CN114879729A - Unmanned aerial vehicle autonomous obstacle avoidance method based on obstacle contour detection algorithm - Google Patents
Unmanned aerial vehicle autonomous obstacle avoidance method based on obstacle contour detection algorithm Download PDFInfo
- Publication number
- CN114879729A CN114879729A CN202210531544.5A CN202210531544A CN114879729A CN 114879729 A CN114879729 A CN 114879729A CN 202210531544 A CN202210531544 A CN 202210531544A CN 114879729 A CN114879729 A CN 114879729A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- image
- obstacle
- pixel
- 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
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 33
- 238000004364 calculation method Methods 0.000 claims abstract description 22
- 238000003708 edge detection Methods 0.000 claims abstract description 9
- 230000000877 morphologic effect Effects 0.000 claims abstract description 6
- 238000001914 filtration Methods 0.000 claims description 22
- 239000011159 matrix material Substances 0.000 claims description 17
- 238000010845 search algorithm Methods 0.000 claims description 9
- 230000010354 integration Effects 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000011897 real-time detection Methods 0.000 claims description 4
- 241000566145 Otus Species 0.000 claims description 3
- 230000001133 acceleration Effects 0.000 claims description 3
- 238000005260 corrosion Methods 0.000 claims description 3
- 230000007797 corrosion Effects 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 238000009825 accumulation Methods 0.000 claims 1
- 238000004088 simulation Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
- G05D1/106—Change initiated in response to external conditions, e.g. avoidance of elevated terrain or of no-fly zones
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an unmanned aerial vehicle autonomous obstacle avoidance method based on an obstacle contour detection algorithm. Then, the converted image is subjected to thresholding and morphological processing. And performing edge detection and contour detection on the processed image, combining the detected result with data calibrated by a camera, and calculating the barycenter coordinate of the obstacle under a world coordinate system, thereby obtaining the position information and contour information of the obstacle. And finally, transmitting the information of the obstacle into a D-obstacle avoidance algorithm to carry out real-time path solving until the autonomous obstacle avoidance function of the unmanned aerial vehicle is completed. The method is high in real-time performance and high in calculation efficiency, and can be popularized to the autonomous obstacle avoidance of the unmanned aerial vehicle under dynamic obstacles and the autonomous obstacle avoidance of the unmanned aerial vehicle under a real three-dimensional scene based on the method.
Description
Technical Field
The invention belongs to the technical field of unmanned aerial vehicles, and particularly relates to an unmanned aerial vehicle autonomous obstacle avoidance method.
Background
Along with the rapid development of the unmanned aerial vehicle industry, the safety of the unmanned aerial vehicle is also paid extensive attention to, and especially under an unknown environment, the autonomous obstacle avoidance of the unmanned aerial vehicle is very important. Due to the lack of prior information of unknown environments, the unmanned aerial vehicle needs to sense and avoid obstacles. Among them, obstacle detection is an important ring. The current obstacle detection methods mainly include: ultrasonic-based detection methods, infrared-based detection methods, laser-based detection methods, and machine-vision-based detection methods. The detection method based on machine vision is to acquire an image by using a camera and process the image by using an image processing algorithm to obtain information such as the outline, the position, the depth and the like of an obstacle. Different from the first three methods, the information acquired by the machine vision-based detection method is richer.
There are also significant differences in the application of obstacle detection algorithms based on machine vision. The interframe difference method is only suitable for detecting dynamic obstacles and cannot meet the requirement of real-time detection; the optical flow estimation method needs to predict the detected area in advance and cannot detect a complete obstacle; the traditional obstacle contour detection algorithm is suitable for feature obstacle detection, but has the defects of large noise influence, poor adaptability and the like. Therefore, designing an obstacle detection algorithm which has good adaptability and small calculation amount and can meet the real-time requirement and completing the autonomous obstacle avoidance function of the unmanned aerial vehicle is a technical problem to be solved by researchers in the field.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides an unmanned aerial vehicle autonomous obstacle avoidance method based on an obstacle contour detection algorithm. Then, the converted image is subjected to thresholding and morphological processing. And performing edge detection and contour detection on the processed image, combining the detected result with data calibrated by a camera, and calculating the barycenter coordinate of the obstacle under a world coordinate system, thereby obtaining the position information and contour information of the obstacle. And finally, transmitting the information of the obstacle into a D-obstacle avoidance algorithm to carry out real-time path solving until the autonomous obstacle avoidance function of the unmanned aerial vehicle is completed. The method is high in real-time performance and high in calculation efficiency, and can be popularized to the autonomous obstacle avoidance of the unmanned aerial vehicle under dynamic obstacles and the autonomous obstacle avoidance of the unmanned aerial vehicle under a real three-dimensional scene based on the method.
