed hovering, and information is sent by digital transmission module and figure transmission module, prompt artificial remote control;
2) compared with the threshold value set_d of setting, unmanned plane carries out selecting following avoidance mode 2 deep:
1. deep enters jogging area when being greater than set_d, unmanned plane adjustment pitch angle suitably reduces speed;
2. deep enters alert zone when being less than or equal to set_d, unmanned plane slows down hovering immediately, and passes through digital transmission module and figure biography
Module sends information, prompts artificial remote control.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910232714.8A CN109857133A (en) | 2019-03-26 | 2019-03-26 | Multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910232714.8A CN109857133A (en) | 2019-03-26 | 2019-03-26 | Multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109857133A true CN109857133A (en) | 2019-06-07 |
Family
ID=66901966
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910232714.8A Withdrawn CN109857133A (en) | 2019-03-26 | 2019-03-26 | Multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109857133A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111638727A (en) * | 2020-05-29 | 2020-09-08 | 西北工业大学 | Multi-rotor aircraft safety navigation control method based on depth image |
CN113271357A (en) * | 2021-05-17 | 2021-08-17 | 南京邮电大学 | Ground-air cooperative networking system and control method |
CN113900443A (en) * | 2021-09-28 | 2022-01-07 | 合肥工业大学 | Unmanned aerial vehicle obstacle avoidance early warning method and device based on binocular vision |
CN118015047A (en) * | 2024-04-08 | 2024-05-10 | 天津所托瑞安汽车科技有限公司 | Multi-target tracking method, device, equipment and storage medium |
-
2019
- 2019-03-26 CN CN201910232714.8A patent/CN109857133A/en not_active Withdrawn
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111638727A (en) * | 2020-05-29 | 2020-09-08 | 西北工业大学 | Multi-rotor aircraft safety navigation control method based on depth image |
CN111638727B (en) * | 2020-05-29 | 2022-09-23 | 西北工业大学 | Multi-rotor aircraft safety navigation control method based on depth image |
CN113271357A (en) * | 2021-05-17 | 2021-08-17 | 南京邮电大学 | Ground-air cooperative networking system and control method |
CN113900443A (en) * | 2021-09-28 | 2022-01-07 | 合肥工业大学 | Unmanned aerial vehicle obstacle avoidance early warning method and device based on binocular vision |
CN113900443B (en) * | 2021-09-28 | 2023-07-18 | 合肥工业大学 | Unmanned aerial vehicle obstacle avoidance early warning method and device based on binocular vision |
CN118015047A (en) * | 2024-04-08 | 2024-05-10 | 天津所托瑞安汽车科技有限公司 | Multi-target tracking method, device, equipment and storage medium |
CN118015047B (en) * | 2024-04-08 | 2024-07-16 | 天津所托瑞安汽车科技有限公司 | Multi-target tracking method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11300413B2 (en) | Systems and methods for auto-return | |
JP5618840B2 (en) | Aircraft flight control system | |
JP5775632B2 (en) | Aircraft flight control system | |
CN109857133A (en) | Multi-rotor unmanned aerial vehicle selectivity avoidance obstacle method based on binocular vision | |
CN107703951B (en) | A kind of unmanned plane barrier-avoiding method and system based on binocular vision | |
US10475209B2 (en) | Camera calibration | |
US10375289B2 (en) | System and method for providing autonomous photography and videography | |
CN108700890B (en) | Unmanned aerial vehicle return control method, unmanned aerial vehicle and machine readable storage medium | |
CN108351654B (en) | System and method for visual target tracking | |
EP3123260B1 (en) | Selective processing of sensor data | |
CN111527463A (en) | Method and system for multi-target tracking | |
CN109782788A (en) | Unmanned plane low latitude obstacle avoidance system and control method based on binocular vision | |
WO2019051832A1 (en) | Movable object control method, device and system | |
US11449076B2 (en) | Method for controlling palm landing of unmanned aerial vehicle, control device, and unmanned aerial vehicle | |
US20210009270A1 (en) | Methods and system for composing and capturing images | |
CN110456811A (en) | Unmanned plane selectivity obstacle avoidance system and method based on binocular vision and three axis holders | |
CN109754420B (en) | Target distance estimation method and device and unmanned aerial vehicle | |
US20230359204A1 (en) | Flight control method, video editing method, device, uav and storage medium | |
WO2020024182A1 (en) | Parameter processing method and apparatus, camera device and aircraft | |
WO2021056139A1 (en) | Method and device for acquiring landing position, unmanned aerial vehicle, system, and storage medium | |
CN214409706U (en) | Indoor unmanned aerial vehicle positioning system based on machine vision | |
CN116416291A (en) | Electronic fence automatic generation method, real-time detection method and device | |
CN213814412U (en) | Double-cradle head unmanned aerial vehicle | |
WO2019127192A1 (en) | Image processing method and apparatus | |
WO2020107487A1 (en) | Image processing method and unmanned aerial 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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190607 |
|
WW01 | Invention patent application withdrawn after publication |