UAV Route Layout Method and Device, Equipment, Medium for Complex Terrain

Abstract

The present disclosure provides a method and device, equipment, and medium for arranging UAV flight routes in complex terrains, relating to the technical field of UAV aerial photogrammetry. The method includes: preliminarily dividing the terrain area to be surveyed to determine multiple regional units; clustering the regional units according to terrain observation data to obtain surveyed sub-regions; determining the survey priorities corresponding to each surveyed sub-region, and matching the corresponding basic flight routes of the UAV according to the survey priorities; splicing different types of basic flight routes corresponding to each surveyed sub-region, and optimizing and adjusting the spliced basic flight routes to obtain the target UAV flight route for surveying the terrain of the terrain area to be surveyed. This technical solution enables the flight route planning of the UAV to adapt to various types of landform features, improves the integrity and accuracy of aerial survey data, and ensures the safety and aerial survey efficiency of the UAV.

Description

Background Art

[0002] In areas with complex terrain and harsh climate conditions (such as high mountainous areas, canyon reservoir slopes, etc.), it brings great challenges to engineering survey work. The traditional method of arranging parallel flight lines is usually applicable to flat areas. However, in areas with large height differences, steep mountains, and deep valleys, this method is difficult to fully adapt to terrain changes, easily resulting in insufficient aerial survey coverage, limited acquisition of image data in some areas, and affecting the overall surveying and mapping accuracy. Especially in weak texture areas of high plateau regions or canyon reservoir slopes (such as ice and snow covered areas or sparse vegetation areas), it is difficult to obtain high-precision data through conventional image matching methods, affecting the accuracy of subsequent data processing and 3D reconstruction.

[0003] Currently, relevant UAV flight line planning technologies mainly target conventional terrain environments and lack adaptability design for special terrains in high mountainous areas. They mainly rely on manual experience judgment for setting, resulting in low efficiency of aerial survey work in high mountainous areas, and the coverage, accuracy, and reliability of aerial survey data are all relatively low, and the quality of the results cannot meet the engineering requirements. At the same time, the instability of airflows in high mountainous areas and the complex distribution of obstacles (such as mountains and cliffs) also pose higher requirements for the safety of UAV flights. There are also many deficiencies in related methods in terms of obstacle avoidance functions and flight path optimization, easily leading to the interruption of flight tasks and even equipment losses.

[0004] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the embodiments of the present disclosure is to provide a UAV flight line arrangement method for complex terrain, a UAV flight line arrangement device for complex terrain, an electronic device, and a computer-readable storage medium, so as to enable the flight line planning of UAVs to adapt to various types of landform features, improve the integrity and accuracy of aerial survey data, and ensure the safety and aerial survey efficiency of UAVs.

[0006] According to the first aspect of the embodiments of the present disclosure, a UAV flight line arrangement method for complex terrain is provided, including:

[0007] Obtain terrain observation data of the terrain area to be surveyed, where the terrain observation data includes digital elevation models, terrain orthophotos, geological disaster data, and remote sensing multispectral data;

[0008] Conduct a preliminary regional division of the terrain area to be surveyed to determine multiple regional units;

[0009] Cluster each of the regional units according to the terrain observation data to obtain surveyed sub-areas;

[0010] Determine the survey priority corresponding to each of the surveyed sub - regions, and match the basic flight route that the drone is to adopt in the surveyed sub - region according to the survey priority;

[0011] Stitch together the different types of basic flight routes corresponding to each of the surveyed sub - regions, and optimize and adjust the stitched - together basic flight route to obtain the target drone flight route, so as to conduct topographic survey on the terrain area to be surveyed through the target drone flight route.

[0012] According to a second aspect of the embodiments of the present disclosure, there is provided a device for laying out drone flight routes for complex terrains, including:

[0013] A terrain observation module, configured to obtain terrain observation data of the terrain area to be surveyed, where the terrain observation data includes a digital elevation model, a terrain orthophoto, geological disaster data, and remote sensing multispectral data;

[0014] A region division module, configured to perform a preliminary regional division on the terrain area to be surveyed to determine a plurality of regional units;

[0015] A region clustering module, configured to cluster each of the regional units according to the terrain observation data to obtain surveyed sub - regions;

[0016] A flight route matching module, configured to determine the survey priority corresponding to each of the surveyed sub - regions, and match the basic flight route that the drone is to adopt in the surveyed sub - region according to the survey priority;

[0017] A flight route stitching module, configured to stitch together the different types of basic flight routes corresponding to each of the surveyed sub - regions, and optimize and adjust the stitched - together basic flight route to obtain the target drone flight route, so as to conduct topographic survey on the terrain area to be surveyed through the target drone flight route.

[0018] According to a third aspect of the embodiments of the present disclosure, there is provided an electronic device, including: a processor; and a memory, where computer - readable instructions are stored on the memory, and when the computer - readable instructions are executed by the processor, the method for laying out drone flight routes for complex terrains in the first aspect is implemented.

[0019] According to a fourth aspect of the embodiments of the present disclosure, there is provided a computer - readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method for laying out drone flight routes for complex terrains in the first aspect is implemented.

[0020] The technical solutions provided by the embodiments of the present disclosure may include the following beneficial effects:

[0021] The method for arranging UAV flight routes for complex terrains in the exemplary embodiments of the present disclosure can provide rich terrain information support for arranging UAV flight routes by acquiring multi-source terrain observation data of the terrain area to be surveyed, enabling the flight route arrangement to be refined and adjusted based on terrain features, avoiding survey blind spots caused by a single data source, and thus improving the integrity and accuracy of aerial survey data. By initially dividing the terrain area to be surveyed and further clustering each regional unit based on terrain observation data, the UAV flight route arrangement can implement differential planning for different terrain types, reducing problems such as repeated or omitted flight routes caused by large differences in terrain features. Especially in weakly textured areas such as high-altitude mountainous areas or canyon reservoir slopes (such as ice and snow covered areas or sparse vegetation areas), through the analysis of the eigenvectors of regional units, accurate division of regional types can be achieved, enabling the flight route planning to adapt to different geomorphic features and improving the reliability of surveying and mapping data and the accuracy of subsequent 3D reconstruction. Combining the division results of the survey sub-areas, further determining the survey priorities of each survey sub-area and matching corresponding basic flight routes can reasonably allocate the flight resources of the UAV according to terrain complexity, avoiding excessive flight time occupied by low-priority areas, and thus improving the overall efficiency of aerial survey work. At the same time, in the face of complex obstacle distributions in high-altitude mountainous areas, the division of different survey priorities can provide a basis for obstacle avoidance strategies, enabling the UAV to reasonably avoid obstacles in a complex environment and reducing the risks of mission interruption and equipment loss caused by unreasonable flight path planning. In the process of flight route splicing and optimization adjustment, by splicing different types of basic flight routes of each survey sub-area and making adaptive adjustments in the splicing process in combination with regional characteristic parameters, the flight redundancy at the flight route splicing can be effectively reduced, the flight route coverage can be optimized, and the accuracy of aerial survey data can be further improved. At the same time, through path smoothing processing and verification of the heading overlap and side overlap, the flight route planning has better adaptability in complex terrain areas, ensuring the high accuracy and consistency of surveying and mapping results.

[0022] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present disclosure and used together with the specification to explain the principles of the present disclosure. Obviously, the accompanying drawings in the following description are only some embodiments of the present disclosure, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0024] Figure 1 The flowchart of the method for arranging UAV flight routes for complex terrains according to some embodiments of the present disclosure is schematically shown.

[0025] Figure 2 Schematically shows a flowchart of dividing to obtain a surveyed sub-region according to some embodiments of the present disclosure.

[0026] Figure 3 Schematically shows a flowchart of constructing to obtain a terrain recognition feature vector according to some embodiments of the present disclosure.

[0027] Figure 4 Schematically shows a flowchart of clustering to obtain a surveyed sub-region according to some embodiments of the present disclosure.

[0028] Figure 5 Schematically shows a flowchart of splicing a basic flight route to obtain a target UAV flight route according to some embodiments of the present disclosure.

[0029] Figure 6 Schematically shows a schematic diagram of a UAV route layout device for complex terrains according to some embodiments of the present disclosure.

[0030] Figure 7 Schematically shows a schematic diagram of a computer system of an electronic device according to some embodiments of the present disclosure.

[0031] Figure 8 Schematically shows a schematic diagram of a computer-readable storage medium according to some embodiments of the present disclosure.

[0032] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts. Detailed Description of the Embodiments

[0033] Here, exemplary embodiments will be described in detail, and examples thereof are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification. On the contrary, they are merely examples of devices and methods consistent with some aspects of this specification as detailed in the appended claims.