The technical scheme adopted by the invention for solving the technical problem comprises the following steps:
step 1: establishing an unmanned aerial vehicle model;
step 1-1: nobodyThe layout structure of the model is X-shaped, namely the included angle between the advancing direction and the adjacent support is 45 degrees; assuming that the model of the unmanned aerial vehicle is a rigid body, the brake of the unmanned aerial vehicle generates a force F and a torque tau, and the force and the torque of the unmanned aerial vehicle at the ith moment are respectively set as F i And τ i The calculation formula is as follows:
in the formula C T And C pow Respectively representing the thrust coefficient and the power coefficient based on the rotor, rho representing the air density, D representing the rotor diameter, omega max Represents the maximum angular velocity of rotation, u i Representing the rotating speed of the motor at the ith moment;
step 1-2: calculating the next motion state of the unmanned aerial vehicle;
let the speed of the unmanned aerial vehicle at the m-1 th moment be v m-1 In the position p m-1 Acceleration of a m-1 Dt, time step, position p at time m m And velocity v m The calculation is as follows:
in the formula, p m Position of the drone at moment m, v m The speed of the unmanned aerial vehicle at the mth moment;
step 2: determining a starting point and a target point of the unmanned aerial vehicle by using a world coordinate system;
and step 3: determining an unmanned aerial vehicle path search algorithm;
selecting a D route search algorithm, and setting an heuristic function as follows:
f(s)=h(s)+g(s) (5)
wherein h(s) represents the cost value from the current node to the target point, and g(s) represents the cost value from the current node to the starting point;
suppose the position coordinate of the current node is (x) s ,y s ) The coordinates of the starting point are (x) start ,y start ) The coordinates of the target point are ((x) goal ,y goal ) Then h(s) and g(s) are respectively expressed as:
generating a path from the current node to the target node through a D-path searching algorithm;
and 4, step 4: establishing an obstacle contour detection algorithm based on color information;
step 4-1: acquiring an environment image through an airborne camera of the unmanned aerial vehicle;
step 4-2: performing smooth filtering processing on the acquired environment image in a mode of combining Gaussian filtering and median filtering, wherein the calculation formula of the Gaussian filtering is as follows:
wherein G (.) is a two-dimensional Gaussian function, (Δ x) 2 +Δy 2 ) Expressed is the sum of squares of the distances between other pixels in the neighborhood and the central pixel, σ is the standard deviation of the two-dimensional normal distribution, (Δ x, Δ y) represents the domain;
performing further filtering on the image subjected to smooth filtering by using median filtering;
step 4-3: converting an environment image of an RGB space into an HSV color space, wherein the calculation formula is as follows:
V=max(R,G,B) (11)
r, G, B respectively represents the values of three color components in the RGB space, H, S, V respectively represents the chromaticity, saturation and brightness in the HSV space;
step 4-4: carrying out binarization operation on the image, and carrying out threshold processing on the image by adopting an Otus algorithm; the specific calculation process is as follows:
step 4-4-1: calculating the zero-order moment of integration of the gray level histogram:
in the formula, histogram I Representing a normalized image grey histogram, histogram I (k) Representing the ratio of pixel points with the gray value equal to k in the image;
step 4-4-2: calculating a first order moment of integration of the gray level histogram:
step 4-4-3: calculating the gray level average value of the image population:
mean=oneCuMo(255) (14)
step 4-4-4: dividing the image into a foreground image and a background image according to the gray characteristic, and calculating a threshold q which can enable the variance of the foreground image and the background image to be maximum; the following metric was used for the measure of variance:
and 4-5: performing morphological processing of expansion and corrosion on the image;
and 4-6: adopting Canny operator edge detection and contour detection:
