Photovoltaic panel inspection unmanned aerial vehicle and inspection method based on big dipper positioning and data fusion

Abstract

The embodiment of the application relates to the technical field of unmanned aerial vehicle detection, and discloses a photovoltaic panel inspection unmanned aerial vehicle based on Beidou positioning and data fusion and an inspection method, the unmanned aerial vehicle comprising: an airborne positioning module, a visible light camera, a thermal imager, a communication module, a storage module, a controller and a driving module, wherein the airborne positioning module, the visible light camera, the thermal imager, the communication module, the storage module and the driving module are electrically connected with the controller. The technical scheme of the application can realize accurate positioning of the photovoltaic panel inspection unmanned aerial vehicle and improve system working stability; the visible light camera and the thermal imager perform double-light fusion, which can not only detect hot spots of the photovoltaic panel, but also detect defects, cracks and other abnormalities of the photovoltaic panel, and the positioning performance and detection performance exceed those of most traditional photovoltaic panel inspection unmanned aerial vehicles.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) inspection technology, specifically to a photovoltaic panel inspection UAV and inspection method based on BeiDou positioning and data fusion. Background Technology

[0002] Drone inspection is a powerful supplement to traditional industrial inspection methods. Compared to manual and robotic inspection, it offers advantages such as mobility, efficiency, speed, precision, low operating costs, wide applicability, and short production cycles. It also has significant advantages in high-resolution, rapid inspection in complex terrain. With the development of the photovoltaic power generation industry, the scale of photovoltaic panel arrays in photovoltaic power plants is increasing, leading to decreased efficiency and higher costs for manual or robotic inspection. Drone-based photovoltaic panel inspection technology has demonstrated its unique advantages, and the integration of drone inspection with the photovoltaic power generation industry has become a new industrial field.

[0003] To meet the needs of different scenarios, various types of photovoltaic panel inspection drones have been developed. Application number CN201910410949.1 proposes a drone and its detection method for photovoltaic power station fault detection, comprising three parts: a hexacopter drone, an automatic detection component, and a ground control terminal. A method for locating hot spots in photovoltaic panels using drones and thermal imaging is proposed. This method involves using a drone to conduct aerial patrols along a predetermined route. The drone's gimbal adjusts the shooting angles of its compound-eye camera, BeiDou positioning system, and thermal infrared imaging detector in real time. The thermal infrared imaging detector, in conjunction with the compound-eye camera and BeiDou positioning system, identifies and locates faulty areas in photovoltaic panels, generating a distribution map of the faulty areas. This solves the problem of low efficiency in troubleshooting hot spots in photovoltaic panels in existing technologies. Application number CN202111319956.4 proposes a method for locating hot spots in photovoltaic panels based on drones and thermal imaging. This method measures the position information of photovoltaic modules using a GPS positioning instrument and stores it in a database. Based on the collected position information, the total number of rows of photovoltaic panels in the photovoltaic module is calculated. The drone is controlled to fly, and during flight, the position information of the four corner points of the photovoltaic modules is used to correct the drone's flight trajectory. Image algorithms are used to identify and segment individual photovoltaic panels in the aerial images and to analyze the light... The method involves locating the number of rows of photovoltaic panels in a photovoltaic module; using a signal receiving device on a drone to activate an infrared camera, real-time detection and processing of hot spots to obtain their location information, fully utilizing drone automatic inspection to achieve automatic hot spot location and correction of the drone's flight trajectory. This method is not only convenient to calculate but also highly efficient, enabling photovoltaic power plants to operate more efficiently and intelligently. Application number CN202010197758.4 discloses a method for identifying solar photovoltaic panel faults based on thermal imaging images. Multiple coordinate points are set within a region, and the drone automatically flies along a flight path formed by these points. Multiple flight paths can be created and saved. The drone's flight distance from the ground can be set. The intelligent drone is highly maneuverable, and considering the vast areas and undulating terrain of photovoltaic power plants, drone inspection is time-saving, labor-saving, and highly efficient. It significantly reduces the number of personnel and inspection time required for photovoltaic power plant inspections, saving manpower maintenance costs and resulting in higher economic benefits.

