Photovoltaic inspection unmanned aerial vehicle path planning method
By constructing precise grid maps and planning paths using multi-objective optimization algorithms, combined with three-dimensional models and no-fly policies, the problem of efficient, low-cost, and safe drone inspection path planning was solved, and efficient, low-cost, and safe drone inspections in complex environments were achieved.
Patent Information
- Application Number
- CN202511123240.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-12
- Publication Date
- 2025-10-17
AI Technical Summary
Existing drone inspection path planning methods fail to fully integrate the three-dimensional terrain characteristics, building height information and drone energy consumption characteristics of the inspection area, resulting in the inability to achieve efficient, low-cost and safe inspection operations in complex environments.
By constructing a precise grid map to obtain building height information, using a multi-objective optimization algorithm to plan the path, combining the three-dimensional model and no-fly strategy to determine the safe area, and comprehensively optimizing the low-cost obstacle avoidance path, we ensure that drones can perform inspection tasks efficiently, with low consumption and safety in complex environments.
It realizes efficient, low-consumption and safe operation of drone inspections in complex environments, meeting the actual needs of photovoltaic inspections.
Smart Images
Figure CN120802996A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to unmanned aerial vehicle inspection, in particular to a photovoltaic inspection unmanned aerial vehicle path planning method. BACKGROUND
[0002] With the rapid development of renewable energy technology, photovoltaic power generation as a clean, sustainable energy form has been widely used around the world. The scale of photovoltaic power stations is expanding, and its operation and management has become a key link to ensure power generation efficiency and safety.
[0003] Unmanned aerial vehicle inspection technology can quickly obtain image data of photovoltaic modules by carrying high-definition cameras, infrared thermal imagers and other sensors, and can realize comprehensive monitoring of the inspection area. However, in actual application, the planning of the unmanned aerial vehicle inspection path has become a problem to be solved. On the one hand, distributed photovoltaic modules are usually set on the top of buildings, and the terrain is complex and changeable, with obstacles such as buildings and trees, which requires the unmanned aerial vehicle inspection path to have an obstacle avoidance function to ensure the safety and reliability of the inspection process. On the other hand, the endurance of the unmanned aerial vehicle is limited, and how to plan an inspection path with the least flight power consumption is of great significance to improve the inspection efficiency and prolong the operation time of the unmanned aerial vehicle.
[0004] At present, although there have been some researches on the planning of the unmanned aerial vehicle inspection path, most of these researches focus on the path planning in two-dimensional plane or only consider simple obstacle avoidance requirements, and fail to fully combine the three-dimensional terrain features of the inspection area, building height information and the energy consumption characteristics of the unmanned aerial vehicle for comprehensive optimization. Therefore, the existing unmanned aerial vehicle inspection path planning method cannot meet the actual needs of photovoltaic inspection, and cannot ensure that the unmanned aerial vehicle can realize efficient, low-consumption and safe inspection operation in complex environments. SUMMARY
[0005] (I) Technical problems solved
[0006] In view of the above shortcomings of the prior art, the present application provides a photovoltaic inspection unmanned aerial vehicle path planning method, which can effectively overcome the defects of the prior art that cannot ensure that the unmanned aerial vehicle can realize efficient, low-consumption and safe inspection operation in complex environments.
[0007] (II) Technical solutions
[0008] In order to achieve the above purpose, the present application is realized by the following technical solutions:
[0009] A photovoltaic inspection unmanned aerial vehicle path planning method, comprising the following steps:
[0010] S1, constructing a precise grid map of the inspection area according to the collected images of the unmanned aerial vehicle;
[0011] S2, acquiring height information of buildings in the inspection area, and planning an inspection path by using a multi-objective optimization algorithm;
[0012] S3, evaluating flight power consumption of different inspection paths, and taking the inspection path with the least flight power consumption as a low-consumption inspection path;
[0013] S4, determining a safe inspection area in the inspection area according to a three-dimensional model of the inspection area and a no-fly strategy;
[0014] S5, determining a task point set according to the inspection plan and the three-dimensional model of the inspection area;
[0015] S6, determining an obstacle-avoiding inspection path according to the safe inspection area, the task point set and the unmanned aerial vehicle information;
[0016] S7, determining a low-consumption obstacle-avoiding optimal inspection path of the unmanned aerial vehicle by combining the low-consumption inspection path and the obstacle-avoiding inspection path.