step 4-6-1: smoothing noise of a non-edge area of the image;
step 4-6-2: calculating the amplitude and direction of the image gradient by using a Sobel operator;
step 4-6-3: traversing the pixel points one by one, judging whether the current pixel point is a point with a maximum gradient value in the gradient direction, if so, keeping the point, and if not, returning the point to zero;
step 4-6-4: carrying out threshold processing by using double thresholds to obtain edge points;
step 4-6-5: fitting the result after the edge detection with the foreground information of the image to approximately obtain the image contour;
and 4-7: determining internal parameter matrix M of airborne camera by adopting Zhangyingyou camera calibration method 1 And an external parameter matrix M 2 ;
And 4-8: solving the barycenter coordinate of the obstacle under the world coordinate system;
step 4-8-1: the calculation formula of the (i + j) order moment of the environment image is as follows:
wherein x and y represent the horizontal and vertical coordinates of the pixel points, and I (x, y) represents the pixel intensity corresponding to the pixel point with the coordinates (x, y);
step 4-8-2: by a zeroth order image moment M 00 And first order moment of image (M) 01 、M 10 ) Calculating the centroid coordinates under the pixel coordinate system:
step 4-8-3: and (3) performing coordinate conversion, and converting the centroid coordinate into a world coordinate system:
wherein u and v are coordinates in a pixel coordinate system, (X) C ,Y C ,Z C ) As coordinates of the camera coordinate system, f x And f y Denotes the physical size of the pixel, u, in the x-axis and y-axis, respectively 0 And v 0 Respectively representing pixel difference values of the central pixel coordinate of the image and the original point pixel coordinate of the image in the x direction and the y direction, wherein f is the focal length of the camera; r is a 3 x 3 rotation matrix, namely a matrix obtained by rotating coordinate axes when a pixel coordinate system is converted into a world coordinate system; t is an offset vector, (X) W ,Y W ,Z W ) Coordinates in a world coordinate system;
and 5: the unmanned aerial vehicle starts to move along the initial path generated in the step 3 from the starting point, meanwhile, the airborne camera adopts the obstacle contour detection algorithm in the step 4 to perform real-time detection, if an unknown obstacle appears on the path, whether the flight of the unmanned aerial vehicle is influenced is judged according to the position information and the contour information of the unmanned aerial vehicle, and if the flight of the unmanned aerial vehicle is influenced, a new autonomous obstacle avoidance path from the current point to the target point is generated by adopting the path search algorithm in the step 3;
and circulating according to the process until the unmanned aerial vehicle reaches a target point, and finishing the whole autonomous obstacle avoidance process.
The invention has the following beneficial effects:
according to the unmanned aerial vehicle obstacle avoidance method, under the condition that the environment is not completely known, the unmanned aerial vehicle can adopt an obstacle detection algorithm to detect and obtain information such as the position and the outline of an unknown obstacle in real time, and the unmanned aerial vehicle can achieve the target of unmanned aerial vehicle autonomous obstacle avoidance under the condition that the environment is unknown by combining with a D-path searching algorithm. The method is high in real-time performance and high in calculation efficiency, and can be popularized to the autonomous obstacle avoidance of the unmanned aerial vehicle under dynamic obstacles and the autonomous obstacle avoidance of the unmanned aerial vehicle under a real three-dimensional scene based on the method.
Drawings
Fig. 1 is a general flow chart of the unmanned aerial vehicle autonomous obstacle avoidance method based on the obstacle contour detection algorithm.
Fig. 2 is a diagram of a simulation model of an unmanned aerial vehicle for use in an AirSim according to an embodiment of the present invention.
FIG. 3 is a top view of a simulation environment built in AirSim according to an embodiment of the present invention.