[0004] In summary, existing photovoltaic panel inspection drones use a single airborne positioning unit, which is easily affected by external factors and may result in incomplete positioning signals, leading to drone deviation or loss of control. Summary of the Invention

[0005] In view of the above problems, embodiments of the present invention provide a photovoltaic panel inspection drone and inspection method based on Beidou positioning and data fusion, which is used to solve the problem that the photovoltaic panel inspection drone uses a single recording and positioning unit in the prior art, which is easily affected by external interference, causing the drone to deviate or lose control.

[0006] According to one aspect of the present invention, a photovoltaic panel inspection drone based on BeiDou positioning and data fusion is provided, comprising: an airborne positioning module, a visible light camera, a thermal imager, a communication module, a storage module, a controller, and a drive module. The airborne positioning module, visible light camera, thermal imager, communication module, storage module, and drive module are all electrically connected to the controller. The communication module is used to receive information from the photovoltaic panel positioning module. The storage module is used to store the photovoltaic panel positioning module information. The controller is used to acquire the photovoltaic positioning module information, airborne positioning module data, and visible light camera data to locate the drone, and compares the drone's real-time position with the stored latitude and longitude coordinates of the photovoltaic panel positioning module to determine if there is a deviation. If a deviation occurs, a correction route command is generated to correct the drone's flight path.

[0007] In one alternative approach, both the airborne positioning module and the photovoltaic positioning module employ a high-precision BeiDou positioning module.

[0008] In an alternative embodiment, a barometric pressure sensor is also included, which is electrically connected to the controller.

[0009] In one alternative approach, the visible light camera is also used to measure the distance between the drone and a photovoltaic panel within a preset range.

[0010] In one alternative approach, the controller employs a PID algorithm to fuse data from the photovoltaic positioning module, the airborne positioning module, and the visible light camera to locate the UAV.

[0011] According to another aspect of the present invention, a method for inspecting photovoltaic panels using a drone based on BeiDou positioning and data fusion is provided, comprising the following steps:

[0012] Step 1: Obtain the location information of the photovoltaic panel positioning module, the latitude and longitude coordinates generated by the photovoltaic panel positioning module, and the location information of the drone inspection path points, and store the latitude and longitude coordinates generated by the photovoltaic panel positioning module;

[0013] Step 2: Create an environmental map based on the latitude and longitude coordinates generated by the photovoltaic panel positioning module;

[0014] Step 3: Generate a navigation route in the environmental map based on the location information of the inspection path points;

[0015] Step 4: Based on the data from the airborne positioning module and the navigation route, traverse all the drone inspection path points, and activate the visible light camera and thermal imager to inspect the photovoltaic panels;

[0016] Step 5: Position the UAV based on the data from the photovoltaic positioning module, the airborne positioning module, and the visible light camera. Compare the UAV's real-time position with the stored latitude and longitude coordinates of the photovoltaic panel positioning module to determine if it is off course. If it is off course, generate a correction command.

[0017] Step Six: Correct the drone's cruise route according to the correction instructions, so that the drone continues to perform inspections along the navigation route.

[0018] In one alternative approach, the visible light camera is also used to measure the distance between the drone and a photovoltaic panel within a preset range.

[0019] In one alternative approach, the visible light camera uses a target detection algorithm to measure the distance between the drone and a photovoltaic panel within a preset range.

[0020] In one alternative approach, the visible light camera uses a target detection algorithm to measure the distance between the drone and the photovoltaic panels within a preset range. Specific steps include:

[0021] Step (1): Perform target detection on the images transmitted back from the visible light camera;

[0022] Step (2): Obtain the pixel value P_bottom of the bottom of the photovoltaic panel under test on the imaging plane;

[0023] Step (3): Visible light camera calibration and adjustment. Obtain the focal length f of the visible light camera and the pixel coordinate P_center of the optical center of the lens in the y direction in the pixel coordinate system;

[0024] Step (4): Obtain the flight altitude H of the UAV using a barometric pressure sensor;

[0025] Step (5): Calculate the distance Dis between the photovoltaic panel and the visible light camera. The specific formula is as follows:

[0026]

[0027] In one alternative approach, step five specifically employs a PID algorithm to fuse data from the photovoltaic positioning module, the airborne positioning module, and the visible light camera to locate the UAV.

