A path planning method, device, equipment and medium for a mountain photovoltaic power station

Abstract

This application relates to the field of inspection of mountain photovoltaic power stations, and in particular to a path planning method, device, equipment, and medium for mountain photovoltaic power stations. The method includes: acquiring road network data of the main road of the mountain photovoltaic power station, static wall data of each zone of the mountain photovoltaic power station, and zone entrance and exit coordinates; acquiring the positions of multiple target nodes, including multiple locations to be inspected and inspection starting points; sorting the multiple targets according to the locations to be inspected and the inspection starting points to obtain a planning order for the multiple targets; for any two first target nodes in the planning order, performing target path planning according to the positions of the two first target nodes, zone entrance and exit coordinates, and first data; and obtaining the inspection path plan based on the target path planning results between every two target nodes in the planning order. This application can accurately plan paths for mountain photovoltaic power stations.

Description

Technical Field

[0001] This application relates to the field of inspection of mountain photovoltaic power stations, and in particular to a path planning method, device, equipment and medium for mountain photovoltaic power stations. Background Technology

[0002] For equipment inspection, current methods primarily rely on satellite positioning for basic location identification, supplemented by electronic maps to plan approximate routes. Additionally, drone inspection technology is frequently used to assist in positioning, marking equipment locations using aerial images and providing feedback to maintenance personnel.

[0003] However, in large-scale mountain photovoltaic power station scenarios, weak signals make satellite positioning difficult, and inaccurate electronic maps result in poor reliability of route planning during inspections. Therefore, how to accurately plan routes for mountain photovoltaic power stations is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a path planning method, device, equipment, and medium for mountain photovoltaic power stations, which can accurately plan paths for mountain photovoltaic power stations.

[0005] Firstly, a path planning method for a mountain photovoltaic power station is provided, comprising: acquiring road network data of the main road of the mountain photovoltaic power station, static wall data of each zone of the mountain photovoltaic power station, and zone entrance and exit coordinates, wherein the static wall data includes: zone range vector data and photovoltaic panel vector data within the zone; acquiring the positions of multiple target nodes, wherein the positions include multiple locations to be inspected and inspection starting points; sorting the multiple targets according to the locations to be inspected and the inspection starting points to obtain a planning order for the multiple targets; for any two first target nodes in the planning order, performing inter-target path planning according to the positions of the two first target nodes, zone entrance and exit coordinates, and first data, wherein if the two first target nodes are both in the same zone, the first data includes: static wall data; if the two first target nodes are not in the same zone, the first data includes: static wall data and road network data; and obtaining the inspection path planning based on the inter-target path planning results between every two target nodes in the planning order.

[0006] In a preferred embodiment, this application can be further configured as follows: sorting multiple objectives based on the location to be inspected and the starting point of the inspection to obtain a planning order for the multiple objectives, including: selecting a preset number of ants, using the starting point of the inspection as the starting position, and determining a location to be inspected for each ant; determining a location to be inspected for each ant until an ant path is obtained for each ant; determining the distance of each ant path, where the distance of an ant path is the sum of the distances between nodes in the ant path; selecting the shortest ant path as the optimal path for the current round based on the distance of each ant path, and performing iterative processing to obtain the final optimal path, thereby determining the planning order for the multiple objectives.

[0007] In a preferred embodiment, this application can be further configured to: determine the distance of each ant path, including: for two adjacent second target nodes of the ant path, determining whether the two adjacent second target nodes are in the same partition; if they are in the same partition, using Euclidean distance to determine the node distance between the two adjacent second target nodes; if they are not in the same partition, using Euclidean distance to determine a first distance between each second target node and its nearest neighbor's entrance / exit location; determining a second distance between two nearest neighbor's entrance / exit locations based on the road network data; determining the node distance between two adjacent second target nodes based on the first distance and the second distance; and determining the distance of the ant path based on the node distance between the two adjacent second target nodes.

[0008] In a preferred embodiment, this application can be further configured to: acquire static wall data of each zone of a mountain photovoltaic power station, including: acquiring drone aerial images of each zone of the mountain photovoltaic power station; compressing the drone aerial images to obtain processed images; and performing edge extraction based on the processed images to obtain static wall data of each zone of the mountain photovoltaic power station.

[0009] In a preferred embodiment, this application can be further configured to: perform inter-target path planning based on the positions of the two first target nodes, the coordinates of the partition entrance and exit, and the first data, including: if the two first target nodes are in the same partition, then perform inter-target path planning pixel by pixel based on the positions of the two first target nodes and the static wall data; if the two first target nodes are not in the same partition, then perform inter-target path planning based on the positions of the two first target nodes, the static wall data, and the road network data.

[0010] In a preferred embodiment, this application can be further configured to: perform inter-target path planning based on the positions of the two first target nodes, the static wall data, and the road network data, including: determining the nearest neighbor entrance / exit positions corresponding to each of the two first target nodes; if the starting point of one of the two first target nodes is within a partition, determining a first sub-path plan between the two first target nodes and their corresponding nearest neighbor entrance / exit positions pixel by pixel based on the position of the starting point, the corresponding nearest neighbor entrance / exit positions, and the static wall data; and determining a second sub-path plan between entrances / exits based on the nearest neighbor entrance / exit positions corresponding to each of the two first target nodes and the road network data. The inter-target path planning includes: a first sub-path planning and a second sub-path planning; if the starting point of the two first target nodes is not within the partition, an intermediate path is determined based on the road network data between the line segment of the starting point in the road network data, the nearest neighbor entrance / exit position of the starting point, and the nearest neighbor entrance / exit position of the endpoint; a third sub-path planning is determined based on the line segment and the intermediate path between the nearest neighbor entrance / exit corresponding to the starting point and the endpoint; a fourth sub-path planning is determined pixel by pixel between the nearest neighbor entrance / exit position of the endpoint and the position of the endpoint, the inter-target path planning includes: the third sub-path planning and the fourth sub-path planning.

[0011] In a preferred embodiment, this application can be further configured to: determine a third sub-path plan between the nearest neighboring entrances and exits corresponding to the starting point and the ending point based on the line segment and the intermediate segment path, including: if the intermediate segment path includes the line segment, then remove the path between the starting point and the nearest neighboring entrance / exit position of the starting point in the intermediate segment path to obtain the third sub-path plan; if the intermediate segment path does not include the line segment, then determine the path between the starting point and the nearest neighboring entrance / exit position of the starting point in the line segment and the intermediate segment path as the third sub-path plan.

