Method and system for obstacle avoidance based on unmanned aerial vehicle inspection of photovoltaic power stations

Abstract

This invention provides a method and system for obstacle avoidance in photovoltaic power plant inspections using unmanned aerial vehicles (UAVs), comprising the following steps: constructing a photovoltaic power plant model based on acquired photovoltaic power plant parameters; constructing a dynamic obstacle model library based on the constructed obstacle database; determining a safe flight zone for the UAV within the photovoltaic power plant model based on acquired real-time obstacle images and the dynamic obstacle model library; planning an obstacle avoidance route for the UAV based on the safe passage zone; comparing the flight parameters of each UAV obstacle avoidance route to determine the optimal flight route; and executing the optimal flight route to complete obstacle avoidance. This method, by constructing an accurate power plant model and a dynamic obstacle model library, combined with real-time obstacle recognition and optimal flight route planning, achieves efficient and safe UAV inspections in complex environments, improving the operation and maintenance efficiency and safety of photovoltaic power plants.

Description

Technical Field

[0001] This invention belongs to the field of energy power plant technology, specifically to a method and system for obstacle avoidance in photovoltaic power plants based on unmanned aerial vehicle (UAV) inspection. Background Technology

[0002] With the rapid development of photovoltaic (PV) power plants, their operation and maintenance management has become increasingly important. To ensure the long-term and stable operation of PV power plants, regular inspections are often necessary to prevent accidents. These inspections primarily involve checking whether the photovoltaic panels are in normal working order, such as checking for detachment, abnormal shading, or cracks, as well as whether the photovoltaic supports are deformed, tilted, or aligned correctly with the light source. Due to the large area occupied by PV power plants and the heavy workload of inspections, traditional manual inspection methods are not only inefficient but also pose safety hazards. Therefore, current PV power plants often utilize drones for inspections. This involves manually controlling drones to navigate along a pre-set inspection route, flying between the installed PV equipment or close to the surface of the photovoltaic panels to complete the inspection. However, during the drone's movement along the pre-set route, if moving obstacles such as birds, wild animals that have strayed between the panels, or vegetation are encountered, operators must handle the situation to prevent collisions, damage, or crashes.

[0003] To facilitate obstacle avoidance during drone flight, existing technologies involve installing a single high-precision measuring device on the drone to acquire information about obstacles along its flight path. This involves mounting a laser transmitter and receiver on the drone. The transmitter emits laser light, and the receiver receives the reflected light from external obstacles, transmitting the signals to the drone's control module. The operator then uses this obstacle information, along with their experience, to guide the drone to avoid obstacles. While this method effectively identifies obstacles through reflected laser signals during drone inspections within photovoltaic power plants, the information obtained from reflected laser signals is limited. Some obstacles alter the direction of laser reflection, preventing the receiver from receiving all emitted laser light. This results in incomplete obstacle information from the control module, leading to inaccurate boundary assessments and uncertainties about the safe flight zone after obstacle avoidance. Consequently, operators may struggle to accurately determine the safe flight area, increasing the risk of collisions and damage to the drone. Summary of the Invention

[0004] To address the problem in existing technologies where a single high-precision measuring device cannot determine the safe flight area for drones during obstacle avoidance when inspecting photovoltaic power plants, leading to collisions between drones and obstacles or other equipment, this invention provides a method and system for obstacle avoidance during drone inspections of photovoltaic power plants.

[0005] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a method for obstacle avoidance during unmanned aerial vehicle (UAV) inspections of photovoltaic power plants, comprising the following steps: A photovoltaic power station model is constructed based on the acquired photovoltaic power station parameters, and a dynamic obstacle model library is constructed based on the constructed obstacle database. Based on the acquired real-time obstacle images and the dynamic obstacle model library, the safe flight area of ​​the UAV is determined in the photovoltaic power station model; Based on the aforementioned drone safety, an obstacle avoidance route for the drone is obtained through regional planning. By comparing the flight parameters of each of the aforementioned drone obstacle avoidance routes, the optimal flight route is determined, and the drone executes the optimal flight route to complete obstacle avoidance.