Fig. 4 is a diagram illustrating the effect of the obstacle contour detection algorithm at a certain time according to an embodiment of the present invention.
Fig. 5 is a path result diagram of autonomous obstacle avoidance by the unmanned aerial vehicle according to the embodiment of the present invention.
Detailed Description
The invention is further illustrated with reference to the following figures and examples.
The invention provides an unmanned aerial vehicle autonomous obstacle avoidance method based on an improved obstacle contour detection algorithm. Firstly, filtering processing and color space conversion are carried out on the acquired image. Then, the converted image is subjected to thresholding and morphological processing. And performing edge detection and contour detection on the processed image, combining the detected result with data calibrated by a camera, and calculating the barycenter coordinate of the obstacle under a world coordinate system, thereby obtaining the position information and contour information of the obstacle. And finally, transmitting the information of the obstacle into a D-obstacle avoidance algorithm to carry out real-time path solving until the autonomous obstacle avoidance function of the unmanned aerial vehicle is completed.
The simulation environment is as follows: windows10 operating system, Airsim emulation platform.
The invention takes into account a three-dimensional map model, the coordinate system being a planar coordinate system. Suppose there are 1 drone with its own vision camera, as shown in fig. 2. The size of the built simulation map is 100m multiplied by 100m, gray obstacles represent obstacles with known positions and contours, orange and white obstacles represent unknown obstacles, and the environment is built as shown in fig. 3. Assume that the position coordinates of the drone are (70, 20) and the coordinates of the target point are (70, 80).
As shown in FIG. 1, the method comprises the following specific steps in AirSim environment:
step 1: establishing an unmanned aerial vehicle model;
step 1-1: the layout structure of the unmanned aerial vehicle model is X-shaped, namely the included angle between the advancing direction and the adjacent support is 45 degrees; assuming that the model of the unmanned aerial vehicle is a rigid body, and the unmanned aerial vehicle can generate a force F and a torque tau by any number of brakes, the force and the torque of the unmanned aerial vehicle at the ith moment are respectively set as F i And τ i The calculation formula is as follows:
in the formula C T And C pow Respectively representing the thrust coefficient and the power coefficient based on the rotor, rho representing the air density, D representing the rotor diameter, omega max Represents the maximum angular velocity of rotation, u i Representing the rotating speed of the motor at the ith moment;
step 1-2: calculating the next motion state of the unmanned aerial vehicle;
let the speed of the unmanned aerial vehicle at the m-1 th moment be v m-1 In the position p m-1 Acceleration of a m-1 Dt, time step, position p at time m m And velocity v m The calculation is as follows:
in the formula, p m Position of the drone at moment m, v m The speed of the unmanned aerial vehicle at the mth moment;
step 2: determining a starting point and a target point of the unmanned aerial vehicle by using a world coordinate system;
and step 3: determining an unmanned aerial vehicle path search algorithm;
selecting a D route search algorithm, and setting an heuristic function as follows:
f(s)=h(s)+g(s) (5)
wherein h(s) represents the cost value from the current node to the target point, and g(s) represents the cost value from the current node to the starting point;
suppose the position coordinate of the current node is (x) s ,y s ) The coordinates of the starting point are (x) start ,y start ) The coordinates of the target point are ((x) goal ,y goal ) Then h(s) and g(s) are respectively expressed as:
generating a path from the current node to the target node through a D-path searching algorithm;
and 4, step 4: establishing an obstacle contour detection algorithm based on color information;
step 4-1: acquiring an environment image through an airborne camera of the unmanned aerial vehicle;
step 4-2: performing smooth filtering processing on the acquired environment image in a mode of combining Gaussian filtering and median filtering, wherein the calculation formula of the Gaussian filtering is as follows:
wherein G (.) is a two-dimensional Gaussian function: (Δx 2 +Δy 2 ) Expressed is the sum of squares of the distances between other pixels in the neighborhood and the central pixel, σ is the standard deviation of the two-dimensional normal distribution, (Δ x, Δ y) represents the domain;