[0017] Preferably, the accurate grid map of the inspection area is constructed according to the collection image of the unmanned aerial vehicle in S1, comprising:
[0018] The unmanned aerial vehicle route is initially positioned manually, and the unmanned aerial vehicle performs an inspection task along the unmanned aerial vehicle route at a preset height to obtain a collection image of the inspection area;
[0019] The planar grid map of the inspection area is constructed by combining the collection image of the inspection area and a map plan;
[0020] The coordinate information of the buildings in the planar grid map of the inspection area is obtained by using image recognition technology, and an accurate grid map containing building height information is obtained.
[0021] Preferably, the planar grid map of the inspection area is constructed by combining the collection image of the inspection area and a map plan, comprising:
[0022] The map plan or satellite image data of the inspection area is obtained, the collection image of the inspection area is processed by removing distortion and adjusting contrast, and the collection image of the inspection area and the map plan are aligned and registered in the same coordinate system by using image registration technology;
[0023] The position and boundary information of the photovoltaic module are obtained by using edge detection and segmentation algorithm for target extraction;
[0024] The extracted photovoltaic module region and the ground object information except the photovoltaic module region are converted into grid data, and the grid data and the map plan of the inspection area after alignment and registration are fused to obtain the planar grid map of the inspection area.
[0025] Preferably, the height information of the building in the inspection area is obtained in S2, and a multi-objective optimization algorithm is used for inspection path planning, including:
[0026] The height information of the building in the inspection area is obtained from the accurate grid map of the inspection area, and the objective function and the corresponding constraint condition of the multi-objective optimization algorithm are determined, and the multi-objective optimization algorithm is used for inspection path planning to obtain a plurality of inspection paths.
[0027] Preferably, the objective function of the multi-objective optimization algorithm includes a first objective function and a second objective function, the first objective function is to minimize the total length of the flight path, and the second objective function is to minimize the number of turns of the flight path. The first objective function and the second objective function are combined into the objective function of the multi-objective optimization algorithm in the form of weighted sum.
[0028] Preferably, the flight power consumption of different inspection paths is evaluated in S3, and the inspection path with the least flight power consumption is taken as the low-consumption inspection path, including:
[0029] According to the change of the building height information in the inspection path, the total length of the flight path is calculated;
[0030] According to the change of the building height information and the azimuth information in the inspection path, the number of turns of the flight path is calculated;
[0031] According to the total length of the flight path and the number of turns of the flight path, the flight power consumption of the inspection path is evaluated, and the inspection path with the least flight power consumption is taken as the low-consumption inspection path.
[0032] Preferably, the total length of the flight path is calculated according to the change of the building height information in the inspection path, including:
[0033] The inspection path is divided into a plurality of path segments, the spatial length of each path segment is calculated according to the change of the building height information in the path segment, and the spatial lengths of all path segments are added to obtain the total length of the flight path.
[0034] Preferably, the number of turns of the flight path is calculated according to the change of the building height information and the azimuth information in the inspection path, including:
[0035] The positions of the turning points in the inspection path are determined in combination with the changes of the building height information and the azimuth information in the inspection path;
[0036] An angle threshold is set, and the number of turning points exceeding the angle threshold is counted, and the number of turning points is taken as the number of turns of the flight path.
[0037] Preferably, the obstacle-avoiding inspection path is determined according to the safe inspection area, the task point set and the unmanned aerial vehicle information in S6, including:
[0038] determine the dangerous degree of the two-side buildings in the safe inspection area to the unmanned aerial vehicle;
[0039] filter all the task points in the task point set according to the data acquisition equipment information on the unmanned aerial vehicle and the unmanned aerial vehicle characteristic information;
[0040] plan an inspection path for the filtered task point set according to the dangerous degree of the two-side buildings in the safe inspection area to the unmanned aerial vehicle and the power information of the unmanned aerial vehicle, and obtain an obstacle-avoiding inspection path.
[0041] Preferably, the unmanned aerial vehicle information comprises the data acquisition equipment information on the unmanned aerial vehicle, the unmanned aerial vehicle characteristic information and the power information of the unmanned aerial vehicle.
