A hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components

Abstract

The present invention discloses a hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components, comprising: a reflector group, which is coated with an ultraviolet film and is used to provide reflected direct sunlight to a hyperspectral UAV payload; a solar tracking device, which is used to track the sun's position in real time and feed back first posture information of the solar tracking device to a control end; an UAV tracking device, which is used to track the UAV payload position in real time and feed back second posture information of the UAV tracking device to the control end; a pan-tilt device, which is used to support the reflector group, the solar tracking device, and the UAV tracking device, and control the rotation of the reflector group based on received control information; a control end, which is used to control the real-time tracking of the solar tracking device and the UAV tracking device, calculate the control information of the reflector group based on the first posture information and the second posture information, and realize high-quality auxiliary remote sensing imaging of ultraviolet band absorption components.

Description

Technical Field

[0001] The invention belongs to the technical field of optical remote sensing, and in particular relates to a hyperspectral unmanned aerial vehicle remote sensing auxiliary system for ultraviolet band absorption components. Background Art

[0002] Global atmospheric environmental problems have always threatened people's lives and health, especially VOCs pollution, which not only irritates the eyes and respiratory tract, causes skin allergies, causes headaches, sore throats and fatigue, but also causes cancer. In order to deal with environmental problems in a targeted manner, it is necessary to effectively monitor pollutant emissions. For industrial areas such as chemical plants and product processing plants with complex and high-concentration atmospheric pollution emissions, current means cannot effectively monitor and trace the source. Specifically, environmental remote sensing satellites that are easily affected by cloud cover have seriously insufficient temporal and spatial resolution, cannot distinguish between different factory areas, and cannot trace the source. The vehicle-mounted mobile VOCs and odorous gas mass spectrometer and pollution source precise locking system with publication number CN108415038A cannot observe deeply inside the factory area, and are easily discovered by industrial parks, causing the park to suspend heavy pollutant emissions, and it is also difficult to trace the source. For example, a remote sensing imaging aerial photography drone disclosed in publication number CN108016628A, the point instrument carried on the drone must also enter the park, and the measurement range is very limited, the efficiency is extremely low and it is difficult to trace the source. Therefore, a hyperspectral remote sensing imaging payload carried on a small UAV is needed, which can perform horizontal observations with meter-level resolution and locate the specific location of emission sources, while quickly monitoring outside the factory and completing the monitoring task before the factory responds.

[0003] However, due to the limited flying altitude of small UAVs, the hyperspectral remote sensing imaging payload (referred to as hyperspectral UAV payload) carried on small UAVs is close to the ground. Therefore, compared with environmental remote sensing satellites, the scattered light path detected by the hyperspectral UAV payload is very limited, the received light flux is very weak, and the signal-to-noise ratio is difficult to meet the requirements of hyperspectral inversion of trace gases, especially for most VOCs with weak spectral absorption and slow absorption cross-section changes. Ultimately, the hyperspectral UAV payload carried on small UAVs cannot obtain effective results when observing the ground at large pitch angles. In order to fill the technical gap in monitoring and tracing atmospheric pollutants in industrial plants as soon as possible and realize miniaturized payloads carried on small UAVs, there is an urgent need for a hyperspectral UAV remote sensing auxiliary system for ultraviolet absorption components such as VOCs. Summary of the invention

[0004] In view of the above, the purpose of the present invention is to provide a hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components, which can increase the luminous flux obtained by the hyperspectral UAV remote sensing payload, so that the payload obtains a larger signal-to-noise ratio and can be inverted normally; make up for the problem that the hyperspectral UAV remote sensing payload cannot perform large-angle ground observation, and realize all-round detection of the payload; the increased luminous flux needs to reduce the integration time of the hyperspectral remote sensing imaging payload, further improving the time resolution of the payload imaging.

[0005] To achieve the above-mentioned purpose of the invention, the embodiment provides a hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components, including:

[0006] A reflector assembly coated with an ultraviolet film for providing solar reflected light after reflecting direct sunlight to the hyperspectral UAV payload;

[0007] A sun tracking device, which is used to track the sun's position in real time and feed back first attitude information of the sun tracking device to the control end;

[0008] A drone tracking device, which is used to track the drone payload position in real time and feed back the second posture information of the drone tracking device to the control end;

[0009] A pan-tilt device, which is used to support the reflector group, the sun tracking device, and the UAV tracking device, and controls the reflector group to rotate based on the received control information so that the hyperspectral UAV payload receives the high-luminous flux solar reflected light reflected by the reflector group;

[0010] The control end is used to control the real-time tracking of the sun tracking device and the UAV tracking device, and calculate the control information of the reflector group based on the first posture information and the second posture information.