[0034] In addition, the drawings are only schematic diagrams and are not necessarily drawn to scale. The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0035] In the present exemplary embodiment, first, a method for arranging drone flight routes for complex terrains is provided. The method for arranging drone flight routes for complex terrains can be applied to a terminal device or a server. The present exemplary embodiment does not make any special limitations in this regard. Subsequently, the case where the server executes this method will be taken as an example for description. Figure 1 Schematically shows a flowchart of a method for arranging drone flight routes for complex terrains according to some embodiments of the present disclosure. Refer to Figure 1 As shown, the method for arranging drone flight routes for complex terrains may include the following steps:

[0036] Step S110, obtaining terrain observation data of the terrain area to be surveyed, where the terrain observation data includes a digital elevation model, a terrain orthophoto, geological disaster data, and remote sensing multispectral data;

[0037] Step S120, performing a preliminary regional division on the terrain area to be surveyed to determine a plurality of regional units;

[0038] Step S130, clustering each of the regional units according to the terrain observation data to obtain surveyed sub-areas;

[0039] Step S140, determining the survey priority corresponding to each of the surveyed sub-areas, and matching the basic flight route to be adopted by the drone in the surveyed sub-areas according to the survey priority;

[0040] Step S150, splicing different types of basic flight routes corresponding to each of the surveyed sub-areas, and optimizing and adjusting the spliced basic flight routes to obtain a target drone flight route, so as to perform terrain survey on the terrain area to be surveyed through the target drone flight route.

[0041] According to the UAV route layout method for complex terrain in this exemplary embodiment, on the one hand, by acquiring multi-source terrain observation data of the terrain area to be surveyed, rich terrain information support can be provided for UAV route layout, enabling the route layout to be finely adjusted based on terrain features, avoiding survey blind spots caused by a single data source, and thus improving the integrity and accuracy of aerial survey data; by initially dividing the terrain area to be surveyed and further clustering each regional unit based on terrain observation data, the UAV route layout can implement differential planning for different terrain types, reducing problems such as route duplication or omission caused by large differences in terrain features. Especially in weakly textured areas such as high mountainous areas or canyon reservoir slopes (such as ice and snow covered areas or sparse vegetation areas), through the analysis of the feature vectors of regional units, accurate division of regional types can be achieved, enabling route planning to adapt to different geomorphic features and improving the reliability of surveying and mapping data and the accuracy of subsequent 3D reconstruction; on the other hand, combining the division results of the survey sub-areas, further determining the survey priorities of each survey sub-area, and matching corresponding basic flight routes, the flight resources of the UAV can be reasonably allocated according to terrain complexity, avoiding excessive flight time occupied by low-priority areas, and thus improving the overall efficiency of aerial survey work. At the same time, in the face of complex obstacle distributions in high mountainous areas, the division of different survey priorities can provide a basis for obstacle avoidance strategies, enabling the UAV to reasonably avoid obstacles in a complex environment and reducing the risks of mission interruption and equipment loss caused by unreasonable flight path planning; on the third hand, in the process of route splicing and optimization adjustment, by splicing different types of basic flight routes of each survey sub-area and making adaptive adjustments in the splicing process by combining regional characteristic parameters, the flight redundancy at the route splicing can be effectively reduced, the route coverage range can be optimized, and the accuracy of aerial survey data can be further improved. At the same time, through path smoothing processing and the verification of heading overlap and side overlap, the route planning has better adaptability in complex terrain areas, ensuring the high accuracy and consistency of surveying and mapping results.

[0042] Next, the UAV route layout method for complex terrain in this exemplary embodiment will be further described.

[0043] In step S110, terrain observation data of the terrain area to be surveyed is acquired, and the terrain observation data includes digital elevation model, terrain orthophoto image, geological disaster data, and remote sensing multispectral data.

[0044] In an exemplary embodiment of the present disclosure, the terrain area to be surveyed refers to the target area that needs to be surveyed by an unmanned aerial vehicle for terrain survey. This area can be a region with complex terrain features. For example, the terrain area to be surveyed can be a plateau mountain area, a canyon reservoir slope, a hilly area, etc. The determination of the terrain area to be surveyed is usually based on surveying and mapping requirements, engineering project planning, and environmental monitoring requirements. The boundary of the terrain area to be surveyed can be set through Geographic Information System (GIS) data. For example, it can be delimited based on administrative region boundaries, natural terrain dividing lines, or existing surveyed areas. Of course, the terrain area to be surveyed can also be determined through remote sensing image analysis or ground survey means. This exemplary embodiment does not make special limitations on the determination method of the terrain area to be surveyed.

[0045] Terrain observation data refers to a data set used to characterize the terrain features of the terrain area to be surveyed. For example, terrain observation data can include, but is not limited to, Digital Elevation Model (DEM), terrain orthophoto image, geological disaster data, and remote sensing multispectral data. The acquisition method of terrain observation data can be based on remote sensing technology, ground measurement technology, and integration of historical data. For example, a digital elevation model can obtain high-precision point cloud data through airborne Light Detection And Ranging (LiDAR) scanning by an unmanned aerial vehicle, while a terrain orthophoto image can be obtained by geometric correction of aerial photography or satellite remote sensing images. Of course, geological disaster data and remote sensing multispectral data in terrain observation data can also be obtained from surveying and mapping departments or geographic information databases. This exemplary embodiment does not make special limitations on the data sources of terrain observation data. In an optional implementation manner, terrain observation data can be managed in a hierarchical storage manner to improve data processing efficiency. For example, it can be hierarchically stored according to regions, time, or data types. Of course, a cloud data storage and processing mode can also be adopted for remote access and sharing. This exemplary embodiment does not make special limitations on the storage method of terrain observation data.

[0046] Digital elevation models can collect high-precision point clouds through airborne lidar technology and perform spatial positioning in combination with the Global Positioning System (GPS) and Inertial Measurement Unit (IMU). Of course, digital elevation models can also be achieved using stereo pair photogrammetry technology or Synthetic Aperture Radar (SAR) mapping technology. This exemplary embodiment does not make special limitations on the construction method of the digital elevation model. Optionally, the data accuracy of the digital elevation model can be adjusted according to survey requirements. For example, the data accuracy of the digital elevation model can be set to a high-precision mode with a 1-meter resolution, suitable for fine mapping requirements. Of course, a medium-precision mode with a 10-meter resolution can also be used, suitable for large-scale terrain surveys. This exemplary embodiment does not make special limitations on the setting method of the data accuracy.

[0047] Terrain orthophotos refer to aerial or satellite images that have undergone orthorectification processing, eliminating image distortion caused by terrain undulation and giving the images a unified scale and geographical coordinates. For example, terrain orthophotos can be obtained by a drone carrying a high-resolution optical camera to collect multi-angle images at a preset flight altitude and perform geometric correction in combination with Ground Control Points (GCPs). Of course, terrain orthophotos can also be obtained through satellite remote sensing, such as using multi-spectral or panchromatic images provided by high-resolution commercial satellites and performing ortho-processing in combination with terrain data. This exemplary embodiment does not make special limitations on the acquisition method of terrain orthophotos. In an optional implementation, terrain orthophotos can adopt a multi-band image synthesis method to enhance the recognition ability of terrain features. For example, visible light, near-infrared, and short-wave infrared images can be combined for fusion to improve the recognition ability of vegetation and geological structures. Of course, single-band panchromatic images can also be used to improve the spatial resolution. This exemplary embodiment does not make special limitations on the image band combination method.

[0048] Geological disaster data refers to geological risk information related to the surveyed area, which may include historical records and predicted risks of disasters such as landslides, collapses, debris flows, etc. For example, geological disaster data can be obtained through historical disaster distribution maps provided by geological departments and combined with geographic information systems for spatial analysis. Of course, geological disaster data can also be used to monitor surface deformation in real time through remote sensing technology. For example, synthetic aperture radar interferometry technology can be used to monitor minute displacements on the surface. The embodiment of this example does not make special limitations on the acquisition method of geological disaster data. In an alternative embodiment, geological disaster data can be dynamically analyzed in combination with meteorological data to predict potential disaster risks. For example, the landslide risk level can be calculated based on historical rainfall, soil moisture, and slope. Of course, the probability of debris flow occurrence can also be predicted based on surface temperature and rainfall trends. The embodiment of this example does not make special limitations on the disaster risk assessment method.

[0049] Remote sensing multispectral data refers to surface reflection information in different bands obtained through multispectral sensors, which is used to identify the composition and distribution of surface materials. For example, remote sensing multispectral data can be obtained by a drone carrying a multispectral camera for multi-angle observations at different flight altitudes, and image processing software can be used to extract features from the multispectral images. Of course, remote sensing multispectral data can also be obtained through hyperspectral imaging technology, such as using satellite remote sensing or airborne remote sensing equipment for large-scale multi-band image acquisition. The embodiment of this example does not make special limitations on the acquisition method of remote sensing multispectral data.

[0050] In step S120, a preliminary regional division is performed on the terrain area to be surveyed, and multiple regional units are determined.

[0051] In an exemplary embodiment of the present disclosure, the preliminary regional division refers to a process of spatially partitioning the area to be surveyed according to terrain observation data to optimize aerial survey planning. For example, the regional division can adopt a method based on a regular grid to divide the area into equal-sized units. Of course, the regional division can also be achieved through adaptive grid segmentation technology, which dynamically adjusts the unit size according to the terrain complexity. The embodiment of this example does not make special limitations on the method of regional division.