[0012] Secondly, a path planning device for a mountain photovoltaic power station is provided, comprising: a first acquisition module for acquiring road network data of the main road of the mountain photovoltaic power station, static wall data of each zone of the mountain photovoltaic power station, and zone entrance and exit coordinates; a second acquisition module for acquiring the positions of multiple target nodes, the positions including multiple locations to be inspected and inspection starting points; a multi-target sorting module for sorting the multiple targets according to the locations to be inspected and the inspection starting points to obtain a planning order of the multiple targets; an inter-target path planning module for performing inter-target path planning for any two first target nodes in the planning order, according to the positions of the two first target nodes, zone entrance and exit coordinates, and first data, wherein if the two first target nodes are both in the same zone, the first data includes static wall data; if the two first target nodes are not in the same zone, the first data includes static wall data and road network data; and a determination module for obtaining an inspection path plan based on the inter-target path planning results between every two target nodes in the planning order.

[0013] Thirdly, an electronic device is provided, the electronic device including a memory and a processor, the memory storing a computer program, the processor executing the path planning method for a mountain photovoltaic power station as described in any one of the first aspects when running the computer program.

[0014] Fourthly, a computer-readable storage medium is provided, wherein at least one piece of program code is stored in the computer-readable storage medium, the program code being loaded and executed by a processor to implement the path planning method for a mountain photovoltaic power station as described in any of the first aspects.

[0015] Fifthly, a computer program product is provided, including a computer program or instructions that, when executed by a processor, implement the path planning method for a mountain photovoltaic power station as described in any of the first aspects.

[0016] In summary, the path planning method for mountain photovoltaic power stations provided in this application has the following beneficial technical effects: It acquires road network data, static wall data, and zone inlet / outlet coordinates, inspection locations, and inspection starting points for the mountain photovoltaic power station. First, it performs multi-objective sorting among nodes to obtain a reasonable planning order. For any two first-objective nodes in the planning order, it performs pixel-level search based on their location, zone inlet / outlet coordinates, and first-order data to plan the path between the targets, ensuring the accuracy and feasibility of the path planning. This is particularly effective when conducting cross-zone inspections, as it fully utilizes road network data. By integrating the path planning results between every two target nodes, a complete inspection path plan is obtained. This application achieves efficient and accurate planning of inspection paths, reducing inspection time and costs, and improving the operation and maintenance efficiency of mountain photovoltaic power stations. Furthermore, this application also provides a path planning device, equipment, and medium for mountain photovoltaic power stations, all of which possess the aforementioned beneficial technical effects. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a path planning method for a mountain photovoltaic power station provided in an embodiment of this application.

[0019] Figure 2 This is a flowchart illustrating a multi-objective sorting algorithm based on ant colony algorithm provided in an embodiment of this application.

[0020] Figure 3 This is a schematic diagram of a process for determining a third sub-path plan provided in an embodiment of this application.

[0021] Figure 4 This is a schematic diagram of a path planning process between targets provided in an embodiment of this application.

[0022] Figure 5a This is a schematic diagram of a single-point navigation within a partition provided in an embodiment of this application.

[0023] Figure 5b This is a schematic diagram of multi-point navigation within a partition provided in an embodiment of this application.

[0024] Figure 6a This is a schematic diagram of a segmented single-point navigation provided in an embodiment of this application.

[0025] Figure 6bThis is a schematic diagram of a segmented multi-point navigation provided in an embodiment of this application.

[0026] Figure 7a This is a schematic diagram of a main road to single-point navigation provided in an embodiment of this application.

[0027] Figure 7b This is a schematic diagram of a main road to multi-point navigation provided in an embodiment of this application.

[0028] Figure 8 This is a schematic diagram of the structure of a path planning device for a mountain photovoltaic power station provided in an embodiment of this application.

[0029] Figure 9 This is a structural diagram of an electronic device provided in an embodiment of the present application. Detailed Implementation

[0030] This specific embodiment is merely an explanation of this application and is not intended to limit it. After reading this specification, those skilled in the art can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they are within the scope of this application.

[0031] It should be noted that, in the optional embodiments of this application, the data related to object information, when applied to specific products or technologies, requires the permission or consent of the object. Furthermore, the collection, use, and processing of this data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. In other words, if the embodiments of this application involve data related to an object, it must be obtained with the permission and consent of the object, the permission and consent of relevant departments, and in accordance with the relevant laws, regulations, and standards of the country and region. If the embodiments involve personal information, the acquisition of all personal information requires the consent of the individual. If sensitive information is involved, the separate consent of the information subject is required. The embodiments also need to be implemented with the permission and consent of the object.

[0032] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0033] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0034] Currently, basic location positioning is mainly achieved through satellite positioning, supplemented by electronic maps to plan approximate routes. In addition, drone inspection technology is also frequently used for auxiliary positioning, marking equipment locations through aerial imagery and providing feedback to power plant operation and maintenance personnel.

[0035] However, these technologies cannot provide a reliable path planning method in the context of large-scale mountain photovoltaic power plants.

[0036] To address the aforementioned issues, this application provides a path planning and navigation technology for maintenance personnel of mountainous photovoltaic power plants based on UAV aerial photography. This invention acquires high-precision base map images of the power plant through UAV aerial photography and extracts the road network using algorithms. It needs to address the problems that vegetation cover, shadow areas, terrain undulations, and uneven lighting can affect road network feature recognition. At the path planning level, mountainous power plant road networks often involve complex terrain such as steep slopes, cliffs, and ravines, requiring comprehensive analysis based on high-precision terrain data and road network information. Traditional path planning algorithms, if not deeply integrated with the terrain details of aerial photographs, are prone to planning infeasible or dangerous paths, failing to meet practical needs. Simultaneously, in terms of navigation, it is necessary to combine the road network and the actual location of equipment to achieve optimal route design for multiple faulty devices. This can significantly improve the efficiency of maintenance personnel in locating faulty equipment and reduce safety risks.

[0037] This application provides a path planning method for a mountain photovoltaic power station, such as... Figure 1 As shown, the method provided in this application embodiment can be executed by an electronic device, which is a server. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet computer, laptop computer, desktop computer, etc., but is not limited to these. The terminal device and electronic device can be directly or indirectly connected through wired or wireless communication. This application embodiment does not impose any restrictions here. The method includes: S101, acquiring the road network data of the main road of the mountain photovoltaic power station, the static wall data of each zone of the mountain photovoltaic power station, and the coordinates of the zone entrance and exit. The static wall data includes: zone range vector data and photovoltaic panel vector data within the zone.

