Photovoltaic power station navigation method, device, computer equipment and storage medium

Abstract

The embodiment of the present invention discloses a photovoltaic power station navigation method, which collects photovoltaic power station site cloud data and constructs a photovoltaic power station map: collecting photovoltaic power station site cloud data based on a drone laser radar system; dividing the photovoltaic power station site cloud data into ground point cloud data of the power station area and photovoltaic panel group point cloud data of the power station area; constructing a photovoltaic power station map including height information based on the ground point cloud data of the power station area; identifying each photovoltaic cell based on the photovoltaic panel group point cloud data of the power station area based on cluster analysis, obtaining the positioning information of each photovoltaic cell, and adding the positioning information of the photovoltaic cell to the photovoltaic power station map. This application provides the positioning information of each photovoltaic cell to the operation and maintenance personnel in the power station based on drone and laser radar technology, which is convenient for simultaneous path planning and navigation. The present invention does not require any modification to the photovoltaic power station, nor does it require the installation of special equipment. It only needs to collect photovoltaic power station site cloud data and a drone with RTK, and the operation is simple. In addition, a photovoltaic power station navigation device, computer equipment and storage medium are also proposed.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power stations, and in particular to a photovoltaic power station navigation method, device, computer equipment and storage medium. Background Art

[0002] At present, the continuous development of smart grids has also put forward higher requirements for the real-time monitoring and inspection of photovoltaic power stations. In the operation and maintenance of photovoltaic power stations, photovoltaic panels are the main inspection and maintenance objects and their number is increasing. The geographical environment of photovoltaic power station sites is also becoming more and more complex. The photovoltaic panels in photovoltaic power stations can no longer meet the actual needs by relying solely on traditional paper records. The traditional way of finding the location of target specific maintenance photovoltaic cells is still very dependent on the experience of operators, or querying records before maintenance, but the queried routes are often different from the actual situation, especially at night, when patrol personnel can only know the approximate location, especially in some areas with complex terrain. When there is an emergency repair task, the operation and maintenance personnel are more likely to miss the target photovoltaic panels, thereby delaying the repair time and reducing the operation and maintenance efficiency.

[0003] Currently, drone technology has been widely used in the inspection of photovoltaic power stations. Drone inspections are efficient, safe, reliable, and intelligent when assisting maintenance personnel in their maintenance tasks. However, the path of drone inspections is not suitable for maintenance personnel. In the existing technology, when maintenance personnel inspect specific photovoltaic cells, they usually use method one or method two.

[0004] Method 1: Use mainstream mobile navigation map software to perform navigation and positioning inside the photovoltaic power station; however, most photovoltaic power stations are located in remote mountainous areas with complex terrain, poor network signals, and mainstream mobile navigation map software does not collect enough information about remote areas. It is also unable to locate individual photovoltaic cells or the positioning accuracy is insufficient, which cannot meet the positioning and navigation needs of operation and maintenance personnel in the power station.

[0005] Method 2: Use point cloud editing software to manually mark the photovoltaic cell point cloud and the traversable routes inside the power station. However, although the photovoltaic cells are neatly arranged, they are huge in number and very densely distributed, so manual editing is labor-intensive and inefficient. Summary of the invention

[0006] Based on this, it is necessary to provide a photovoltaic power station navigation method, device, computer equipment and storage medium to address the above problems.

[0007] In a first aspect, an embodiment of the present invention provides a photovoltaic power station navigation method, the method comprising collecting photovoltaic power station cloud data and constructing a photovoltaic power station map:

[0008] The method of collecting photovoltaic power station cloud data and constructing a photovoltaic power station map includes the following steps:

[0009] Collect photovoltaic power station cloud data based on UAV laser radar system; divide photovoltaic power station cloud data into ground point cloud data of power station area and photovoltaic panel group point cloud data of power station area;

[0010] Construct a PV power station map including height information based on ground point cloud data of the power station area;

[0011] Based on cluster analysis, each photovoltaic cell is identified according to the point cloud data of the photovoltaic panel group in the power station area, the positioning information of each photovoltaic cell is obtained, and the positioning information of the photovoltaic cell is added to the photovoltaic power station map.

[0012] In one embodiment, the step of identifying each photovoltaic cell based on the point cloud data of photovoltaic panel groups in the power station area based on cluster analysis includes:

[0013] The flatness and inclination of each cluster of the photovoltaic panel point cloud data in the power station area are detected. Based on the Euclidean distance clustering method, clusters with flatness greater than a preset threshold and similar inclination are identified as photovoltaic cells.

[0014] In one embodiment, obtaining the positioning information of each photovoltaic cell includes:

[0015] Each identified photovoltaic cell is numbered, and the 3D coordinates of the center point of each photovoltaic cell are calculated, and the number is associated with the 3D coordinates as positioning information.