[0011] Preferably, the reflector group includes two reflectors, there is a gap of less than 10 cm between the reflectors, the size of a single reflector is greater than 0.5m×1m, the radius of curvature of the reflector is greater than 50m, the coating band of each reflector covers at least 280nm-450nm, and the difference in reflectivity of the ultraviolet film at different wavelengths is less than 5%.

[0012] Preferably, the solar tracking device comprises a polarizing plate group, a first camera, a first pitch angle servo motor, a first azimuth angle servo motor, a first pitch angle sensor, a first azimuth angle sensor, and a first positioning module, wherein the polarizing plate group is arranged on the incident light path of the first camera to realize polarization extinction of direct sunlight, the first camera has a field of view angle less than 60° and a resolution greater than 720P, the first pitch angle servo motor and the first azimuth angle servo motor control the rotation of the first camera, the first pitch angle sensor and the first azimuth angle sensor collect the first pitch angle and the first azimuth angle of the solar tracking device to form the first posture information, and the first positioning module is used for spatial positioning of the solar tracking device;

[0013] The control end controls the sun tracking device to track in real time, including:

[0014] Coarse adjustment: obtaining the latitude and longitude information, the first attitude information and the time information of the spatial positioning of the first positioning module, and calculating the current solar altitude angle and the solar azimuth angle according to the obtained information, and controlling the first pitch angle servo motor and the first azimuth angle servo motor according to the current solar altitude angle and the solar azimuth angle to turn the first camera to the position of the sun to receive direct sunlight;

[0015] Fine adjustment: obtaining a first image captured by a first camera, identifying the sun based on the first image, aiming to make the center of the sun in the first image coincide with the geometric center of the first image, and achieving real-time tracking of the sun by rotating a first pitch angle servo motor and a first azimuth angle servo motor;

[0016] When tracking is not possible based on the first image, the process jumps to the coarse adjustment process.

[0017] Preferably, the polarizer group includes a first polarizer, a second polarizer, and a third polarizer arranged along the direction of the incident light, wherein the first polarizer and the second polarizer are fixed polarizers, which are used to weaken the intensity of direct sunlight, and the degree of weakening is more than 95%, and the third polarizer is a rotatable polarizer, which rotates according to the real-time feedback of direct sunlight intensity, so that the brightness of the direct sunlight spot received by the first camera remains relatively uniform in real time.

[0018] Preferably, identifying the sun based on the first image comprises:

[0019] First, the image is grayed out, then the shape is judged to identify the sun, and finally the outline and center of mass of the identified sun are drawn.

[0020] Preferably, the drone tracking device comprises a second camera, a second pitch angle servo motor, a second azimuth angle servo motor, a second pitch angle sensor, a second azimuth angle sensor, and a second positioning module, wherein the second camera has a small field of view angle of less than 30° and a resolution of more than 2K, the second pitch angle servo motor and the second azimuth angle servo motor control the rotation of the second camera, the second pitch angle sensor and the second azimuth angle sensor collect the second pitch angle and the second azimuth angle of the sun tracking device to form the second posture information, and the second positioning module is used for spatial positioning of the drone tracking device;

[0021] The control end controls the drone tracking device to track in real time, including:

[0022] Coarse adjustment: obtain the altitude and longitude and latitude of the hyperspectral UAV payload, as well as the longitude and latitude information of the second positioning module, calculate the relative position between the hyperspectral UAV payload and the UAV tracking device, and control the second pitch angle servo motor and the second azimuth angle servo motor according to the relative position to turn the second camera to the direction of the hyperspectral UAV payload;

[0023] Fine adjustment: obtain the second image captured by the second camera, identify the load center through the second image, aim to make the load center in the second image coincide with the geometric center of the second image, and realize real-time tracking of the hyperspectral UAV payload through the rotation of the second pitch angle servo motor and the second azimuth angle servo motor;

[0024] When tracking is not possible based on the second image, the process jumps to the coarse adjustment process.

[0025] Preferably, identifying the load center through the second image includes:

[0026] Color recognition is used to identify the hyperspectral UAV payload in the second image that is different from the background color;

[0027] or, using motion recognition to identify the hyperspectral UAV payload in the second image;

[0028] Alternatively, a recognition model built based on deep learning is used to identify the hyperspectral UAV payload in the second image.

[0029] Preferably, the pan-tilt device includes a servo motor turntable, a servo motor electric cylinder, a reflector support frame, a reflector fixing frame and a tracking device fixing frame.