[0038] In this context, a mountain photovoltaic power station refers to a photovoltaic power generation system built on mountainous terrain, utilizing solar energy resources and underutilized land to convert solar energy into electrical energy. In this embodiment, the mountain photovoltaic power station is divided into multiple zones, each containing multiple photovoltaic panels. Each zone is surrounded by a fence, and roads between zones form a road network. The main road network data represents the layout information of roads within the mountain photovoltaic power station area, including road location, direction, and connection relationships, indicating the distribution of roads within the power station. The static wall data for each zone of the mountain photovoltaic power station refers to a fixed set of data related to the zone, which can be represented as a set of obstacle data, indicating locations where maintenance personnel cannot pass. This static wall data may include: zone range vector data and photovoltaic panel vector data within the zone. The zone range vector data represents the boundary range of each zone, accurately describing the outline of the zone in vector form, facilitating the determination of the zone's size and shape. The photovoltaic panel vector data within the zone represents the distribution of photovoltaic panels within the zone. Each zone corresponds to one or more zone entrances / exits. The zone entrance / exit coordinates represent the specific location of each zone entrance / exit in the coordinate system, allowing for accurate determination of the entrance / exit location.

[0039] In this embodiment, since complete road network data within the mountain photovoltaic power station is not easily obtained directly, even manual annotation cannot be entirely accurate. Therefore, in this embodiment, the mountain photovoltaic power station is divided into multiple zones, each with fixed entrances and exits, and only these entrances and exits are accessible. Road network data for the main roads outside the zones can be obtained through manual annotation. Then, within each zone, a pixel-by-pixel navigation method is used for path planning.

[0040] In this embodiment, the included path planning algorithm requires main road network data, zone range vector data, zone entrance and exit coordinate data, and photovoltaic panel vector data within the zone. The main road network data is primarily used for main road path planning; the photovoltaic panel vector data and zone range vector data are primarily used for static wall generation in the path planning within the zone.

[0041] The purpose of static walls is to provide prior data information for the path planning algorithm stage; that is, the coordinates of the walls are the coordinates of impassable points, thus determining the navigation route. This step is mainly used during pixel-by-pixel navigation within a partition. Therefore, wall data consists of multiple coordinate points. It is understandable that the edges of the photovoltaic panels and the edges of the partitions are designed as walls. To better adapt to the pixel-by-pixel navigation algorithm, the static walls are transformed from the geographic coordinate system to the pixel coordinate system.

[0042] To further reduce the size of the offline file, the number of static walls needs to be further controlled. In practice, in addition to edge extraction (using pixel count as static walls), the image is also compressed to reduce its pixel resolution. Specifically, aerial images of photovoltaic power plants taken by drones are acquired and compressed to reduce their pixel resolution; road network data, zoning range vector data, and zoning entrance / exit coordinates are labeled in the photovoltaic power plant images; and photovoltaic panels are identified in the images to obtain photovoltaic panel vector data; based on the photovoltaic panel vector data and zoning range vector data, static wall data is determined.

[0043] S102. Obtain the location of multiple target nodes, including multiple locations to be inspected and the starting point of the inspection.

[0044] Among them, the multi-target node consists of multiple target points and the inspection start point, with the inspection start point being its current location; the multiple locations to be inspected are the locations that need to be inspected, which can be the locations of maintenance targets, such as the locations of photovoltaic modules, inverters, brackets, and other equipment.

[0045] After completing steps S101 and S102, in this embodiment, the path planning algorithm is the core of path planning and navigation technology, and due to its versatility, it can be reused for navigation in different photovoltaic power plants. A two-stage algorithm scheme is designed to address the navigation application requirements of multiple targets. The first stage is to sort the navigation of multiple navigation targets (maintenance targets, multiple photovoltaic panels) and determine the shortest route target navigation order, i.e., step S103; the second stage is to perform path planning between two targets based on the navigation order, i.e., step S104.

[0046] S103. Sort the multiple targets according to the location to be inspected and the starting point of the inspection to obtain the planning order of the multiple targets.

[0047] In some embodiments, multi-target node sorting can be implemented in several ways: Optionally, an ant colony algorithm can be used, with the Euclidean distance between any two points as the cost function, to sort multiple target nodes. Optionally, a genetic algorithm or particle swarm optimization can be used to sort multiple target nodes. It is understood that other methods can also be used to implement multi-target node sorting, which are not limited here.

[0048] This sorting process ensures that in subsequent path planning between nodes, the path planning between nodes can be performed according to the sorted node order, that is, the priority order.

[0049] S104. For any two first target nodes in the planning sequence, perform target path planning according to the positions of the two first target nodes, the coordinates of the zone entrance and exit and the first data.

[0050] If the two first target nodes are both in the same partition, the first data includes: static wall data. If the two first target nodes are not in the same partition, the first data includes: static wall data and road network data.

[0051] This step allows for the planning of paths between nodes, yielding the target path planning results between every two target nodes in the planning sequence.

[0052] S105. Based on the target path planning results between every two target nodes in the planning sequence, the inspection path planning is obtained.

[0053] According to the planning sequence, the path planning results between every two target nodes are connected to obtain the inspection path plan.

[0054] As can be seen, in this embodiment, road network data, static wall data, and zone entrance / exit coordinates, inspection locations, and inspection starting points of the mountain photovoltaic power station are acquired. First, multi-target sorting is performed between nodes to obtain a reasonable planning order. For any two first target nodes in the planning order, pixel-level search is used based on their location, zone entrance / exit coordinates, and the first data to perform target path planning, ensuring the accuracy and feasibility of the path planning. Especially during cross-zone inspections, road network data is fully utilized. By integrating the target path planning results between every two target nodes, a complete inspection path plan is obtained. This application achieves efficient and accurate inspection path planning, reducing inspection time and costs, and improving the operation and maintenance efficiency of the mountain photovoltaic power station.