[0052] A regional unit refers to the basic measurement unit obtained after preliminary regional division of the terrain area to be surveyed. The division principle is usually based on the similarity of terrain features, the requirements of survey accuracy, and the consideration of operation efficiency. The regional unit can be obtained by using the regular grid method to divide the terrain area to be surveyed into sub-regions of equal size according to a preset resolution. For example, it can be divided according to a fixed geographical coordinate grid. Of course, the size of the regional unit can also be automatically adjusted according to the change of terrain slope to adapt to complex terrain environments. In this exemplary embodiment, there is no special limitation on the division method of the regional unit. In an alternative embodiment, the division of the regional unit can be combined with the automatic segmentation algorithm of remote sensing images to improve the accuracy and rationality of the division. For example, regions with similar terrain features can be divided into the same unit based on the clustering analysis algorithm. Of course, the unit boundary can also be manually adjusted to meet special survey requirements. In this exemplary embodiment, there is no special limitation on the adjustment method of the regional unit.

[0053] In step S130, each of the regional units is clustered according to the terrain observation data to obtain survey sub-regions.

[0054] In an exemplary embodiment of the present disclosure, regional clustering refers to the process of merging regional units with similar features based on the similarity of terrain features for differential aerial survey planning. Its principle is to classify and group terrain features through a machine learning algorithm to form several survey sub-regions with different terrain features. For example, the K-means algorithm can be used for clustering to calculate the similarity of regional units according to terrain feature vectors. Of course, the density clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, DBSCAN) can also be used for clustering to automatically identify the clustering boundary through spatial density. In this exemplary embodiment, there is no special limitation on the clustering method.

[0055] A survey sub-region refers to a set of regions with similar terrain features obtained by clustering the similarity of regional units. The formation of the survey sub-region is based on the terrain recognition feature vectors extracted from the terrain observation data, which may include terrain slope features, terrain undulation features, surface coverage and occlusion features, and geological disaster distribution features.

[0056] In step S140, the survey priority corresponding to each of the survey sub-regions is determined, and the basic flight route to be adopted by the unmanned aerial vehicle in the survey sub-region is matched according to the survey priority.

[0057] In an exemplary embodiment of the present disclosure, the survey priority refers to the priority data for allocating different aerial survey resources to different regions according to factors such as terrain complexity, risk level, data requirements, etc. Its principle is to rank the importance of regions based on terrain features and, in combination with the aerial survey capabilities of unmanned aerial vehicles (UAVs), reasonably allocate flight tasks. For example, the survey priority can be set according to parameters such as terrain steepness and geological disaster risk index. Of course, the survey priority can also be adjusted based on the requirements of aerial survey accuracy and data coverage. This exemplary embodiment does not make any special limitations on the priority setting method. For example, the survey priority can be divided into highest, high, relatively high, medium, and low to adapt to different survey requirements. Of course, the specific division of the survey priority can be customized according to the actual situation, and this embodiment does not make any special limitations on this.

[0058] The basic flight route refers to the initial UAV flight path pre-determined according to different terrain features in the survey sub-regions of different terrain types. The design of the basic flight route takes into account parameters such as heading, side overlap, flight altitude, and shooting angle. Different terrain types or survey sub-regions with different survey priorities adopt different basic flight routes. The design of the basic flight route can be based on terrain complexity and survey accuracy requirements. For example, in flat areas (low survey priority), a parallel sparse route layout method is adopted, while in steep mountainous areas (high survey priority), a winding dense route layout method along the contour lines is adopted. Of course, the basic flight route can also be adjusted according to meteorological conditions to ensure flight safety. This exemplary embodiment does not make any special limitations on the layout method of the basic flight route.

[0059] In step S150, different types of basic flight routes corresponding to each of the survey sub-regions are spliced, and the spliced basic flight route is optimized and adjusted to obtain the target UAV flight route, so as to conduct terrain survey on the terrain area to be surveyed through the target UAV flight route.

[0060] In an exemplary embodiment of the present disclosure, route splicing refers to the processing process of merging the basic flight routes of different survey sub-regions into a complete flight path. Route splicing can be based on the regional adjacency relationship and priority order to sequentially connect route nodes. For example, the shortest path algorithm can be used for route splicing to ensure the minimization of the flight path. Route splicing can also optimize the route turning points through a route smoothing algorithm (such as a Bezier curve). Of course, the obstacle avoidance algorithm can also be combined during the splicing process to ensure the safety and stability of UAV flight. This exemplary embodiment does not make any special limitations on the splicing method.

[0061] The target UAV flight route refers to the final flight path obtained by splicing and optimizing the basic flight routes of each survey sub-region. This path ensures that the UAV covers all areas to be surveyed during the survey mission and meets the data collection accuracy requirements. The generation of the target UAV flight route can be based on a route splicing algorithm to seamlessly connect the flight paths of adjacent regions. For example, the shortest path algorithm can be used to optimize the route connection order. Of course, the target UAV flight route can also be smoothed to reduce track deviation and improve flight stability. The optimization method of the target UAV flight route in this exemplary embodiment is not specifically limited. It can be understood that the target UAV flight route can be dynamically adjusted according to real-time flight data to cope with complex terrain changes. For example, the flight height can be automatically adjusted in high wind speed areas to maintain stability. Of course, the flight route can also be monitored in real time through the ground station and manually intervened. The adjustment method of the flight route in this exemplary embodiment is not specifically limited.

[0062] By determining the survey priorities of the survey sub-regions and matching the corresponding basic flight routes, and then splicing the basic flight routes to obtain the target UAV flight route, the layout efficiency of the target UAV flight route can be effectively improved. At the same time, the flight resources of the UAV can be reasonably allocated according to the terrain complexity, avoiding excessive flight time occupied by low-priority areas, thereby improving the overall efficiency of the aerial survey work. At the same time, in the face of the complex obstacle distribution in plateau mountainous areas, the division of different survey priorities can provide a basis for the obstacle avoidance strategy, enabling the UAV to reasonably avoid obstacles in a complex environment, reducing the risk of mission interruption and equipment loss caused by unreasonable flight path planning, and improving the accuracy, accuracy and reliability of the aerial survey results.

[0063] The content in steps S110 to S150 will be described in detail below.

[0064] In an exemplary embodiment of the present disclosure, step S130 of clustering each regional unit according to the terrain observation data to obtain survey sub-regions can be implemented through the steps in Figure 2 As shown in Figure 2 , it specifically may include:

[0065] Step S210, determining the terrain recognition feature vector corresponding to each regional unit according to the terrain observation data;

[0066] Step S220, performing similarity clustering on the regional units based on the terrain recognition feature vector to obtain survey sub-regions with different terrain types.

[0067] Among them, the terrain recognition feature vector is a feature representation of regional units in different dimensions of terrain observation data, which is used for subsequent similarity clustering of regional units. The terrain recognition feature vector can be obtained by extracting and quantifying the core information in the terrain observation data into a unified feature representation through multi-dimensional data analysis methods, so that the similarity between regional units can be evaluated mathematically in the subsequent processing process. The construction of the terrain recognition feature vector depends on key parameters in the terrain observation data, such as terrain slope features, terrain undulation features, surface coverage and occlusion features, and geological disaster distribution features, and data normalization processing is carried out based on these features to ensure that data from different sources are calculated at the same scale. For example, the terrain recognition feature vector can extract the corresponding terrain slope features and terrain undulation features of each regional unit from the digital elevation model and terrain orthophoto image, calculate the trend of regional slope change through the boundary extraction algorithm, and use the filtering method to reduce data noise. Of course, the terrain slope features can also analyze the terrain spectrum features through Fourier transform to reflect the detailed features of surface undulation. In this exemplary embodiment, no special limitation is imposed on the feature extraction method.

[0068] The terrain recognition feature vector can also adopt methods such as Principal Components Analysis (PCA) or Factor Analysis (FA) for feature dimensionality reduction to reduce redundant data and improve calculation efficiency. For example, in mountainous areas, the focus can be on slope and elevation differences, while in flat areas, surface coverage features can be emphasized. Of course, deep learning feature extraction methods can also be used to automatically identify the most discriminative terrain features. In this exemplary embodiment, no special limitation is imposed on the feature dimensionality reduction method.

[0069] Similarity clustering refers to classifying regional units with similar terrain features into the same survey sub-region based on the terrain recognition feature vector of the regional units through a specific clustering algorithm for targeted data collection and flight path planning. Similarity clustering can form survey sub-regions with high cohesion and low separation by calculating the feature vector distance or similarity index between regional units. For example, the similarity between each of the regional units can be calculated using Euclidean distance or cosine similarity in a multi-dimensional feature space, and a clustering algorithm can be used to process the similarity matrix to form survey sub-regions. Taking the K-means clustering algorithm as an example, K clustering centers can be initialized, and then the regional units can be classified into the nearest clustering center according to the minimum distance principle. Of course, the positions of the clustering centers can be updated iteratively to minimize the within-cluster variance. In this exemplary embodiment, no special limitation is imposed on the selection of the clustering algorithm.