[0030] The reflector support frame is fixed at both ends of the servo motor turntable, and is used to support the reflector fixing frame, which is used to fix the reflector group. The reflector support frame has a bearing, and the center of gravity of the reflector fixing frame and the reflector is on the center line of the reflector support frame bearing, so that the reflector group can rotate in the vertical direction;

[0031] The servo motor electric cylinder is fixed on the servo motor turntable, and its function is to convert the rotation of the servo motor into translation, and push the reflector fixing frame and the reflector group upward, so that the reflector group rotates in the vertical direction, that is, to control the pitch angle of the reflector group;

[0032] The servo motor turntable is fixed on the bracket, and drives the reflector group to rotate in the horizontal direction through the reflector support frame, that is, controls the azimuth angle of the reflector group;

[0033] The tracking device fixing frame passes through the hollow hole of the servo motor turntable and is directly fixed on the bracket, which is used to support and fix the solar tracking device and the drone tracking device.

[0034] Preferably, the control end calculates control information of the reflector group based on the first posture information and the second posture information, including:

[0035] The first attitude information includes a first pitch angle α 1 and the first azimuth angle β 1 , the second attitude information includes the second pitch angle α 2 and the second azimuth β 2 The calculated control information includes the third pitch angle α of the control mirror group 3 and the third position angle β 3 :

[0036]

[0037] in, is the normal vector of the mirror's reflection plane.

[0038] Preferably, the control terminal further includes:

[0039] When it is considered that the obtained α 3 and β 3 When correct, the normal vector of the reflector reflection plane is obtained

[0040]

[0041] Normal vector of the reflector's reflection plane obtained by vector operation

[0042]

[0043] when and If they are inconsistent, let β 3 =β 3 +π, to achieve error correction.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] A solar tracking device and its real-time tracking control method are designed, so that the solar tracking device can track the exact position of the sun in real time, and return the first attitude information of the sun relative to the reflector group to the control end; a drone tracking device and its real-time tracking control method are designed, so that the drone tracking device can accurately track the exact position of the drone and its hanging load in real time, and return the second attitude information of the drone relative to the reflector group to the control end; a gimbal device and its real-time tracking control method are designed, and after the control end calculates the first attitude information and the second attitude information, the reflector is turned to the appropriate position according to the control information, so that the direct sunlight can be reflected into the drone payload, changing the original way of receiving the sun's scattered light, and finally achieving the increase of the luminous flux of the hyperspectral drone remote sensing payload, so that the payload obtains a larger signal-to-noise ratio and can be inverted normally.

[0046] The present invention reflects direct sunlight into the instrument of the remote sensing drone payload, which is different from the previous instrument that receives scattered sunlight. The luminous flux of direct light is very large. If the instrument follows the previous form of scattered sunlight, the spectrometer in the instrument will be directly saturated, so it is necessary to significantly reduce the integration time and improve the time resolution to normally receive direct sunlight. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 1 is a schematic diagram of the structure of a hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components provided in an embodiment;

[0049] Figure 2 is a schematic diagram of a drone tracking device and a sun tracking device provided in an embodiment;

[0050] Figure 3 is a schematic structural diagram of a pan / tilt device provided in an embodiment;

[0051] Figure 4 is a schematic diagram of a fixing method of a reflector provided in an embodiment;

[0052] Figure 5 is a schematic diagram of calculating the rotation angle of the reflector provided in the embodiment; DETAILED DESCRIPTION

[0053] To make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific implementation methods described herein are only used to explain the present invention and do not limit the scope of protection of the present invention.

[0054] The inventive concept of the present invention is: to help hyperspectral UAV payloads solve the key technical problems of poor timeliness, poor effectiveness, and poor flexibility in existing pollution source monitoring, the embodiments of the present invention provide a hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components such as VOCs, which can reflect direct sunlight to the hyperspectral UAV payload. Since the sunlight signal received by the hyperspectral UAV payload is changed from the original ground scattered light to the reflected direct sunlight, the luminous flux obtained by the payload can be increased tenfold or even a hundredfold. While the payload greatly improves the signal-to-noise ratio, it must also greatly reduce the integration time of the spectrometer in the payload to prevent the saturation of the light signal received by the spectrometer. The lower integration time reduces the integration time of the hyperspectral payload from about 10s to about 0.3s, and the time resolution is greatly improved.

[0055] Based on the above inventive concept, the embodiment provides a hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components, such as Figure 1 As shown, it includes a reflector group 1, a solar tracking device 2, an unmanned aerial vehicle tracking device 3, a pan-tilt device 4, and a control terminal 5, wherein the reflector group 1 is used to provide solar reflected light after reflecting direct solar light to the hyperspectral unmanned aerial vehicle payload; the solar tracking device 2 is used to track the position of the sun in real time, and feed back first posture information of the solar tracking device to the control terminal 5; the unmanned aerial vehicle tracking device 3 is used to track the position of the unmanned aerial vehicle payload in real time, and feed back second posture information of the unmanned aerial vehicle tracking device to the control terminal 5; the pan-tilt device 4 is used to support the reflector group 1, the solar tracking device 2, and the unmanned aerial vehicle tracking device 3, and control the rotation of the reflector group 1 based on the received control information so that the hyperspectral unmanned aerial vehicle payload receives high-luminous flux solar reflected light reflected by the reflector group; the control terminal 5 is used to control the real-time tracking of the solar tracking device 2 and the unmanned aerial vehicle tracking device 3, and calculate the control information of the reflector group 1 based on the first posture information and the second posture information.