[0006] Preferably, the acquisition of the photovoltaic power station parameters includes: Acquire images of the terrain and vegetation where the photovoltaic power station is located; Feature extraction is performed on the terrain and vegetation images to obtain elevation data and vegetation height data; Select the lowest altitude data from the given altitude data to obtain the lowest altitude data; Using the lowest altitude data as the baseline for model construction of the area where the photovoltaic power station is located, the altitude data and the vegetation height data are corrected to obtain corrected altitude data and corrected vegetation height data. The parameters of the first photovoltaic power station are obtained by integrating the model construction baseline, the corrected altitude data, and the corrected vegetation height data; Acquire images of photovoltaic panel arrays, roads, buildings, and fences; Feature extraction is performed on the photovoltaic panel array image, the road image, the building image, and the fence image to obtain the parameters of the second photovoltaic power station; The photovoltaic power station parameters are obtained by integrating the parameters of the first photovoltaic power station and the parameters of the second photovoltaic power station.

[0007] Preferably, the process of constructing a photovoltaic power station model based on photovoltaic power station parameter data includes: A basic model of the photovoltaic power station is constructed using the model construction baseline and the corrected altitude data in the parameters of the first photovoltaic power station. A vegetation model is constructed in the basic model of the photovoltaic power station using the corrected vegetation height data to obtain a preliminary model of the photovoltaic power station. Using the parameters of the second photovoltaic power station, a photovoltaic panel array model, a road model, a building model, and a fence model are constructed in the preliminary model of the photovoltaic power station to obtain the photovoltaic power station model.

[0008] Preferably, the construction of the obstacle model based on the constructed obstacle database includes: Images of obstacles encountered by drones during flight within a photovoltaic power station are used as training samples; Manually label the obstacle types in the training samples, and extract features from the labeled obstacle types of the training samples to generate an obstacle database; A dynamic obstacle model library was constructed based on the obstacle database.

[0009] Preferably, determining the safe flight area of ​​the UAV in the photovoltaic power station model based on the acquired real-time obstacle images and dynamic obstacle model library includes: Acquire real-time obstacle images, input the real-time obstacle images into the dynamic obstacle model library for comparison, and extract the dynamic obstacle model corresponding to the real-time obstacle image; Obtain the real-time environmental data and real-time location of the obstacle; An obstacle environment model is constructed based on the environmental data, and the obstacle is located in the photovoltaic power station model to determine the real-time obstacle position of the obstacle environment model in the photovoltaic power station model. The dynamic obstacle model is input into the real-time obstacle position in the photovoltaic power station model to determine the drone obstacle avoidance area corresponding to the real-time obstacle position; Set a safe flight distance for the drone during flight; Based on the flight safety distance, the outline region of the object in the photovoltaic power station model is determined, and the flight safety outline region of the UAV is obtained. The safe flight zone for humans and drones is determined based on the flight safety contour area and the drone obstacle avoidance area.

[0010] Preferably, the step of planning multiple drone flight routes based on the safe passage area of ​​the drone includes: Acquire real-time flight environment data of the UAV's real-time location, and construct a real-time flight environment model based on the real-time flight environment data; The real-time flight environment model is input into the photovoltaic power station model for comparison, and the real-time position of the UAV in the photovoltaic power station model is located to obtain the real-time position of the UAV. Based on the constructed drone model, the flight area at the drone's real-time location is simulated to obtain the drone's flight area at its real-time location. A no-entry zone for the drone itself is constructed in the flight area to obtain a safe flight space for the drone. Starting from the center point of the safe flight space of the UAV and ending at the center point of the safe passage area of ​​the UAV corresponding to the real-time obstacle position, multiple UAV flight curves are planned to avoid all object outline areas in the photovoltaic power station model, resulting in multiple UAV flight routes. Simulate each of the aforementioned drone flight paths, filter the drone flight paths, and obtain the drone obstacle avoidance path.

[0011] Preferably, simulating each of the drone flight paths and filtering the drone flight paths includes: The safe flight space of the UAV is simulated along each UAV flight path. It is determined whether the safe flight space of the UAV overlaps with the outline area of ​​the object. If they overlap, the UAV is rotated around the flight direction of the UAV at that position. If an overlapping area still appears after the rotation, the UAV flight path is discarded. If the UAV flight path does not overlap with the outline area of ​​the object in the photovoltaic power station model after the rotation or without rotation, the UAV flight path is retained. Integrate all retained drone flight paths to obtain drone obstacle avoidance routes; and record the flight parameters of each drone flight path. Preferably, determining the optimal flight route by comparing the flight parameters of each of the UAV obstacle avoidance routes includes: The flight time, flight path length, flight speed, flight altitude, drone attitude adjustment angle, and deviation value between the obstacle avoidance end point and the preset inspection route are obtained for each of the aforementioned drone flight paths. By comparing the flight time, flight path length, flight speed, flight altitude, drone attitude adjustment angle, and deviation value between the end point and the preset inspection route for each drone flight path, the drone obstacle avoidance route with the shortest flight path length, the smallest drone attitude adjustment angle, and the largest drone flight distance from all objects is determined. Based on the drone's current battery level, an optimal flight path is determined, and then the drone is controlled to avoid obstacles.