[0028] This invention utilizes high-precision BeiDou positioning module data from photovoltaic panel arrays to create an environmental map. Data fusion is then performed using an airborne high-precision BeiDou positioning module and a visible light camera for monocular ranging, enabling precise positioning of the photovoltaic panel inspection drone and improving system stability. The dual-light fusion of the visible light camera and thermal imager not only detects hot spots on the photovoltaic panels but also detects defects, cracks, and other anomalies, exceeding the positioning and detection performance of most traditional photovoltaic panel inspection drones. Furthermore, using a visible light camera for ranging of the photovoltaic panels eliminates the need for additional equipment such as lidar when implementing ranging and collision avoidance functions with traditional drones, reducing the drone's load, lowering costs, and simplifying its structure.

[0029] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0030] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0031] Figure 1 The flowchart of the photovoltaic panel inspection drone of the present invention is shown.

[0032] Figure 2 The diagram illustrates the data fusion positioning control principle of the photovoltaic panel inspection drone of the present invention.

[0033] Figure 3 The diagram illustrates the monocular ranging principle of the photovoltaic panel inspection drone of the present invention. Detailed Implementation

[0034] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0035] Example:

[0036] A photovoltaic panel inspection drone based on BeiDou positioning and data fusion includes: an airborne positioning module, a visible light camera, a thermal imager, a communication module, a storage module, a controller, and a drive module. The airborne positioning module, visible light camera, thermal imager, communication module, storage module, and drive module are all electrically connected to the controller.

[0037] The communication module receives information from the photovoltaic panel positioning module. This information includes the photovoltaic panel positioning module's location information and latitude and longitude coordinates generated by the module. In this embodiment, an environmental map is specifically built based on the latitude and longitude coordinates generated by the photovoltaic panel positioning module, resulting in higher positioning accuracy during UAV navigation. The photovoltaic panel positioning module information is used to compare and determine whether the UAV deviates from its course during cruise.

[0038] Both the airborne positioning module and the photovoltaic positioning module adopt the high-precision Beidou positioning module, which has many advantages such as high sensitivity, low power consumption, and miniaturization.

[0039] The storage module is used to store photovoltaic panel positioning module information. The controller retrieves the photovoltaic panel positioning module information from the storage module to determine whether the drone deviates from its course during cruise.

[0040] The drive module includes a drone propeller drive motor and a rotor. The output shaft of the drone propeller drive motor is fixedly connected to the rotor by fixing bolts, and the controller is connected to the drone propeller drive motor via telecommunications.

[0041] It also includes a power module, which supplies power to the controller, the drone propeller drive motor, the airborne positioning module, the communication module, and the storage module.

[0042] The controller acquires information from the photovoltaic positioning module, airborne positioning module, and visible light camera to locate the drone. It compares the drone's real-time position with the stored latitude and longitude coordinates of the photovoltaic positioning module to determine if it is deviating from its course. If deviation is detected, a correction command is generated to adjust the drone's route. The controller uses a PID algorithm to fuse data from the photovoltaic positioning module, airborne positioning module, and visible light camera to locate the drone. See attached details. Figure 2 As shown, the airborne Beidou high-precision positioning module detects the real-time coordinates of the photovoltaic panel inspection drone, and the memory unit stores the coordinates as a given value u1. The photovoltaic panel inspection drone compares the positioning information of the high-precision Beidou positioning module on the photovoltaic panel with that of the high-precision Beidou positioning module on the photovoltaic panel to obtain the photovoltaic panels within a 5-meter radius of the drone. The Beidou high-precision positioning module within a 5-meter radius of the drone generates a correction signal u2. The generated positioning data e1 is then fused with the data from the monocular ranging of the visible light camera after passing through feedforward compensation de1 / dt and the first-stage PID controller to generate fused data e2. After passing through feedforward control and the second-stage PID controller, a PWM signal is generated and transmitted to the drone's propeller drive motor. During this process, the PID controller automatically adjusts the parameters, calculates the PWM control signal required to restore the predetermined operating route of the photovoltaic panel inspection drone, and transmits it to the drone's propeller drive motor.