[0056] In the embodiment, the reflector group 1 provides the hyperspectral UAV payload with reflected sunlight, while the ultraviolet band absorption components such as VOCs are ultraviolet light and a small part of visible light. Therefore, the reflector group 1 must use an ultraviolet coated reflector, and the coating band should cover at least 280nm-450nm, and the difference in reflectivity of the ultraviolet film at different wavelengths should be less than 5%, so as to ensure that the reflectivity at different wavelengths should be as consistent as possible. In addition, in order to make it easier for the reflector group 1 to reflect direct sunlight to the UAV payload, the area of ​​the reflector group 1 should be as large as possible, and the reflector can be given an appropriate small curvature. Specifically, the diameter of a single reflector should be more than 0.5m×1m, and the curvature radius of the reflector needs to be greater than 50m. In order to prevent the reflector from touching the two tracking devices and the tracking device fixing frame, the reflector group 1 uses two reflectors, and there is a gap of less than 10cm between the reflectors. When the UAV is far away, the diffuse light spot reflected by the reflector will make up for this gap.

[0057] In the embodiment, Figure 2 As shown, the solar tracking device 2 and the drone tracking device 3 are fixed on the tracking device fixing frame 45, and the solar tracking device 2 includes a polarizing plate group 21, a first camera 22, a first pitch angle servo motor 23, a first azimuth angle servo motor 24, a first pitch angle sensor 25, a first azimuth angle sensor 26, and a first positioning module 27.

[0058] Among them, the polarizer group 21 is arranged on the incident light path of the first camera to realize polarization extinction of direct sunlight, that is, to prevent excessive sunlight from damaging the camera and affecting the normal operation of the camera by extinction of the polarizer. Specifically, the polarizer group 21 includes a first polarizer, a second polarizer, and a third polarizer arranged along the direction of the incident light, wherein the first polarizer and the second polarizer are fixed polarizers, which are used to weaken the intensity of direct sunlight, and the degree of weakening is more than 95%. Due to the different intensity of sunlight in different seasons and weather, in order to facilitate the camera and computer to perform image recognition, a third polarizer is required, specifically a rotatable polarizer, which rotates according to the real-time direct sunlight intensity fed back, so that the brightness of the direct sunlight spot received by the first camera remains relatively uniform in real time, thereby increasing the success rate of recognition and tracking of the sun tracking device 2.

[0059] More specifically, since sunlight is natural light that does not directly show polarization and has an intensity of I, the sunlight becomes linearly polarized light with an intensity of 0.5I after passing through the first polarizer. According to Malus's law, by adjusting the relative angle between the second polarizer and the first polarizer, the sunlight after passing through the second polarizer can be converted into linearly polarized light with an intensity less than 0.05I.

[0060] Since the volume of the sun is very large and its position changes very slowly, the first camera 22 of the solar tracking device 1 can use a larger field of view, which is less than 60° and has a resolution higher than 720P, for example, a field of view of 45° and a resolution of 4K. The first pitch angle servo motor 23 and the first azimuth angle servo motor 24 control the rotation of the first camera 21 to achieve control of the pitch angle and azimuth angle. At the same time, the first pitch angle sensor 25 and the first azimuth angle sensor 26 collect the first pitch angle and the first azimuth angle of the solar tracking device 2 in real time, form the first posture information and feed it back to the control terminal 5. The first positioning module 27 is used for spatial positioning of the solar tracking device 2, and can be set on the tracking device fixing frame 45.

[0061] The real-time tracking control of the sun tracking device 2 is implemented by the control terminal 5. The specific control process includes:

[0062] Coarse adjustment: obtain the latitude and longitude information, first attitude information and time information of the spatial positioning of the first positioning module, and calculate the current solar altitude angle and solar azimuth angle according to the obtained information, and control the first pitch angle servo motor 23 and the first azimuth angle servo motor 24 according to the current solar altitude angle and solar azimuth angle to turn the first camera 21 to the position of the sun to receive direct sunlight. Due to the limited accuracy of the calculation, the next step is fine adjustment;

[0063] Fine adjustment: obtaining a first image captured by a first camera, and identifying the sun based on the first image, with the goal of coinciding the center of the sun in the first image with the geometric center of the second image, and achieving real-time tracking of the sun by rotating the first pitch angle servo motor and the first azimuth angle servo motor;

[0064] When tracking is not possible based on the first image, the process jumps to the coarse adjustment process.

