A method of visual target positioning for UAV based on DEM

Abstract

The present invention discloses a method for visual target positioning of an unmanned aerial vehicle based on a DEM, which comprises the following steps: obtaining the coordinates of a geodetic coordinate system of a solution site corresponding to a target; obtaining the angle between the height direction of the unmanned aerial vehicle and the line connecting the unmanned aerial vehicle and the target, and the angle between the height direction of the unmanned aerial vehicle and the line connecting the unmanned aerial vehicle and the orthographic projection site; correcting the position of the orthographic projection site according to the difference in the angle; obtaining the longitude and latitude of the correction site; obtaining the angle between the height direction of the unmanned aerial vehicle and the line connecting the unmanned aerial vehicle and the correction site; when the difference in the angle meets a preset condition, outputting the position information of the corresponding correction site. The present invention can obtain the elevation information of the theoretical solution site based on a DEM query, and use this to obtain the position information of the orthographic projection site of the solution site on the surface, and further, through a continuous iteration method, the position of the orthographic projection site is continuously approached to the target, so as to avoid the influence of the surface undulation on the calculation result of the target longitude and latitude.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) positioning, and in particular to a DEM-based UAV visual target positioning method. Background Art

[0002] In the prior art, drones mostly calculate the position of targets through onboard positioning modules. However, when a drone is flying in the air, it can only obtain the altitude of the drone itself. In the case of uneven terrain, the drone cannot perceive its actual distance from the ground, and thus cannot accurately obtain the actual height of the ground target, which affects the target positioning result. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention provides a DEM-based UAV visual target positioning method, which can obtain the elevation information of the theoretical solution site based on DEM query, and use this to solve the position information of the orthographic projection site of the site on the surface, and further based on the principle that the angle between the UAV height and the line connecting the UAV and the target remains unchanged, through continuous iteration, the position of the orthographic projection site is continuously approached to the target, so as to avoid the influence of surface undulation on the calculation results of the target latitude and longitude.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A DEM-based UAV visual target positioning method is provided, which includes the following steps:

[0006] Obtain the geodetic coordinate system coordinates of the solution point P1 corresponding to the target P, and query the elevation information of the solution point P1 through DEM;

[0007] Obtain the angle α between the height direction of the UAV and the line connecting the UAV and the target P, and the angle β between the height direction of the UAV and the line connecting the UAV and the orthographic projection point P2;

[0008] The position of the orthographic projection point P2 is corrected according to the difference between the angle α and the angle β to obtain a corrected point P3 of the orthographic projection point P2;

[0009] Obtain the latitude and longitude of the correction point P3, and obtain the elevation H2 of the correction point P3 through DEM query;

[0010] Obtain the angle β1 between the altitude direction of the UAV and the line connecting the UAV and the correction point P3;

[0011] Take β=β1 and repeat the above steps until the difference between the angles α and β meets the preset conditions, output the position information of the corrected point P3 corresponding to the angle β1 at this time, and use the longitude and latitude information of the corrected point P3 as the longitude and latitude information of the target P.

[0012] Preferably, the geodetic coordinate system coordinates of the solution point P1 are obtained by:

[0013] Obtain the current position information of the drone through the drone's flight control system;

[0014] The current attitude information of the gimbal is obtained through the servo control system of the gimbal mounted on the UAV;

[0015] An image containing the target P is acquired by an imaging device carried on the UAV;

[0016] Based on the current posture information of the UAV and the current posture information of the gimbal, the pixel coordinates of the target P are converted into geographic coordinate system coordinates through the pinhole imaging principle and the coordinate system conversion principle;

[0017] The geographic coordinate system coordinates of the target P are converted into the geodetic coordinate system coordinates of the solution point P1 corresponding to the target P.

[0018] Preferably, the current posture information of the UAV includes one or more of heading angle, roll angle, pitch angle, longitude, latitude and altitude; the current posture information of the gimbal includes one or more of yaw angle, pitch angle and roll angle.

[0019] Preferably, if β≥α+γ, the position of the orthographic projection site P2 is corrected to obtain a corrected site P3 of the orthographic projection site P2, wherein the corrected site P3 is a site on the ground with a horizontal distance L2 from the drone, and L2=L1-s×L1, s is a distance attenuation factor with a value range of 0<s<1; γ is the angle error threshold.

[0020] Preferably, if β≤α-γ, the position of the orthographic projection site P2 is corrected to obtain a corrected site P3 of the orthographic projection site P2, wherein the corrected site P3 is a site on the ground with a horizontal distance L2 from the drone, and L2=L1+s×L1, s is a distance attenuation factor with a value range of 0<s<1; γ is the angle error threshold.

