A method, device, storage medium and electronic device for arranging an unmanned aerial vehicle sensing device

Abstract

This application relates to the field of unmanned aerial vehicle (UAV) technology, and specifically provides a method, device, storage medium, and electronic device for arranging UAV sensing devices. The method may include: determining a plurality of arrangement points based on the arrangement parameters of the UAV sensing devices; randomly generating a set of arrangement plans based on the plurality of arrangement points, where the set of arrangement plans includes a plurality of to-be-arranged plans; using a multi-objective arrangement optimization algorithm to evaluate the plurality of to-be-arranged plans and determine a target arrangement plan, where the multi-objective arrangement optimization algorithm is related to the global coverage rate, key area coverage rate, sensing device performance, and number of sensing devices of each to-be-arranged plan; and the target arrangement plan represents the arrangement points for arranging the UAV sensing devices. Some embodiments of this application can improve the efficiency of formulating the arrangement of UAV sensing devices.

Description

Technical Field

[0001] This application relates to the technical field of unmanned aerial vehicles, and more particularly, to a method, device, storage medium, and electronic device for arranging unmanned aerial vehicle sensing devices. Background Art

[0002] Unmanned aerial vehicle sensing devices refer to various sensors and technologies used to help unmanned aerial vehicles sense the surrounding environment, achieve autonomous flight, and avoid obstacles. These devices play a crucial role in the flight control, navigation, and obstacle avoidance of unmanned aerial vehicles.

[0003] Currently, there are mainly two methods for the layout control of unmanned aerial vehicle sensing devices. One is the method based on electromagnetic simulation, and the other is the method based on machine learning. The method based on electromagnetic simulation usually uses building information modeling (BIM) or City Geography Markup Language (CityGML) to construct a three-dimensional urban model, and combines electromagnetic wave propagation simulation (such as ray tracing or finite element analysis) to determine the layout positions of sensing devices. However, this method requires complex electromagnetic simulation calculations. The method based on machine learning usually predicts the layout points of sensing devices through a machine learning model (such as supervised learning). However, this method involves model training and also requires a large amount of computing time and resources.

[0004] Therefore, how to provide a technical solution for arranging unmanned aerial vehicle sensing devices with lower complexity has become a technical problem that needs to be solved urgently. Summary of the Invention

[0005] Some embodiments of this application aim to provide a method, device, storage medium, and electronic device for arranging unmanned aerial vehicle sensing devices. Through the technical solutions of the embodiments of this application, the computational complexity of formulating a layout plan for unmanned aerial vehicle sensing devices can be reduced, without occupying a large amount of computing resources, and the efficiency is relatively high.

[0006] In a first aspect, some embodiments of this application provide a method for arranging unmanned aerial vehicle sensing devices, including: determining a plurality of layout points based on the layout parameters of the unmanned aerial vehicle sensing devices, where the layout parameters include: the size of the layout area, the range of the layout airspace, the spatial position information of geographical elements within the layout area, key element information, and the communication distance between the sensing devices; randomly generating a set of layout plans based on the plurality of layout points, where the set of layout plans includes a plurality of to-be-arranged plans; using a multi-objective layout optimization algorithm to evaluate the plurality of to-be-arranged plans and determine a target layout plan, where the multi-objective layout optimization algorithm is related to the global coverage rate, key area coverage rate, sensing device performance, and the number of sensing devices of each to-be-arranged plan; the target layout plan represents the layout points of the unmanned aerial vehicle sensing devices.

[0007] In some embodiments of the present application, a plurality of layout points are first determined based on the layout parameters of the UAV sensing device, and then points are randomly selected from the plurality of layout points to generate a plurality of layout schemes to be arranged; finally, a multi-objective layout optimization algorithm is used to evaluate the plurality of layout schemes to be arranged, and a target layout scheme is determined. Some embodiments of the present application can accurately select the target layout scheme only by using the multi-objective layout optimization algorithm, reduce the computational complexity, improve the computational efficiency, and save computational resources.

[0008] In some embodiments, determining a plurality of layout points based on the layout parameters of the UAV sensing device includes: selecting a plurality of spatial points from the geographical elements to form a spatial point set, wherein the distance between two adjacent spatial points is not less than the point distance threshold; dividing the plane space corresponding to the layout area into a plurality of grids; taking the center of each grid in the plurality of grids as a plane point to form a plane point set; and obtaining the plurality of layout points by analyzing the distance between the plane points in the plane point set and each spatial point in the plurality of spatial points.