[0065] The image recognition of the solar tracking device 2 should adopt spot recognition, which specifically includes: firstly graying the first image, then making a shape judgment, identifying the sun, and drawing the outline and center of mass of the sun, and finally, through the rotation of the servo motor, making the center of mass coincide with the center of the sun and the center of the image to realize real-time tracking.

[0066] In the embodiment, Figure 2 As shown, the drone tracking device 3 includes a second camera 31, a second pitch angle servo motor 32, a second azimuth angle servo motor 33, a second pitch angle sensor 34, a second azimuth angle sensor 35, and a second positioning module 36. Since the flight altitude of the drone is often above 100m, in order to enable the second high-definition camera 31 to see the drone clearly, that is, the drone image occupies more pixels in the overall camera image, the second camera 31 needs to have a small field of view of less than 30° and a resolution higher than 2K, for example, a field of view of 25° and a resolution of 4K.

[0067] The second pitch angle servo motor 32 and the second azimuth angle servo motor 33 control the rotation of the second camera 31 to respectively control the second pitch angle and the second azimuth angle. Since the object to be tracked is far away from the system, a very high-precision motor should be used, such as a servo motor with an accuracy of 0.02°. All servo motors mentioned in the present invention include appropriate reducers. The second pitch angle sensor 34 and the second azimuth angle sensor 35 collect the second pitch angle and the second azimuth angle of the solar tracking device in real time, and form the second posture information to be fed back to the control terminal 5. The second positioning module 36 is used for spatial positioning of the drone tracking device 3, and can also be set on the tracking device fixing frame 45, and can be the same module as the first positioning module 27, that is, the simultaneous positioning of the solar tracking device 2 and the drone tracking device 3 is achieved through one positioning module. The first positioning module 27 and the second positioning module 36 can be GPS modules.

[0068] The real-time tracking control of the drone tracking device 3 is implemented by the control terminal 5. The specific control process includes:

[0069] Coarse adjustment: obtain the altitude and longitude and latitude of the hyperspectral UAV payload, as well as the longitude and latitude information of the second positioning module 36, calculate the relative position between the hyperspectral UAV payload and the UAV tracking device 3, and control the second pitch angle servo motor 32 and the second azimuth angle servo motor 33 according to the relative position to turn the second camera 31 to the direction of the hyperspectral UAV payload. Due to the limited accuracy of GPS, the next step is fine adjustment;

[0070] Fine adjustment: obtaining the second image captured by the second camera 31, identifying the load center through the second image, aiming to make the load center in the second image coincide with the geometric center of the second image, and realizing real-time tracking of the hyperspectral UAV load through the rotation of the second pitch angle servo motor 32 and the second azimuth angle servo motor 33;

[0071] When the drone flies too fast and cannot be tracked based on the second image, the process jumps to the coarse adjustment process.

[0072] The payload center is identified through the second image, including color recognition, motion recognition and target detection. That is, color recognition is used to identify the hyperspectral UAV payload that is different from the background color in the second image; or, motion recognition is used to identify the hyperspectral UAV payload in the second image; or, a recognition model based on deep learning is used to identify the hyperspectral UAV payload in the second image.

[0073] Specifically, since the drone and the payload in the sky are black, which is quite different from the color of the sky background, color recognition is used to identify the black drone; further, since only color recognition will misidentify other black objects, motion recognition is also needed to extract the foreground (moving object) by differentiating the current frame of the picture with the previous frame to identify the moving drone; further, only using the above two methods, there is still the possibility of misidentification, such as identifying a flying bird, so the target detection algorithm of computer vision is used to accurately identify the drone and the payload. The implementation of the target detection algorithm first requires creating a dataset containing drone pictures and common flying bird pictures, and then using the dataset to train the YOLOv5 model. By adjusting the parameters of the model, the model recognition accuracy exceeds 85%. Finally, the trained model is used to accurately identify the drone in the picture, and the outline and center of mass of the identified drone target are drawn. After the drone is identified, the center of mass and the center of the image are overlapped by the rotation of the servo motor to track the drone.

[0074] In the embodiment, Figure 3 As shown, the pan-tilt device 4 is the core part of the auxiliary system, which includes a servo motor turntable 41, a servo motor electric cylinder 42, a reflector support frame 43, a reflector fixing frame 44 and a tracking device fixing frame 45.

[0075] The reflector support frame 43 is fixed on both ends of the servo motor turntable 41, and is used to support the reflector fixing frame 44. The reflector fixing frame 44 is used to fix the reflector group 1. In order to make the reflector group 1 can rotate more easily in the vertical direction, the reflector support frame 43 is provided with bearings. On the other hand, the center of gravity of the reflector fixing frame 44 and the reflector group 1 is on the center line of the two reflector support frame bearings, so that the reflector group 1 can rotate in the vertical direction.