[0012] Preferably, the method further includes: acquiring first motion information of the drone and second motion information of the obstacle; determining whether the obstacle is tracking the drone based on the first motion information and the second motion information; and if so, sending the tracking information to a preset smart terminal.

[0013] This invention proposes a system for obstacle avoidance in photovoltaic power station inspection based on unmanned aerial vehicles (UAVs). Based on the above-mentioned method, it includes an input module, a model building module, a path planning module, a filtering module, and an obstacle avoidance execution module. The input module is used to input photovoltaic power station parameters and an obstacle database. The model building module is used to build a photovoltaic power station model based on the acquired photovoltaic power station parameters and to build a dynamic obstacle model library based on the constructed obstacle database. The path planning module is used to determine the safe flight area of ​​the UAV in the photovoltaic power station model based on the acquired real-time obstacle images and the dynamic obstacle model library, and to plan the obstacle avoidance route of the UAV based on the safe passage area of ​​the UAV. The filtering module is used to compare the flight parameters of each UAV obstacle avoidance route to determine the optimal flight route; The obstacle avoidance module is used to control the UAV to execute the optimal flight path to complete obstacle avoidance.

[0014] Compared with the prior art, the present invention has the following beneficial technical effects: This invention proposes a method for obstacle avoidance in photovoltaic power station inspections using unmanned aerial vehicles (UAVs). This method, through the precise construction of a photovoltaic power station model and a dynamic obstacle model library, can comprehensively and in real-time reflect the complex environment of the photovoltaic power station, including fixed power station facilities, terrain features, and potential temporary obstacles. This provides the UAV with detailed environmental information. By matching real-time obstacle images with the dynamic obstacle model library, obstacles in the current environment can be quickly identified and located. Furthermore, a safe flight zone for the UAV can be precisely delineated within the photovoltaic power station model. This not only improves the automation level of UAV inspections but also significantly enhances its ability to respond to emergencies, ensuring the continuity and safety of inspection tasks. Based on the planning of the safe flight zone, multiple obstacle avoidance routes can be generated for the UAV to choose from. Through comparison... The flight parameters of these routes, such as flight time, energy consumption, and safety, can intelligently determine the optimal flight route, optimizing the flight efficiency of the UAV and improving the accuracy and reliability of the inspection task. When the UAV performs the inspection task according to the optimal flight route, it can effectively avoid all obstacles, ensuring the safe and smooth progress of the inspection process, reducing the risk of collisions that may occur during the inspection, and also reducing blind spots caused by improper obstacle avoidance, thereby improving the overall operation and maintenance level of the photovoltaic power station. The obstacle avoidance method for UAV inspection of photovoltaic power stations in this invention, by constructing an accurate power station model and a dynamic obstacle model library, combined with real-time obstacle recognition and optimal flight route planning, realizes efficient and safe inspection of UAVs in complex environments, improving the operation and maintenance efficiency and safety of photovoltaic power stations.

[0015] Furthermore, this method acquires and extracts features from high-resolution terrain and vegetation images to accurately obtain altitude and vegetation height information for the photovoltaic power station area. Correction is performed using the lowest altitude data as a benchmark, ensuring the accuracy and consistency of subsequent model construction. By combining terrain and vegetation feature data with specific facility information such as photovoltaic panel arrays, roads, buildings, and fences, a comprehensive and three-dimensional photovoltaic power station model is constructed. This model not only reflects the natural environmental characteristics of the power station but also accurately depicts the layout of the main facilities within the power station. Based on this refined photovoltaic power station model, drones can more accurately identify and avoid obstacles during inspections. Whether it's terrain undulations, vegetation cover, or specific facilities within the power station, the drone can make timely and accurate obstacle avoidance decisions based on precise data from the model, greatly improving the safety and efficiency of inspection tasks. By integrating multiple data sources and constructing a refined model, the intelligent and automated obstacle avoidance system for photovoltaic power station inspection drones is achieved. This not only reduces the burden of manual operation but also improves the accuracy and coverage of inspections, providing strong support for the long-term stable operation of photovoltaic power stations.