[0016] In one of the embodiments, the method further includes identifying a walkable area of ​​the photovoltaic power station based on the photovoltaic power station cloud data;

[0017] The method of identifying the walkable area of ​​the photovoltaic power station based on the photovoltaic power station map includes the following steps:

[0018] Performing a two-dimensional planar projection on the photovoltaic power station map to obtain a projection image, then performing edge detection on each photovoltaic cell in the projection image, identifying the spacing area between adjacent photovoltaic cells, and expanding the photovoltaic cells at the edge according to the spacing distance of the spacing area to obtain an extended area; the spacing area and the extended area are used as walkable areas;

[0019] Based on the polygonal geometric relationship features, all walkable areas are detected for intersections, and the walkable areas are segmented according to the intersections. The segmented walkable areas correspond to each photovoltaic cell.

[0020] In one of the embodiments, the identifying the walkable area of ​​the photovoltaic power station further includes: obtaining a flight path of the drone based on the walkable area.

[0021] In one of the embodiments, the photovoltaic power station navigation method further includes: photovoltaic power station inspection path planning;

[0022] The photovoltaic power station inspection path planning includes the following steps:

[0023] At least one corner point is set in the walkable area of ​​the photovoltaic power station; a corner point is selected as the starting point of the task path, and the target photovoltaic cells are sorted in order from small to large according to the distance between the target photovoltaic cells and the starting point; starting from the starting point, the walkable area is moved forward in the order of the target photovoltaic cells, and the task path is obtained.

[0024] In one of the embodiments, the photovoltaic power station navigation method further includes: photovoltaic power station inspection path navigation;

[0025] According to the planning of the inspection task path, the flight path is obtained;

[0026] The drone is guided along the flight path.

[0027] In a second aspect, an embodiment of the present invention provides a photovoltaic power station navigation device, the device comprising: a photovoltaic power station cloud data acquisition module for collecting photovoltaic power station cloud data based on a drone laser radar system;

[0028] The photovoltaic power station cloud data classification module is used to divide the photovoltaic power station cloud data into ground point cloud data of the power station area and photovoltaic panel group point cloud data of the power station area;

[0029] A map construction module, used to construct a map including height information based on ground point cloud data of the power station area;

[0030] The photovoltaic cell identification and positioning module is used to identify each photovoltaic cell according to the photovoltaic panel group point cloud data in the power station area based on cluster analysis, obtain the positioning information of each photovoltaic cell, and add the positioning information of the photovoltaic cell to the photovoltaic power station map.

[0031] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps:

[0032] Collect photovoltaic power station cloud data based on UAV laser radar system; divide photovoltaic power station cloud data into ground point cloud data of power station area and photovoltaic panel group point cloud data of power station area;

[0033] Construct a map including height information based on ground point cloud data of the power station area;

[0034] Based on cluster analysis, each photovoltaic cell is identified according to the point cloud data of the photovoltaic panel group in the power station area, the positioning information of each photovoltaic cell is obtained, and the positioning information of the photovoltaic cell is added to the photovoltaic power station map.

[0035] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the following steps:

[0036] Collect photovoltaic power station cloud data based on UAV laser radar system; divide photovoltaic power station cloud data into ground point cloud data of power station area and photovoltaic panel group point cloud data of power station area;

[0037] Construct a map including height information based on ground point cloud data of the power station area;

[0038] Based on cluster analysis, each photovoltaic cell is identified according to the point cloud data of the photovoltaic panel group in the power station area, the positioning information of each photovoltaic cell is obtained, and the positioning information of the photovoltaic cell is added to the photovoltaic power station map.

[0039] Implementing the embodiments of the present invention will have the following beneficial effects:

[0040] The above-mentioned photovoltaic power station navigation method, device, computer equipment and storage medium, through the unmanned aerial vehicle laser radar technology, quickly obtain the 3D point cloud model of the photovoltaic power station, realize the identification, segmentation and arrangement and positioning of the point cloud of a single photovoltaic cell, and generate a walkable area according to the edge detection of the photovoltaic cell to obtain a vector map of the photovoltaic power station. At the same time, the flight route of the drone is calculated according to the center line of the walkable area. The drone flies according to the flight route to guide the operation and maintenance personnel in the power station to reach the target photovoltaic cell to be inspected. The path planning is to avoid the operation and maintenance personnel from finding the wrong equipment or getting lost, thereby improving the operation and maintenance efficiency of the photovoltaic power station. At the same time, the present invention does not require any modification to the photovoltaic power station, nor does it require the installation of special equipment. It only needs to collect the photovoltaic power station site cloud data and a drone with RTK. It is simple to operate and has low operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] in:

[0043] Figure 1A flow chart of collecting photovoltaic power station cloud data and constructing a photovoltaic power station map in one embodiment.