[0076] The servo motor cylinder 42 is fixed on the servo motor turntable 41. Its function is to convert the rotation of the servo motor into translation and push the reflector fixing frame 44 and the reflector group 1 upward to rotate the reflector group 1 in the vertical direction, that is, to control the pitch angle of the reflector group 1. The accuracy of the servo motor cylinder 42 must be higher than 0.05mm.

[0077] The servo motor turntable 41 is fixed on a bracket 46 (such as a tripod), and drives the reflector assembly 1 to rotate in the horizontal direction through the reflector support frame 43, that is, controls the azimuth angle of the reflector assembly 1. The accuracy of the servo motor turntable 41 should be higher than 0.05°. It is worth noting that the servo motor turntable 41 must be hollow, that is, have a hollow hole.

[0078] The tracking device fixing frame 45 passes through the hollow hole of the servo motor turntable 41 and is directly fixed on the bracket 46, which is used to support and fix the solar tracking device 2 and the drone tracking device 3.

[0079] The fixing method of the reflector group 1 carried by the pan-tilt device 4 is as follows: Figure 4 As shown, the reflector is fixed from the outside to the inside by the reflector clamping device A, the reflector supporting device B, and the reflector center device C. The three devices are connected by a hollow carbon fiber tube D. The reflector clamping device A is used to clamp the reflector to prevent the reflector from moving in the normal direction of the reflective surface. The reflector center device C is used to fix the reflector clamping device on the side. The four reflector clamping devices A and the reflector center device C can ensure that the reflector will not move in the tangent direction of the reflective surface. Since the reflector area is large, several reflector supporting devices are required to hold up the B reflector ( Figure 4 Only four are shown in the figure) to prevent the reflector from breaking due to uneven force, the contact surface area between the reflector support device B and the reflector should be larger. The reflector support device B is also fixed on the carbon fiber tube, one side contacts the reflector, and the other side is fixed to the reflector fixing frame 44.

[0080] The design of the pan-tilt device 4 is different from fixing the reflector directly to the ordinary pan-tilt, and has five advantages:

[0081] (1) The hollow servo motor turntable 41 allows the tracking device fixing frame 45 to be directly fixed on the bracket, so that the rotation of the reflector is completely independent of the rotation of the two tracking devices. Compared with directly fixing the tracking device and the reflector on the gimbal, on the one hand, the accuracy calculation and error analysis between the three devices, the sun tracking device 2, the drone tracking device 3, and the gimbal device 4, are independent of each other, which makes it easier to zero-position calibrate, and there will be no error accumulation between the three devices; on the other hand, the rotation algorithms of the three devices are also independent of each other, which greatly reduces the amount of calculation.

[0082] (2) The pitch angle control of the reflector is achieved by the translation of the servo motor cylinder 42 in the vertical direction. Since the cylinder precision is higher than 0.05mm, the precision of the pitch angle of the reflector after the distance conversion angle is higher than 0.03°. This precision can be further improved by adjusting the distance from the servo motor cylinder 42 to the center of the turntable. Compared with the ordinary pan-tilt head that converts the motor rotation into the pan-tilt angle change, its precision is greatly improved.

[0083] (3) The position of the reflector support frame 43 needs to be in the middle of the servo motor turntable 41. In this way, the reflector will have a static point whose spatial position does not change with the azimuth rotation or pitch angle rotation. For the reflector group 1, this static point is the center (center of gravity) position of the reflector group 1; for the gimbal device 4, it is the intersection of the center line of the reflector support frame 43 bearing and the rotation center axis of the servo motor turntable 41. The meaning of the static point is: in the ideal model, when the sun, the drone, and the auxiliary system are regarded as three points for calculating the reflection angle, the rotation of the auxiliary system will not affect the position of the auxiliary system under the ideal model, and this point is the static point. If the reflector is directly fixed to an ordinary gimbal, it is difficult to achieve this static point.

[0084] (4) The solar tracking device 2 and the UAV tracking device 3 are basically at the center of the reflector group 1. Therefore, when performing the vector calculation of the reflector angle, the distance between the reflector group 1, the UAV tracking device 2 and the solar tracking device 3 can be ignored.