[0016] Furthermore, this method, by constructing a comprehensive obstacle database and a dynamic obstacle model library, combined with real-time obstacle images and environmental data, achieves accurate identification and positioning of obstacles during UAV inspections. By setting safe flight distances and determining safe flight contour areas, it ensures safe flight of UAVs in complex environments, improves the efficiency and accuracy of UAV inspections, and significantly reduces safety risks during flight.

[0017] This invention proposes a system for obstacle avoidance in photovoltaic power plant inspections based on unmanned aerial vehicles (UAVs). This system automates the entire process from data input to obstacle avoidance execution, significantly improving the efficiency and accuracy of photovoltaic power plant inspections. The input module can flexibly receive photovoltaic power plant parameters and obstacle databases, providing rich data sources for subsequent model building and path planning. The model building module, based on this data, constructs accurate photovoltaic power plant models and a dynamic obstacle model library, providing a solid foundation for UAV obstacle avoidance decisions in complex environments. The path planning module is the core of the system; it can process real-time obstacle images and compare them with the dynamic obstacle model library to quickly determine the safe flight area for the UAV within the photovoltaic power plant model. Based on this, the module can plan multiple obstacle avoidance routes and intelligently evaluate these routes through a screening module to ultimately determine the optimal flight path. This improves the flexibility of UAV obstacle avoidance and ensures the safety and efficiency of the UAV during inspection. The obstacle avoidance execution module, as the system's execution terminal, accurately controls the UAV to perform inspections along the optimal flight path, effectively avoiding various obstacles. This not only reduces the burden of manual operation but also improves the automation level and reliability of inspection tasks. Through real-time feedback and adjustments, the system can continuously optimize obstacle avoidance strategies and improve overall obstacle avoidance performance. The obstacle avoidance system for UAV inspection of photovoltaic power plants proposed in this invention integrates multiple functional modules to achieve intelligent management of the entire process from data input to obstacle avoidance execution. This system not only improves the efficiency and accuracy of photovoltaic power plant inspections but also ensures the safety and reliability of UAVs in complex environments, providing strong support for the long-term stable operation of photovoltaic power plants. Attached Figure Description

[0018] Figure 1 This is a flowchart of the method for drone inspection and obstacle avoidance of photovoltaic power stations proposed in this invention. Detailed Implementation