[0044] Figure 2 is a flow chart of a photovoltaic power station navigation method in one embodiment;

[0045] Figure 3 A flowchart for identifying a walkable area of ​​a photovoltaic power station in one embodiment;

[0046] Figure 4 is a schematic diagram of a photovoltaic power station navigation method in one embodiment;

[0047] Figure 5 A flowchart of inspection route planning for a photovoltaic power station in one embodiment;

[0048] Figure 6 A flowchart of a photovoltaic power station inspection route navigation in one embodiment;

[0049] Figure 7 A schematic diagram of a navigation device for a photovoltaic power station in one embodiment;

[0050] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] like Figure 1 As shown, a photovoltaic power station navigation method is proposed, wherein the photovoltaic power station includes at least one photovoltaic cell, and the method includes step 1 of collecting photovoltaic power station cloud data to construct a photovoltaic power station map:

[0053] The method of collecting photovoltaic power station cloud data and constructing a photovoltaic power station map specifically includes the following steps:

[0054] Step 101, collecting photovoltaic power station cloud data based on the UAV laser radar system; dividing the photovoltaic power station cloud data into ground point cloud data of the power station area and photovoltaic panel group point cloud data of the power station area;

[0055] Among them, the photovoltaic power station point cloud data is collected based on the drone lidar system, including two parts: field flight and indoor point cloud solution.

[0056] Among them, the field flight includes the following steps: design the take-off, landing, and route planning scheme of the drone according to the design requirements; find a location with a wide field of view in the power station area to assume a GPS base station; determine whether it meets the flight requirements of the drone; when the take-off requirements are met, perform lidar data collection according to the planned route to obtain the original data collected by the lidar. The design requirements are information such as the topography around the photovoltaic power station, the distribution of photovoltaic panels, and the point cloud overlap requirements between each flight strip. The take-off requirements of the drone mean that before taking off, all equipment and devices are checked to confirm that the equipment and devices are in normal condition, and the flight environment (such as wind speed, wind direction, etc.) meets the flight requirements of the drone.

[0057] Among them, the in-house point cloud solution includes the following steps: based on the original data collected by the lidar, use the GNSS / INS post-processing software to solve the laser point cloud POS trajectory during the flight; then combine the laser point cloud POS trajectory with the original data collected by the lidar to solve and obtain the 3D point cloud coordinates; filter the original data collected by the lidar to remove noise data, and manually crop the non-power station area through the interactive software to obtain point cloud data containing the entire photovoltaic power station.

[0058] Among them, the photovoltaic power station site cloud data is divided into the ground point cloud data of the power station area and the photovoltaic panel group point cloud data of the power station area, including the following steps: based on the point cloud filtering algorithm, extract the ground points of the power station area to obtain the ground point cloud data of the power station area. The point cloud filtering algorithm is the CSF point cloud ground filtering algorithm in the Kaiyuan software CloudCompare. In this embodiment, the ground point cloud data extracted from the power station area can be used to construct the digital elevation model DEM of the area where the photovoltaic power station is located. Since most photovoltaic power stations are distributed in mountainous areas with complex terrain, by extracting the ground point cloud data of the power station area to construct the digital elevation model DEM, auxiliary information can be provided for operation and maintenance personnel to select transportation tools and photovoltaic power stations, and a basis is also provided for the flight altitude estimation of drones in the power station. By using the CSF point cloud ground filtering algorithm in the Kaiyuan software CloudCompare, the ground point cloud of the power station area is automatically extracted. In this embodiment, the ground point cloud of the power station area can also be obtained by the point cloud plane detection method. The photovoltaic panel group point cloud data of the power station area refers to the part of the point cloud data of the entire photovoltaic power station that does not belong to the ground point cloud data of the power station area.

[0059] Step 102, construct a photovoltaic power station map including height information based on the ground point cloud data of the power station area; identify each photovoltaic cell based on the photovoltaic panel group point cloud data of the power station area based on cluster analysis, obtain the positioning information of each photovoltaic cell, and add the positioning information of the photovoltaic cell to the photovoltaic power station map.

[0060] The extracted ground point cloud data of the power station area is used to construct a digital elevation model (DEM), which can provide auxiliary information for operation and maintenance personnel to choose transportation and photovoltaic power stations, and also provide a basis for estimating the flight altitude of drones in the power station. The positioning information of each photovoltaic cell is obtained to facilitate navigation and inspection.

[0061] In one embodiment, the cluster analysis-based identification of each photovoltaic cell based on the point cloud data of the photovoltaic panel group in the power station area includes the following steps: detecting the flatness and inclination of each cluster of the point cloud data of the photovoltaic panel group in the power station area, and based on the Euclidean distance clustering method, identifying the clusters with a flatness greater than a preset threshold and a similar inclination as photovoltaic cells.