[0055] One possible implementation of this application embodiment is as follows: S103 sorts multiple objectives based on the location to be inspected and the starting point of the inspection to obtain the planning order of the multiple objectives, including: selecting a preset number of ants, taking the starting point of the inspection as the starting position, and determining a location to be inspected for each ant; determining a location to be inspected for each ant until the ant path constructed by each ant is obtained; determining the distance of each ant path, where the distance of the ant path is the sum of the distances between nodes in the ant path; selecting the ant path with the shortest distance as the best path for the current round based on the distance of each ant path, and performing iterative processing to obtain the final best path, thereby determining the planning order of the multiple objectives.

[0056] The ant colony algorithm can be explained as follows: The feasible solutions to the problem to be optimized are represented by the walking paths of the ants. All paths of the entire ant colony constitute the solution space of the problem. Since ants release pheromones during their movement, and these pheromones evaporate over time, ants on shorter paths release more pheromones, and therefore tend to choose paths with higher pheromone levels. As time progresses, the pheromone concentration on shorter paths gradually increases, and more and more ants choose those paths. Ultimately, under the influence of positive feedback, all the ants converge on the optimal path, which corresponds to the optimal solution to the problem.

[0057] See Figure 2 , Figure 2 This application provides a flowchart of a multi-objective sorting process based on an ant colony algorithm, including: data preprocessing; initialization parameters, including a preset number of ants m, a pheromone evaporation coefficient, and the number of iterations; placing m ants at preset m points to be inspected; determining a current candidate road set for each ant, which is all locations to be inspected except the inspection starting point and the current ant's location; selecting the next movement position for each ant based on probability within the candidate road set; and determining whether the path has been completed, i.e., whether the ant has completed all nodes. An ant starts from the starting point and visits a series of nodes sequentially. List the target location; if it's not a completed path, jump back to the step of determining the current candidate road set for each ant; if it is, select the best path among m ants. In each round (iteration), select the shortest path among all current ant paths as the best path for the current round. At this point, distances between nodes are generally calculated using Euclidean distance. Determine if the maximum number of iterations has been reached. If so, determine the best path; otherwise, update the pheromones, including pheromone evaporation and pheromone enhancement. Pheromones evaporation is the process of the pheromone concentration on each path gradually decreasing automatically, while pheromone enhancement is the enhancement of the pheromone for the determined best path. After reaching the maximum number of iterations, based on the paths obtained in the final round, select the shortest path as the final best path to determine the planning order for multiple objectives.

[0058] As can be seen, in this embodiment of the application, planning the target order using the ant colony algorithm can further improve the planning accuracy.

[0059] Furthermore, considering that in scenarios with complex road conditions and varied terrain, factors such as actual road directions are not taken into account, sorting based solely on Euclidean distance may lead to insufficient sorting accuracy, affecting the accuracy and practicality of path planning. In this embodiment, the cost function for Euclidean distance can be modified, using a distance more closely resembling reality as the new cost function. This mainly involves calculating the Euclidean distances from the starting and ending points to the nearest entrances and exits within their respective partitions, determining the actual distance between these two entrances and exits, and using the sum of these two distances as the cost function for the ant colony algorithm, thereby significantly improving the accuracy of multi-objective sorting.

[0060] Based on this, one possible implementation of this application embodiment is to determine the distance of each ant path, including: for two adjacent second target nodes of the ant path, determining whether the two adjacent second target nodes are in the same partition; if they are in the same partition, then using Euclidean distance to determine the node distance between the two adjacent second target nodes; if they are not in the same partition, then using Euclidean distance to determine a first distance between each second target node and its nearest neighbor's entrance / exit location; determining a second distance between two nearest neighbor's entrance / exit locations based on road network data; determining the node distance between two adjacent second target nodes based on the first distance and the second distance; and determining the distance of the ant path based on the node distance between the two adjacent second target nodes.

[0061] For two adjacent target nodes u and v in the ant path, the distance is calculated as follows: Case 1: Nodes u and v are in the same partition, so the Euclidean distance between them is directly calculated. Case 2: Nodes u and v are not in the same partition. The nearest neighbor entry / exit location is found. For node u, all entry / exit locations in its partition are found, and the Euclidean distance from node u to each entry / exit location is calculated. The nearest entry / exit location eu is selected. Similarly, for node v, all entry / exit locations in its partition are found, and the Euclidean distance from node v to each entry / exit location is calculated. The nearest entry / exit location ev is selected. The Euclidean distance d1 from node u to eu is calculated. The Euclidean distance d2 from node v to ev is calculated. Based on the road network data, the actual road distance droad between eu and ev is found. Therefore, the total distance between nodes u and v is: dtotal = d1 + road + d2.

[0062] Then, after obtaining the distances between all adjacent nodes, they are summed up to obtain the distance of the ant's path.

[0063] As can be seen, in this embodiment of the application, by combining partition, entrance and exit locations and road network data, the actual driving distance between cross-partition nodes can be calculated more accurately.

[0064] In one possible implementation of this application embodiment, the static wall data further includes: terrain obstacle vector data.

[0065] One possible implementation of this application embodiment is that obtaining static wall data of each zone of the mountain photovoltaic power station in S101 includes: obtaining drone aerial images of each zone of the mountain photovoltaic power station; compressing the drone aerial images to obtain processed images; and performing edge extraction based on the processed images to obtain static wall data of each zone of the mountain photovoltaic power station.

[0066] The process involves acquiring aerial images of various sections of a mountain photovoltaic power station using drones, followed by image compression, specifically lossless compression, to significantly reduce file size while maintaining visual quality. Due to the reduced data volume, less data is obtained when further performing edge extraction to determine static wall data.

[0067] When performing edge extraction, algorithms such as Canny edge detection, Sobel edge detection, and LoG can be used to extract edges, and then edge concatenation can be performed to remove noisy edges.

[0068] Furthermore, when using drones for aerial photography, the drone's shooting altitude and angle can be adjusted to ensure sufficient image overlap for image stitching, thereby improving image accuracy. Additionally, if the lighting is low, techniques such as increasing the ISO and decreasing the shutter speed to increase light intake can be employed.

[0069] As can be seen, in this embodiment of the application, by compressing the image, the pixel resolution can be reduced, thereby controlling the number of static walls.