[0019] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0020] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0021] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0022] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a communication connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0023] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" of the second feature includes the first feature being directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature being directly above or diagonally above the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0024] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0025] See Figure 1 This invention proposes a method for unmanned aerial vehicle (UAV) inspection and obstacle avoidance in photovoltaic power plants. The method includes the following steps: A photovoltaic power station model is constructed based on the acquired photovoltaic power station parameters, and a dynamic obstacle model library is constructed based on the constructed obstacle database. Specifically, obtain the photovoltaic power station parameters; Specifically, the topographic images of the photovoltaic power station and the vegetation images of the area are obtained by surveying drones. The topographic images and vegetation images of the area are manually marked to mark the clear and blurry areas in the topographic images and vegetation images of the area. The surveying drones fly at low altitude to the corresponding areas to collect data again, and the image data corresponding to the unclear areas in the topographic images and vegetation images of the area are collected. All image data are input into the SSD network model for feature extraction to obtain the elevation data of all terrain features in the area where the photovoltaic power station is located and the elevation data of all vegetation in the area where the photovoltaic power station is located. Among them, the data with the smallest elevation data is selected to obtain the lowest elevation data. The model construction baseline for the photovoltaic power station area is based on the lowest altitude data. The altitude data of all terrain features and vegetation height data are corrected. The corrected altitude data and vegetation height data are sorted out to obtain corrected altitude data and corrected vegetation height data. The model construction baseline, corrected altitude data and corrected vegetation height data are integrated to obtain the parameters of the first photovoltaic power station. Acquire images of photovoltaic panel arrays, roads, buildings, and fences. Input these images into the SSD network model for feature extraction to obtain parameters for the second photovoltaic power station. The parameters of the photovoltaic power station are obtained by integrating the parameters of the first photovoltaic power station and the parameters of the second photovoltaic power station. Specifically, a photovoltaic power station model is constructed based on photovoltaic power station parameter data; A basic model of the photovoltaic power station is constructed using the model of the first photovoltaic power station parameters, the baseline line, and the corrected altitude data. A vegetation model is constructed in the basic model of the photovoltaic power station using the corrected vegetation height data, resulting in a preliminary model of the photovoltaic power station. A photovoltaic panel array model, a road model, a building model, and a fence model are constructed in the preliminary model of the photovoltaic power station using the parameters of the second photovoltaic power station, resulting in the photovoltaic power station model. Images of obstacles encountered by drones during flight within photovoltaic power stations are obtained from the internet and used as training samples. The types of obstacles in the training samples are then manually labeled. The labeled training samples are then input into the SSD network model for feature extraction to generate an obstacle database. A dynamic obstacle model library is then constructed using the obstacle database. The obstacle images include dynamic obstacle images from various perspectives (such as top view, bottom view, side view, and eye view). The types of dynamic obstacles in the obstacle images are manually labeled to obtain dynamic obstacle type images. The dynamic obstacle type images are then input into the SSD network model for training to build an obstacle database. A dynamic obstacle model library is then built based on the obstacle database. Dynamic obstacles include birds, animals, and swaying vegetation. In the above method, obstacle images of drones during flight are obtained from the network, and the types of dynamic obstacles are manually labeled. Then, the SSD network module is used to extract and train the obstacle database, which makes the constructed obstacle database comprehensive and high-resolution. At the same time, the obstacle database classifies the types of obstacles, so that the dynamic obstacle types constructed based on the obstacle database can be applied to drones to safely avoid multiple obstacles during the inspection of photovoltaic power stations, thereby improving the safety of drone inspection.

[0026] Based on the acquired real-time obstacle images and dynamic obstacle model library, the safe flight area of ​​the UAV is determined in the photovoltaic power station model; Specifically, real-time obstacle images are acquired, and the real-time obstacle images are input into a dynamic obstacle model library for comparison, and the dynamic obstacle model corresponding to the real-time obstacle image is extracted. Acquire real-time obstacle environment data and real-time location of obstacles, construct obstacle environment model based on real-time obstacle environment data, locate the obstacle in the photovoltaic power station model, determine the real-time obstacle position of the obstacle environment model in the photovoltaic power station model, input the dynamic obstacle model into the real-time obstacle position in the photovoltaic power station model, and determine the drone obstacle avoidance area corresponding to the real-time obstacle position. A safe flight distance is set during the drone's flight. Based on this safe flight distance, the outline region of objects in the photovoltaic power station model is determined. That is, a safe region is added to the boundary of the outline region of objects in the photovoltaic power station model to obtain the safe flight outline region of the drone. The safe flight zone for humans and drones is determined based on the flight safety contour area and the drone obstacle avoidance area.

[0027] Based on the constructed real-time flight environment model in the photovoltaic power station model, the drone's real-time position in the photovoltaic power station model is located. Based on the drone's real-time position, the constructed drone model and the real-time obstacle position, the drone safely passes through the area and multiple drone flight routes are planned. Specifically, real-time flight environment data of the UAV's real-time location is obtained, a real-time flight environment model is constructed using the real-time flight environment data, and the real-time flight environment model is input into the photovoltaic power station model for comparison. That is, the real-time flight environment model is compared with a local area model of the photovoltaic power station model to locate the UAV's real-time position in the photovoltaic power station model, thus obtaining the UAV's real-time position in the photovoltaic power station model. Obtain the size parameters of the drone and construct a drone model based on the size parameters; input the drone model into the real-time position of the drone in the photovoltaic power station model, simulate the drone's flight at the real-time position, and obtain the flight area of ​​the drone at the real-time position. A no-entry zone for the drone itself is constructed in the flight area to obtain a safe flight space for the drone. Taking the center point of the safe flight space as the starting point and the center point of the safe passage area for the drone corresponding to the real-time obstacle position as the ending point, multiple drone flight curves are planned to avoid the outline areas of all objects in the photovoltaic power station model, thus obtaining multiple drone flight routes. By simulating movement along each drone's flight path within the safe flight space area, drone flight paths are filtered to obtain drone obstacle avoidance routes. At the same time, the flight parameters of each drone obstacle avoidance route are recorded. Specifically, the safe flight space of the UAV is simulated along each UAV flight path. During the simulation, when the safe flight space of the UAV overlaps with the outline of an object in the photovoltaic power station model, the safe flight space of the UAV is rotated around the flight direction of the UAV at that position as the rotation axis. Then, it is determined whether the safe flight space of the UAV overlaps with the outline of the object. If it still overlaps, it is rotated again. During the rotation, the safe flight space of the UAV rotates by 10° each time until it still overlaps after one full rotation. If it still overlaps, the UAV flight path is discarded. If it does not overlap with the outline of the object in the photovoltaic power station model after rotation or without rotation, the UAV flight path is retained. All retained UAV flight paths are integrated to obtain the UAV obstacle avoidance path. At the same time, during the model flight, the flight parameters of each UAV flight path are recorded, namely, flight time, flight path length, flight distance of the UAV from all objects, UAV attitude adjustment angle, and deviation value of technical points from the preset inspection path.