[0042] (III) Advantages
[0043] Compared with the prior art, the photovoltaic inspection unmanned aerial vehicle path planning method has the following advantages:
[0044] 1) The precise grid map of the inspection area is constructed according to the collected images of the unmanned aerial vehicle, the height information of the buildings in the inspection area is obtained, the multi-objective optimization algorithm is used for inspection path planning, the total length of the flight path is calculated according to the change of the height information of the buildings in the inspection path, the number of turns of the flight path is calculated according to the change of the height information and the azimuth information of the buildings in the inspection path, the flight power consumption of the inspection path is evaluated according to the total length of the flight path and the number of turns of the flight path, and the inspection path with the least flight power consumption is taken as the low-consumption inspection path, so that the low-consumption inspection path can be obtained by fully combining the building height information and the energy consumption characteristics of the unmanned aerial vehicle for comprehensive optimization;
[0045] 2) The safe inspection area in the inspection area is determined according to the three-dimensional model of the inspection area and the no-fly strategy, the task point set is determined according to the inspection plan and the three-dimensional model of the inspection area, the dangerous degree of the two-side buildings in the safe inspection area to the unmanned aerial vehicle is determined, all the task points in the task point set are filtered according to the data acquisition equipment information on the unmanned aerial vehicle and the unmanned aerial vehicle characteristic information, the inspection path for the filtered task point set is planned according to the dangerous degree of the two-side buildings in the safe inspection area to the unmanned aerial vehicle and the power information of the unmanned aerial vehicle, and the obstacle-avoiding inspection path is obtained, so that the obstacle-avoiding inspection path can be obtained by fully combining the three-dimensional terrain characteristics of the inspection area and the dangerous degree of the buildings to the unmanned aerial vehicle for comprehensive optimization;
[0046] 3) combine the low-consumption inspection path and the obstacle avoidance inspection path to determine the low-consumption obstacle avoidance optimal inspection path of the unmanned aerial vehicle, so that the building height information, the energy consumption characteristics of the unmanned aerial vehicle, the three-dimensional terrain characteristics of the inspection area and the dangerous degree of the building to the unmanned aerial vehicle are comprehensively optimized to obtain the low-consumption obstacle avoidance optimal inspection path, effectively meet the actual needs of photovoltaic inspection, and ensure that the unmanned aerial vehicle realizes efficient, low-consumption and safe inspection operation in a complex environment. BRIEF DESCRIPTION OF DRAWINGS
[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.
[0048] Figure 1 The flowchart of the present application is shown in the figure.
[0049] Figure 2 The flowchart of obtaining the low-consumption inspection path in the present application is shown in the figure.
[0050] Figure 3 The flowchart of obtaining the obstacle avoidance inspection path in the present application is shown in the figure. DETAILED DESCRIPTION
[0051] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the present application.
[0052] A photovoltaic inspection unmanned aerial vehicle path planning method, as shown in Figure 1 and Figure 2 S1, according to the collection image of the unmanned aerial vehicle, constructs the accurate grid map of the inspection area, specifically including:
[0053] The initial manual positioning of the unmanned aerial vehicle route is performed, and the unmanned aerial vehicle performs the inspection task along the unmanned aerial vehicle route at a preset height to obtain the collection image of the inspection area;
[0054] Combined with the collection image and the plan map of the inspection area, the plan grid map of the inspection area is constructed;
[0055] The coordinate information of the building in the plan grid map of the inspection area is obtained by using image recognition technology, and the accurate grid map containing the building height information is obtained.
[0056] Specifically, a planar grid map of the inspection area is constructed in combination with the collected images and the map plan of the inspection area, comprising:
[0057] Obtaining the map plan or satellite image data of the inspection area, removing distortion and adjusting contrast of the collected images of the inspection area, and aligning and registering the collected images and the map plan of the inspection area in the same coordinate system by using image registration technology;
[0058] Using edge detection and segmentation algorithm for target extraction to obtain the position and boundary information of the photovoltaic components;
[0059] Converting the extracted photovoltaic component region and the ground object information other than the photovoltaic component region into grid data, and fusing the grid data and the aligned and registered map plan of the inspection area to obtain the planar grid map of the inspection area.
[0060] S2, obtaining the height information of the buildings in the inspection area, and planning the inspection path by using a multi-objective optimization algorithm, specifically comprising:
[0061] Obtaining the height information of the buildings in the inspection area from the accurate grid map of the inspection area, determining the objective function and the corresponding constraint condition of the multi-objective optimization algorithm, and planning the inspection path by using the multi-objective optimization algorithm to obtain a plurality of inspection paths.
[0062] In the technical solution of the present application, the objective function of the multi-objective optimization algorithm includes a first objective function and a second objective function, the first objective function is to minimize the total length of the flight path, and the second objective function is to minimize the number of turns of the flight path, and the first objective function and the second objective function are combined into the objective function of the multi-objective optimization algorithm in the form of weighted sum.