performing further filtering on the image subjected to smooth filtering by using median filtering, so as to furthest retain the contour information of the image while eliminating noise; (ii) a
Step 4-3: converting an environment image of an RGB space into an HSV color space, wherein the calculation formula is as follows:
V=max(R,G,B) (11)
r, G, B respectively represents the values of three color components in RGB space, H, S, V respectively represents the chroma, saturation and brightness in HSV space;
step 4-4: carrying out binarization operation on the image, and carrying out threshold processing on the image by adopting an Otus algorithm; the specific calculation process is as follows:
step 4-4-1: calculating the zero-order moment of integration of the gray level histogram:
in the formula, histogram I Representing a normalized image grey histogram, histogram I (k) Representing the ratio of pixel points with the gray value equal to k in the image;
step 4-4-2: calculating a first order moment of integration of the gray level histogram:
step 4-4-3: calculating the gray level average value of the image population:
mean=oneCuMo(255) (14)
step 4-4-4: dividing the image into a foreground image and a background image according to the gray characteristic, and calculating a threshold q which can enable the variance of the foreground image and the background image to be maximum; the following metric was used for the measure of variance:
and 4-5: performing morphological processing of expansion and corrosion on the image;
and 4-6: adopting Canny operator edge detection and contour detection:
step 4-6-1: smoothing noise of a non-edge area of the image;
step 4-6-2: calculating the amplitude and direction of the image gradient by using a Sobel operator;
step 4-6-3: traversing the pixel points one by one, judging whether the current pixel point is a point with a maximum gradient value in the gradient direction, if so, keeping the point, and if not, returning the point to zero;
step 4-6-4: carrying out threshold processing by using double thresholds to obtain edge points;
step 4-6-5: fitting the result after the edge detection with the foreground information of the image to approximately obtain the image contour;
and 4-7: determining internal parameter matrix M of airborne camera by adopting Zhangyingyou camera calibration method 1 And an external parameter matrix M 2 ;
And 4-8: solving the barycenter coordinate of the obstacle under the world coordinate system;
step 4-8-1: the calculation formula of the (i + j) order moment of the environment image is as follows:
wherein x and y represent the horizontal and vertical coordinates of the pixel points, and I (x, y) represents the pixel intensity corresponding to the pixel point with the coordinates (x, y);
step 4-8-2: by a zeroth order image moment M 00 And first order moment of image (M) 01 、M 10 ) Calculating the centroid coordinates under the pixel coordinate system:
step 4-8-3: and performing coordinate conversion, and converting the centroid coordinate into a world coordinate system:
wherein u and v are coordinates in a pixel coordinate system, (X) C ,Y C ,Z C ) As coordinates of the camera coordinate system, f x And f y Denotes the physical size of the pixel, u, in the x-axis and y-axis, respectively 0 And v 0 Respectively representing pixel difference values of the central pixel coordinate of the image and the original point pixel coordinate of the image in the x direction and the y direction, wherein f is the focal length of the camera; r is a 3 x 3 rotation matrix, namely a matrix obtained by rotating coordinate axes when a pixel coordinate system is converted into a world coordinate system; t is an offset vector, (X) W ,Y W ,Z W ) For coordinates in the world coordinate system, matrix M 1 Is an internal reference matrix of the camera, matrix M 2 The external parameter matrix is an external parameter matrix of the camera and can be measured through calibration of the camera;
the unmanned aerial vehicle can obtain the position information and the contour information of an unknown obstacle;
and 5: the unmanned aerial vehicle starts to move along the initial path generated in the step 3 from the starting point, meanwhile, the airborne camera adopts the obstacle contour detection algorithm in the step 4 to perform real-time detection, if an unknown obstacle appears on the path, whether the flight of the unmanned aerial vehicle is influenced is judged according to the position information and the contour information of the unmanned aerial vehicle, and if the flight of the unmanned aerial vehicle is influenced, a new autonomous obstacle avoidance path from the current point to the target point is generated by adopting the path search algorithm in the step 3;
and circulating according to the process until the unmanned aerial vehicle reaches a target point, and finishing the whole autonomous obstacle avoidance process.