[0070] Furthermore, for path planning between two nodes, the following can be used: Algorithms such as Dijkstra's algorithm and D* algorithm are used for pixel-level path search. Algorithm example The algorithm flow includes: 1. Add the starting point to the open list. 2. Traverse the open list, find the node with the smallest F value, and make it the current node to be processed. Move this node to the closed list. F = G + H, where G is the cost of moving from the starting point to the specified point, and H is the estimated cost of moving from the specified point to the destination. 3. For the adjacent nodes of the current node. If they are unreachable or in the closed list, ignore them. Otherwise, do the following: If the node is not in the open list, add it to the open list, set the current cell as its parent, and record the F, G, and H values ​​of the node. If it is already in the open list, check if this path (i.e., reaching it via the current cell) is better, using the G value as a reference. A smaller G value indicates that this is a better path. If so, set its parent node as the current node and recalculate the G and F values. 4. Repeat the above steps until the destination is added to the open list, at which point the path has been found, or the search for the destination fails and the open list is empty, at which point there is no path. 5. Save the path. Starting from the end point, each square moves along its parent node back to the starting point to find the shortest path.

[0071] One possible implementation of this application embodiment is as follows: S104 performs inter-target path planning based on the positions of the two first target nodes, the coordinates of the partition entrance and exit, and the first data, including:

[0072] If the two first target nodes are in the same partition, then the path planning between the targets is performed pixel by pixel based on the positions of the two first target nodes and the static wall data;

[0073] If the two primary target nodes are not in the same partition, then path planning between the targets is performed based on the locations of the two primary target nodes, static wall data, and road network data.

[0074] Since the route is planned starting from the inspection start point, and this route includes multiple locations to be inspected, the locations of two adjacent first target nodes can be the inspection start point and the first location to be inspected, or two locations to be inspected in between (the two locations to be inspected may be in the same zone or different zones). The inspection start point can be within a zone or in the road network outside the zone, while the locations to be inspected are generally within a zone.

[0075] Based on this, if the two first target nodes are within the same partition, path planning between the targets can be performed pixel-by-pixel based on the A* algorithm / Dijkstra algorithm / D* algorithm. Specifically, taking the A* algorithm as an example, the positions of the two first target nodes are obtained and set as the starting and ending positions, and static wall data, i.e., obstacle pixel-level information, is also obtained. The starting point is added to the open list, and its F=G+H value is recorded (G is the actual cost, and H is the heuristically estimated cost). The specific calculation method is not limited in this embodiment of the application, and users can set it according to actual needs. The pixel with the smallest F is taken from the open list as the current node. Neighboring pixels (usually 8-neighborhood): if it is a walkable area (not a wall and not visited), its G and H values ​​are calculated. If the pixel is in the closed list and the new G value is better, it is updated. This continues until the ending point is found, thus realizing the path planning between the targets.

[0076] If the two first target nodes are not in the same partition, there are two situations: one is that the starting point is in the road network and the ending point is in the partition; the other is that the starting point and the ending point are not in the same partition. In this case, path planning between the targets can be performed based on the positions of the two first target nodes, static wall data, and road network data.

[0077] In one feasible approach, if the two primary target nodes are not in the same partition, the methods for inter-target path planning include:

[0078] SA1. Determine the nearest neighbor import / export locations for the two first target nodes.

[0079] In one feasible approach, when determining the nearest neighbor entrance / exit location corresponding to the first target node, the distance between the first target node and each entrance / exit location in its partition can be determined using Euclidean distance. Then, the entrance / exit location with the shortest distance can be selected as the nearest neighbor entrance / exit location for the first target node. Alternatively, based on the first target node's location, each entrance / exit location in its partition, and the partition's static wall data, the distance can be determined using... Algorithm / Dijkstra's Algorithm / The algorithm performs a pixel-by-pixel search to plan paths between targets, ensuring that obstacles are avoided and the path is as short as possible. Then, it determines the distance of the path corresponding to each entrance and exit, and selects the entrance and exit position corresponding to the shortest distance as the nearest neighbor entrance and exit position of the first target node.

[0080] SA2. If the starting point of one of the two first target nodes is within the partition, then the first sub-path planning between the two first target nodes and their corresponding nearest neighbor entrance / exit positions is determined pixel by pixel based on the position of the starting point, the corresponding nearest neighbor entrance / exit positions, and the static wall data; the second sub-path planning between entrances / exits is determined based on the nearest neighbor entrance / exit positions and road network data corresponding to the two first target nodes respectively; the inter-target path planning includes: the first sub-path planning and the second sub-path planning.

[0081] In this embodiment, if two first target nodes are located in two partitions respectively, then the path between the two first target nodes includes: a first sub-path plan between the starting point and the first entry / exit, a second sub-path plan between the first entry / exit and the second entry / exit corresponding to the destination, and a first sub-path plan between the second entry / exit corresponding to the destination and the destination. The first sub-path plan can be implemented using... Algorithm / Dijkstra's Algorithm / The algorithm performs a pixel-by-pixel search to plan paths between targets, while the second sub-path planning combines the actual paths of the road network for planning.

[0082] SA3. If the starting point of one of the two first target nodes is not within the partition, determine the intermediate path between the line segment of the starting point in the road network data, the nearest neighbor entrance / exit position of the starting point, and the nearest neighbor entrance / exit position of the destination based on the road network data; determine the third sub-path planning between the nearest neighbor entrance / exit corresponding to the starting point and the destination based on the line segment and the intermediate path; determine the fourth sub-path planning between the nearest neighbor entrance / exit position of the destination and the position of the destination pixel by pixel. The inter-target path planning includes: the third sub-path planning and the fourth sub-path planning.

[0083] In this embodiment of the application, if the starting point of the two first target nodes is not in the partition, it means that the starting point is likely to be the inspection starting point. At this time, the middle section path between the line segment of the starting point in the road network data, the nearest neighbor of the starting point and the nearest neighbor of the endpoint can be determined based on the road network data.

[0084] Then, the third and fourth sub-path plans are determined. For the third sub-path plan, if the intermediate path includes the line segment, the path between the nearest neighboring entrance / exit position of the starting point is removed from the intermediate path to obtain the third sub-path plan; if the intermediate path does not include the line segment, the path between the nearest neighboring entrance / exit position of the starting point in the line segment and the path therein are determined as the third sub-path plan.

[0085] For example, see Figure 3Where a, b, and e are the target point locations, q and p are the entrances and exits, c is the intersection of L1 and L2 in the road network data, and d is the intersection of L1 and L3 in the road network data. If a is the starting point location and e is the ending point location, then the line segment of the starting point in the road network data is a→c, and the two nearest entrances and exits are q and p. The intermediate path between the nearest entrance and exit locations of the starting point and the nearest entrance and exit locations of the ending point is q→c→d→p. At this time, a→c is not within q→c→d→p, so the line segment is divided to obtain a→q. The final third sub-path planning is a→q→c→d→p.