[0070] Optionally, the similarity clustering of regional units can also be achieved through the density clustering algorithm. Specifically, survey sub-regions can be formed in the feature space based on the density threshold. This method is relatively flexible in terrain boundary recognition and is applicable to mountainous terrains with irregular boundaries. Of course, the hierarchical clustering method can also be used to refine the regional division layer by layer at different scales for refined survey. The use of the clustering method in this exemplary embodiment is not specifically limited.

[0071] Through the construction of the terrain recognition feature vector, the terrain attributes of regional units can be accurately expressed, providing a scientific basis for the subsequent clustering process and improving the accuracy and consistency of the survey area division; through similarity clustering, regions with similar terrain features can be automatically identified, improving the pertinence of the UAV flight path layout, reducing duplicate coverage, and enhancing the effectiveness and accuracy of data collection.

[0072] Optionally, it can be achieved through Figure 3 the steps in to determine the terrain recognition feature vector corresponding to each regional unit according to the terrain observation data in step S210. Referring to Figure 3 shown, it can specifically include:

[0073] Step S310, extracting the terrain slope feature and terrain undulation feature corresponding to each regional unit from the digital elevation model and the terrain orthophoto;

[0074] Step S320, extracting the surface coverage and occlusion feature corresponding to each regional unit from the terrain orthophoto and the remote sensing multispectral data;

[0075] Step S330, determining the geological hazard distribution feature corresponding to the regional unit according to the geological hazard data;

[0076] Step S340, constructing a terrain recognition feature vector based on the terrain slope feature, the terrain undulation feature, the surface coverage and occlusion feature, and the geological hazard distribution feature.

[0077] Among them, the terrain slope feature refers to the change of the surface slope within a regional unit, which is used to reflect the steepness of the terrain. It can be calculated based on the elevation data in the digital elevation model and combined with the image slope feature points in the terrain orthoimage. By using the differential calculation method, the elevation change rate or slope feature of each point within the regional unit can be calculated, and the continuous slope distribution can be obtained through the spatial interpolation algorithm. For example, the gradient calculation method, the raster data based on the digital elevation model, the image slope feature points in the terrain orthoimage, the third-order finite difference method or the Sobel operator can be used to calculate the terrain slope matrix, that is, the terrain slope feature. Of course, the terrain slope feature can also estimate the average slope within the regional unit by combining the local plane fitting method with the digital elevation model and the terrain orthoimage to smooth the local terrain changes. This exemplary embodiment does not make special limitations on the slope calculation method. In some optional embodiments, the extraction of the terrain slope feature can be combined with the terrain classification algorithm to further identify steep slopes, gentle slopes and flat areas. For example, the regional unit can be divided into different slope levels based on the threshold classification method. Of course, a machine learning model can also be used to automatically extract the slope feature to improve the accuracy of feature recognition. This exemplary embodiment does not make special limitations on the classification method of the slope feature.

[0078] The terrain undulation feature refers to the degree of fluctuation of the terrain elevation within a regional unit, which reflects the complexity of the terrain undulation. It can be calculated based on the digital elevation model and the terrain orthoimage to calculate the elevation mean value, the maximum and minimum elevation difference and the elevation variation coefficient of the regional unit, so as to quantify the undulation degree of the terrain. For example, based on the moving window calculation method, a 3×3 or 5×5 window can be used to statistically analyze the elevation data or image texture within the regional unit to calculate the elevation standard deviation and the undulation index. Of course, the terrain undulation feature can also extract the terrain spectrum information through Fourier transform to distinguish gentle and severely undulating terrains. This exemplary embodiment does not make special limitations on the calculation method of the terrain undulation feature. The terrain undulation feature can also be comprehensively analyzed in combination with the slope feature to more accurately describe the complexity of the terrain. For example, the terrain roughness index (Terrain Ruggedness Index, TRI) can be used to measure the terrain complexity. Of course, the terrain surface curvature (TSC) can also be used for multi-scale analysis. This exemplary embodiment does not make special limitations on the calculation method of the undulation feature.

[0079] The surface coverage occlusion feature refers to the degree of occlusion of the aerial survey line of sight by surface objects within a regional unit, which affects the effectiveness of UAV image acquisition. Based on topographic orthophoto images and remote sensing multispectral data, the distribution, types, and heights of surface objects can be analyzed to identify occlusion factors that may affect aerial surveys. For example, based on image classification algorithms, such as using Support Vector Machine (SVM) or Random Forest (RF) classification models to classify surface features and extract the types and distributions of occluders. Of course, the surface coverage occlusion feature can also calculate the vegetation density through the Normalized Difference Vegetation Index (NDVI) to identify high-occlusion areas. The identification method of the occlusion feature in this exemplary embodiment is not specifically limited. In some alternative embodiments, the surface coverage occlusion feature can be combined with lidar point cloud data to obtain ground object height information for more accurately assessing the occlusion risk. For example, a point-by-point classification algorithm can be used to distinguish ground and non-ground points. Of course, it is also possible to automatically detect occlusion areas based on a deep learning model. This embodiment is not limited thereto.

[0080] The geological hazard distribution feature refers to the types of geological hazards that may exist within a regional unit and their spatial distribution, which is used to evaluate the safety of aerial survey operations. Based on geological hazard data, the occurrence frequency, distribution range, and risk level of geological hazards within the regional unit can be analyzed to provide a safety reference for UAV aerial survey planning. For example, the geological hazard database can be utilized, combined with remote sensing images to analyze the occurrence trend of geological hazards, and a geological model can be used to construct a regional risk map. Of course, the geological hazard distribution feature can also be comprehensively evaluated in combination with environmental factors such as rainfall and soil moisture to improve the accuracy of prediction. The geological hazard assessment method in this exemplary embodiment is not specifically limited. In some embodiments, the geological hazard distribution feature can be dynamically predicted through a machine learning model. For example, a Random Forest model can be used to combine historical hazard data for risk assessment, or a spatial interpolation method can be used to refine the prediction of the geological hazard distribution.

[0081] A terrain recognition feature vector can be constructed based on the terrain slope feature, terrain undulation feature, surface coverage occlusion feature, and geological hazard distribution feature. The terrain recognition feature vector uniformly represents various terrain features for subsequent similarity clustering processing.

[0082] Accurately characterize the slope changes within the region through terrain slope features, provide reliable slope information for the layout of UAV flight routes, and avoid uneven aerial survey coverage caused by excessive slopes; by extracting terrain undulation features, the UAV flight altitude strategy can be optimized to ensure the uniformity of route coverage and image quality; by accurately identifying surface occlusion features, the UAV flight altitude and route layout can be adjusted to ensure the integrity of data collection; by determining the distribution features of geological disasters, high-risk areas can be avoided during UAV aerial surveys to ensure flight safety; by constructing a terrain recognition feature vector, the accuracy of regional unit classification can be improved to ensure the rationality of the division of survey sub-regions.

[0083] Optionally, it can be achieved through the Figure 4 steps in to perform similarity clustering on the regional units based on the terrain recognition feature vector in step S220, obtaining survey sub-regions with different terrain types. Referring to Figure 4 as shown, it can specifically include:

[0084] Step S410, construct a multi-dimensional feature space based on the terrain slope feature, terrain undulation feature, surface coverage occlusion feature, and geological disaster distribution feature included in the terrain recognition feature vector;

[0085] Step S420, obtain the pre-determined density radius parameter and the minimum number of neighborhood points;

[0086] Step S430, calculate the similarity data between the terrain recognition feature vectors corresponding to each regional unit in the multi-dimensional feature space;

[0087] Step S440, mark the regional units with similarity data less than the density radius parameter and the number of neighborhood points greater than or equal to the minimum number of neighborhood points as core regions;

[0088] Step S450, cluster the adjacent core regions and the boundary regions associated with the core regions based on the terrain recognition feature vector to obtain survey sub-regions with different terrain types, where the terrain type is determined by the terrain recognition feature vector of the core region.

[0089] Among them, the multi-dimensional feature space refers to mapping each component element of the terrain recognition feature vector into a high-dimensional space for similarity analysis of regional units. The construction principle of the multi-dimensional feature space lies in mapping multiple feature variables into a unified coordinate system in a mathematical form, so that each regional unit can be compared and classified in the same feature space. Specifically, various features in the terrain recognition feature vector can be standardized to eliminate the scale differences between different features, and principal component analysis (PCA) can be used to reduce the dimension of the terrain recognition feature vector to reduce data redundancy. Of course, the multi-dimensional feature space can also adopt the method of weighted feature fusion, and different weights can be assigned according to the importance of each feature to form a feature space that better conforms to the terrain complexity. The present exemplary embodiment does not make special limitations on the setting method of feature weights.

[0090] In an optional implementation manner, the multi-dimensional feature space can be represented by an Euclidean space or a cosine space. For example, the Euclidean distance is used to measure the similarity of regional units in the Euclidean space. Of course, the cosine similarity can also be used for comparative analysis of feature vectors in the cosine space. The present exemplary embodiment does not make special limitations on the measurement method of the multi-dimensional feature space. By constructing the multi-dimensional feature space, the terrain features of regional units can be more accurately described, the accuracy of similarity analysis can be ensured, and the rationality of the division of the surveyed sub-regions can be improved.