[0009] Some embodiments of the present application select a plurality of spatial points based on geographical elements and select a plane point set based on the plane space; finally, a plurality of layout points are determined according to the plane point set and the plurality of spatial points, so as to achieve accurate selection and positioning of the layout points.

[0010] In some embodiments, obtaining the plurality of layout points by analyzing the distance between the plane points in the plane point set and each spatial point in the plurality of spatial points includes: selecting a target plane point that is the farthest from each spatial point in the plane point set; confirming that the distance between the target plane point and the adjacent spatial point is not less than the point distance threshold, and then moving the target plane point into the spatial point set to obtain a layout point set; wherein the layout point set includes the plurality of layout points.

[0011] Some embodiments of the present application select a target plane point that meets the requirements from the plane point set and add it to the spatial point set to obtain a layout point set, so as to achieve accurate selection of the layout points.

[0012] In some embodiments, evaluating the multiple to-be-arranged schemes by using the multi-objective layout optimization algorithm to determine the target layout scheme includes: repeatedly performing the following operations until it is confirmed that the fitness value of any to-be-arranged scheme in the multiple to-be-arranged schemes remains the maximum value unchanged, and then using the any to-be-arranged scheme as the target layout scheme: calculating the fitness value of each to-be-arranged scheme in the layout scheme set by using the multi-objective layout optimization algorithm; classifying and randomly recombining the multiple to-be-arranged schemes based on the fitness values of each to-be-arranged scheme to obtain multiple new layout schemes; and adding the multiple new layout schemes to the layout scheme set.

[0013] In some embodiments of the present application, the optimal target layout scheme is selected from the layout scheme set through iterative operations, with high efficiency and low computational complexity.

[0014] In some embodiments, the global coverage rate is obtained by the following method: dividing the layout airspace within the layout airspace range into voxels according to a preset size to obtain multiple voxels; calculating the coverage index of each voxel in the multiple voxels, where the coverage index represents the number of the central points of each voxel covered by the UAV sensing device; and determining the global coverage rate based on the coverage index of each voxel, where the global coverage rate represents the ratio of the voxels covered by at least one UAV sensing device.

[0015] In some embodiments of the present application, by determining the coverage index of each voxel after dividing the layout airspace, the global coverage rate is obtained, providing a basis for constructing the multi-objective layout optimization algorithm.

[0016] In some embodiments, the key area coverage rate is obtained by the following method: mapping the key element information to multiple voxels to obtain key voxels; and determining the key area coverage rate based on the coverage index of each voxel in the key voxels, where the key area coverage rate represents the ratio of the voxels covered by at least two UAV sensing devices.

[0017] In some embodiments of the present application, by calculating the coverage index of the key voxels, the key area coverage rate is obtained, providing a basis for constructing the multi-objective layout optimization algorithm.

[0018] In some embodiments, the sensing device performance is obtained by the following method: randomly removing a preset number of the UAV sensing devices in each to-be-arranged scheme multiple times to obtain the global coverage rate each time; and taking the mean value of the global coverage rates obtained multiple times as the sensing device performance.

[0019] Some embodiments of the present application determine the performance of the UAV sensing device by removing the UAV sensing device, achieving accurate performance testing and providing a basis for constructing a multi-target layout optimization algorithm.

[0020] In some embodiments, the formula of the multi-target layout optimization algorithm is as follows:

[0021]

[0022] In the formula, f is the fitness value, , , , are preset coefficients respectively, CR total is the global coverage rate, CR vital is the key area coverage rate, Res is the performance of the sensing device, and V is the number of sensing devices.

[0023] In a second aspect, some embodiments of the present application provide a device for arranging UAV sensing devices, including: a point determination module for determining a plurality of arrangement points based on the arrangement parameters of the UAV sensing device; wherein the arrangement parameters include: the size of the arrangement area, the airspace range of the arrangement, the spatial position information of geographical elements in the arrangement area, key element information, and the communication distance between the sensing devices; a scheme generation module for randomly generating a set of arrangement schemes based on the plurality of arrangement points, where the set of arrangement schemes includes a plurality of to-be-arranged schemes; an optimal scheme evaluation module for evaluating the plurality of to-be-arranged schemes by using the multi-target layout optimization algorithm to determine the target arrangement scheme, where the multi-target layout optimization algorithm is related to the global coverage rate, key area coverage rate, performance of the sensing device, and number of sensing devices of each to-be-arranged scheme; the target arrangement scheme represents the arrangement points of the UAV sensing device arrangement.

