A device and method for measuring the landing visual pose of an unmanned aerial vehicle (UAV).

Abstract

This invention provides a visual pose measurement device and method for UAV landing, comprising a UAV, a monocular camera, a vision processor, and n targets. The n targets are fixed on the UAV landing runway, and each target is equipped with several infrared LEDs. The monocular camera is mounted below the UAV to acquire original target scene images of the landing area. The monocular camera is connected to the vision processor, which transmits the acquired original target scene images to the vision processor. The vision processor processes the original target scene images, extracts target features from the target scene images, and calculates the relative pose of the UAV relative to the landing area based on the target features. This invention improves the accuracy of UAV pose determination and ensures the safety of UAV landing.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) technology, specifically relating to a UAV landing visual pose measurement device and method. Background Technology

[0002] Traditional UAV autonomous positioning methods include inertial measurement unit (IMU) positioning, radar positioning, and high-precision satellite positioning. The effectiveness of IMU positioning largely depends on the performance of the IMU itself. High-performance IMUs are expensive and bulky, while miniature IMUs with built-in gyroscopes suffer from time drift. Radar positioning has low accuracy and is complex. High-precision satellite positioning relies on satellite signals, making it susceptible to interference and difficult to guarantee reliability during landing. Visual pose measurement positioning, on the other hand, primarily uses monocular or multi-view cameras to capture images of the UAV's landing area. A visual processor processes this image information, establishes an optimization model based on a perspective projection model, and calculates the relative pose of the UAV to its landing area using an optimization objective function, thus achieving autonomous positioning. However, multi-view pose measurement systems require high camera installation accuracy, are complex, and involve tedious initial calibration. Monocular visual pose measurement systems, in contrast, have a simpler structure, are easier to install, and require less initial calibration.

[0003] Currently, targets for monocular vision UAV landing pose measurement are typically planar black-and-white markers with specific patterns or infrared LEDs. In low light conditions, the vision processor cannot recognize these planar black-and-white markers, limiting UAV landings at night. When using infrared LEDs for pose measurement on monocular vision UAVs, typically four infrared LEDs are arranged, and the 3D-2D coordinate matching relationship of the targets is determined based on their relative positions. However, due to camera field of view limitations, when the UAV deviates significantly from the landing area, it cannot simultaneously capture all targets, causing the pose measurement system to malfunction. Increasing the number of infrared LEDs, relying solely on the relative positions of the targets may lead to 3D-2D coordinate matching errors, affecting pose measurement accuracy. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a device and method for measuring the visual pose of a drone during landing. The solution of this invention can solve the problems existing in the prior art.

[0005] The technical solution of this invention:

[0006] According to a first aspect, a visual pose measurement device for UAV landing is provided, comprising a UAV, a monocular camera, a vision processor, and n targets. The n targets are fixed on the UAV landing runway, and each target is equipped with several infrared LEDs. The monocular camera is installed below the UAV to acquire original target scene images of the landing area. The monocular camera is connected to the vision processor and transmits the acquired original target scene images to the vision processor. The vision processor processes the original target scene images, extracts target features from the target scene images, and calculates the relative pose of the UAV relative to the landing area based on the target features.

[0007] Furthermore, n is greater than 10.

[0008] Furthermore, the heights of the targets are not uniform.

[0009] Furthermore, a light-diffusing plate is installed outside the infrared LED lamp.

[0010] Furthermore, an infrared filter is installed on the lens of the monocular camera.

[0011] According to the second aspect, a method for measuring the visual pose of a UAV during landing is provided, including the following steps:

[0012] After the drone flies to the effective area above the landing zone, the monocular camera captures the original target scene image;

[0013] The original target scene image is processed to extract the edge contour of the target in the target scene image;

[0014] The obtained edge contours are processed to obtain graphics that are the same as or similar to the candidate targets;

[0015] The obtained candidate identical or similar graphics are filtered, and the candidate identical or similar graphics whose size meets the set range are selected as the target graphics.