[0091] The density radius parameter refers to the neighborhood range for determining the similarity of regional units in the multi-dimensional feature space, and this parameter directly affects the tightness of clustering; the minimum number of neighborhood points refers to the minimum number of data points that must be included within the set density radius to ensure the reliability of clustering. The density radius parameter and the minimum number of neighborhood points can be set for different terrain types by statistical analysis of historical data in combination with factors such as terrain complexity and data distribution density. Specifically, the multi-dimensional feature space can be initially scanned, and the K-distance Graph can be used to analyze the inflection point position to determine the appropriate density radius parameter, and the threshold of the minimum number of neighborhood points can be set based on experimental statistics. The density radius parameter can also be dynamically adjusted by an adaptive optimization algorithm to adapt to the terrain features of different regions. The present exemplary embodiment does not make special limitations on the parameter adjustment method. Of course, in some optional implementation manners, the density radius parameter can also be dynamically set according to the variation ranges of the terrain slope feature and the terrain undulation feature. For example, in a region with large terrain undulation, the density radius is increased to cover more similar units. Of course, in a region with large changes in surface coverage, the density radius can be reduced to improve the clustering accuracy. By reasonably setting the density radius and the minimum number of neighborhood points, the accuracy of clustering can be improved, and the reasonable division of the surveyed sub-regions can be ensured.

[0092] Similarity data refers to the result of quantifying the similarity of regional units in a multi-dimensional feature space through a mathematical measurement method. The principle of similarity calculation lies in evaluating the similarity between each regional unit through a specific mathematical model, thereby providing a basis for the subsequent clustering process. For example, the Euclidean distance can be used to represent the similarity data between the terrain recognition feature vectors corresponding to each regional unit, and the calculation can be performed through the following relational expression:

[0093] ;

[0094] where, can represent the similarity data between the terrain recognition feature vectors corresponding to each regional unit, and can respectively represent the eigenvalues on the th feature dimension of the terrain recognition feature vector, can represent the total number of feature dimensions of the terrain recognition feature vector.

[0095] In some optional embodiments, the similarity calculation can combine a weighted similarity index to assign different weights to different terrain features. For example, in areas with a high risk of geological disasters, the weight of geological disaster features can be increased. Of course, in areas with dense vegetation, the weight of surface coverage features can also be increased. This embodiment does not make special limitations on this. By accurately calculating the similarity data, the reliability of the clustering result can be improved, and the reasonable grouping of regional units can be ensured.

[0096] The core area refers to a regional unit with a sufficiently dense neighborhood of points in the multi-dimensional feature space, indicating that this area has a high terrain consistency. The core area marking can be based on the basic principles of density clustering, by detecting the local density of regional units to identify the core part of the surveyed sub-region. Specifically, the similarity data can be sorted, and the number of neighborhood points of the regional units with similarity data less than the density radius parameter can be calculated in sequence and compared with the minimum number of neighborhood points. The regional units that meet the conditions are marked as the core area; the marking of the core area can also be further optimized through a boundary detection algorithm to eliminate isolated points. This exemplary embodiment does not make special limitations on the identification method of the core area. Of course, the marking of the core area can be adjusted in combination with the regional morphological features. For example, in areas with a drastic change in slope, the density radius parameter of the core area can be increased. Of course, in flat areas, the density radius parameter can also be decreased to ensure the accurate coverage of key areas. This exemplary embodiment does not make special limitations on the adjustment method of the core area marking. By accurately marking the core area, the stability and scientific nature of the subsequent clustering result can be ensured, and the division accuracy of the sub-region can be improved.

[0097] Optionally, the density radius parameter and the minimum number of neighborhood points can be determined through the following steps, which can specifically include:

[0098] The density radius parameter can be determined according to the variance change range of the terrain slope feature and the terrain undulation feature in all terrain recognition feature vectors; the minimum number of neighborhood points can be determined according to the density degree of the surface coverage occlusion feature and the geological disaster distribution feature in all terrain recognition feature vectors.

[0099] The variance change range refers to the distribution of the terrain undulation degree reflected after statistical analysis of the terrain slope feature and the terrain undulation feature in each regional unit, which is used to provide a quantitative reference for the terrain complexity to assist in the reasonable setting of the density radius parameter. The variance change range can evaluate the dispersion degree of the terrain data by calculating the mean value and its deviation of the slope and undulation feature data, so as to provide data support for the subsequent parameter setting. The variance calculation can be based on the sliding window analysis, and different-sized windows are set to evaluate the local and overall change trends of the terrain slope feature and the terrain undulation feature. For example, a smaller window is used to finely evaluate the local slope change in steep mountainous areas. Of course, a larger window can be used in flat areas to obtain the global terrain trend. Through the analysis of the variance change range, the density radius parameter can be dynamically adjusted to make the division of regional units more in line with the terrain features and improve the clustering accuracy of the surveyed sub-regions.

[0100] The density degree refers to the spatial distribution density of surface coverages (such as vegetation, buildings) and geological disaster potential points in the area to be surveyed, which can be used to measure the distribution of obstacles in the surveyed area to optimize the setting of the number of neighborhood points in the clustering process. Spatial statistical analysis can be carried out on the surface coverage data and the geological disaster data to evaluate the feature point density per unit area, so as to provide a basis for setting the minimum number of neighborhood points. Specifically, the surface coverage occlusion feature can be classified, and remote sensing image classification algorithms (such as support vector machines, random forests) can be used to classify and label coverages such as vegetation, water bodies, and buildings, and calculate the spatial distribution density of each category. Of course, the density degree of the surface coverage occlusion feature can also be calculated based on the lidar point cloud data to describe the obstacle density more accurately. The density degree of the geological disaster distribution feature can be estimated by spatial interpolation methods. For example, the Kriging interpolation method can be used to interpolate the existing geological disaster data to generate a disaster density distribution map. Of course, machine learning models can also be combined to predict future potential disaster occurrence areas to dynamically adjust the minimum number of neighborhood points. Through the analysis of the density degree, the setting of the minimum number of neighborhood points can accurately reflect the spatial distribution characteristics of surface coverage and geological disasters, thus improving the scientificity and adaptability of the clustering results.

[0101] By combining the variance range of the terrain slope characteristics and terrain undulation characteristics with the density of the surface coverage occlusion characteristics and geological disaster distribution characteristics, optimize the parameter settings in the clustering process, so that the settings of the density radius parameter and the minimum number of neighborhood points can take into account both the complexity of the terrain characteristics and the distribution of surface characteristics, and comprehensively consider the mutual influence of various terrain and surface elements to ensure the rationality and scientific nature of the clustering parameters. Optionally, a multi-factor weight analysis model can also be constructed, and the influence weights of various characteristics can be calculated through a weighted comprehensive analysis method, which can be used as the basis for setting the density radius parameter and the minimum number of neighborhood points. Of course, parameter optimization can also be performed through intelligent optimization techniques such as genetic algorithms and particle swarm optimization algorithms to achieve the optimal parameter combination. The method for parameter optimization in this exemplary embodiment is not specifically limited. In an optional implementation manner, a sub-region parameter setting strategy can be adopted, that is, according to the differences in terrain and geological characteristics of the region, different density radius parameters and minimum number of neighborhood points are set respectively. For example, a smaller density radius and a higher number of neighborhood points are adopted for high-steep slope regions to ensure the survey accuracy. Of course, the density radius can be appropriately increased in flat regions to reduce the computational complexity. The method for setting regional parameters in this exemplary embodiment is not specifically limited. By combining multi-factor analysis and optimizing parameter settings, the accuracy of the clustering process is improved, and the division of the survey sub-regions is made more scientific and efficient.

[0102] By combining the terrain recognition feature vector to achieve similarity clustering of regional units, the terrain feature analysis can be further refined, making the classification of terrain data more accurate, and ensuring that factors such as slope, undulation, surface coverage occlusion characteristics, and geological disaster distribution characteristics are fully considered during regional division; by constructing a multi-dimensional feature space to quantify terrain features from multiple dimensions, the similarity between terrain units can be scientifically evaluated in a high-dimensional space; by the pre-determined density radius parameter and the minimum number of neighborhood points, it is ensured that the boundary range of each regional unit can be reasonably defined during the density clustering process, improving the division accuracy of the survey sub-regions; using the similarity data to perform density clustering on regional units can automatically identify terrain change areas, reduce overlapping operations in aerial surveys, and optimize flight path planning; the identification of the core area helps to determine the main survey area, improving the efficiency and integrity of data acquisition.

[0103] In an exemplary embodiment of the present disclosure, the determination of the survey priority corresponding to each survey sub-region can be achieved through the following steps, which can specifically include:

[0104] The mapping relationship between the pre-set terrain types and survey priorities can be obtained. The survey priority is set based on the survey difficulty of different terrain types, and the survey priority is in a direct proportional relationship with the survey difficulty. The survey priority corresponding to the survey sub-region is matched from the mapping relationship according to the terrain type corresponding to the survey sub-region.