[0062] The method based on Euclidean distance clustering is the Eu-distance method. The cluster refers to the point cloud data of the photovoltaic panel group in the power station area to be clustered. Photovoltaic panels have the characteristics of plane and tilt, so the cluster is tested for flatness and tilt, and the clusters that meet the requirements of having a high flatness score and a similar tilt of the neighboring point cloud are identified as photovoltaic cells. Through this embodiment, the point cloud data of the photovoltaic panel group in the power station area is further classified to separate the individual photovoltaic cells.

[0063] In one embodiment, obtaining the positioning information of each photovoltaic cell comprises the following steps: numbering each identified photovoltaic cell, calculating the 3D coordinates of the center point of each photovoltaic cell, and associating the number with the 3D coordinates as the positioning information.

[0064] Wherein, the numbering includes row numbers and column numbers. First, calculate the main plane direction of one of the photovoltaic panels, use this direction as the row direction for row numbering, and use the direction perpendicular to the row direction as the column direction for column numbering. Then calculate the 3D coordinates of the center point of each photovoltaic panel, associate the row number, column number and 3D coordinates as positioning information, and perform row and column numbering on the photovoltaic panels according to the 3D coordinates of these center points and the row and column direction information. The row number and column number are associated with the 3D coordinates obtained by locating the center point of the corresponding photovoltaic cell, which facilitates the identity correspondence of the photovoltaic cell. The photovoltaic cells are neatly arranged and have similar placement directions, with a relatively regular row and column distribution pattern. Row and column numbering of the individual photovoltaic cells that have been segmented can assist in the positioning, identity confirmation and subsequent navigation of the individual photovoltaic cells.

[0065] like Figure 2 As shown, in one embodiment, the photovoltaic power station navigation method further includes step 2, identifying a navigable area of ​​the photovoltaic power station based on the photovoltaic power station map;

[0066] like Figure 3As shown, the step 2 includes the following steps:

[0067] Step 201, performing a two-dimensional projection of the photovoltaic power station map to obtain a projection image, then performing edge detection on each photovoltaic cell in the projection image, identifying the interval area between adjacent photovoltaic cells, and expanding the photovoltaic cells at the edge according to the interval distance of the interval area to obtain an extended area; the interval area and the extended area are used as walkable areas;

[0068] Step 202, based on the polygonal geometric relationship features, all walkable areas are detected for intersections, and the walkable areas are segmented according to the intersections. The segmented walkable areas correspond to the photovoltaic cells.

[0069] Among them, the spacing area is an area where there is a certain spacing between photovoltaic cells and can be used as a walking area for staff during inspection and maintenance. The extended area is an area that is expanded outward according to the spacing distance of the spacing area. For the photovoltaic cells that have been arranged, the photovoltaic power station map is first projected in a two-dimensional plane, and then the edges of the photovoltaic cells in the same column and the same row are detected, and the spacing area between adjacent photovoltaic cells is calculated as the walking area for operation and maintenance personnel. For the areas located at the head and tail edges of the entire photovoltaic cell arrangement, the walkable area can be directly expanded by shifting outward from the outer edge of the photovoltaic cell according to the spacing distance of the spacing area or the commonly used photovoltaic cell spacing distance. At the same time, polygonal geometric relationship features are used to perform intersection detection on all walkable areas. The intersection detected is used as the route turning execution area, and the intersection is as follows. Figure 4 The red area is shown. The feasible walking area is divided according to the intersection, and the row and column numbers of the segmented walkable areas are associated with the relevant photovoltaic cells.

[0070] like Figure 3 As shown, in one embodiment, step 2 also includes step S203, obtaining the flight path of the drone based on the walkable area.

[0071] Among them, the flight path is obtained by fitting the center line of the movable area, and the row and column walkable areas are scanned according to the length and width directions, respectively, and the corresponding center points are calculated, and the center lines are fitted according to the center points. Each segmented walkable area corresponds to a flight route, which serves as a calculation unit during path planning; and the midpoint of the flight route is taken as the drone hovering site corresponding to the photovoltaic cell. Since photovoltaic power stations may be distributed in some areas with undulating terrain, the flight height adjustment parameters of the drone on the flight route trajectory are calculated according to the ground point cloud data corresponding to the flight route, while ensuring the safe flight of the drone imitating the terrain, it is also necessary to ensure that the operation and maintenance personnel on the ground can judge the walking route according to the guidance of the drone flying along the flight route. This embodiment realizes a more effective provision of high-precision navigation maps for drones performing guidance work, while ensuring that the operation and maintenance personnel can correctly find the target photovoltaic panel quickly through the walking route guided by the drone.