[0016] Based on the number of target graphics n' obtained, determine whether the processing of the original target image is reasonable. If it is reasonable, proceed to the next step; otherwise, return to reprocess the original target scene image.

[0017] Based on the coordinates of the target center in the UAV landing area {(X i ,Y i Z i )|i=1,2,3...n}、The coordinates of the center of the target graphic in the image coordinate system {(x i ,y i Given the values ​​of |i = 1, 2, 3, ..., n'} and the focal length f of the monocular camera, establish a global objective function E:

[0018]

[0019] Where, m jk For the corresponding relationship weight coefficients, M = s(R1, T) x ), N = s(R2, T y ), s=f / T Z The translation matrix between the UAV landing area coordinate system and the image coordinate system is T = [T x T y T z ] T The rotation matrix between the UAV landing area coordinate system and the image coordinate system is R = [R1 R2 R3]. T w k (k = 1, 2, 3…n) = 1, S k =(X k ,Y k Z k ) T ;

[0020] The corresponding weight coefficients m are obtained by solving the global objective function. jk The rotation matrix R between the UAV landing area coordinate system and the image coordinate system; the translation matrix T between the UAV landing area coordinate system and the image coordinate system.

[0021] Based on the rotation matrix R' and translation matrix T' between the camera coordinate system and the UAV coordinate system, and the corresponding weighting coefficient m jk Given the rotation matrix R between the UAV landing area coordinate system and the image coordinate system, and the translation matrix T between the UAV landing area coordinate system and the image coordinate system, solve for the UAV landing pose.

[0022] Furthermore, the method for processing the original target scene image to extract the edge contour of the target in the target scene image includes the following steps:

[0023] Linear distortion of the original scene image is corrected to obtain target scene image 1;

[0024] Find the threshold H of the target scene image 1, and use the threshold H to binarize the target scene image 1 to obtain the target scene image 2;

[0025] The edge contour of the target in the target scene image 2 is presented.

[0026] Furthermore, the method for processing the obtained edge contours to obtain graphics identical or similar to the target is as follows:

[0027] Fit the obtained edge contour of the target to obtain a graphic that is the same as or similar to the target shape, and use the graphic with similar size and center position as the candidate graphic.

[0028] Furthermore, the dimensions satisfy the set range as follows: R smin <R s <R smax , where R smin = (LED light graphic dimensions / track width) × image width × 0.5, R smax = (LED light graphic size / track width) × image width × 1.5.

[0029] Furthermore, the standard for judging whether the processing of the original target image is reasonable is: if 4≤n'≤n, then the processing of the original target image is qualified; otherwise, it is unqualified.

[0030] The beneficial effects of this invention compared to the prior art are as follows:

[0031] (1) This invention only requires simple installation of a monocular camera, and does not require extensive calibration in the early stage, saving time;

[0032] (2) This invention establishes matching coefficients for multiple target 3D-2D coordinates and continuously updates them during the iterative solution of the optimization model to avoid pose measurement failure due to incorrect matching of target 3D-2D coordinates, thereby improving the accuracy of UAV pose determination and ensuring the safety of UAV landing. Attached Figure Description

[0033] The accompanying drawings, which form part of this specification, are provided to further illustrate embodiments of the invention and, together with the textual description, explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0034] Figure 1 A schematic diagram of a UAV landing visual pose measurement device according to an embodiment of the present invention is shown.

[0035] Figure 2 The diagram illustrates the steps of a UAV landing visual pose measurement method according to an embodiment of the present invention.

[0036] Figure 3 A schematic diagram of a target provided according to a specific embodiment of the present invention is shown;

[0037] Figure 4 A schematic diagram of a UAV landing visual pose measurement device provided according to a specific embodiment of the present invention is shown.