[0021] Preferably, the value range of s is 0.1-0.5.

[0022] Preferably, the value range of γ is [0.5°, 5°].

[0023] Preferably, β=β1, repeat the above steps, and when |β-α|<γ, output the position information of the correction point P3 corresponding to the angle β1 at this time, and use the longitude and latitude information of the correction point P3 as the longitude and latitude information of the target P.

[0024] The present invention can obtain the elevation information of the theoretical solution point P1 based on DEM query, and use this to solve the position information of the orthographic projection point P2 of the point P1 on the surface. Further, based on the principle that the angle α between the height of the UAV and the line connecting the UAV and the target P remains unchanged, through continuous iteration, the position of the orthographic projection point P2 is continuously approached to the target P, and when the iterative calculation result meets the conditions, the longitude and latitude information of the corrected point P3 is output as the longitude and latitude information of the target P, so as to improve the accuracy of the longitude and latitude of the target P and avoid the influence of surface undulation on the calculation result of the longitude and latitude of the target P. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a flowchart of the steps of the DEM-based UAV visual target positioning method in the present invention;

[0026] Figure 2 It is a schematic diagram of the positional relationship among the elevation H1 of the target P, the solution point P1, and the orthographic projection point P2 in the present invention;

[0027] Figure 3 It is a schematic diagram of the positional relationship of the elevation H1 of the target P, the solution point P1, and the orthographic projection point P2 when the orthographic projection point P2 is on the right side of the target P in the present invention;

[0028] Figure 4 It is a schematic diagram of the positional relationship among the elevation H1 of the target P, the solution point P1, and the orthographic projection point P2 when the orthographic projection point P2 is on the left side of the target P in the present invention. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] Embodiment 1:

[0031] like Figure 1 As shown, this embodiment provides a DEM-based UAV visual target positioning method, which includes the following steps:

[0032] S1, obtaining the geodetic coordinate system coordinates of the solution point P1 corresponding to the target P through the current posture information of the UAV and the current posture information of the gimbal carried on the UAV, wherein the geodetic coordinate system coordinates of the solution point P1 include one or more of the longitude, latitude and altitude of the solution point P1;

[0033] Specifically, the geodetic coordinate system coordinates of the solution point P1 are obtained in the following manner:

[0034] Acquiring the current posture information of the drone through the flight control system of the drone, wherein the current posture information includes one or more of heading angle, roll angle, pitch angle, longitude, latitude and altitude;

[0035] Acquiring current attitude information of the gimbal through a servo control system of the gimbal carried on the unmanned aerial vehicle, wherein the current attitude information includes one or more of a yaw angle, a pitch angle, and a roll angle;

[0036] An image containing the target P is acquired by an imaging device (such as a visible light camera, etc.) carried on the UAV;

[0037] Based on the current posture information of the UAV and the current posture information of the gimbal, the pixel coordinates of the target P are converted into geographic coordinate system coordinates through the pinhole imaging principle and the coordinate system conversion principle;

[0038] The geographic coordinate system coordinates of the target P are converted into the geodetic coordinate system coordinates of the solution point P1 corresponding to the target P by combining the major and minor axes of the earth and the eccentricity of the earth;

[0039] like Figure 1 As shown in the figure, the target P to be positioned is actually located on an undulating surface. Since the UAV cannot know the height change of the surface during flight, the position information calculated based on the current posture information of the UAV and the current posture information of the gimbal (i.e., the geodetic coordinate system coordinates of the calculated position point P1) is actually a theoretical calculation result and cannot actually reflect the real position of the target P.

[0040] S2, querying the elevation information of the solution point P1 through DEM (Digital Elevation Model) to determine the elevation H1 of the orthographic projection point P2 of the solution point P1 on the surface according to the elevation information, and the longitude and latitude of the solution point P1 and the orthographic projection point P2 are the same;

[0041] S3, such as Figure 2 As shown, the angle α between the height direction of the drone UVA and the line connecting the drone UVA and the target P, and the angle β between the height direction of the drone UVA and the line connecting the drone UVA and the positive projection position P2 are obtained, and the angle α=arctan(L1 / H0), and the angle β=arctan(L1 / (H0-H1));

[0042] Among them, H0 is the height of the drone UVA; L1 is the horizontal distance between the drone UVA and the solution site P1, that is, the horizontal distance between the drone UVA and the orthographic projection site P2, which can be obtained by taking the absolute value after subtracting the latitude and longitude of the drone UVA and the solution site P1;