[0063] S3, evaluating the flight power consumption of different inspection paths, and taking the inspection path with the least flight power consumption as the low-consumption inspection path, specifically comprising:
[0064] According to the change of the building height information in the inspection path, the total length of the flight path is calculated;
[0065] According to the change of the building height information and the azimuth information in the inspection path, the number of turns of the flight path is calculated;
[0066] According to the total length of the flight path and the number of turns of the flight path, the flight power consumption of the inspection path is evaluated, and the inspection path with the least flight power consumption is taken as the low-consumption inspection path.
[0067] 1) According to the change of the building height information in the inspection path, the total length of the flight path is calculated, comprising:
[0068] The inspection path is divided into multiple path segments, the spatial length of each path segment is calculated according to the change of the building height information in the path segment, and the total length of the flight path is obtained by adding the spatial lengths of all path segments.
[0069] 2) According to the change of the building height information and the orientation information in the inspection path, the number of turns of the flight path is calculated, including:
[0070] According to the change of the building height information and the orientation information in the inspection path, the position of the turning point in the inspection path is determined;
[0071] An angle threshold is set, and the number of turning points exceeding the angle threshold is counted, and the number of turning points is taken as the number of turns of the flight path.
[0072] The above technical solution constructs a precise grid map of the inspection area according to the collected images of the unmanned aerial vehicle, obtains the height information of the buildings in the inspection area, plans the inspection path by using a multi-objective optimization algorithm, calculates the total length of the flight path according to the change of the building height information in the inspection path, calculates the number of turns of the flight path according to the change of the building height information and the orientation information in the inspection path, evaluates the flight power consumption of the inspection path according to the total length of the flight path and the number of turns of the flight path, and takes the inspection path with the least flight power consumption as the low-consumption inspection path, so that the low-consumption inspection path can be obtained by fully combining the building height information and the energy consumption characteristics of the unmanned aerial vehicle for comprehensive optimization.
[0073] As shown in Figure 1 and Figure 3 S4, according to the three-dimensional model of the inspection area and the no-fly strategy, the safe inspection area in the inspection area is determined.
[0074] S5, according to the inspection plan and the three-dimensional model of the inspection area, a task point set is determined.
[0075] S6, according to the safe inspection area, the task point set and the unmanned aerial vehicle information, an obstacle avoidance inspection path is determined, specifically including:
[0076] Determine the dangerous degree of the two-sided buildings in the safe inspection area to the unmanned aerial vehicle;
[0077] According to the data acquisition equipment information on the unmanned aerial vehicle and the unmanned aerial vehicle feature information, all task points in the task point set are filtered;
[0078] According to the dangerous degree of the two-sided buildings in the safe inspection area to the unmanned aerial vehicle and the unmanned aerial vehicle power information, the filtered task point set is planned for the inspection path to obtain the obstacle avoidance inspection path.
[0079] In the technical solution of this application, drone information includes data collection equipment information on the drone, drone feature information and drone power information.
[0080] The above technical solution determines the safe inspection area within the inspection area based on the three-dimensional model of the inspection area and the no-fly policy, determines the task point set based on the inspection plan and the three-dimensional model of the inspection area, determines the degree of danger of the buildings on both sides of the safe inspection area to the drone, filters all the task points in the task point set based on the data acquisition equipment information on the drone and the drone feature information, plans the inspection path for the filtered task point set based on the degree of danger of the buildings on both sides of the safe inspection area to the drone and the drone power information, and obtains an obstacle avoidance inspection path, thereby being able to fully combine the three-dimensional terrain characteristics of the inspection area and the degree of danger of the buildings to the drone for comprehensive optimization to obtain an obstacle avoidance inspection path.
[0081] like Figure 1 As shown, S7, combining the low-cost inspection path and the obstacle avoidance inspection path, determines the low-cost obstacle avoidance optimal inspection path of the UAV.
[0082] The above technical solution combines the low-consumption inspection path and the obstacle avoidance inspection path to determine the drone's optimal low-consumption obstacle avoidance inspection path. It can fully combine the building height information, the drone's energy consumption characteristics, the three-dimensional terrain characteristics of the inspection area and the danger level of the building to the drone for comprehensive optimization, and obtain the low-consumption obstacle avoidance optimal inspection path, effectively meeting the actual needs of photovoltaic inspections and ensuring that drones can achieve efficient, low-consumption and safe inspection operations in complex environments.
[0083] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.