[0038] The above figures include the following reference numerals:

[0039] 1. Infrared LED light; 2. Light dome; 3. Target; 4. Coordinate system of UAV landing area; 5. UAV; 6. Monocular camera; 7. Camera coordinate system; 8. UAV coordinate system. Detailed Implementation

[0040] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0042] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0043] like Figure 1As shown, according to an embodiment of the present invention, a visual pose measurement device for landing of a UAV is provided according to a first aspect, including a UAV 5, a monocular camera, a vision processor, and n targets 3. The n targets 3 are fixed on the UAV landing runway, and each target 3 is equipped with a plurality of infrared LEDs 1. The monocular camera is installed below the UAV 5 to acquire the original target scene image of the landing area. The monocular camera is connected to the vision processor and transmits the acquired original target scene image to the vision processor. The vision processor processes the original target scene image, extracts the target 3 features in the target scene image, and calculates the relative pose of the UAV 5 relative to the landing area based on the target 3 features.

[0044] In a further embodiment, the number of targets 3, n, is greater than 10. This avoids the situation where the number of targets in the captured image is less than four due to small changes in the monocular field of view, which would prevent the pose information of the UAV 5 and the UAV landing area from being calculated. Increasing the number of targets involved in the pose calculation effectively improves the pose measurement accuracy.

[0045] In another embodiment, the target 3 has different heights. By designing targets with different heights, richer 3D coordinates can be obtained, thereby improving the accuracy of pose measurement.

[0046] Further in one embodiment, such as Figure 3 As shown, a light-diffusing plate 2 is installed outside the infrared LED light 1, which effectively reduces the area of ​​irregular halos around the target in the image captured by the monocular camera. This facilitates more accurate positioning of the target's center image coordinates, reduces the complexity of the image processing, and effectively improves the program's running speed.

[0047] In one further embodiment, an infrared filter is installed on the lens of the monocular camera to effectively filter out light other than infrared light from entering the camera, thereby effectively improving the contrast between the target and the surrounding environment in the captured image and making it more adaptable to nighttime environments with low ambient light.

[0048] According to the second aspect of the embodiment, such as Figure 2 As shown, a method for measuring the visual pose of a UAV during landing is provided, including the following steps:

[0049] Step 1: After the drone flies to the effective area above the landing area, the monocular camera captures the original target scene image;

[0050] Step 2: Process the original target scene image to extract the edge contour of the target in the target scene image;

[0051] In a further embodiment, a method for processing the original target scene image to extract the edge contour of the target in the target scene image includes the following steps:

[0052] S1.1 Corrects the linear distortion of the original scene image to obtain the target scene image 1;

[0053] S1.2 Calculate the threshold H of the target scene image 1, and use the threshold H to binarize the target scene image 1 to obtain the target scene image 2;

[0054] S1.3 presents the edge contour of the target in the target scene image 2.

[0055] Step 3: Process the obtained edge contours to obtain graphics that are the same as or similar to the candidate targets;

[0056] In a further embodiment, the method for processing the obtained edge contours to obtain graphics identical or similar to the target is as follows:

[0057] Fit the obtained edge contour of the target to obtain a graphic that is the same as or similar to the target shape, and use the graphic with similar size and center position as the candidate graphic.

[0058] Step 4: Filter the obtained candidate identical or similar graphics, and select the candidate identical or similar graphics whose size meets the set range as the target graphics.

[0059] In another embodiment, the size satisfies the set range: R smin <R s <R smax , where R smin = (LED light graphic dimensions / track width) × image width × 0.5, R smax = (LED light graphic size / track width) × image width × 1.5.

[0060] Step 5: Based on the number of target images n' obtained, determine whether the processing of the original target image is reasonable. If it is reasonable, proceed to the next step; otherwise, return to Step 1 to reprocess the original target scene image.

[0061] In a further embodiment, the criterion for determining whether the processing of the original target image is reasonable is: if 4≤n'≤n, then the processing of the original target image is qualified; otherwise, it is unqualified.