[0024] In a third aspect, some embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the method described in any embodiment of the first aspect can be implemented.

[0025] In a fourth aspect, some embodiments of the present application provide an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program, the method described in any embodiment of the first aspect can be implemented.

[0026] In a fifth aspect, some embodiments of the present application provide a computer program product, which includes a computer program. When the computer program is executed by a processor, the method described in any one of the embodiments of the first aspect can be implemented. Description of the Drawings

[0027] To more clearly illustrate the technical solutions of some embodiments of the present application, the drawings required for use in some embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative efforts.

[0028] Figure 1 System diagram of the layout of UAV sensing devices provided by some embodiments of the present application;

[0029] Figure 2 One of the method flowcharts of the layout of UAV sensing devices provided by some embodiments of the present application;

[0030] Figure 3 Another method flowchart of the layout of UAV sensing devices provided by some embodiments of the present application;

[0031] Figure 4 Block diagram of the device for the layout of UAV sensing devices provided by some embodiments of the present application;

[0032] Figure 5 Schematic diagram of an electronic device provided by some embodiments of the present application. Detailed Embodiments

[0033] Next, the technical solutions in some embodiments of the present application will be described with reference to the drawings in some embodiments of the present application.

[0034] It should be noted that similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, the terms "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0035] In the related art, the layout optimization of UAV sensing devices is mainly divided into two methods, namely the method based on electromagnetic simulation and the method based on machine learning. The method based on electromagnetic simulation often requires complex electromagnetic simulations (such as ray tracing), and the computational cost is extremely high. Especially when dealing with large-scale urban areas, it takes a long time and requires a large amount of resources. The method based on machine learning needs to train the machine learning model, which may also consume a large amount of computing time and computing resources. Moreover, the method based on electromagnetic simulation often requires a very accurate 3D urban model (such as BIM or CityGML) to construct a high-precision urban geometric structure. These data are generally generated by architectural design or construction units, but they usually only target individual buildings or building groups and generally cannot cover the entire urban area. The main difficulty of the method based on machine learning lies in obtaining sufficient historical data of UAV flight trajectories. Generally speaking, in addition to cooperative UAVs (UAVs that will actively report their flight paths), there are also non-cooperative UAVs (UAVs that will not actively report information). Therefore, during the process of training the subsequent machine learning model, the prediction of the device layout points may be inaccurate or fall into a local optimal solution due to incomplete input data (trajectory information).

[0036] In view of this, some embodiments of the present application provide a method for arranging UAV sensing devices. This method first determines multiple arrangement points in advance through arrangement parameters; then, based on the multiple arrangement points, multiple arrangement schemes are randomly generated; finally, the constructed multi-objective layout optimization algorithm is used to evaluate and analyze the multiple arrangement schemes, and the target arrangement scheme is selected. Some embodiments of the present application do not require complex urban geometric modeling and electromagnetic propagation simulation, only rely on easily obtainable basic spatial data, simplify the point generation process, reduce data dependence and computational complexity, and significantly improve computational efficiency and applicability, especially suitable for small and medium-sized cities with scarce data. Through global search rather than local optimization, it ensures the overall optimality of the layout scheme in terms of coverage and robustness, and at the same time does not rely on UAV flight history data, overcomes the shortcomings of insufficient coverage optimization and poor robustness of traditional methods, provides higher monitoring reliability, and can effectively reduce blind spots especially in complex urban environments and adapt to dynamic changes.

[0037] The following combines the attached Figure 1 Exemplarily elaborates the overall composition structure of the system for arranging UAV sensing devices provided by some embodiments of the present application.

[0038] Such as Figure 1As shown in the figure, some embodiments of the present application provide a system diagram for deploying UAV sensing devices. The system for deploying UAV sensing devices may include: a terminal 100 and a server 200. Among them, the terminal 100 can obtain the deployment parameters of the UAV sensing devices, and the deployment parameters can be input by the user on the terminal 100. The terminal 100 sends the deployment parameters to the server 200. The server 200 can determine multiple deployment points based on the deployment parameters; then, based on the multiple deployment points, randomly generate multiple deployment plans; finally, use the constructed multi-objective deployment optimization algorithm to evaluate and analyze the multiple deployment plans, and select the target deployment plan.