[0043] As one embodiment of the present invention, the real-time coordinates of the photovoltaic panel inspection drone are y = (x... t ,y t In the predetermined flight path of the photovoltaic panel inspection drone, u1 = (x1, y1). Assume there are n photovoltaic panels within a 5m radius of the drone. The calibration signal... After the addition operation e1 = u1 + u2 - y, it is input together with the feedforward compensation de1 / dt into the first-stage PID controller for positional PID calculation: Where K p T is the proportionality coefficient. I T is the integration constant. D Here is a differential constant. The above signal is fused with the signal obtained from monocular ranging to obtain e2, and then position-based PID calculation is performed by a second-stage PID controller. The magnitude of the acquired signal is proportional to the pulse width of the PWM control signal, thereby adjusting the propeller speed of the photovoltaic panel inspection drone and consequently adjusting the real-time coordinates y = (x - y) of the drone. t ,y t This allows for the precise positioning of the photovoltaic panel inspection drone.

[0044] It also includes a barometric pressure sensor, which is electrically connected to the controller. The barometric pressure sensor is mainly used to measure the pressure of gas. A barometric pressure sensor with one atmosphere range is usually used to measure weather changes and to measure altitude by utilizing the correlation between air pressure and altitude. The flight altitude of the UAV can be obtained through the barometric pressure sensor.

[0045] The visible light camera is also used to measure the distance between the drone and the photovoltaic panels within a preset range. See attached details. Figure 3 As shown, the UAV controller performs target detection using images transmitted from the visible light camera and obtains the pixel value P_bottom of the bottom of the photovoltaic panel on the imaging plane. Through visible light camera calibration and adjustment, the focal length f of the visible light camera and the pixel coordinate P_center of the lens's optical center in the y-direction of the pixel coordinate system are calculated. The UAV's flight altitude H can be obtained from the barometric pressure sensor, and the distance Dis between the photovoltaic panel and the visible light camera can be calculated. The specific calculation method is as follows:

[0046]

[0047] The focal length f mentioned above, after camera calibration, yields the following intrinsic parameter matrix:

[0048]

[0049] The focal length is (fx, fy), and the optical center pixel coordinates are (Cx, Cy). As shown in the diagram above, we are using data along the Y-axis as the basis for our calculations. Therefore:

[0050]

[0051] For example, after calibration, the camera has a focal length of (0.05m, 0.15m) and a light center pixel coordinate of (0.2m, 0.2m). Then, according to formula (3), the focal length in the y-axis direction is f = f y =0.15m, the pixel coordinates of the optical center of the lens in the y-direction of the pixel coordinate system are P_center = 0.2m. The height H = 2m obtained by the barometric pressure sensor of the photovoltaic panel inspection drone, and the pixel value P_bottom = 0.25m of the bottom of the photovoltaic panel on the imaging plane obtained by the camera through the target detection algorithm, then according to formula (1):

[0052]

[0053]

[0054] The distance Dis between the photovoltaic panel and the visible light camera can be obtained as 6m, thus enabling the monocular ranging algorithm to measure the distance between the photovoltaic panel and the visible light camera.

[0055] This solution also provides the above-mentioned photovoltaic panel inspection drone method based on BeiDou positioning and data fusion, including the following steps:

[0056] S1: Obtain the location information of the photovoltaic panel positioning module, the latitude and longitude coordinates generated by the photovoltaic panel positioning module, and the location information of the drone inspection path points, and store the latitude and longitude coordinates generated by the photovoltaic panel positioning module;

[0057] S2: Create an environmental map based on the latitude and longitude coordinates generated by the photovoltaic panel positioning module;

[0058] S3: Generate navigation routes in the environment map based on the location information of inspection path points;

[0059] S4: Based on the airborne positioning module data and navigation route, traverse all UAV inspection path points and activate the visible light camera and thermal imager to inspect the photovoltaic panels;

[0060] S5: The drone is positioned based on the data from the photovoltaic positioning module, the airborne positioning module, and the visible light camera. The drone's real-time position is compared with the latitude and longitude coordinates of the stored photovoltaic positioning module to determine whether it is deviating from its course. If it is deviating from its course, a correction command is generated.