[0062] Step 6: Based on the coordinates of the target center in the UAV landing area {(X... i ,Y i Z i )|i=1,2,3...n}、The coordinates of the center of the target graphic in the image coordinate system {(x i ,y i Given the values ​​of |i = 1, 2, 3, ..., n'} and the focal length f of the monocular camera, establish a global objective function E:

[0063]

[0064] Where, m jk For the corresponding relationship weight coefficients, M = s(R1, T) x ), N = s(R2, T y ), s=f / T Z The translation matrix between the UAV landing area coordinate system and the camera coordinate system is T = [T x T y T z ] T The rotation matrix between the UAV landing area coordinate system and the camera coordinate system is R = [R1 R2 R3]. T w k (k = 1, 2, 3…n) = 1, S k =(X k ,Y k Z k ) T .

[0065] Step 7: Iteratively optimize the global objective function E using the Levenberg-Marquardt algorithm, and finally output the corresponding relational weight coefficients {m}. jk |i=1, 2, 3, ..., n, j=1, 2, 3, ..., n}, the rotation matrix R between the UAV landing area coordinate system and the camera coordinate system, and the translation matrix T between the UAV landing area coordinate system and the camera coordinate system.

[0066] Step 8: Given that R' and T' are the rotation and translation matrices between the camera coordinate system and the UAV coordinate system, respectively. Based on the rotation matrix R from the UAV landing area coordinate system to the camera coordinate system and the translation matrix T from the UAV landing area coordinate system to the camera coordinate system, solve for the final UAV pose parameters: the rotation matrix R from the UAV landing area coordinate system to the UAV coordinate system. Z The translation matrix T from the UAV landing area coordinate system to the UAV coordinate system. Z :

[0067]

[0068] After obtaining the pose of the UAV in the UAV coordinate system, the translation matrix T can be used to determine the pose. Z By obtaining the pose of the UAV relative to the coordinate system of the landing area, the accuracy and safety of UAV landing can be improved.

[0069] To gain a better understanding of the UAV landing visual pose measurement device and method provided by the present invention, a detailed description is provided below with reference to specific examples and accompanying drawings.

[0070] A method for measuring the landing visual pose of a drone, using ten circular targets as an example, such as... Figure 4 As shown, a light-diffusing plate 2 is set on the LED lights, and the targets are scattered on the UAV's landing runway at different positions. In this embodiment, the targets appear in pairs. The coordinate system of the UAV landing area is set as XYZ, the camera coordinate system is X1Y1Z1, and the UAV coordinate system is X2Y2Z2.

[0071] A method for measuring the visual pose of a drone during landing includes the following steps:

[0072] Step 1: After the drone flies to the effective area above the landing area, the monocular camera captures the original target scene image;

[0073] Step 2: Process the original target scene image to extract the edge contour of the target in the target scene image;

[0074] In a further embodiment, a method for processing the original target scene image to extract the edge contour of the target in the target scene image includes the following steps:

[0075] S1.1 Corrects the linear distortion of the original scene image to obtain the target scene image 1;

[0076] S1.2. The threshold H of the target scene image 1 is obtained by using the OTSU algorithm, and the target scene image 1 is binarized using the threshold H to obtain the target scene image 2.

[0077] S1.3 presents the edge contour of the target in the target scene image 2.

[0078] Step 3: Process the obtained edge contours to obtain alternative circular or elliptical shapes;

[0079] In a further embodiment, the method for processing the obtained edge contour to obtain candidate circular or elliptical shapes is as follows:

[0080] The edge contours of the target scene image 2 are extracted using the EDLines algorithm. The angle θ1 between each pair of edge contour segments is calculated. Edge contour segments with θ1 less than 60 degrees are connected to obtain candidate circles or elliptical arc segments. Candidate circles or elliptical arc segments with similar radii and center positions are fitted as candidate circles or ellipses.

[0081] Step 4: Filter the obtained blank circles or ellipses and select circles or ellipses whose size meets the set range as the target circles or ellipses.

[0082] In another embodiment, the size satisfies the set range: R smin <R s <R smax , where Rsmin = (Radius of the LED light graphic / Track width) × Image width × 0.5, R smax = (radius of the LED light graphic / track width) × image width × 1.5.