[0086] If b is the starting point and e is the ending point, then the line segment of the starting point in the road network data is b→c, and the two nearest neighboring entrances and exits are q and p. The intermediate path between the nearest neighboring entrance / exit of the starting point and the nearest neighboring entrance / exit of the ending point is q→c→d→p (subdivided into q→b→c→d→p). Since the intermediate path includes the line segment b→c, we remove q→b from the intermediate path to obtain b→c→d→p as the third sub-path planning.

[0087] For the fourth sub-path planning, based on the nearest neighbor entrance / exit positions of the destination and the location of the destination, the A* algorithm / Dijkstra algorithm / D* algorithm is used to search pixel by pixel to plan the path between the targets.

[0088] Based on the above, see Figure 4 , Figure 4 This is a flowchart illustrating a path planning process between targets provided in an embodiment of this application, where s and e represent the start and end points, respectively; snd and end represent the nearest entrance / exit to the start point (tartNearDoor) and the nearest entrance / exit to the destination (endNearDoor), respectively; line represents the line segment where the start point s is located; and way represents the intermediate path between snd and end.

[0089] If the starting point s and the ending point e are in the same partition, the A* algorithm is used directly to find the optimal path pixel by pixel.

[0090] If the starting point s and the ending point e are not in the same partition, then it is necessary to determine whether the starting point s is in the same partition.

[0091] If the starting point is within the partition, the path planning is divided into three stages: path planning between the starting point and its nearest neighbor, path planning between the starting point's nearest neighbor and the ending point's nearest neighbor, and path planning between the ending point's nearest neighbor and the ending point. The path planning for the first two stages uses the pixel-wise search A* algorithm, while the path planning between intermediate nodes (between the two entrances / exits) uses a node-wise search algorithm.

[0092] If the starting point is not within the partition, it is necessary to find the line segment containing the starting point and the path (way) between the two nearest neighbors. If the path includes the line segment, it is necessary to segment the path, remove multiple line segments between the starting point and its nearest neighbor, and then add them to the planned path, along with the path from the ending point to its nearest neighbor. If the path does not contain the line segment, it is necessary to segment the line, select multiple line segments between the starting point and its nearest neighbor, add them to the path, and simultaneously add the path (way) between the two nearest neighbors and the path from the ending point to its nearest neighbor. When the starting point is not within the partition, and the nearest neighbors of the starting point and the ending point are the same point, it is only necessary to segment the line and add it to the path, along with the path from the ending point to its nearest neighbor.

[0093] Furthermore, in the path planning algorithm, since the static "wall" offline file used records pixel coordinates, the planned route is also in pixel coordinates and needs to be converted to the WGS84 (EPSG:4326) coordinate system for use. Because the inter-target path planning uses a pixel-by-pixel search algorithm, the planned path may suffer from severe distortion, which can be addressed by simplifying the path trajectory. One possible implementation of this application, after obtaining the inspection path planning based on the inter-target path planning results between every two target nodes in the planning sequence, further includes: performing multi-segment smoothing processing on the inspection path planning. To address this issue, the Bezier spline algorithm / Douglaspoker algorithm / B-spline algorithm are used to smooth the multi-segment lines.

[0094] Furthermore, it is understandable that in existing technologies, satellite positioning is significantly affected by mountainous terrain and multipath effects, resulting in signal interruptions or large deviations, making it difficult to meet the precise positioning requirements of equipment in complex mountainous environments. Electronic maps often lack detailed markings of subtle terrain features, hidden paths, and densely populated equipment areas within mountainous power plants, leading to navigation route planning that deviates from the actual terrain and significantly reducing practicality. In addition, there is a lack of intelligent identification and avoidance capabilities for obstacles such as mountain vegetation cover, steep cliffs, and fences, making it easy for maintenance personnel to find themselves in dead-end or dangerous situations when following navigation. Moreover, it is impossible to effectively combine the real-time operating status of power plant equipment (such as the priority of faulty equipment) for dynamic navigation optimization, resulting in positioning and navigation efficiency and reliability that cannot meet actual maintenance needs.

[0095] In this application, it is not necessary to obtain all the road networks, but only the road networks on the main road. Path planning is performed within the power station zone using photovoltaic panels as obstacles, saving the initial data acquisition cost. The multi-objective node sorting algorithm adopts a combination of Euclidean distance within the zone and the actual distance on the main road, which speeds up path planning while maintaining a certain level of accuracy. The path planning scheme between two nodes is designed with different methods: intra-zone navigation and inter-zone navigation. Intra-zone navigation uses photovoltaic panels as obstacles and adopts pixel-by-pixel navigation for path planning. Inter-zone navigation uses node-by-node navigation between entrances and exits for path planning, which speeds up navigation efficiency.

[0096] For the proposed solution, the offline navigation system for mountain photovoltaic power stations was tested from three perspectives: navigation within a zone, navigation between different zones, and navigation from the main road to the target zone.

[0097] For navigation within a zone, see [link / reference]. Figure 5a and Figure 5b , Figure 5a This is a schematic diagram of single-point navigation within a partition provided in an embodiment of this application, wherein the white triangle represents the starting point and the white circle represents the target point, and the curve between the two represents path planning. Figure 5b This is a schematic diagram of multi-point navigation within a partition provided in an embodiment of this application, with the curve representing path planning. Rigorous testing was conducted on photovoltaic panel navigation within the same partition, including path planning for single photovoltaic panel targets located in the same partition as the starting point, as well as for multiple photovoltaic panel targets. Experimental results show that the system can accurately plan routes.

[0098] For section-based navigation, see [link / reference]. Figure 6a and Figure 6b . Figure 6a This is a schematic diagram of a segmented single-point navigation provided in an embodiment of this application, wherein the white triangle represents the starting point and the white circle represents the target point, and the curve between the two represents the path planning. Figure 6b This is a schematic diagram of a segmented multi-point navigation provided in an embodiment of this application, where the curve represents path planning.

[0099] In addition to testing photovoltaic panel navigation within the same zone, we also conducted in-depth research on photovoltaic panel navigation in different zones. Experimental results show that the system can quickly and accurately plan paths for both single and multiple targets. This further demonstrates the system's superior performance and efficiency in handling targets across different zones.