[0105] Among them, the mapping relationship between terrain types and survey priorities refers to associating different terrain types with corresponding survey priorities, so as to allocate reasonable UAV survey resources according to terrain complexity during route planning. The classification of terrain types can be based on factors such as terrain geomorphic features, surface coverage conditions, and geological disaster risk levels. For example, the terrain can be classified into categories such as flat areas, hilly areas, steep slope areas, and high-risk geological disaster areas.

[0106] The principle for determining survey priorities is based on the positive correlation between terrain complexity and survey difficulty, that is, the more complex the terrain and the greater the mapping difficulty, the higher the survey priority. Specifically, a terrain type classification standard can be constructed. Based on the weighted scores of slope, undulation, surface coverage occlusion features, and geological disaster distribution features in the terrain recognition feature vector, a terrain type classification model is formed, and the classification standard is optimized using expert knowledge or historical mapping data; the mapping relationship of terrain types can also automatically extract the survey priority patterns of historical aerial survey data through big data analysis methods to generate a mapping relationship table; of course, the mapping relationship between terrain types and survey priorities can also be set using a multi-level scoring system. For example, using the terrain complexity index as an evaluation indicator, different terrains are divided into five or more priority levels. Of course, a fuzzy logic method can also be used to divide priorities according to the uncertainty of the survey area characteristics to improve the adaptability to complex terrains. The classification method of the mapping relationship in this exemplary embodiment is not specifically limited. For example, the mapping relationship between survey priorities and terrain types can be referred to as shown in Table 1:

[0107] Table 1 Mapping relationship between survey priorities and terrain types

[0108]

[0109] It can be understood that the mapping relationship between survey priorities and terrain types in Table 1 is only formed based on a single combination of features such as slope, undulation, surface coverage, and geological disaster risk. In the actual application process, more types of mapping combination relationships between survey priorities and terrain types can be combined according to the above features and can be customized according to the actual application situation. The mapping relationship in Table 1 should not cause special limitations to this embodiment.

[0110] The matching process refers to automatically or semi - automatically assigning appropriate survey priorities to each survey sub - area based on the established mapping relationship between terrain types and survey priorities, so as to ensure the reasonable allocation of UAV aerial survey resources. Priority matching can adopt a hierarchical processing strategy, that is, manual review and correction are carried out after preliminary matching to ensure accurate surveys in key areas. For example, the priority can be manually increased for geological disaster risk areas. Of course, an automated adjustment mechanism can also be adopted to dynamically correct the priority allocation scheme according to real - time mapping feedback data. This exemplary embodiment does not make special limitations on the matching method.

[0111] In the process of determining the survey priorities corresponding to each survey sub - area, based on the corresponding relationship between terrain types and survey difficulties, the survey priorities can be reasonably divided to ensure priority surveys in areas with high geological disaster risks and drastic terrain changes, avoiding problems such as missing or inaccurate mapping data caused by improper flight line layout; the division of survey priorities helps to reasonably allocate UAV aerial survey resources, ensuring high - precision aerial surveys in high - priority areas with limited time and flight resources, and improving the scientific nature of the overall aerial survey task; by obtaining the mapping relationship between terrain types and survey priorities, a reasonable aerial survey mode can be automatically matched according to terrain complexity, avoiding waste of resources and missed surveys caused by human intervention.

[0112] In an exemplary embodiment of the present disclosure, the basic flight lines can be obtained by adopting different flight line layout strategies according to the terrain characteristics of different terrain types. For example, the flight line layout strategy of the basic flight lines can be to adopt a parallel sparse flight line layout in flat areas (low survey priority), while in steep mountainous areas (high survey priority), a meandering dense flight line layout along the contour lines is adopted. Specifically, different flight line layout strategies can be customized according to the terrain characteristics of different terrain types. This embodiment is not limited thereto.

[0113] It can be achieved through Figure 5 the steps in to splice different types of basic flight lines corresponding to each survey sub - area, and optimize and adjust the spliced basic flight lines to obtain the target UAV flight line. Referring to Figure 5 as shown, it specifically can include:

[0114] Step S510, determining the regional parameters corresponding to the survey sub - area, and adaptively adjusting the basic flight line according to the regional parameters to obtain the adapted basic flight lines of each survey sub - area;

[0115] Step S520, according to the survey priorities and spatial position relationships of each survey sub - area, splicing the starting points and ending points of different adapted basic flight lines head - to - tail in sequence to obtain the spliced basic flight line;

[0116] In step S530, a Bessel curve is used to perform path smoothing on the spliced basic flight route, and the heading overlap degree and side overlap degree of the smoothed basic flight route are verified to obtain the target UAV flight route.

[0117] Among them, the regional parameter refers to the key index used to describe the geometric characteristics and spatial attributes of the surveyed sub-region. For example, the regional parameter can include the area size, shape, dimensions, etc. of the region to ensure that the basic flight route can cover the entire surveyed area and adapt to its specific spatial layout. The regional parameter can analyze the geometric morphological characteristics of the surveyed sub-region through geographic information system (GIS) analysis or remote sensing data processing, and combine the flight mission requirements to determine the optimal aerial survey layout. Specifically, the boundary of the surveyed sub-region can be vectorized to calculate the area size of the region, and the minimum bounding rectangle method can be used to determine the length and width of the region. At the same time, the complexity of the region can be analyzed using shape factors, such as calculating the aspect ratio, perimeter-area ratio, etc. to evaluate the regularity of the region. Of course, the determination of the regional parameter can also be through manual input to manually correct the parameters of the key region. The present exemplary embodiment does not make special limitations on the determination method of the regional parameter. In some embodiments, the regional parameter can be optimized in combination with the UAV flight ability. For example, the flight line spacing can be appropriately increased in a large area to improve the aerial survey efficiency, and the flight line direction can be adaptively adjusted in a region with a complex shape. It can also dynamically adjust the regional parameter in combination with real-time environmental data to adapt to the changes in the actual survey conditions. Through precise setting of the regional parameter, the flight line layout is made more reasonable, avoiding aerial survey blind spots and optimizing the UAV operation efficiency.

[0118] Flight line splicing refers to sequentially connecting the starting points and ending points of the basic flight routes of each surveyed sub-region based on the survey priority and spatial position relationship of the surveyed sub-region according to the principle of the optimal path to ensure the continuity and efficiency of the aerial survey operation. The greedy algorithm can be used for splicing. The flight line splicing order is sequentially selected in the order from high to low according to the priority, and the path optimization algorithm (such as A* or Dijkstra algorithm) is used to calculate the shortest connection path. Of course, the flight line splicing can also be manually intervened to manually adjust the splicing order in the key region to ensure that the important region is surveyed first.

[0119] It can be understood that the flight line splicing can also be adjusted in combination with real-time flight data. For example, when it is found during the flight that a certain region cannot complete the aerial survey due to environmental changes, the splicing order can be automatically re-planned. Of course, it can also be manually adjusted in combination with the remote instructions of the ground station to cope with emergencies. The present exemplary embodiment does not make special limitations on the adjustment method of the flight line splicing.

[0120] Path smoothing refers to the curve optimization of the spliced flight path to reduce the turning points of the flight path and improve the flight stability and energy-saving efficiency of the UAV. The Bezier curve can adjust the flight path parameters of the spliced basic flight path through control points, enabling the basic flight path to form a smooth curve in space, reducing sharp turns during UAV flight, lowering energy consumption, and enhancing the endurance of aerial survey. Specifically, control points can be inserted at the turning points of the flight path, and smoothing calculations can be performed using cubic or quintic Bezier curves to ensure the continuity of the flight path. Of course, path smoothing can also be combined with B-spline curves to further optimize the smoothness of the flight path. This exemplary embodiment does not make special limitations on the path smoothing algorithm.

[0121] In some alternative embodiments, path smoothing can also be optimized in combination with the dynamic model of the UAV. For example, the curve parameters can be adjusted considering the minimum turning radius of the UAV. Of course, the smoothness can also be dynamically adjusted according to the flight speed to reduce data distortion caused by overly large curves at high speeds. This exemplary embodiment does not make special limitations on the optimization method of the smoothing parameters.

[0122] The forward overlap degree refers to the overlapping degree of adjacent captured images in the flight direction of the UAV, usually expressed as a percentage. Its purpose is to ensure that when the UAV continuously captures images on the same flight path, there is a sufficient overlapping area (such as 70%) between the front and rear images for image stitching, 3D modeling, and subsequent image matching processing. The forward overlap degree can determine the overlapping ratio of images in the flight direction based on the UAV flight speed, camera shooting interval, ground resolution, and image coverage range to ensure that each target area is captured and covered at least twice or multiple times. The ground projection length of a single image can be calculated using the focal length of the UAV camera, ground resolution, shooting frequency, and flight speed, and the forward overlap degree can be calculated through the following formula:

[0123] Forward overlap degree = [1 - (UAV flight distance × shooting interval) / image ground projection length] × 100%

[0124] Among them, the image ground projection length refers to the coverage range of a single image captured by the UAV camera on the ground; the calculation of the forward overlap degree can also consider the terrain undulation and adopt a variable interval shooting strategy to adapt to different terrain changes. Of course, the forward overlap degree can also be dynamically adjusted in combination with real-time flight data. For example, in areas with strong winds, the shooting interval can be shortened to increase the overlap rate and ensure image quality, or the overlap degree can be appropriately reduced in flat areas to reduce data redundancy and improve operation efficiency. This exemplary embodiment does not make special limitations on the adjustment method of the forward overlap degree.