[0083] Step 5: Based on the number n' of the obtained target images, determine whether the processing of the original target image is reasonable. The standard for determining whether the processing of the original target image is reasonable is: 4≤n'≤n, then the processing of the original target image is qualified; otherwise, it is unqualified. If it is qualified, proceed to the next step; if it is unqualified, return to step 1 to reprocess the original target scene image. If n'<4, decrease H; if n'>n, increase H, and repeat steps S1.1, S1.2, and S1.3.

[0084] Step 6: Based on the coordinates of the target center in the UAV landing area {(X... i ,Y i Z i )|i=1,2,3...n}、The coordinates of the center of the target graphic in the image coordinate system {(x i ,y i Given the values ​​of |i = 1, 2, 3, ..., n'} and the focal length f of the monocular camera, establish a global objective function E:

[0085]

[0086] Where, m jk For the corresponding relationship weight coefficients, M = s(R1, T) x ), N = s(R2, T y ), s=f / T Z The translation matrix between the UAV landing area coordinate system and the camera coordinate system is T = [T x T y T z ] T The rotation matrix between the UAV landing area coordinate system and the camera coordinate system is R = [R1 R2 R3]. T w k (k = 1, 2, 3…n) = 1, S k =(X k ,Y k Z k ) T .

[0087] Step 7: Iteratively optimize the global objective function E using the Levenberg-Marquardt algorithm, and finally output the corresponding relational weight coefficients {m}. jk|i=1, 2, 3, ..., n, j=1, 2, 3, ..., n}, the rotation matrix R between the UAV landing area coordinate system and the camera coordinate system, and the translation matrix T between the UAV landing area coordinate system and the camera coordinate system.

[0088] Step 8: Given that R' and T' are the rotation and translation matrices between the camera coordinate system and the UAV coordinate system, respectively. Based on the rotation matrix R from the UAV landing area coordinate system to the camera coordinate system and the translation matrix T from the UAV landing area coordinate system to the camera coordinate system, solve for the final UAV pose parameters: the rotation matrix R from the UAV landing area coordinate system to the UAV coordinate system. Z The translation matrix T from the UAV landing area coordinate system to the UAV coordinate system. Z :

[0089]

[0090] After obtaining the pose of the UAV in the UAV coordinate system, the translation matrix T can be used to determine the pose. Z Obtain the pose of the UAV relative to the coordinate system of the landing area.

[0091] In summary, the UAV landing visual pose measurement device and method provided by this invention have at least the following advantages compared to the prior art:

[0092] (1) This invention only requires simple installation of a monocular camera, and does not require extensive calibration in the early stage, saving time;

[0093] (2) This invention establishes matching coefficients for multiple target 3D-2D coordinates and continuously updates them during the iterative solution of the optimization model. This avoids pose measurement failure due to incorrect matching of target 3D-2D coordinates.

[0094] (3) The present invention is simple to implement, has low engineering difficulty, and has high practical value.

[0095] For ease of description, spatial relative terms such as "above," "on top of," "on the upper surface of," "above," etc., are used herein to describe the spatial positional relationship of a device or feature as shown in the figures to other devices or features. It should be understood that spatial relative terms are intended to encompass different orientations in use or operation beyond the orientation of the device as described in the figures. For example, if the device in the figures were inverted, a device described as "above" or "on top of" other devices or structures would subsequently be positioned as "below" or "under" other devices or structures. Thus, the exemplary term "above" can include both "above" and "below." The device may also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatial relative descriptions used herein will be interpreted accordingly.

[0096] Furthermore, it should be noted that the use of terms such as "first" and "second" to define components is merely for the purpose of distinguishing the corresponding components. Unless otherwise stated, the above terms have no special meaning and therefore should not be construed as limiting the scope of protection of this invention.