[0039] In some embodiments of the present application, the terminal 100 can be a mobile terminal or a non-portable computer terminal, and the embodiments of the present application do not make specific limitations here.

[0040] In some other embodiments of the present application, if the terminal 100 has the functions of determining the deployment points in the server 200, generating the deployment plan, and selecting the optimal target deployment plan, the server 200 may not be set at this time. Specifically, it can be flexibly selected according to the actual application scenario, and the embodiments of the present application are not limited thereto.

[0041] The following combines the attached Figure 2 Exemplarily elaborates the implementation process of the deployment of UAV sensing devices executed by the server 200 provided by some embodiments of the present application.

[0042] Please refer to the attached Figure 2 , Figure 2 is a flowchart of a method for deploying UAV sensing devices provided by some embodiments of the present application. The method for deploying UAV sensing devices may include:

[0043] S210, based on the deployment parameters of the UAV sensing devices, determine multiple deployment points; wherein, the deployment parameters include: the size of the deployment area, the range of the deployment airspace, the spatial position information of the geographical elements in the deployment area, the key element information, and the communication distance between the sensing devices.

[0044] For example, in some embodiments of the present application, first obtain the deployment parameters. The deployment parameters mainly include: the size of the target area (as a specific example of the size of the deployment area, for example, an area of 50 square kilometers), the digital surface model DSM of the target area (as a specific example of the spatial position information of the geographical elements in the deployment area), the point of interest data POI of the target area (as a specific example of the key element information), the communication distance R of the sensing devices, and the detection airspace height range (as a specific example of the range of the deployment airspace).

[0045] Specifically, the digital surface model (DSM) is a two-dimensional grid data that represents the height of geographical elements at each location in the city, including terrain and buildings. It is usually stored in the form of a grid, for example, each grid has a resolution of 10m×10m, which can be represented by {(X D , Y D , H D )} to represent DSM data, (X D , Y D , H D ) represents the information of a DSM grid, X D With Y D represent the coordinates of the center point of the DSM grid, and H D Represents the height of the DSM grid. DSM data is mainly used to calculate potential deployment points and consider possible occlusion problems in deployment point optimization.

[0046] Point of interest data POI mainly considers important places in the city, which is a set of two-dimensional points {(X poi ,Y poi , L poi )}, where X poi and Y poi represents the coordinates of a POI point, and L poi The label representing the POI point (such as "hospital", "power plant", etc.). The purpose of introducing POI data is mainly to consider the redundancy of deployment in important areas (for example, the airspace above important locations is covered by multiple sensing devices).

[0047] The sensing device communication distance R (for example, 1 km) refers to the maximum distance of the device for wireless detection and communication; at the same time, the detection airspace altitude range H min and H max Need to be pre-defined (for example, 100m and 300m).

[0048] Through the above layout parameters, multiple potential layout points can be determined.

[0049] According to the above deployment parameters, under the premise of considering the spatial coverage and distance constraints, it is necessary to first calculate the potential deployment points (referring to the locations where the equipment may be deployed). Generally speaking, the sensing equipment is either deployed on the ground (if it is an open space and there are no buildings here) or on the top of the building, so the method for generating potential deployment points is to select points evenly in space while retaining the high points as much as possible. Generally, it is necessary to preset the minimum threshold d of the distance between potential deployment points. min (As a specific example of a point distance threshold, such as 500 meters), this ensures that potential deployment points are not too dense.

[0050] In some embodiments of the present application, S210 may include:

[0051] S211, selecting multiple spatial points from the geographical elements to form a set of spatial points, wherein the distance between two adjacent spatial points is not less than the point distance threshold.

[0052] For example, in some embodiments of the present application, first, M high points in cities (as a specific example of multiple spatial points) are selected. Generally, it needs to be set in advance. For example, M is 100 points. According to the Digital Surface Model (DSM) of the city, the grids in the DSM are sequentially selected in the order of height from high to low as the high points in the city. For example, first, the highest DSM grid in the city is selected, then the second-highest DSM grid in the city is selected, and so on. At the same time, it is ensured that the distance between each newly selected high point in the city and all the previously selected high points cannot be less than d min . And so on, finally, M high points in the city will be obtained, represented by a set C init (as a specific example of the set of spatial points), and the distance between any two points is greater than d min . The set C init is as follows:

[0053]

[0054] S212, dividing the planar space corresponding to the deployment area into multiple grids. The center of each grid in the multiple grids is used as a planar point to form a set of planar points.