In summary, the invention determines the profile information and the position information of the unknown obstacle by using the obstacle profile detection algorithm, and fig. 4 is a profile result graph at a certain moment in the obstacle detection process, so that the required information of the unknown obstacle can be provided for the unmanned aerial vehicle autonomous obstacle avoidance algorithm quickly and accurately. When the unmanned aerial vehicle encounters an unknown obstacle during the process of searching and detecting, a path from a current unknown node to a target node can be generated well, as shown in fig. 5, a feasible path generated by the unmanned aerial vehicle through an autonomous obstacle avoidance algorithm verifies the real-time performance and feasibility of the algorithm. For autonomous obstacle avoidance of the unmanned aerial vehicle, the method is simple, the real-time robustness is high, and autonomous obstacle avoidance of the unmanned aerial vehicle is achieved.
Claims (1)
1. An unmanned aerial vehicle autonomous obstacle avoidance method based on an obstacle contour detection algorithm is characterized by comprising the following steps:
step 1: establishing an unmanned aerial vehicle model;
step 1-1: the layout structure of the unmanned aerial vehicle model is X-shaped, namely the included angle between the advancing direction and the adjacent support is 45 degrees; assuming that the model of the unmanned aerial vehicle is a rigid body, the brake of the unmanned aerial vehicle generates a force F and a torque tau, and the force and the torque of the unmanned aerial vehicle at the ith moment are respectively set as F i And τ i The calculation formula is as follows:
in the formula C T And C pow Respectively representing the thrust coefficient and the power coefficient based on the rotor, rho representing the air density, D representing the rotor diameter, omega max Represents the maximum angular velocity of rotation, u i Representing the rotating speed of the motor at the ith moment;
step 1-2: calculating the next motion state of the unmanned aerial vehicle;
let the speed of the unmanned aerial vehicle at the m-1 th moment be v m-1 In the position p m-1 Acceleration of a m-1 Dt, time step, position p at time m m And velocity v m The calculation is as follows:
in the formula, p m Position of the drone at moment m, v m The speed of the unmanned aerial vehicle at the mth moment;
step 2: determining a starting point and a target point of the unmanned aerial vehicle by using a world coordinate system;
and step 3: determining an unmanned aerial vehicle path search algorithm;
selecting a D route search algorithm, and setting an heuristic function as follows:
f(s)=h(s)+g(s) (5)
wherein h(s) represents the cost value from the current node to the target point, and g(s) represents the cost value from the current node to the starting point;
suppose the position coordinate of the current node is (x) s ,y s ) The coordinates of the starting point are (x) start ,y start ) The coordinates of the target point are ((x) goal ,y goal ) Then h(s) and g(s) are respectively expressed as:
generating a path from the current node to the target node through a D-path searching algorithm;
and 4, step 4: establishing an obstacle contour detection algorithm based on color information;
step 4-1: acquiring an environment image through an airborne camera of the unmanned aerial vehicle;
step 4-2: performing smooth filtering processing on the acquired environment image in a mode of combining Gaussian filtering and median filtering, wherein the calculation formula of the Gaussian filtering is as follows:
wherein G (.) is a two-dimensional Gaussian function, (Δ x) 2 +Δy 2 ) Expressed is the sum of squares of the distances between other pixels in the neighborhood and the central pixel, σ is the standard deviation of the two-dimensional normal distribution, (Δ x, Δ y) represents the domain;
performing further filtering on the image subjected to smooth filtering by using median filtering;
step 4-3: converting an environment image of an RGB space into an HSV color space, wherein the calculation formula is as follows:
V=max(R,G,B) (11)
r, G, B respectively represents the values of three color components in the RGB space, H, S, V respectively represents the chromaticity, saturation and brightness in the HSV space;
step 4-4: carrying out binarization operation on the image, and carrying out threshold processing on the image by adopting an Otus algorithm; the specific calculation process is as follows:
step 4-4-1: calculating the zero-order moment of integration of the gray level histogram:
in the formula, histogram I Representing a normalized image grey histogram, histogram I (k) Representing the ratio of pixel points with the gray value equal to k in the image;
step 4-4-2: calculating a first order moment of accumulation of the gray level histogram:
step 4-4-3: calculating the gray level average value of the image population:
mean=oneCuMo(255) (14)
step 4-4-4: dividing the image into a foreground image and a background image according to the gray characteristic, and calculating a threshold q which can enable the variance of the foreground image and the background image to be maximum; the following metric was used for the measure of variance:
and 4-5: performing morphological processing of expansion and corrosion on the image;
and 4-6: adopting Canny operator edge detection and contour detection:
step 4-6-1: smoothing noise of a non-edge area of the image;
step 4-6-2: calculating the amplitude and direction of the image gradient by using a Sobel operator;
step 4-6-3: traversing the pixel points one by one, judging whether the current pixel point is a point with a maximum gradient value in the gradient direction, if so, keeping the point, and if not, returning the point to zero;
step 4-6-4: carrying out threshold processing by using double thresholds to obtain edge points;
step 4-6-5: fitting the result after the edge detection with the foreground information of the image to approximately obtain the image contour;
and 4-7: determining internal parameter matrix M of airborne camera by adopting Zhangyingyou camera calibration method 1 And an external parameter matrix M 2 ;
And 4-8: solving the barycenter coordinate of the obstacle under the world coordinate system;
step 4-8-1: the calculation formula of the (i + j) order moment of the environment image is as follows:
wherein x and y represent the horizontal and vertical coordinates of the pixel points, and I (x, y) represents the pixel intensity corresponding to the pixel point with the coordinates (x, y);
step 4-8-2: by a zeroth order image moment M 00 And first order moment of image (M) 01 、M 10 ) Calculating the centroid coordinates under the pixel coordinate system:
step 4-8-3: and (3) performing coordinate conversion, and converting the centroid coordinate into a world coordinate system:
wherein u and v are coordinates in a pixel coordinate system, (X) C ,Y C ,Z C ) As coordinates of the camera coordinate system, f x And f y Denotes the physical size of the pixel, u, in the x-axis and y-axis, respectively 0 And v 0 Respectively representing pixel difference values of the central pixel coordinate of the image and the original point pixel coordinate of the image in the x direction and the y direction, wherein f is the focal length of the camera; r is a 3 x 3 rotation matrix, namely a matrix obtained by rotating coordinate axes when a pixel coordinate system is converted into a world coordinate system; t is an offset vector, (X) W ,Y W ,Z W ) Coordinates in a world coordinate system;
and 5: the unmanned aerial vehicle starts to move along the initial path generated in the step 3 from the starting point, meanwhile, the airborne camera adopts the obstacle contour detection algorithm in the step 4 to perform real-time detection, if an unknown obstacle appears on the path, whether the flight of the unmanned aerial vehicle is influenced is judged according to the position information and the contour information of the unmanned aerial vehicle, and if the flight of the unmanned aerial vehicle is influenced, a new autonomous obstacle avoidance path from the current point to the target point is generated by adopting the path search algorithm in the step 3;
and circulating according to the process until the unmanned aerial vehicle reaches a target point, and finishing the whole autonomous obstacle avoidance process.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210531544.5A CN114879729A (en) | 2022-05-16 | 2022-05-16 | Unmanned aerial vehicle autonomous obstacle avoidance method based on obstacle contour detection algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210531544.5A CN114879729A (en) | 2022-05-16 | 2022-05-16 | Unmanned aerial vehicle autonomous obstacle avoidance method based on obstacle contour detection algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114879729A true CN114879729A (en) | 2022-08-09 |