[0043] The latitude and longitude of the UAV UVA and the height H0 can be directly obtained by the UAV flight control system. When the positions of the UAV, UVA and the target P remain unchanged, the angle α is a constant value;

[0044] S4, such as Figure 3 As shown, the position of the orthographic projection point P2 is corrected according to the difference between the angle α and the angle β to obtain a corrected point P3 of the orthographic projection point P2;

[0045] Specifically, if β≥α+γ, it means that the orthographic projection point P2 is on the right side of the target P, then the position of the orthographic projection point P2 is corrected to obtain a corrected point P3 of the orthographic projection point P2, wherein the corrected point P3 is a point on the ground surface and has a horizontal distance L2 from the drone UVA, and L2=L1-s×L1;

[0046] Wherein, s is the distance attenuation factor, and its value range is 0<s<1; γ is the angle error threshold, which can be set according to the positioning accuracy requirement. For example, in this embodiment, the value range of s is 0.1-0.5, and the value range of γ is [0.5°, 5°];

[0047] For example, in this embodiment, when the angle α=30° and γ=1°, 31° (i.e., α+γ) is the allowable deviation angle. If β=35° at this time, it means that the positive projection point P2 deviates too much from the target P and needs to be corrected to make it closer to the target P. Therefore, the corrected point P3 is obtained by further reducing the horizontal distance L1.

[0048] S5. Obtain the longitude and latitude of the correction point P3 according to the longitude and latitude of the drone UVA and the horizontal distance L2, and obtain the elevation H2 of the correction point P3 through DEM query based on the longitude and latitude of the correction point P3;

[0049] and obtaining an angle β1 between the height direction of the drone UVA and the line connecting the drone UVA and the correction point P3, wherein the angle β1=arctan(L2 / (H0-H2));

[0050] For example, in this embodiment, since the horizontal distance L2 can be obtained by taking the absolute value of the difference between the longitude and latitude of the drone UVA and the correction point P3, the longitude and latitude of the correction point P3 can be calculated back under the premise that the longitude and latitude of the drone UVA and the horizontal distance L2 are known;

[0051] In this embodiment, the "horizontal distance" is the vertical distance from the site (including the calculated site P1, the orthographic projection site P2, and the corrected site P3) to the height direction of the drone;

[0052] S6. Take β=β1 and repeat steps S4-S5 until |β-α|<γ, output the position information of the correction point P3 corresponding to the angle β1 at this time, and use the longitude and latitude information of the correction point P3 as the longitude and latitude information of the target P.

[0053] When steps S4-S5 are repeated, the correction point P3 will continue to approach the target P. When |β-α|<γ is satisfied, it means that the position deviation of the correction point P3 and the target P is within the allowable range, and the two are very close. At this time, the longitude and latitude of the correction point P3 can be directly used as the longitude and latitude of the target P.

[0054] Therefore, this embodiment can obtain the elevation information of the theoretical solution site P1 based on the DEM query, and use this to solve the position information of the orthographic projection site P2 of the site P1 on the surface, and further based on the principle that the angle α between the UAV height H0 and the line connecting the UAV and the target P is constant, through continuous iteration, the position of the orthographic projection site P2 is continuously approached to the target P (that is, the process of continuously solving the longitude and latitude of the correction site P3), and when the iterative calculation result meets the condition (that is, |β-α|<γ), the longitude and latitude information of the correction site P3 at this time is output as the longitude and latitude information of the target P, so as to improve the accuracy of the longitude and latitude, and avoid the influence of the surface undulation on the longitude and latitude calculation results of the target P.

[0055] Embodiment 2:

[0056] The difference between this embodiment and embodiment 1 is that Figure 4 As shown in the figure, due to different values ​​of s, the correction point P3 may appear on the left side of the target P. At this time, it is also necessary to perform iterative calculation to make the correction point P3 approach the target P continuously.

[0057] Therefore, the step S4 also includes the following steps:

[0058] If β≤α-γ, it means that the correction point P3 is on the left side of the target P, and the position of the orthographic projection point P2 is corrected to obtain the correction point P3 of the orthographic projection point P2. The correction point P3 is a point on the ground and has a horizontal distance L2 from the drone, and L2=L1+s×L1. For example, in this embodiment, when the angle α=30° and γ=1°, 29° (i.e., α-γ) is also the allowable deviation angle. If β=26° at this time, it means that the correction point P3 also deviates too much from the target P and also needs to be corrected to make it closer to the target P. Therefore, the correction point P3 is further brought closer to the target P by increasing the horizontal distance L1.