[0072] like Figure 2 As shown, in one embodiment, the photovoltaic power station navigation method further includes step 3, photovoltaic power station inspection path planning;

[0073] like Figure 5 Specifically, the step 3 of photovoltaic power station inspection route planning includes the following steps:

[0074] Step 301, at least one corner point is set in the walkable area of ​​the photovoltaic power station; a corner point is selected as the starting point of the task path;

[0075] The corner points are several important entry points to the photovoltaic cell installation area of ​​the photovoltaic power station. The surveying staff will carry out high-precision measurement of the coordinates of these entry points in advance to facilitate navigation.

[0076] Step 302 , sort the target photovoltaic cells in ascending order of distance from the starting point; starting from the starting point, move forward in the walkable area in the order of sorting the target photovoltaic cells to obtain a task path.

[0077] Among them, the distances between all target photovoltaic cells and the starting point of the task path are sorted to determine the route guidance sequence of the target photovoltaic cells.

[0078] In one embodiment, selecting a corner point as the starting point of the task path includes obtaining the coordinates of the current position and selecting the corner point with the smallest distance from the current position as the starting point of the task path.

[0079] Among them, by extracting a single photovoltaic cell point cloud, the coordinates of the center point of the photovoltaic cell point cloud are obtained, and each photovoltaic cell cluster is automatically arranged and matched according to the main direction and the two-dimensional projection relationship of the center point. At the same time, it is also matched and associated with the corresponding walkable area and flight route. The starting point of the inspection task is mainly fixed at several important entrance points in the photovoltaic cell installation area of ​​the photovoltaic power station. The measurement staff will measure the coordinates of these entrance points with high precision in advance. When the operation and maintenance personnel obtain the inspection fault results, they match the row and column numbers bound to the target photovoltaic cell with the center point coordinates, sort the distances of all target photovoltaic cells from the starting point of the task, and determine the route guidance sequence of the target photovoltaic cells.

[0080] like Figure 2 As shown, in one embodiment, the photovoltaic power station navigation method further includes step 4, photovoltaic power station inspection path navigation;

[0081] like Figure 6 As shown, the step 4, the inspection route navigation of the photovoltaic power station includes the following steps:

[0082] Step 401, planning of task path;

[0083] Among them, the planning of task paths includes the following two situations:

[0084] Path navigation from the starting point: According to the coordinates of the starting point, search for the edge corners of the photovoltaic cells around it and calculate the plane distance. The row and column numbers of the photovoltaic cells corresponding to the edge corners with the smallest distance are identified as the row and column numbers of the starting point. By comparing the difference in the row and column numbers between the starting point and the target photovoltaic cell, the number of walking routes in the row and column directions starting from the starting point is obtained. The row route is walked first, and then the column route is walked to realize the planning of the task path.

[0085] Path navigation between adjacent target photovoltaic cells: Since the row and column numbers of the target photovoltaic cells are already known, the row and column number difference between the two target photovoltaic cells can be directly calculated to obtain the number of walking routes in the row and column directions in the guidance route. The row route is walked first, and then the column route is walked to realize the planning of the task path.

[0086] Step 402, drone route guidance;

[0087] According to the task path planned in step 401, the corresponding flight route is extracted. It is converted into the input format of the UAV task path planning, and the UAV flight path is automatically generated, so that the UAV can automatically fly according to the flight route, and the UAV in the photovoltaic power station can guide the route of the operation and maintenance personnel. At the same time, the UAV can be equipped with lighting equipment to illuminate the route for the operation and maintenance personnel during night guidance, and it can also be carried out normally in the night environment.

[0088] In this embodiment, the UAV continues to fly while the heading remains unchanged. When it reaches a turning point where the direction of the row or column changes, it is set to hover and wait at the intersection, waiting for the operation and maintenance personnel to follow the guidance to complete the route in one direction; when the operation and maintenance personnel arrive at the turning point of the route, they confirm through the remote control operation, and the UAV continues to start route guidance in the next direction.

[0089] In this embodiment, when the UAV reaches the location of the target photovoltaic cell, it hovers and waits at the hovering station of the flight route corresponding to the target photovoltaic cell, waiting for the operation and maintenance personnel to arrive at the target location. According to the maintenance workload of the target photovoltaic cell, the UAV can choose between hovering and waiting and return charging modes.

[0090] When the maintenance workload is not large, the drone will continue to guide the route to the next target photovoltaic cell after the maintenance personnel have performed a brief operation; when the maintenance workload is large, the drone will directly return to charge. When the staff completes the maintenance work, the remote control confirms that the drone returns to the maintenance point and starts the route guidance to the next target location.