[0061] S6: Correct the drone's cruise route according to the correction command, so that the drone can continue to carry out inspections along the navigation route.

[0062] In one alternative approach, the visible light camera is also used to measure the distance between the drone and a photovoltaic panel within a preset range.

[0063] In one alternative approach, the visible light camera uses a target detection algorithm to measure the distance between the drone and a photovoltaic panel within a preset range.

[0064] In one alternative approach, the visible light camera uses a target detection algorithm to measure the distance between the drone and the photovoltaic panels within a preset range. Specific steps include:

[0065] Step (1): Perform target detection on the images transmitted back from the visible light camera;

[0066] Step (2): Obtain the pixel value P_bottom of the bottom of the photovoltaic panel under test on the imaging plane;

[0067] Step (3): Visible light camera calibration and adjustment. Obtain the focal length f of the visible light camera and the pixel coordinate P_center of the optical center of the lens in the y direction in the pixel coordinate system;

[0068] Step (4): Obtain the flight altitude H of the UAV using a barometric pressure sensor;

[0069] Step (5): Calculate the distance Dis between the photovoltaic panel and the visible light camera, specifically by referring to formula (1).

[0070] In one alternative approach, step S5 specifically employs a PID algorithm to fuse data from the photovoltaic positioning module, the airborne positioning module, and the visible light camera to locate the UAV.

[0071] This invention utilizes high-precision BeiDou positioning module data on photovoltaic panel arrays to create an environmental map, resulting in higher positioning accuracy during UAV navigation. Furthermore, it employs data fusion based on an onboard high-precision BeiDou positioning module and a visible light camera for monocular ranging, solving the problem of easy velocities in traditional single-positioning UAVs, accelerating response speed, achieving precise positioning of photovoltaic panel inspection UAVs, and improving system stability. The dual-light fusion of a visible light camera and a thermal imager not only detects hot spots on photovoltaic panels but also detects defects, cracks, and other anomalies, exceeding the positioning and detection performance of most traditional photovoltaic panel inspection UAVs. Simultaneously, using a visible light camera for photovoltaic panel ranging eliminates the need for additional equipment such as lidar when implementing ranging and collision avoidance functions with traditional UAVs, reducing UAV load, lowering costs, and simplifying the UAV structure.

[0072] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. Similarly, for the sake of brevity and to aid in understanding one or more aspects of the invention, in the description of exemplary embodiments of the invention above, various features of the embodiments are sometimes grouped together in a single embodiment, figure, or description thereof. The claims, which follow the detailed description, are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0073] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components, except that at least some of such features and / or processes or units are mutually exclusive.

[0074] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A photovoltaic panel inspection drone based on BeiDou positioning and data fusion, characterized in that, include: The system includes an airborne positioning module, a visible light camera, a thermal imager, a communication module, a storage module, a controller, and a drive module, all of which are electrically connected to the controller. The communication module is used to receive information from the photovoltaic panel positioning module; the storage module is used to store the photovoltaic panel positioning module information; the controller is used to acquire photovoltaic positioning module information, airborne positioning module data and visible light camera data to locate the UAV, and compare the UAV's real-time position with the stored latitude and longitude coordinates of the photovoltaic panel positioning module to determine whether it is deviating from its course. If it is deviating from its course, a correction route instruction is generated so that the UAV continues to inspect along the navigation route. Both the airborne positioning module and the photovoltaic positioning module adopt high-precision Beidou positioning modules; the visible light camera is also used to measure the distance between the UAV and the photovoltaic panels within a preset range; the controller uses a PID algorithm to fuse data from the photovoltaic positioning module, the airborne positioning module, and the visible light camera to locate the UAV. The specific methods for generating route correction instructions include: The airborne positioning module detects the real-time coordinates of the drone, and stores the latitude and longitude coordinates of the photovoltaic panel positioning module as u1; The controller compares the real-time coordinates of the drone with the positioning information of the photovoltaic positioning module to obtain the photovoltaic panels within a 5-meter radius of the drone. The correction signal generated by the photovoltaic positioning module within a 5-meter radius of the drone is u2. The generated positioning data e1 is fed forward compensation de1 / dt and the first-stage PID controller, and then fused with the data from the monocular ranging of the visible light camera to generate fused data e2. After passing through the feedforward control and the second-stage PID controller, a PWM signal is generated and transmitted to the drone propeller drive motor. During this process, the PID controller automatically adjusts the parameters, calculates the PWM control signal required to restore the predetermined running route of the photovoltaic panel inspection drone, and transmits it to the drone propeller drive motor.