[0059] Therefore, when the correction point P3 jumps horizontally on the left and right sides of the target P, the horizontal distance L2 can be corrected to make the correction point P3 approach the target P continuously, and the longitude and latitude information of the correction point P3 can be obtained through iterative calculation as the longitude and latitude information of the target P.

[0060] In summary, the present invention can obtain the elevation information of the theoretical solution point P1 based on DEM query, and use this to solve the position information of the orthographic projection point P2 of the point P1 on the surface, and further based on the principle that the angle α between the UAV height H0 and the line connecting the UAV and the target P is constant, through continuous iteration, the position of the orthographic projection point P2 is continuously approached to the target P, and when the iterative calculation result meets the conditions, the longitude and latitude information of the corrected point P3 is output as the longitude and latitude information of the target P, so as to improve the accuracy of the longitude and latitude, and avoid the influence of the surface undulation on the longitude and latitude calculation results of the target P.

[0061] It should be noted that the technical features in the above-mentioned embodiments 1-2 can be combined arbitrarily, and the combined technical solutions all belong to the protection scope of the present application. In this article, terms such as "including", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the presence of other identical elements in the process, method, article or equipment including the elements.

[0062] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A DEM-based UAV visual target positioning method, characterized in that: The steps include: Obtain the geodetic coordinate system coordinates of the solution point P1 corresponding to the target P, and query the elevation information of the solution point P1 through DEM; Obtain the angle α between the height direction of the UAV and the line connecting the UAV and the target P, and the angle β between the height direction of the UAV and the line connecting the UAV and the orthographic projection point P2; The position of the orthographic projection point P2 is corrected according to the difference between the angle α and the angle β to obtain a corrected point P3 of the orthographic projection point P2; Obtain the latitude and longitude of the correction point P3, and obtain the elevation H2 of the correction point P3 through DEM query; Obtain the angle β1 between the altitude direction of the UAV and the line connecting the UAV and the correction point P3; Take β=β1 and repeat the above steps until the difference between the angles α and β meets the preset conditions, output the position information of the corrected point P3 corresponding to the angle β1 at this time, and use the longitude and latitude information of the corrected point P3 as the longitude and latitude information of the target P.

2. The UAV visual target positioning method according to claim 1, characterized in that: The geodetic coordinate system coordinates of the solution point P1 are obtained in the following way: Obtain the current position information of the drone through the drone's flight control system; The current attitude information of the gimbal is obtained through the servo control system of the gimbal mounted on the UAV; An image containing the target P is acquired by an imaging device carried on the UAV; Based on the current posture information of the UAV and the current posture information of the gimbal, the pixel coordinates of the target P are converted into geographic coordinate system coordinates through the pinhole imaging principle and the coordinate system conversion principle; The geographic coordinate system coordinates of the target P are converted into the geodetic coordinate system coordinates of the solution point P1 corresponding to the target P.

3. The UAV visual target positioning method according to claim 2, characterized in that: The current posture information of the drone includes one or more of the heading angle, roll angle, pitch angle, longitude, latitude and altitude; the current posture information of the gimbal includes one or more of the yaw angle, pitch angle and roll angle.

4. The UAV visual target positioning method according to claim 1, characterized in that: If β≥α+γ, the position of the orthographic projection site P2 is corrected to obtain the corrected site P3 of the orthographic projection site P2. The corrected site P3 is a site on the ground with a horizontal distance L2 from the drone, and L2=L1-s×L1, s is the distance attenuation factor, and the value range is 0<s<1; γ is the angle error threshold.

5. The UAV visual target positioning method according to claim 1, characterized in that: If β≤α-γ, the position of the orthographic projection site P2 is corrected to obtain the corrected site P3 of the orthographic projection site P2. The corrected site P3 is a site on the ground with a horizontal distance L2 from the drone, and L2=L1+s×L1, s is the distance attenuation factor, and the value range is 0<s<1; γ is the angle error threshold.

6. The unmanned aerial vehicle visual target positioning method according to claim 4 or 5, characterized in that: The value range of s is 0.1-0.

5.

7. The UAV visual target positioning method according to claim 4 or 5, characterized in that: The value range of γ is [0.5°, 5°].

8. The unmanned aerial vehicle visual target positioning method according to claim 1, characterized in that: Take β=β1, repeat the above steps, and when |β-α|<γ, output the position information of the corrected point P3 corresponding to the angle β1 at this time, and use the longitude and latitude information of the corrected point P3 as the longitude and latitude information of the target P.

Images