[0125] The side overlap degree refers to the image overlap degree in the horizontal direction between adjacent flight lines, usually expressed as a percentage. Its purpose is to ensure the complete coverage of the surveyed area and reduce the missing of surveying and mapping data caused by terrain undulation, UAV attitude change or environmental factors. The side overlap degree can be calculated by calculating the ground coverage width between adjacent flight lines and comparing and analyzing parameters such as the camera field of view angle and flight altitude to determine the overlap ratio of aerial survey images in the horizontal direction.

[0126] The focal length, resolution and flight altitude of the UAV aerial camera can be used to calculate the ground projection width of a single image. Based on the preset side overlap degree requirement (such as 60%-80%), the optimal distance between adjacent flight lines can be calculated to ensure sufficient overlap in the edge area of the image. The calculation of the side overlap degree can also be combined with terrain elevation difference data and adjusted in sections to adapt to terrain changes. This exemplary embodiment does not make special limitations on the calculation method of the side overlap degree. In some embodiments, the side overlap degree can also be combined with an automatic optimization algorithm to dynamically adjust the flight line distance based on the regional terrain characteristics before the UAV aerial survey task is executed. For example, the overlap degree can be reduced in flat terrain areas to improve operation efficiency, while the overlap degree can be increased in areas with large terrain undulations to ensure the image stitching quality. Of course, real-time flight adjustment technology can also be used to adjust the side overlap degree according to the real-time attitude of the UAV. This exemplary embodiment does not make special limitations on the adjustment method of the side overlap degree.

[0127] In this embodiment, the layout strategy of the basic flight lines enables the UAV to adopt targeted flight line layouts in different surveyed sub-areas, thereby improving the surveying and mapping accuracy and data coverage integrity; the basic flight lines are adaptively adjusted according to the regional parameters corresponding to the surveyed sub-areas, so that the flight line parameters can be flexibly adjusted under different terrain characteristics to ensure the aerial survey accuracy. The determination of the regional parameters enables the aerial survey task to be optimally adapted to various regional characteristics such as area size, shape and dimensions, improving the efficiency and data integrity of the aerial survey. Through flight line stitching and optimization adjustment, the flight range of the UAV can be effectively reduced, energy consumption can be reduced, and the efficiency of the surveying and mapping operation can be improved.

[0128] By using Bessel curves to perform path smoothing on the spliced basic flight route, the inflection points of the route can be reduced, the flight stability of the UAV can be improved, image acquisition distortion or data loss caused by sudden route changes can be avoided, and at the same time, the endurance of the UAV can be enhanced; by verifying the heading overlap and side overlap, sufficient overlap of image data can be ensured, the requirements of 3D reconstruction can be met, and the integrity and consistency of data acquisition can be optimized. Adjusting the side overlap helps ensure the lateral splicing accuracy of aerial survey images, while optimizing the heading overlap can ensure a reasonable time interval for data acquisition, reduce data loss, and improve surveying accuracy; the finally generated target UAV flight route can best adapt to complex terrain features, meet the comprehensiveness and high-precision requirements of aerial survey data, and ensure the efficient implementation of the survey task.

[0129] It should be noted that although the steps of the methods in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be executed in that specific order, or that all the steps shown must be executed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc.

[0130] In addition, in the present exemplary embodiment, a UAV route layout device for complex terrain is also provided. Referring to Figure 6 as shown, the UAV route layout device 600 for complex terrain includes: a terrain observation module 610, a region division module 620, a region clustering module 630, a flight route matching module 640, and a flight route splicing module 650. Among them:

[0131] The terrain observation module 610 is used to obtain terrain observation data of the terrain area to be surveyed, and the terrain observation data includes digital elevation model, terrain orthophoto image, geological disaster data, and remote sensing multispectral data;

[0132] The region division module 620 is used to perform preliminary region division on the terrain area to be surveyed to determine multiple region units;

[0133] The region clustering module 630 is used to cluster each of the region units according to the terrain observation data to obtain surveyed sub-regions;

[0134] The flight route matching module 640 is used to determine the survey priorities corresponding to each of the surveyed sub-regions, and match the basic flight routes to be adopted by the UAV in the surveyed sub-regions according to the survey priorities;

[0135] The flight route stitching module 650 is configured to stitch different types of basic flight routes corresponding to each of the surveyed sub-regions, and optimize and adjust the stitched basic flight routes to obtain a target UAV flight route, so as to perform topographic survey on the terrain area to be surveyed through the target UAV flight route.

[0136] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the region clustering module 630 is configured to: determine a terrain recognition feature vector corresponding to each of the region units according to the terrain observation data; perform similarity clustering on the region units based on the terrain recognition feature vector to obtain surveyed sub-regions with different terrain types.

[0137] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the region clustering module 630 is configured to: extract a terrain slope feature and a terrain undulation feature corresponding to each of the region units from the digital elevation model and the terrain orthoimage; extract a surface coverage occlusion feature corresponding to each of the region units from the terrain orthoimage and the remote sensing multispectral data; determine a geological hazard distribution feature corresponding to the region unit according to the geological hazard data; construct a terrain recognition feature vector based on the terrain slope feature, the terrain undulation feature, the surface coverage occlusion feature, and the geological hazard distribution feature.

[0138] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the region clustering module 630 is configured to: construct a multi-dimensional feature space based on the terrain slope feature, the terrain undulation feature, the surface coverage occlusion feature, and the geological hazard distribution feature included in the terrain recognition feature vector; obtain a pre-determined density radius parameter and a minimum number of neighborhood points; calculate similarity data between the terrain recognition feature vectors corresponding to each of the region units in the multi-dimensional feature space; mark the region units with similarity data less than the density radius parameter and the number of neighborhood points greater than or equal to the minimum number of neighborhood points as core regions; cluster adjacent core regions and boundary regions associated with the core regions based on the terrain recognition feature vector to obtain surveyed sub-regions with different terrain types, where the terrain type is determined by the terrain recognition feature vector of the core region.

[0139] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the UAV route layout device 600 for complex terrain further includes a clustering parameter determination unit, which is configured to: determine the density radius parameter according to the variance change range of the terrain slope feature and the terrain undulation feature in all the terrain recognition feature vectors; determine the minimum number of neighborhood points according to the density of the surface coverage occlusion feature and the geological hazard distribution feature in all the terrain recognition feature vectors.

[0140] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the flight route matching module 640 is configured to: obtain the mapping relationship between the preset terrain types and the survey priorities, where the survey priorities are set based on the survey difficulties of different terrain types, and the survey priorities are in a direct proportional relationship with the survey difficulties; match the survey priority corresponding to the survey sub-region from the mapping relationship according to the terrain type corresponding to the survey sub-region.

[0141] In some exemplary embodiments of the present disclosure, based on the foregoing solution, the basic flight route is obtained by adopting different route layout strategies according to the terrain characteristics of different terrain types; the flight route splicing module 650 is configured to: determine the regional parameters corresponding to the survey sub-region, and adaptively adjust the basic flight route according to the regional parameters to obtain the adapted basic flight routes of each survey sub-region; splice the starting points and ending points of different adapted basic flight routes in sequence from head to tail according to the survey priorities and spatial position relationships of each survey sub-region to obtain the spliced basic flight route; perform path smoothing processing on the spliced basic flight route by using a Bessel curve, and perform heading overlap degree and side overlap degree verification on the smoothed basic flight route to obtain the target UAV flight route.

[0142] The specific details of each module of the above UAV route layout device for complex terrain have been described in detail in the corresponding UAV route layout method for complex terrain, so they will not be repeated here.

[0143] It should be noted that although several modules or units of the UAV route layout device for complex terrain are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0144] In addition, in the exemplary embodiments of the present disclosure, an electronic device capable of implementing the above UAV route layout method for complex terrain is also provided.

[0145] Those skilled in the art of the relevant technical field can understand that various aspects of the present disclosure can be implemented as a system, a method, or a program product. Therefore, various aspects of the present disclosure can be specifically implemented in the following forms, namely: a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as "circuitry", "module", or "system" here.

[0146] Refer to the followingFigure 7 to describe an electronic device 700 according to such an embodiment of the present disclosure. Figure 7 The illustrated electronic device 700 is merely an example and should not impose any limitation on the functions and scope of use of the embodiments of the present disclosure.

[0147] As Figure 7 shown, the electronic device 700 is presented in the form of a general-purpose computing device. The components of the electronic device 700 may include, but are not limited to: at least one of the above-mentioned processing units 710, at least one of the above-mentioned storage units 720, a bus 730 connecting different system components (including the storage unit 720 and the processing unit 710), and a display unit 740.