Claims
1. A photovoltaic inspection drone path planning method, characterized by: The following steps are involved: S1. Construct an accurate grid map of the inspection area based on the images collected by the drone; S2. Obtain the height information of buildings in the inspection area and use a multi-objective optimization algorithm to plan the inspection path; S3. Evaluate the flight power consumption of different inspection routes and select the inspection route with the lowest flight power consumption as the low-consumption inspection route; S4. Determine the safe inspection area within the inspection area based on the three-dimensional model of the inspection area and the no-fly policy; S5. Determine the set of task points based on the inspection plan and the three-dimensional model of the inspection area; S6. Determine the obstacle avoidance inspection path based on the safety inspection area, mission point set, and drone information; S7. Combine the low-cost inspection path and the obstacle avoidance inspection path to determine the optimal low-cost obstacle avoidance inspection path for the UAV.
2. The photovoltaic inspection drone path planning method according to claim 1 is characterized by: S1 builds a precise grid map of the inspection area based on the images collected by the drone, including: Manually locate the drone route for the first time, so that the drone can perform inspection tasks along the drone route at a preset height and obtain collected images of the inspection area; Combining the collected images and the map plan of the inspection area, a planar grid map of the inspection area is constructed; Image recognition technology is used to obtain the coordinate information of buildings in the plane grid map of the inspection area, and an accurate grid map containing building height information is obtained.
3. The photovoltaic inspection drone path planning method according to claim 2 is characterized by: The step of combining the collected images and the map plan of the inspection area to construct a planar grid map of the inspection area includes: Obtain a map plan or satellite image data of the inspection area, remove distortion and adjust contrast of the collected image of the inspection area, and use image registration technology to align the collected image of the inspection area and the map plan in the same coordinate system; Use edge detection and segmentation algorithms to extract targets and obtain the location and boundary information of photovoltaic modules; The extracted PV module area and ground feature information other than the PV module area are converted into raster data, and the raster data are fused with the aligned map plan of the inspection area to obtain a planar raster map of the inspection area.
4. The photovoltaic inspection drone path planning method according to claim 2 is characterized by: S2 obtains the height information of buildings in the inspection area and uses a multi-objective optimization algorithm to plan the inspection path, including: The height information of buildings in the inspection area is obtained from the precise grid map of the inspection area, and the objective function and corresponding constraints of the multi-objective optimization algorithm are determined. The multi-objective optimization algorithm is used to plan the inspection path and obtain multiple inspection paths.
5. The photovoltaic inspection drone path planning method according to claim 4 is characterized by: The objective function of the multi-objective optimization algorithm includes a first objective function and a second objective function. The first objective function is to minimize the total length of the flight path, and the second objective function is to minimize the number of turns in the flight path. The first objective function and the second objective function constitute the objective function of the multi-objective optimization algorithm in the form of a weighted sum.
6. The photovoltaic inspection drone path planning method according to claim 5 is characterized by: In S3, the flight power consumption of different inspection paths is evaluated, and the inspection path with the lowest flight power consumption is selected as the low-consumption inspection path, including: Calculate the total length of the flight path based on the changes in building height information along the inspection path; Calculate the number of turns in the flight path based on changes in building height and orientation information along the inspection path; The flight power consumption of the inspection path is evaluated according to the total length of the flight path and the number of turns of the flight path, and the inspection path with the least flight power consumption is regarded as the low-consumption inspection path.
7. The photovoltaic inspection drone path planning method according to claim 6, characterized in that: The calculation of the total length of the flight path according to the change of building height information in the inspection path includes: The inspection path is divided into multiple path segments. The spatial length of the path segment is calculated according to the change of building height information in each path segment, and the total length of the flight path is obtained by adding up the spatial lengths of all path segments.
8. The photovoltaic inspection drone path planning method according to claim 7, characterized in that: The calculation of the number of turns of the flight path according to changes in building height information and orientation information in the inspection path includes: Determine the location of the turning point in the inspection route based on the changes in building height and orientation information along the inspection route; An angle threshold is set, and the number of turning points exceeding the angle threshold is counted, and the number of turning points is used as the number of turns in the flight path.
9. The photovoltaic inspection drone path planning method according to claim 1, characterized in that: In S6, the obstacle avoidance inspection path is determined based on the safety inspection area, task point set, and drone information, including: Determine the danger level of buildings on both sides of the safety inspection area to drones; Filter all mission points in the mission point set based on the data collection equipment information and the UAV feature information on the UAV; According to the danger level of the buildings on both sides of the safety inspection area to the drone and the drone's power information, the inspection path is planned for the filtered task point set to obtain the obstacle avoidance inspection path.
10. The photovoltaic inspection drone path planning method according to claim 9, characterized in that: The drone information includes data collection equipment information on the drone, drone feature information, and drone battery information.