[0100] For navigation from main roads to the target zone, see [link / reference]. Figure 7a and Figure 7b , Figure 7a This is a schematic diagram illustrating a main road to single-point navigation provided in an embodiment of this application. Figure 7b This is a schematic diagram of a main road to multi-point navigation provided in an embodiment of this application. Besides considering navigation where the starting point is within a zone, we have also fully considered various scenarios in practical applications, especially the case where the starting point is on a main road. For this scenario, detailed testing was conducted, and the following results were obtained: Figure 7a and Figure 7b The test results shown further validate the stability and reliability of the system under different scenarios, providing users with a more comprehensive and efficient photovoltaic panel navigation service.

[0101] The following describes a path planning device for a mountain photovoltaic power station provided in an embodiment of this application. The path planning device for the mountain photovoltaic power station described below can be referred to in correspondence with the path planning method for the mountain photovoltaic power station described above. In this embodiment, the path planning device for the mountain photovoltaic power station is installed in an electronic device. Figure 8 , Figure 8 This is a structural block diagram of a path planning device for a mountain photovoltaic power station according to one embodiment of this application, including: a first acquisition module 210, used to acquire road network data of the main road of the mountain photovoltaic power station, static wall data of each zone of the mountain photovoltaic power station, and zone entrance and exit coordinates; a second acquisition module 220, used to acquire the positions of multiple target nodes, including multiple locations to be inspected and inspection starting positions; a multi-target sorting module 230, used to sort the multiple targets according to the locations to be inspected and inspection starting positions to obtain the planning order of the multiple targets; an inter-target path planning module 240, used to perform inter-target path planning for any two first target nodes in the planning order, according to the positions of the two first target nodes, zone entrance and exit coordinates, and first data. If the two first target nodes are both in the same zone, the first data includes static wall data; if the two first target nodes are not in the same zone, the first data includes static wall data and road network data; and a determination module 250, used to obtain the inspection path planning based on the inter-target path planning results between every two target nodes in the planning order.

[0102] In one feasible approach, the multi-objective sorting module 230 is used to: select a preset number of ants, using the inspection starting point as the starting position, and determine an inspection position for each ant; determine an inspection position for each ant until the ant path constructed by each ant is obtained; determine the distance of each ant path, where the distance of the ant path is the sum of the distances between nodes in the ant path; and, based on the distance of each ant path, select the ant path with the shortest distance as the best path for the current round, and perform iterative processing to obtain the final best path, thereby determining the planning order of the multi-objectives.

[0103] In one feasible implementation, the multi-objective sorting module 230 is used to: determine whether two adjacent second target nodes of an ant path are in the same partition; if they are in the same partition, determine the node distance between the two adjacent second target nodes using Euclidean distance; if they are not in the same partition, determine the first distance between each second target node and its nearest neighbor's entrance / exit location using Euclidean distance; determine the second distance between the two nearest neighbor's entrance / exit locations based on road network data; determine the node distance between two adjacent second target nodes based on the first and second distances; and determine the distance of the ant path based on the node distance between the two adjacent second target nodes.

[0104] In one feasible approach, the first acquisition module 210 is configured to: acquire drone aerial images of each section of the mountain photovoltaic power station; compress the drone aerial images to obtain processed images; and perform edge extraction based on the processed images to obtain static wall data of each section of the mountain photovoltaic power station.

[0105] In one possible implementation, the inter-target path planning module 240 is used to: if two first target nodes are in the same partition, perform inter-target path planning pixel by pixel based on the positions of the two first target nodes and static wall data; if two first target nodes are not in the same partition, perform inter-target path planning based on the positions of the two first target nodes, static wall data and road network data.

[0106] In one feasible implementation, the inter-target path planning module 240 is used to: determine the nearest neighbor entrance / exit positions corresponding to two first target nodes; if the starting point of the two first target nodes is within a partition, determine a first sub-path plan between the two first target nodes and their corresponding nearest neighbor entrance / exit positions pixel by pixel based on the starting point position, the corresponding nearest neighbor entrance / exit positions, and static wall data; determine a second sub-path plan between entrances / exits based on the nearest neighbor entrance / exit positions corresponding to the two first target nodes and road network data; the inter-target path planning includes: the first sub-path plan and the second sub-path plan; if the starting point of the two first target nodes is not within a partition, determine the intermediate path between the line segment of the starting point in the road network data, the nearest neighbor entrance / exit position of the starting point, and the nearest neighbor entrance / exit position of the ending point based on the road network data; determine a third sub-path plan between the nearest neighbor entrance / exit positions corresponding to the starting point and the ending point based on the line segment and the intermediate path; and determine a fourth sub-path plan between the nearest neighbor entrance / exit position of the ending point and the position of the ending point pixel by pixel; the inter-target path planning includes: the third sub-path plan and the fourth sub-path plan.

[0107] In one possible implementation, the inter-target path planning module 240 is configured to: if the intermediate path includes the line segment, remove the path between the nearest neighboring entrance / exit positions from the starting point to the starting point in the intermediate path to obtain a third sub-path plan; if the intermediate path does not include the line segment, determine the path between the nearest neighboring entrance / exit positions from the starting point to the starting point in the line segment and the intermediate path as the third sub-path plan.

[0108] This application provides an electronic device, such as... Figure 9 As shown, Figure 9 The illustrated electronic device 300 includes: at least one processor 301 ( Figure 9 The image shows one processor 301 and one memory 303. The processor 301 and memory 303 are connected, for example, via a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that in practical applications, the transceiver 304 is not limited to one, and the structure of the electronic device 300 does not constitute a limitation on the embodiments of this application.

[0109] Processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 301 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0110] Bus 302 may include a pathway for transmitting information between the aforementioned components. Bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 302 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 9 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0111] The memory 303 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0112] The memory 303 is used to store application code that executes the solution of this application, and its execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0113] Figure 9 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0114] This application provides a computer-readable storage medium storing at least one piece of program code that, when run on a computer, enables the computer to execute the corresponding content in the aforementioned method embodiments.

[0115] This application provides a computer program product, including a computer program or instructions, which, when executed by a processor, implement the corresponding content in the aforementioned method embodiments.