2. The inspection drone according to claim 1, characterized in that, It also includes a pressure sensor, which is electrically connected to the controller.

3. A photovoltaic panel inspection method using unmanned aerial vehicles (UAVs) based on BeiDou positioning and data fusion, the method comprising: Step 1: Obtain the location information of the photovoltaic panel positioning module, the latitude and longitude coordinates generated by the photovoltaic panel positioning module, and the location information of the drone inspection path points, and store the latitude and longitude coordinates generated by the photovoltaic panel positioning module; Step 2: Create an environmental map based on the latitude and longitude coordinates generated by the photovoltaic panel positioning module; Step 3: Generate a navigation route in the environmental map based on the location information of the inspection path points; Step 4: Based on the data from the airborne positioning module and the navigation route, traverse all the drone inspection path points, and activate the visible light camera and thermal imager to inspect the photovoltaic panels; Step 5: Position the UAV based on the data from the photovoltaic positioning module, the airborne positioning module, and the visible light camera. Compare the UAV's real-time position with the stored latitude and longitude coordinates of the photovoltaic panel positioning module to determine if it is off course. If it is off course, generate a correction route instruction. Step Six: Correct the drone's cruise route according to the corrected route instructions, so that the drone continues to perform inspections along the navigation route; The visible light camera is also used to measure the distance between the drone and the photovoltaic panels within a preset range; in step five, a PID algorithm is specifically used to fuse the data from the photovoltaic positioning module, the airborne positioning module, and the visible light camera to locate the drone; The method for generating the corrected flight route instruction in step five specifically includes: The airborne positioning module detects the real-time coordinates of the drone, and stores the latitude and longitude coordinates of the photovoltaic panel positioning module as u1; The real-time coordinates of the drone are compared with the positioning information of the photovoltaic positioning module to obtain the photovoltaic panels within a 5-meter radius of the drone. The correction signal generated by the photovoltaic positioning module within a 5-meter radius of the drone is u2. The generated positioning data e1 is fed forward compensation de1 / dt and the first-level PID controller, and then fused with the data from the monocular ranging of the visible light camera to generate fused data e2. After passing through the feedforward control and the second-level PID controller, a PWM signal is generated and transmitted to the drone propeller drive motor.

4. The method according to claim 3, characterized in that, The visible light camera uses a target detection algorithm to measure the distance between the drone and the photovoltaic panels within a preset range.

5. The method according to claim 4, characterized in that, The visible light camera uses a target detection algorithm to measure the distance between the drone and the photovoltaic panels within a preset range. Specific steps include: Step (1): Target detection is performed on the images transmitted back from the visible light camera; Step (2): Obtain the pixel value P_bottom of the bottom of the photovoltaic panel under test on the imaging plane; Step (3): Visible light camera calibration and adjustment. Obtain the focal length f of the visible light camera and the pixel coordinate P_center of the optical center of the lens in the y direction in the pixel coordinate system; Step (4): Obtain the UAV's flight altitude H using a barometric pressure sensor; Step (5): Calculate the distance Dis between the photovoltaic panel and the visible light camera. The specific formula is as follows: 。

Images