[0097] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for measuring the visual pose of a UAV during landing, characterized in that, The method employs a UAV landing visual pose measurement device comprising a UAV, a monocular camera, a vision processor, and n targets. The n targets are fixed on the UAV landing runway, and each target is equipped with several infrared LEDs. The monocular camera is mounted below the UAV to acquire the original target scene image of the landing area. The monocular camera is connected to the vision processor and transmits the acquired original target scene image to the vision processor. The vision processor processes the original target scene image, extracts the target features in the target scene image, and calculates the relative pose of the UAV relative to the landing area based on the target features. The method includes the following steps: After the drone flies to the effective area above the landing zone, the monocular camera captures the original target scene image; The original target scene image is processed to extract the edge contour of the target in the target scene image; The obtained edge contours are processed to obtain graphics that are the same as or similar to the candidate targets; Filter the obtained candidate identical or similar graphics, and select the candidate identical or similar graphics whose size meets the set range as the target graphics; Based on the number of target graphics n' obtained, determine whether the processing of the original target image is reasonable. If it is reasonable, proceed to the next step; otherwise, return to reprocess the original target scene image. Based on the coordinates of the target center in the UAV landing area {(X i ,Y i Z i ) |i=1,2,3...n}、The coordinates of the center of the target graphic in the image coordinate system {(x i ,y i Given |i=1,2,3...n'} and the focal length f of the monocular camera, establish a global objective function E: , where m jk For the corresponding relationship weight coefficients, M=s(R1, T) x ), N=s(R2, T y ), s=f / T Z The translation matrix between the UAV landing area coordinate system and the image coordinate system is T=[T x T y T z ] T The rotation matrix between the UAV landing area coordinate system and the image coordinate system is R = [R1 R2 R3]. T w k (k=1,2,3…n)=1, S k =( X k ,Y k Z k ) T ; The corresponding weight coefficients m are obtained by solving the global objective function. jk The rotation matrix R between the UAV landing area coordinate system and the image coordinate system; the translation matrix T between the UAV landing area coordinate system and the image coordinate system. Based on the rotation matrix R' and translation matrix T' between the camera coordinate system and the UAV coordinate system, and the corresponding weighting coefficient m jk Given the rotation matrix R between the UAV landing area coordinate system and the image coordinate system, and the translation matrix T between the UAV landing area coordinate system and the image coordinate system, solve for the UAV landing pose.

2. The method for measuring the visual pose of a UAV landing as described in claim 1, characterized in that, The n mentioned is greater than 10.

3. The method for measuring the visual pose of a UAV landing as described in claim 2, characterized in that, The heights of the targets are not the same.

4. A method for measuring the visual pose of a UAV landing as described in claim 2 or 3, characterized in that, The infrared LED light is equipped with a light-diffusing plate.

5. A method for measuring the visual pose of a UAV landing as described in claim 2 or 3, characterized in that, An infrared filter is installed on the lens of the monocular camera.

6. The method for measuring the visual pose of a UAV landing as described in claim 1, characterized in that, The method for processing the original target scene image and extracting the edge contour of the target in the target scene image includes the following steps: Linear distortion of the original scene image is corrected to obtain target scene image 1; Find the threshold H of the target scene image 1, and use the threshold H to binarize the target scene image 1 to obtain the target scene image 2. The edge contour of the target in the target scene image 2 is presented.

7. The method for measuring the visual pose of a UAV landing as described in claim 1, characterized in that, The method for processing the obtained edge contours to obtain graphics that are the same as or similar to the target is as follows: Fit the obtained edge contour of the target to obtain a graphic that is the same as or similar to the target shape, and use the graphic with similar size and center position as the candidate graphic.

8. The method for measuring the visual pose of a UAV landing as described in claim 7, characterized in that, The size meets the set range: R smin <R s < R smax , where R smin = (LED light graphic dimensions / track width) × image width × 0.5, R smax = (LED light graphic size / track width) × image width × 1.

5.

9. The method for measuring the visual pose of a UAV landing as described in claim 1, characterized in that, The standard for judging whether the processing of the original target image is reasonable is: if 4≤n'≤n, the processing of the original target image is qualified; otherwise, it is unqualified.

Images