[0085] The control information received by the pan / tilt device 4 for controlling the rotation of the reflector group is calculated by the control terminal 5. The specific calculation diagram is shown in FIG. Figure 5 As shown in Figure 1, when the hyperspectral UAV payload is far away from the auxiliary system, the UAV and payload, the center of the sun, and the reflector group can be regarded as three points. The reflector group is taken as the origin, and the line connecting the reflector group and the sun is recorded as the sun direction vector. The line connecting the reflector group and the UAV is recorded as the UAV direction vector The normal direction of the reflective surface of the reflector is the reflector direction vector Define the sun direction vector and the drone direction vector as unit vectors, Although the concept of vector is used here, the direction vectors here have a common starting point, that is, the static point (origin) of the reflector. According to the law of reflection, the direction vector of the reflector should be equal to the vector sum of the direction vector of the sun and the direction vector of the drone. That is:

[0086]

[0087] Since the rotation of the servo motor and the signal of the sensor are all in angle, the vector operation must be converted into angle operation. Convert the vector operation into the pitch angle α of the sun tracking device, drone tracking device, and gimbal device 1 , α 2 and α 3 and azimuth β 1 , β 2 and β 3 ,like Figure 5 As shown, the range of the pitch angle is 0 to 90° (horizontal to vertical upward), and the range of the azimuth angle is 0 to 360°. Then:

[0088]

[0089] The calculation target is through α 1 , α 2 , β 1 and β 2 Get α 3 and β 3 , then:

[0090]

[0091] However, due to α 3 and β 3 Involving multiple angle operations, so in actual operation, β 3 There will be an error of π. Through the reverse process of the result, it is considered that the α obtained above 3 and β 3 Correct, then you can get a

[0092]

[0093] When the angle is calculated correctly, Should be equal to

[0094] When the angle is calculated incorrectly, Not equal to When 3 =β 3 +π. In this way, the correct direction vector of the reflector can be obtained, and the two servo motors controlling the pitch angle and azimuth angle of the reflector are adjusted according to α 3 and β 3 Just turn it.

[0095] The above is the hyperspectral UAV remote sensing auxiliary system provided by the embodiment. By designing the pan-tilt device, the UAV tracking device and the sun tracking device, the direct sunlight can be reflected to the hyperspectral UAV payload, providing sufficient light flux for the payload, so that the hyperspectral UAV remote sensing payload can perform normal spectral inversion while allowing the payload to have a greater signal-to-noise ratio and a shorter integration time, and make up for the problem that the hyperspectral UAV remote sensing payload cannot perform large elevation angle ground observation, so that the payload can achieve all-round detection. Ultimately, the hyperspectral UAV remote sensing imaging payload can perform all-round high-temporal and spatial resolution remote sensing imaging inside key industrial parks.

[0096] The specific implementation methods described above provide a detailed description of the technical solutions and beneficial effects of the present invention. It should be understood that the above is only the most preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, supplements and equivalent substitutions made within the scope of the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components, characterized in that: include: A reflector assembly coated with an ultraviolet film for providing solar reflected light after reflecting direct sunlight to the hyperspectral UAV payload; A sun tracking device, which is used to track the sun's position in real time and feed back first attitude information of the sun tracking device to the control end; A drone tracking device, which is used to track the drone payload position in real time and feed back the second posture information of the drone tracking device to the control end; A pan-tilt device, which is used to support the reflector group, the sun tracking device, and the UAV tracking device, and controls the reflector group to rotate based on the received control information so that the hyperspectral UAV payload receives the high-luminous flux solar reflected light reflected by the reflector group; A control end, which is used to control the real-time tracking of the sun tracking device and the drone tracking device, and calculate the control information of the reflector group based on the first posture information and the second posture information; The pan-tilt device includes a servo motor turntable, a servo motor electric cylinder, a reflector support frame, a reflector fixing frame and a tracking device fixing frame, wherein the reflector support frame is fixed on both ends of the servo motor turntable, and is used to support the reflector fixing frame, and the reflector fixing frame is used to fix the reflector group, and there is a bearing on the reflector support frame, and the center of gravity of the reflector fixing frame and the reflector is on the center line of the reflector support frame bearing, so that the reflector group rotates in the vertical direction; the servo motor electric cylinder is fixed on the servo motor turntable, and its function is to convert the rotation of the servo motor into translation, and push the reflector fixing frame and the reflector group upward, so that the reflector group rotates in the vertical direction, that is, to control the pitch angle of the reflector group; the servo motor turntable is fixed on the bracket, and it drives the reflector group to rotate in the horizontal direction through the reflector support frame, that is, to control the azimuth angle of the reflector group; the tracking device fixing frame is directly fixed on the bracket through the hollow hole of the servo motor turntable, and is used to support and fix the sun tracking device and the drone tracking device.

2. The hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components according to claim 1 is characterized in that: The reflector group includes two reflectors, there is a gap of less than 10 cm between the reflectors, the size of a single reflector is greater than 0.5m×1m, the curvature radius of the reflector is greater than 50m, the coating band of each reflector covers at least 280nm-450nm, and the difference in reflectivity of the ultraviolet film at different wavelengths is less than 5%.