[0028] By comparing the flight parameters of each drone flight path, the optimal flight path is determined, and the drone executes the optimal flight path to complete obstacle avoidance.

[0029] Specifically, the flight time, flight path length, flight speed, flight altitude, drone attitude adjustment angle, and deviation values ​​of technical points from the preset inspection route are obtained for each drone flight path. The flight time, flight path length, flight speed, flight altitude, drone attitude adjustment angle, and deviation values ​​of the obstacle avoidance end point from the preset inspection route are compared to determine the drone obstacle avoidance route with the shortest flight path length, the smallest attitude adjustment angle, and the largest flight distance of the drone from all objects. Then, the drone's current battery information is obtained, and an optimal flight path is determined based on the drone's battery information. The drone executes the optimal flight path to complete obstacle avoidance.

[0030] Preferably, the method further includes: acquiring first motion information of the drone and second motion information of the obstacle; determining whether the obstacle is tracking the drone based on the first motion information and the second motion information; and if so, sending the tracking information to a preset smart terminal.

[0031] The process involves extracting motion parameters from the first motion information and the second motion information to obtain multiple first motion parameter values ​​and second motion parameter values. The motion parameters also include the distance between the drone and the obstacle. This distance is the interval between the drone and the obstacle at two adjacent moments after the drone completes obstacle avoidance. By comparing the interval between two adjacent moments, it is determined whether the obstacle is moving towards the drone, thereby obtaining a preliminary judgment on whether the obstacle is tracking the drone. If so, the tracking information is sent to a preset smart terminal. The first motion parameter value includes: the real-time motion direction of the UAV and the real-time speed of the UAV; the second motion parameter value includes: the real-time motion direction of the obstacle and the real-time speed of the obstacle. A first motion state vector of the first motion information is constructed based on the first motion parameter value, and a second motion state vector of the second motion information is constructed based on the second motion parameter value. The first motion state vector and the second motion state vector are both motion paths constructed from real-time motion direction and data; Obtain a preset tracking event, extract the third motion state vector of the tracked target and the fourth motion state vector of the tracked target from the tracking event, compare the first motion state vector with the third motion state vector, the fourth motion state vector and the second motion state vector to obtain the comparison value; If the comparison value is within the preset comparison threshold range, it is determined that the obstacle avoidance device is tracking the drone, and the tracking information is sent to the preset smart terminal; if the comparison value exceeds the preset comparison threshold range, it is determined that the obstacle avoidance device is not tracking the drone.

[0032] The preset tracking events are specifically: historical aerial tracking records, which include: the motion curve and motion state vector of the tracked target constructed by its direction of motion and speed of motion, and the motion curve and motion state vector of the tracked target constructed by its direction of motion and speed of motion.

[0033] In this invention, the distance between the obstacle and the drone at two adjacent moments is compared to initially determine whether the obstacle is tracking the drone. Then, the first motion state information of the drone is compared with the second motion state information of the obstacle to further determine whether the obstacle is tracking the drone. By using the above two methods, it is possible to accurately determine whether the obstacle is tracking the drone, thus improving the rationality and accuracy of the judgment.