Family
ID=82674949
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210531544.5A Pending CN114879729A (en) | 2022-05-16 | 2022-05-16 | Unmanned aerial vehicle autonomous obstacle avoidance method based on obstacle contour detection algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114879729A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106681353A (en) * | 2016-11-29 | 2017-05-17 | 南京航空航天大学 | Unmanned aerial vehicle (UAV) obstacle avoidance method and system based on binocular vision and optical flow fusion |
CN106708084A (en) * | 2016-11-24 | 2017-05-24 | 中国科学院自动化研究所 | Method for automatically detecting and avoiding obstacles for unmanned aerial vehicle under complicated environments |
WO2019015158A1 (en) * | 2017-07-21 | 2019-01-24 | 歌尔科技有限公司 | Obstacle avoidance method for unmanned aerial vehicle, and unmanned aerial vehicle |
CN110032211A (en) * | 2019-04-24 | 2019-07-19 | 西南交通大学 | Multi-rotor unmanned aerial vehicle automatic obstacle-avoiding method |
-
2022
- 2022-05-16 CN CN202210531544.5A patent/CN114879729A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106708084A (en) * | 2016-11-24 | 2017-05-24 | 中国科学院自动化研究所 | Method for automatically detecting and avoiding obstacles for unmanned aerial vehicle under complicated environments |
CN106681353A (en) * | 2016-11-29 | 2017-05-17 | 南京航空航天大学 | Unmanned aerial vehicle (UAV) obstacle avoidance method and system based on binocular vision and optical flow fusion |
WO2019015158A1 (en) * | 2017-07-21 | 2019-01-24 | 歌尔科技有限公司 | Obstacle avoidance method for unmanned aerial vehicle, and unmanned aerial vehicle |
CN110032211A (en) * | 2019-04-24 | 2019-07-19 | 西南交通大学 | Multi-rotor unmanned aerial vehicle automatic obstacle-avoiding method |
Non-Patent Citations (1)
Title |
---|
林涛;: "无人机视觉定位与避障子系统研究", 机械工程师, no. 03, 10 March 2020 (2020-03-10) * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109345588B (en) | Tag-based six-degree-of-freedom attitude estimation method | |
US10288418B2 (en) | Information processing apparatus, information processing method, and storage medium | |
WO2020151109A1 (en) | Three-dimensional target detection method and system based on point cloud weighted channel feature | |
US20200334843A1 (en) | Information processing apparatus, control method for same, non-transitory computer-readable storage medium, and vehicle driving support system | |
KR101595537B1 (en) | Networked capture and 3d display of localized, segmented images | |
CN108171715B (en) | Image segmentation method and device | |
Munoz-Banon et al. | Targetless camera-lidar calibration in unstructured environments | |
CN108447094B (en) | Method and system for estimating attitude of monocular color camera | |
JP6817742B2 (en) | Information processing device and its control method | |
CN112164117A (en) | V-SLAM pose estimation method based on Kinect camera | |
Fan et al. | Dynamic objects elimination in SLAM based on image fusion | |
Hochdorfer et al. | 6 DoF SLAM using a ToF camera: The challenge of a continuously growing number of landmarks | |
CN108596947B (en) | Rapid target tracking method suitable for RGB-D camera | |
CN112184765A (en) | Autonomous tracking method of underwater vehicle based on vision | |
CN114179788A (en) | Automatic parking method, system, computer readable storage medium and vehicle terminal | |
CN114919584A (en) | Motor vehicle fixed point target distance measuring method and device and computer readable storage medium | |
US20230376106A1 (en) | Depth information based pose determination for mobile platforms, and associated systems and methods | |
CN110207702A (en) | The method and device of target positioning | |
CN112733678A (en) | Ranging method, ranging device, computer equipment and storage medium | |
Tsai et al. | Vision-Based Obstacle Detection for Mobile Robot in Outdoor Environment. | |
Petrovai et al. | Obstacle detection using stereovision for Android-based mobile devices | |
CN114879729A (en) | Unmanned aerial vehicle autonomous obstacle avoidance method based on obstacle contour detection algorithm | |
CN114972491A (en) | Visual SLAM method, electronic device, storage medium and product | |
Zhang et al. | Feature regions segmentation based RGB-D visual odometry in dynamic environment | |
Yang et al. | A new algorithm for obstacle segmentation in dynamic environments using a RGB-D sensor |
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 |