3. The hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components according to claim 1 is characterized in that: The solar tracking device includes a polarizing plate group, a first camera, a first pitch angle servo motor, a first azimuth angle servo motor, a first pitch angle sensor, a first azimuth angle sensor, and a first positioning module, wherein the polarizing plate group is arranged on the incident light path of the first camera to realize polarization extinction of direct sunlight, the first camera has a field of view angle less than 60° and a resolution greater than 720P, the first pitch angle servo motor and the first azimuth angle servo motor control the rotation of the first camera, the first pitch angle sensor and the first azimuth angle sensor collect a first pitch angle and a first azimuth angle of the solar tracking device to form first posture information, and the first positioning module is used for spatial positioning of the solar tracking device; The control end controls the sun tracking device to track in real time, including: Coarse adjustment: obtaining the latitude and longitude information, the first attitude information and the time information of the spatial positioning of the first positioning module, and calculating the current solar altitude angle and the solar azimuth angle according to the obtained information, and controlling the first pitch angle servo motor and the first azimuth angle servo motor according to the current solar altitude angle and the solar azimuth angle to turn the first camera to the position of the sun to receive direct sunlight; Fine adjustment: obtaining a first image captured by a first camera, identifying the sun based on the first image, aiming to make the center of the sun in the first image coincide with the geometric center of the first image, and achieving real-time tracking of the sun by rotating a first pitch angle servo motor and a first azimuth angle servo motor; When tracking is not possible based on the first image, the process jumps to the coarse adjustment process.

4. The hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components according to claim 3 is characterized in that: The polarizer group includes a first polarizer, a second polarizer, and a third polarizer arranged along the direction of the incident light, wherein the first polarizer and the second polarizer are fixed polarizers, which are used to weaken the intensity of direct sunlight, and the degree of weakening is more than 95%, and the third polarizer is a rotatable polarizer, which rotates according to the real-time feedback of direct sunlight intensity, so that the brightness of the direct sunlight spot received by the first camera remains relatively uniform in real time.

5. The hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components according to claim 3 is characterized in that: Identify the sun based on the first image, including: First, the image is grayed out, then the shape is judged to identify the sun, and finally the outline and center of mass of the identified sun are drawn.

6. The hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components according to claim 1 is characterized in that: The drone tracking device includes a second camera, a second pitch angle servo motor, a second azimuth angle servo motor, a second pitch angle sensor, a second azimuth angle sensor, and a second positioning module, wherein the second camera has a field of view angle less than 30° and a resolution greater than 2K, the second pitch angle servo motor and the second azimuth angle servo motor control the rotation of the second camera, the second pitch angle sensor and the second azimuth angle sensor collect the second pitch angle and the second azimuth angle of the drone tracking device to form second posture information, and the second positioning module is used to perform spatial positioning of the drone tracking device; The control end controls the drone tracking device to track in real time, including: Coarse adjustment: obtain the altitude and longitude and latitude of the hyperspectral UAV payload, as well as the longitude and latitude information of the second positioning module, calculate the relative position between the hyperspectral UAV payload and the UAV tracking device, and control the second pitch angle servo motor and the second azimuth angle servo motor according to the relative position to turn the second camera to the direction of the hyperspectral UAV payload; Fine adjustment: obtain the second image captured by the second camera, identify the load center through the second image, aim to make the load center in the second image coincide with the geometric center of the second image, and realize real-time tracking of the hyperspectral UAV payload through the rotation of the second pitch angle servo motor and the second azimuth angle servo motor; When tracking is not possible based on the second image, the process jumps to the coarse adjustment process.

7. The hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components according to claim 6 is characterized in that: Identify the load center through the second image, including: Color recognition is used to identify the hyperspectral UAV payload in the second image that is different from the background color; or, using motion recognition to identify the hyperspectral UAV payload in the second image; Alternatively, a recognition model built based on deep learning is used to identify the hyperspectral UAV payload in the second image.

8. The hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components according to claim 1 is characterized in that: The control end calculates control information of the reflector group based on the first posture information and the second posture information, including: The first posture information includes a first pitch angle α1 and a first azimuth angle β1, the second posture information includes a second pitch angle α2 and a second azimuth angle β2, and the calculated control information includes a third pitch angle α3 and a third azimuth angle β3 for controlling the reflector group: in, is the normal vector of the mirror's reflection plane.

9. The hyperspectral UAV remote sensing auxiliary system for ultraviolet band absorption components according to claim 8 is characterized in that: Also includes: When the obtained α3 ​​and β3 are considered correct, the normal vector of the reflection plane of the reflector is obtained. Normal vector of the reflector's reflection plane obtained by vector operation when and When they are inconsistent, set β3=β3+π to achieve error correction.

Images