[0116] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0117] The above are only some embodiments of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A path planning method for a mountain photovoltaic power station, characterized in that, include: The system acquires road network data of the main road of the mountain photovoltaic power station, static wall data of each zone of the mountain photovoltaic power station, and coordinates of the entrance and exit of each zone. The static wall data includes: zone range vector data and photovoltaic panel vector data within the zone; wherein, the zone range vector data represents the boundary range of each zone and accurately describes the outline of the zone in vector form. Obtain the location of multiple target nodes, including multiple locations to be inspected and the inspection start point location; Based on the locations to be inspected and the starting point of the inspection, multiple targets are sorted to obtain the planning order of the multiple targets; For any two first target nodes in the planning sequence, a path planning between the targets is performed according to the positions of the two first target nodes, the coordinates of the zone entrance and exit, and the first data. If the two first target nodes are both in the same zone, the first data includes: static wall data. If the two first target nodes are not in the same zone, the first data includes: static wall data and road network data. Based on the target path planning results between every two target nodes in the planning sequence, the inspection path planning is obtained, and multi-segment smoothing processing is performed on the inspection path planning. Based on the positions of the two first target nodes, the coordinates of the zone entrance and exit, and the first data, target path planning is performed, including: If the two first target nodes are in the same partition, then target path planning is performed pixel by pixel based on the positions of the two first target nodes and the static wall data; If the two first target nodes are not in the same partition, then target path planning is performed based on the positions of the two first target nodes, the static wall data, and the road network data.

2. The path planning method for a mountain photovoltaic power station according to claim 1, characterized in that, Based on the locations to be inspected and the starting point of the inspection, multiple targets are sorted to obtain the planning order of the multiple targets, including: Select a preset number of ants, take the inspection starting point as the starting position, and determine an inspection position for each ant; Determine the location to be inspected for each ant until the ant path constructed by each ant is obtained; Determine the distance of each ant path; the distance of an ant path is the sum of the distances between the nodes along the ant path. Based on the distance of each ant path, the shortest ant path is selected as the best path for the current round, and the process is iterated to obtain the final best path, so as to determine the planning order of multi-objectives.

3. The path planning method for mountain photovoltaic power stations according to claim 2, characterized in that, Determine the distance of each ant's path, including: For two adjacent second target nodes in the ant path, determine whether the two adjacent second target nodes in the ant path are in the same partition; If they are in the same partition, the distance between two adjacent second target nodes is determined by Euclidean distance. If they are not in the same partition, the first distance between each second target node and its nearest neighbor's entrance / exit location is determined using Euclidean distance; the second distance between two nearest neighbor entrance / exit locations is determined based on the road network data; and the node distance between two adjacent second target nodes is determined based on the first distance and the second distance. The distance of the ant path is determined based on the distance between two adjacent second target nodes.

4. The path planning method for a mountain photovoltaic power station according to claim 1, characterized in that, Obtain static wall data for each zone of the mountain photovoltaic power station, including: Acquire drone aerial images of various sections of the mountain photovoltaic power station; The aerial images captured by the UAV are compressed to obtain the processed images; Edge extraction is performed on the processed image to obtain static wall data for each section of the mountain photovoltaic power station.

5. The path planning method for a mountain photovoltaic power station according to claim 1, characterized in that, Based on the locations of the two first target nodes, the static wall data, and the road network data, target path planning is performed, including: Determine the nearest neighbor entry / exit locations for the two first target nodes respectively; If the starting point of one of the two first target nodes is within the partition, a first sub-path plan is determined pixel by pixel between the two first target nodes and their corresponding nearest neighbor entrance / exit positions based on the position of the starting point, the position of the corresponding nearest neighbor entrance / exit, and the static wall data; a second sub-path plan is determined between the entrances / exits based on the nearest neighbor entrance / exit positions of the two first target nodes and the road network data; the inter-target path planning includes: the first sub-path plan and the second sub-path plan; If the starting point of the two first target nodes is not within the partition, the intermediate path between the line segment of the starting point in the road network data, the nearest neighbor entrance / exit position of the starting point, and the nearest neighbor entrance / exit position of the destination is determined according to the road network data; a third sub-path plan between the nearest neighbor entrance / exit corresponding to the starting point and the destination is determined according to the line segment and the intermediate path; a fourth sub-path plan between the nearest neighbor entrance / exit position of the destination and the position of the destination is determined pixel by pixel, and the inter-target path plan includes: the third sub-path plan and the fourth sub-path plan.

6. The path planning method for a mountain photovoltaic power station according to claim 5, characterized in that, Based on the line segment and the intermediate path, a third sub-path plan is determined between the nearest neighboring entrances and exits from the starting point to the ending point, including: If the intermediate path includes the line segment, then remove the path between the starting point and the nearest neighboring entrance / exit position of the starting point in the intermediate path to obtain the third sub-path plan; If the intermediate path does not include the line segment, then the path from the starting point of the line segment to the nearest neighboring entrance / exit location of the starting point and the intermediate path are determined as the third sub-path plan.

7. A path planning device for a mountain photovoltaic power station, characterized in that, include: The first acquisition module is used to acquire the road network data of the main road of the mountain photovoltaic power station, the static wall data of each zone of the mountain photovoltaic power station, and the coordinates of the zone entrance and exit. The static wall data includes: partition range vector data and photovoltaic panel vector data within the partition; wherein, the partition range vector data represents the boundary range of each partition, accurately describing the outline of the partition in vector form; the second acquisition module is used to acquire the position of multiple target nodes, the position including multiple locations to be inspected and the inspection starting point position; The multi-target sorting module is used to sort multiple targets according to the location to be inspected and the starting point of the inspection, so as to obtain the planned order of the multiple targets; The inter-target path planning module is used to perform inter-target path planning for any two first target nodes in the planning sequence, according to the positions of the two first target nodes, the coordinates of the partition entrance and exit, and the first data. If the two first target nodes are both in the same partition, the first data includes: static wall data; if the two first target nodes are not in the same partition, the first data includes: static wall data and road network data. The determination module is used to obtain the inspection path plan based on the target path planning results between every two target nodes in the planning sequence, and to perform multi-segment smoothing processing on the inspection path plan. The multi-objective sorting module is used for: If the two first target nodes are in the same partition, then target path planning is performed pixel by pixel based on the positions of the two first target nodes and the static wall data; If the two first target nodes are not in the same partition, then target path planning is performed based on the positions of the two first target nodes, the static wall data, and the road network data.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and the processor executes the path planning method for a mountain photovoltaic power station according to any one of claims 1 to 6 when running the computer program.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one piece of program code, which is loaded and executed by a processor to implement the path planning method for a mountain photovoltaic power station as described in any one of claims 1 to 6.

Images