[0034] This invention proposes a system for obstacle avoidance in photovoltaic power station inspections based on unmanned aerial vehicles (UAVs). Based on the above-mentioned method, the system is characterized by including an input module, a model building module, a path planning module, a filtering module, and an obstacle avoidance execution module. The system includes an input module for inputting photovoltaic power station parameters and an obstacle database; a model building module for constructing a photovoltaic power station model based on the acquired parameters and a dynamic obstacle model library based on the constructed obstacle database; a path planning module for determining a safe flight area for the UAV within the photovoltaic power station model based on the acquired real-time obstacle images and the dynamic obstacle model library, and for planning an obstacle avoidance route for the UAV based on the safe passage area; a filtering module for comparing the flight parameters of each UAV obstacle avoidance route to determine the optimal flight route; and an obstacle avoidance execution module for controlling the UAV to execute the optimal flight route to complete obstacle avoidance.

[0035] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0036] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A method for obstacle avoidance during unmanned aerial vehicle (UAV) inspections of photovoltaic power stations, characterized in that: Includes the following steps: A photovoltaic power station model is constructed based on the acquired photovoltaic power station parameters, and a dynamic obstacle model library is constructed based on the constructed obstacle database. Based on the acquired real-time obstacle images and the dynamic obstacle model library, the safe flight area of ​​the UAV is determined in the photovoltaic power station model; Based on the aforementioned drone safety, an obstacle avoidance route for the drone is obtained through regional planning. By comparing the flight parameters of each of the aforementioned drone obstacle avoidance routes, the optimal flight route is determined, and the drone executes the optimal flight route to complete obstacle avoidance.

2. The method for obstacle avoidance in photovoltaic power station inspection based on unmanned aerial vehicle (UAV) as described in claim 1, characterized in that, The acquisition of the photovoltaic power station parameters includes: Acquire images of the terrain and vegetation where the photovoltaic power station is located; Feature extraction is performed on the terrain and vegetation images to obtain elevation data and vegetation height data; Select the lowest altitude data from the given altitude data to obtain the lowest altitude data; Using the lowest altitude data as the baseline for model construction of the area where the photovoltaic power station is located, the altitude data and the vegetation height data are corrected to obtain corrected altitude data and corrected vegetation height data. The parameters of the first photovoltaic power station are obtained by integrating the model construction baseline, the corrected altitude data, and the corrected vegetation height data; Acquire images of photovoltaic panel arrays, roads, buildings, and fences; Feature extraction is performed on the photovoltaic panel array image, the road image, the building image, and the fence image to obtain the parameters of the second photovoltaic power station; The photovoltaic power station parameters are obtained by integrating the parameters of the first photovoltaic power station and the parameters of the second photovoltaic power station.

3. The method for obstacle avoidance in photovoltaic power station inspection based on unmanned aerial vehicle (UAV) as described in claim 2, characterized in that, The construction of the photovoltaic power station model based on photovoltaic power station parameter data includes: A basic model of the photovoltaic power station is constructed using the model construction baseline and the corrected altitude data in the parameters of the first photovoltaic power station. A vegetation model is constructed in the basic model of the photovoltaic power station using the corrected vegetation height data to obtain a preliminary model of the photovoltaic power station. Using the parameters of the second photovoltaic power station, a photovoltaic panel array model, a road model, a building model, and a fence model are constructed in the preliminary model of the photovoltaic power station to obtain the photovoltaic power station model.

4. The method for obstacle avoidance in photovoltaic power station inspection based on unmanned aerial vehicle (UAV) as described in claim 1, characterized in that, The construction of the obstacle model based on the constructed obstacle database includes: Images of obstacles encountered by drones during flight within a photovoltaic power station are used as training samples; Manually label the obstacle types in the training samples, and extract features from the labeled obstacle types of the training samples to generate an obstacle database; A dynamic obstacle model library was constructed based on the obstacle database.

5. The method for obstacle avoidance in photovoltaic power station inspection based on unmanned aerial vehicle (UAV) as described in claim 1, characterized in that, The process of determining the safe flight zone for the UAV within the photovoltaic power station model based on the acquired real-time obstacle images and dynamic obstacle model library includes: Acquire real-time obstacle images, input the real-time obstacle images into the dynamic obstacle model library for comparison, and extract the dynamic obstacle model corresponding to the real-time obstacle image; Obtain the real-time environmental data and real-time location of the obstacle; An obstacle environment model is constructed based on the environmental data, and the obstacle is located in the photovoltaic power station model to determine the real-time obstacle position of the obstacle environment model in the photovoltaic power station model. The dynamic obstacle model is input into the real-time obstacle position in the photovoltaic power station model to determine the drone obstacle avoidance area corresponding to the real-time obstacle position; Set a safe flight distance for the drone during flight; Based on the flight safety distance, the outline region of the object in the photovoltaic power station model is determined, and the flight safety outline region of the UAV is obtained. The safe flight zone for humans and drones is determined based on the flight safety contour area and the drone obstacle avoidance area.