[0091] like Figure 7 As shown, an embodiment of the present invention provides a navigation device for a photovoltaic power station, the device comprising:

[0092] Photovoltaic power station cloud data collection module 701, used to collect photovoltaic power station cloud data based on the UAV laser radar system;

[0093] The photovoltaic power station cloud data classification module 702 is used to classify the photovoltaic power station cloud data into ground point cloud data of the power station area and photovoltaic panel group point cloud data of the power station area;

[0094] A map construction module 703 is used to construct a map including height information based on the ground point cloud data of the power station area;

[0095] The photovoltaic cell identification and positioning module 704 is used to identify each photovoltaic cell according to the photovoltaic panel group point cloud data in the power station area based on cluster analysis, obtain the positioning information of each photovoltaic cell, and add the positioning information of the photovoltaic cell to the photovoltaic power station map.

[0096] In one embodiment, the cluster analysis-based identification of each photovoltaic cell based on the point cloud data of the photovoltaic panel group in the power station area includes: the photovoltaic power station site cloud data classification module 702 is used to detect the flatness and inclination of each cluster of the point cloud data of the photovoltaic panel group in the power station area, and based on the Euclidean distance clustering method, the clusters with flatness greater than a preset threshold and similar inclination are identified as photovoltaic cells.

[0097] In one embodiment, obtaining the positioning information of each photovoltaic cell includes: the photovoltaic cell identification and positioning module 704 is used to number each identified photovoltaic cell, calculate the 3D coordinates of the center point of each photovoltaic cell, and associate the number with the 3D coordinates as the positioning information.

[0098] In one embodiment, the device further includes a module 705 for identifying a walkable area of ​​a photovoltaic power station, which is used to:

[0099] Performing a two-dimensional planar projection on the photovoltaic power station map to obtain a projection image, then performing edge detection on each photovoltaic cell in the projection image, identifying the spacing area between adjacent photovoltaic cells, and expanding the photovoltaic cells at the edge according to the spacing distance of the spacing area to obtain an extended area; the spacing area and the extended area are used as walkable areas;

[0100] Based on the polygonal geometric relationship features, all walkable areas are detected for intersections, and the walkable areas are segmented according to the intersections. The segmented walkable areas correspond to each photovoltaic cell.

[0101] In one embodiment, the module for identifying the walkable area of ​​the photovoltaic power station is also used to obtain the flight path of the drone based on the walkable area.

[0102] In one embodiment, the photovoltaic power station navigation device further includes a photovoltaic power station inspection path planning module 706, which is used to:

[0103] At least one corner point is set in the walkable area of ​​the photovoltaic power station; a corner point is selected as the starting point of the task path, and the target photovoltaic cells are sorted in order from small to large according to the distance between the target photovoltaic cells and the starting point; starting from the starting point, the walkable area is moved forward in the order of the target photovoltaic cells, and the task path is obtained.

[0104] In one embodiment, the photovoltaic power station navigation device further includes a photovoltaic power station inspection path navigation module 707, which is used to:

[0105] According to the planning of the inspection task path, the flight path is obtained;

[0106] The drone is guided along the flight path.

[0107] Figure 8The figure shows the internal structure of a computer device in one embodiment.

[0108] The computer device may be a terminal. Figure 8 As shown, the computer device includes a processor, a memory and a network interface connected via a system bus. The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor may implement a photovoltaic power station navigation method. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor may implement a photovoltaic power station navigation method. The network interface is used to communicate with the outside world. Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0109] In one embodiment, the photovoltaic power station navigation method provided in the present application can be implemented in the form of a computer program. The computer program can be Figure 8 The computer device shown in the figure is run. The memory of the computer device can store various program templates constituting the photovoltaic power station navigation method. For example, the photovoltaic power station cloud data acquisition module 701, the photovoltaic power station cloud data classification module 702, the map construction module 703, and the photovoltaic cell identification and positioning module 704.

[0110] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps: the photovoltaic power station navigation method includes collecting photovoltaic power station site cloud data and constructing a photovoltaic power station map: collecting photovoltaic power station site cloud data based on an unmanned aerial vehicle laser radar system; dividing the photovoltaic power station site cloud data into ground point cloud data of the power station area and photovoltaic panel group point cloud data of the power station area; constructing a photovoltaic power station map including height information based on the ground point cloud data of the power station area; identifying each photovoltaic cell based on the photovoltaic panel group point cloud data of the power station area based on cluster analysis, obtaining the positioning information of each photovoltaic cell, and adding the positioning information of the photovoltaic cell to the photovoltaic power station map.

[0111] In one embodiment, the cluster analysis-based identification of each photovoltaic cell based on the point cloud data of the photovoltaic panel group in the power station area includes: detecting the flatness and inclination of each cluster of the point cloud data of the photovoltaic panel group in the power station area, and based on the Euclidean distance clustering method, identifying the clusters with a flatness greater than a preset threshold and a similar inclination as photovoltaic cells.