[0148] Among them, the storage unit stores program codes, and the program codes can be executed by the processing unit 710, so that the processing unit 710 executes the steps according to various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of this specification. For example, the processing unit 710 may execute steps S110 as Figure 1 shown in, obtain topographic observation data of a terrain area to be surveyed, where the topographic observation data includes a digital elevation model, a topographic orthophoto, geological disaster data, and remote sensing multispectral data; step S120, perform a preliminary area division on the terrain area to be surveyed to determine a plurality of area units; step S130, cluster each of the area units according to the topographic observation data to obtain surveyed sub-areas; step S140, determine the survey priority corresponding to each of the surveyed sub-areas, and match the basic flight routes to be adopted by the unmanned aerial vehicle in the surveyed sub-areas according to the survey priority; step S150, splice different types of basic flight routes corresponding to each of the surveyed sub-areas, and optimize and adjust the spliced basic flight routes to obtain a target unmanned aerial vehicle flight route, so as to perform topographic survey on the terrain area to be surveyed through the target unmanned aerial vehicle flight route.

[0149] The storage unit 720 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 721 and / or a cache storage unit 722, and may further include a read-only storage unit (ROM) 723.

[0150] The storage unit 720 may further include a program / utilities 724 having a set (at least one) of program modules 725. Such program modules 725 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include the implementation of a network environment.

[0151] The bus 730 can represent one or more of several types of bus architectures, including a memory bus or memory controller, a peripheral bus, an Accelerated Graphics Port, a processor unit, or a local bus using any of the various bus architectures.

[0152] The electronic device 700 can also communicate with one or more external devices 770 (such as a keyboard, a pointing device, a Bluetooth device, etc.), and can also communicate with one or more devices that enable a user to interact with the electronic device 700, and / or communicate with any device that enables the electronic device 700 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication can be carried out through the input / output (I / O) interface 750. Moreover, the electronic device 700 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 760. As shown in the figure, the network adapter 760 communicates with other modules of the electronic device 700 through the bus 730. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device 700, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0153] Through the description of the above embodiments, those skilled in the art can easily understand that the exemplary embodiments described herein can be implemented by software, or can be implemented by the way of software combined with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure can be embodied in the form of a software product, and the software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0154] In an exemplary embodiment of the present disclosure, there is also provided a computer-readable storage medium, on which a program product capable of implementing the above method of this specification is stored. In some possible embodiments, various aspects of the present disclosure can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary embodiments of the present disclosure described in the above "Exemplary Method" section of this specification.

[0155] Reference Figure 8As shown, a program product 800 for implementing the above-described method for laying out UAV flight paths in complex terrains according to an embodiment of the present disclosure is described. It can be in the form of a portable compact disc read-only memory (CD-ROM) and includes program code, and can run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited to this. In this document, a readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0156] The program product can adopt any combination of one or more readable media. The readable media can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0157] The computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, in which the readable program code is carried. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable signal medium can also be any readable medium other than the readable storage medium, and this readable medium can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0158] The program code contained on the readable medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination of the above.

[0159] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user device, executed as a stand-alone software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or alternatively, may be connected to an external computing device (e.g., by connecting through the Internet using an Internet service provider).

[0160] In addition, the above-mentioned drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, rather than for limiting purposes. It is easy to understand that the processes shown in the above-mentioned drawings do not indicate or limit the chronological order of these processes. Additionally, it is also easy to understand that these processes may be executed synchronously or asynchronously in, for example, multiple modules.

[0161] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software or in a manner of software combined with necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0162] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present disclosure. This application is intended to cover any variations, uses, or adaptations of the present disclosure, which follow the general principles of the present disclosure and include well-known knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present disclosure are pointed out by the claims.

[0163] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A method for laying out a UAV route in complex terrain, characterized in that: include: Acquire terrain observation data of the terrain area to be surveyed, wherein the terrain observation data includes digital elevation model, terrain orthophoto, geological disaster data and remote sensing multispectral data; Performing preliminary regional division on the terrain area to be surveyed to determine a plurality of regional units; Clustering each of the regional units according to the terrain observation data to obtain a survey sub-region; Acquire a mapping relationship between a preset terrain type and a survey priority, wherein the survey priority is set based on the survey difficulty of different terrain types, and the survey priority is in direct proportion to the survey difficulty; match the survey priority from the mapping relationship according to the terrain type corresponding to the survey sub-area; and match a basic flight route to be adopted by the UAV in the survey sub-area according to the survey priority, wherein the basic flight route is obtained by adopting different route layout strategies according to the terrain characteristics of different terrain types; Determine the regional parameters corresponding to the survey sub-area, and adaptively adjust the basic flight route according to the regional parameters to obtain the adapted basic flight route of each survey sub-area; according to the survey priority and spatial position relationship of each survey sub-area, the starting points and end points of the different adapted basic flight routes are spliced ​​head to tail in sequence to obtain the spliced ​​basic flight route; use Bezier curves to perform path smoothing on the spliced ​​basic flight route, and perform heading overlap and lateral overlap verification on the smoothed basic flight route to obtain the target UAV flight route, so as to conduct terrain survey on the terrain area to be surveyed through the target UAV flight route.

2. The method for laying out a UAV route according to claim 1, characterized in that: The step of clustering the regional units according to the terrain observation data to obtain survey sub-regions includes: Determine the terrain recognition feature vector corresponding to each of the regional units according to the terrain observation data; Similarity clustering is performed on the area units based on the terrain recognition feature vector to obtain survey sub-areas with different terrain types.

3. The method for laying out a UAV route according to claim 2, characterized in that: The determining, according to the terrain observation data, a terrain identification feature vector corresponding to each of the regional units comprises: Extracting terrain slope characteristics and terrain relief characteristics corresponding to each of the regional units from the digital elevation model and the terrain orthophoto; Extracting the surface cover shielding features corresponding to each of the regional units from the terrain orthophoto and the remote sensing multispectral data; Determine the geological disaster distribution characteristics corresponding to the regional unit according to the geological disaster data; A terrain recognition feature vector is constructed based on the terrain slope feature, the terrain undulation feature, the surface cover shielding feature and the geological disaster distribution feature.

4. The method for laying out a UAV route according to claim 3, characterized in that: Based on the terrain recognition feature vector, the regional units are clustered by similarity to obtain survey sub-regions with different terrain types, including: Constructing a multidimensional feature space based on the terrain slope feature, terrain undulation feature, surface cover shielding feature and geological disaster distribution feature contained in the terrain recognition feature vector; Obtain the predetermined density radius parameter and the minimum number of neighborhood points; In the multi-dimensional feature space, calculating similarity data between the terrain recognition feature vectors corresponding to the area units; Marking the area unit whose similarity data is less than the density radius parameter and whose number of neighborhood points is greater than or equal to the minimum number of neighborhood points as a core area; The adjacent core areas and boundary areas associated with the core areas are clustered based on the terrain recognition feature vectors to obtain survey sub-areas with different terrain types, where the terrain types are determined by the terrain recognition feature vectors of the core areas.

5. The method for laying out a UAV route according to claim 4, characterized in that: The method further comprises: Determining the density radius parameter according to the variance variation range of the terrain slope characteristics and the terrain relief characteristics in all the terrain recognition feature vectors; The minimum number of neighborhood points is determined based on the density of the surface cover shielding features and the geological disaster distribution features in all the terrain recognition feature vectors.

6. A drone route layout device for complex terrain, characterized in that: include: A terrain observation module is used to obtain terrain observation data of the terrain area to be surveyed, wherein the terrain observation data includes digital elevation model, terrain orthophoto, geological disaster data and remote sensing multispectral data; A region division module is used to perform preliminary region division on the terrain region to be surveyed and determine a plurality of region units; A regional clustering module, used for clustering each of the regional units according to the terrain observation data to obtain a survey sub-region; A flight route matching module is used to obtain a mapping relationship between a preset terrain type and a survey priority, wherein the survey priority is set based on the survey difficulty of different terrain types, and the survey priority is in direct proportion to the survey difficulty; the survey priority is matched from the mapping relationship according to the terrain type corresponding to the survey sub-area, and a basic flight route to be adopted by the UAV in the survey sub-area is matched according to the survey priority, and the basic flight route is obtained by adopting different route layout strategies according to the terrain characteristics of different terrain types; The flight route splicing module is used to determine the regional parameters corresponding to the survey sub-area, and adaptively adjust the basic flight route according to the regional parameters to obtain the adapted basic flight route of each survey sub-area; according to the survey priority and spatial position relationship of each survey sub-area, the starting points and end points of the different adapted basic flight routes are spliced ​​head to tail in sequence to obtain the spliced ​​basic flight route; the Bezier curve is used to smooth the path of the spliced ​​basic flight route, and the heading overlap and lateral overlap of the smoothed basic flight route are checked to obtain the target UAV flight route, so as to conduct terrain survey on the terrain area to be surveyed through the target UAV flight route.

7. An electronic device, characterized in that: include: processor; as well as A memory having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by the processor, implement the method for laying out a UAV route for complex terrain as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the method for laying out a UAV route for complex terrain as described in any one of claims 1 to 5 is implemented.

Images