6. The method for obstacle avoidance in photovoltaic power station inspection based on unmanned aerial vehicle (UAV) according to claim 1, characterized in that, The planning of multiple drone flight routes based on the safe passage of the drone through the area includes: Acquire real-time flight environment data of the UAV's real-time location, and construct a real-time flight environment model based on the real-time flight environment data; The real-time flight environment model is input into the photovoltaic power station model for comparison, and the real-time position of the UAV in the photovoltaic power station model is located to obtain the real-time position of the UAV. Based on the constructed drone model, the flight area at the drone's real-time location is simulated to obtain the drone's flight area at its real-time location. A no-entry zone for the drone itself is constructed in the flight area to obtain a safe flight space for the drone. Starting from the center point of the safe flight space of the UAV and ending at the center point of the safe passage area of ​​the UAV corresponding to the real-time obstacle position, multiple UAV flight curves are planned to avoid all object outline areas in the photovoltaic power station model, resulting in multiple UAV flight routes. Simulate each of the aforementioned drone flight paths, filter the drone flight paths, and obtain the drone obstacle avoidance path.

7. The method for obstacle avoidance in photovoltaic power station inspection based on unmanned aerial vehicle (UAV) according to claim 6, characterized in that, The process of simulating each of the drone flight paths and filtering the drone flight paths includes: The safe flight space of the UAV is simulated along each UAV flight path. It is determined whether the safe flight space of the UAV overlaps with the outline area of ​​the object. If they overlap, the UAV is rotated around the flight direction of the UAV at that position. If an overlapping area still appears after the rotation, the UAV flight path is discarded. If the UAV flight path does not overlap with the outline area of ​​the object in the photovoltaic power station model after the rotation or without rotation, the UAV flight path is retained. Integrate all retained drone flight paths to obtain drone obstacle avoidance routes; and record the flight parameters of each drone flight path.

8. The method for obstacle avoidance in photovoltaic power station inspection based on unmanned aerial vehicle (UAV) according to claim 7, characterized in that, The process of comparing the flight parameters of each UAV obstacle avoidance route to determine the optimal flight route includes: The flight time, flight path length, flight speed, flight altitude, drone attitude adjustment angle, and deviation value between the obstacle avoidance end point and the preset inspection route are obtained for each drone flight path. By comparing the flight time, flight path length, flight speed, flight altitude, drone attitude adjustment angle, and deviation value between the end point and the preset inspection route for each drone flight path, the drone obstacle avoidance route with the shortest flight path length, the smallest drone attitude adjustment angle, and the largest drone flight distance from all objects is determined. Based on the drone's current battery level, an optimal flight path is determined, and then the drone is controlled to avoid obstacles.

9. The method for obstacle avoidance in photovoltaic power station inspection based on unmanned aerial vehicle (UAV) according to claim 1, characterized in that, The method further includes: acquiring first motion information of the drone and second motion information of the obstacle; determining whether the obstacle is tracking the drone based on the first motion information and the second motion information; and if so, sending the tracking information to a preset smart terminal.

10. A system for obstacle avoidance in photovoltaic power station inspections based on unmanned aerial vehicles (UAVs), based on the method described in any one of claims 1 to 9, characterized in that, It includes an input module, a model building module, a path planning module, a filtering module, and an obstacle avoidance execution module; The input module is used to input photovoltaic power station parameters and an obstacle database. The model building module is used to build a photovoltaic power station model based on the acquired photovoltaic power station parameters and to build a dynamic obstacle model library based on the constructed obstacle database. The path planning module is used to determine the safe flight area of ​​the UAV in the photovoltaic power station model based on the acquired real-time obstacle images and the dynamic obstacle model library, and to plan the obstacle avoidance route of the UAV based on the safe passage area of ​​the UAV. The filtering module is used to compare the flight parameters of each UAV obstacle avoidance route to determine the optimal flight route; The obstacle avoidance module is used to control the UAV to execute the optimal flight path to complete obstacle avoidance.

Images