[0112] In one embodiment, obtaining the positioning information of each photovoltaic cell includes: numbering each identified photovoltaic cell, calculating the 3D coordinates of the center point of each photovoltaic cell, and associating the number with the 3D coordinates as the positioning information.

[0113] In one embodiment, the method also includes identifying the walkable area of ​​the photovoltaic power station based on the photovoltaic power station site cloud data: performing a two-dimensional projection on the photovoltaic power station map to obtain a projection image, and then performing edge detection on each photovoltaic cell in the projection image to identify the spacing area between adjacent photovoltaic cells, and expanding the edge photovoltaic cells according to the spacing distance of the spacing area to obtain an extended area; the spacing area and the extended area are used as walkable areas; performing intersection detection on all walkable areas based on polygonal geometric relationship features, and segmenting the walkable areas according to the intersections, and the segmented walkable areas correspond to each photovoltaic cell.

[0114] In one embodiment, the identifying of the walkable area of ​​the photovoltaic power station further includes: obtaining a flight path of the drone based on the walkable area.

[0115] In one embodiment, the photovoltaic power station navigation method also includes photovoltaic power station inspection path planning: setting at least one corner point in the walkable area of ​​the photovoltaic power station; selecting a corner point as the starting point of the task path, and sorting the target photovoltaic cells in order of the distance from the starting point to the target photovoltaic cells from small to large; starting from the starting point, moving forward in the walkable area in the order of sorting of the target photovoltaic cells to obtain the task path.

[0116] In one embodiment, the photovoltaic power station navigation method further includes: photovoltaic power station inspection path navigation: obtaining a flight path according to the planning of the inspection task path; and guiding the drone along the flight path.

[0117] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the processor executes the following steps: collecting photovoltaic power station site cloud data and constructing a photovoltaic power station map: collecting photovoltaic power station site cloud data based on an unmanned aerial vehicle laser radar system; dividing the photovoltaic power station site cloud data into ground point cloud data of the power station area and photovoltaic panel group point cloud data of the power station area; constructing a photovoltaic power station map including height information based on the ground point cloud data of the power station area; identifying each photovoltaic cell based on the photovoltaic panel group point cloud data of the power station area based on cluster analysis, obtaining the positioning information of each photovoltaic cell, and adding the positioning information of the photovoltaic cell to the photovoltaic power station map.

[0118] In one embodiment, the cluster analysis-based identification of each photovoltaic cell based on the point cloud data of the photovoltaic panel group in the power station area includes: detecting the flatness and inclination of each cluster of the point cloud data of the photovoltaic panel group in the power station area, and based on the Euclidean distance clustering method, identifying the clusters with a flatness greater than a preset threshold and a similar inclination as photovoltaic cells.

[0119] In one embodiment, obtaining the positioning information of each photovoltaic cell includes: numbering each identified photovoltaic cell, calculating the 3D coordinates of the center point of each photovoltaic cell, and associating the number with the 3D coordinates as the positioning information.

[0120] In one embodiment, the method also includes identifying the walkable area of ​​the photovoltaic power station based on the photovoltaic power station site cloud data: performing a two-dimensional projection on the photovoltaic power station map to obtain a projection image, and then performing edge detection on each photovoltaic cell in the projection image to identify the spacing area between adjacent photovoltaic cells, and expanding the edge photovoltaic cells according to the spacing distance of the spacing area to obtain an extended area; the spacing area and the extended area are used as walkable areas; performing intersection detection on all walkable areas based on polygonal geometric relationship features, and segmenting the walkable areas according to the intersections, and the segmented walkable areas correspond to each photovoltaic cell.

[0121] In one embodiment, the identifying of the walkable area of ​​the photovoltaic power station further includes: obtaining a flight path of the drone based on the walkable area.

[0122] In one embodiment, the photovoltaic power station navigation method also includes photovoltaic power station inspection path planning: setting at least one corner point in the walkable area of ​​the photovoltaic power station; selecting a corner point as the starting point of the task path, and sorting the target photovoltaic cells in order of the distance from the starting point to the target photovoltaic cells from small to large; starting from the starting point, moving forward in the walkable area in the order of sorting of the target photovoltaic cells to obtain the task path.

[0123] In one embodiment, the photovoltaic power station navigation method further includes: photovoltaic power station inspection path navigation: obtaining a flight path according to the planning of the inspection task path; and guiding the drone along the flight path.

[0124] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0125] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A photovoltaic power station navigation method, wherein the photovoltaic power station comprises at least one photovoltaic cell, It is characterized in that The method includes collecting photovoltaic power station cloud data and constructing a photovoltaic power station map: The method of collecting photovoltaic power station cloud data and constructing a photovoltaic power station map includes the following steps: Photovoltaic power station cloud data is collected based on the UAV laser radar system; the photovoltaic power station cloud data is divided into ground point cloud data of the power station area and photovoltaic panel group point cloud data of the power station area. The ground point cloud data of the power station area is used to construct a digital elevation model DEM, which provides auxiliary information for operation and maintenance personnel to choose transportation and photovoltaic power stations, and also provides a basis for estimating the flight altitude of the UAV in the power station; Construct a PV power station map including height information based on ground point cloud data of the power station area; Based on cluster analysis, each photovoltaic cell is identified according to the photovoltaic panel group point cloud data in the power station area, the positioning information of each photovoltaic cell is obtained, and the positioning information of the photovoltaic cell is added to the photovoltaic power station map; The method further includes identifying a walkable area of ​​the photovoltaic power station based on the photovoltaic power station cloud data; The method of identifying the walkable area of ​​the photovoltaic power station based on the photovoltaic power station map includes the following steps: Performing a two-dimensional planar projection on the photovoltaic power station map to obtain a projection image, then performing edge detection on each photovoltaic cell in the projection image, identifying the spacing area between adjacent photovoltaic cells, and expanding the photovoltaic cells at the edge according to the spacing distance of the spacing area to obtain an extended area; the spacing area and the extended area are used as walkable areas; Based on the polygonal geometric relationship features, all walkable areas are detected for intersections, and the walkable areas are segmented according to the intersections. The segmented walkable areas correspond to each photovoltaic cell.

2. The method according to claim 1, Features: The method of identifying each photovoltaic cell according to the photovoltaic panel group point cloud data in the power station area based on cluster analysis includes: The flatness and inclination of each cluster of the photovoltaic panel point cloud data in the power station area are detected. Based on the Euclidean distance clustering method, clusters with flatness greater than a preset threshold and similar inclination are identified as photovoltaic cells.

3. The method according to claim 1, Features: The obtaining of the positioning information of each photovoltaic cell includes: Each identified photovoltaic cell is numbered, and the 3D coordinates of the center point of each photovoltaic cell are calculated, and the number is associated with the 3D coordinates as positioning information.

4. The method according to claim 1, It is characterized in that The identifying of the walkable area of ​​the photovoltaic power station further includes: obtaining a flight path of the drone based on the walkable area.

5. The method according to claim 1, It is characterized in that The photovoltaic power station navigation method further includes: photovoltaic power station inspection path planning; The photovoltaic power station inspection path planning includes the following steps: At least one corner point is set in the walkable area of ​​the photovoltaic power station; a corner point is selected as the starting point of the task path, and the target photovoltaic cells are sorted in order from small to large according to the distance between the target photovoltaic cells and the starting point; starting from the starting point, the walkable area is moved forward in the order of the target photovoltaic cells, and the task path is obtained.

6. The method according to claim 5, It is characterized in that The photovoltaic power station navigation method further includes: photovoltaic power station inspection path navigation; According to the planning of the inspection task path, the flight path is obtained; The drone is guided along the flight path.

7. A photovoltaic power station navigation device, It is characterized in that The device comprises: Photovoltaic power station cloud data acquisition module, used to collect photovoltaic power station cloud data based on drone lidar system; The photovoltaic power station cloud data classification module is used to divide the photovoltaic power station cloud data into ground point cloud data of the power station area and photovoltaic panel group point cloud data of the power station area. The ground point cloud data of the power station area is used to construct a digital elevation model DEM, which provides auxiliary information for operation and maintenance personnel to choose transportation tools and photovoltaic power stations, and also provides a basis for estimating the flight altitude of drones in the power station. A map construction module, used to construct a map including height information based on ground point cloud data of the power station area; Photovoltaic cell identification and positioning module, used to identify each photovoltaic cell based on the point cloud data of photovoltaic panel groups in the power station area based on cluster analysis, obtain the positioning information of each photovoltaic cell, and add the positioning information of the photovoltaic cell to the photovoltaic power station map; A module for identifying the walkable area of ​​a photovoltaic power station is used to perform a two-dimensional projection of the photovoltaic power station map to obtain a projection image, and then perform edge detection on each photovoltaic cell in the projection image to identify the interval area between adjacent photovoltaic cells, and expand the edge photovoltaic cells according to the interval distance of the interval area to obtain an extended area; the interval area and the extended area are used as the walkable area; Based on the polygonal geometric relationship features, all walkable areas are detected for intersections, and the walkable areas are segmented according to the intersections. The segmented walkable areas correspond to each photovoltaic cell.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor is caused to execute the steps of the method according to any one of claims 1 to 6.

Images