[0055] For example, in some embodiments of the present application, secondly, several ordinary points (as a specific example of planar points) are selected. The urban planar space (as a specific example of the planar space corresponding to the deployment area) is evenly grid-divided. The dimension of the grid needs to refer to the communication distance R of the sensing device. Usually, the side length of each grid cannot be greater than R, that is, the side length of the grid can be any value greater than zero and less than or equal to R. For example, the grid can be constructed with a dimension of R×R. The center point of each grid can be selected as an ordinary point. Assuming that in the corresponding urban area (i.e., the deployment area), N grids and the corresponding N center points are divided, represented by a set C grid (as a specific example of the set of planar points):

[0056]

[0057] S213, obtaining the multiple deployment points by analyzing the distance between the planar points in the set of planar points and each spatial point in the multiple spatial points.

[0058] For example, in some embodiments of the present application, the nearest neighbor distance is defined , for measuring C grid the distance from any planar point position i in to a certain point in the set C init . By calculating , a plurality of layout point positions are determined, where the calculation formula of is as follows:

[0059]

[0060] Take the minimum value of the distances from any planar point position i to each point in C init as the nearest neighbor distance from any planar point position i to C init . That is, through this formula, the distance between each planar point position and can be calculated. C init In some embodiments of the present application, S213 may include: selecting a target planar point position that is the farthest from each spatial point position in the set of planar point positions; confirming that the distance between the target planar point position and the adjacent spatial point position is not less than the point position distance threshold, and then moving the target planar point position into the set of spatial point positions to obtain a set of layout point positions; where the set of layout point positions includes the plurality of layout point positions.

[0062] For example, in some embodiments of the present application, C what is calculated is the distance from any planar point position i to the point in the set C that is the closest to it in C init . Then, from C grid , each time select a target point (as a specific example of the target planar point position) that is the farthest (i.e., the maximum distance) from C init ; if the distance between the target point and the nearest point (i.e., the adjacent spatial point position) in C init is not less than d min , then move the target point from C grid to C init ; otherwise, remove the target point from C grid (i.e., discard the target point). Such selection can ensure that the layout point positions are not concentrated as much as possible in space, but are dispersed. Subsequently, continuously select from Cgrid Into C init add appropriate points until C grid it is empty. At this time, C init contains K points (M < K < M + N), and these K points are the potential multiple deployment points at this time. C init ={P1, P2, ..., P K}.

[0063] S220. Based on the multiple deployment points, randomly generate a set of deployment plans, where the set of deployment plans includes multiple to-be-deployed plans.

[0064] For example, in some embodiments of the present application, with different numbers of deployed sensing devices, for example, the numbers of 1 / 5K, 2 / 5K, 3 / 5K, 4 / 5K are randomly selected from the potential multiple deployment points to correspond to the corresponding number of points, and the deployment device is used to generate multiple batches of deployment plans Plan (as a specific example of the multiple to-be-deployed plans). Each to-be-deployed plan can be represented as a vector of length K (because there are K potential deployment points), and the elements in the vector take values of 1 or 0 (1 indicates deployment, 0 indicates non-deployment). For example, Plan1 = [1, 0, 0, 1, ..., 1], which means deploying sensing devices at the 1st and 4th potential deployment points. Plan2 = [0, 0, 1, 1, ..., 1], which represents deploying sensing devices at the 3rd and 4th potential deployment points. Similar to Plan1 and Plan2, many sets of to-be-deployed plans are randomly generated. For example, first generate 100 sets of plans (as a specific example of the multiple to-be-deployed plans), which is called a population.

[0065] It can be understood that before executing S230, it is first necessary to construct a multi-objective deployment optimization algorithm. The multi-objective deployment optimization algorithm is related to the global coverage rate, key area coverage rate, sensing device performance, and number of sensing devices of each to-be-deployed plan. Among them, corresponding weight coefficients are set for the global coverage rate, key area coverage rate, sensing device performance, and number of sensing devices respectively. By performing weighted summation with the weight coefficients and the global coverage rate, key area coverage rate, sensing device performance, and number of sensing devices, the fitness value output by the multi-objective deployment optimization algorithm is obtained.

[0066] In some embodiments of the present application, the formula of the multi-objective deployment optimization algorithm is as follows:

[0067]

[0068] In the formula, f is the fitness value, , , , are preset coefficients (i.e., the weight coefficients in the above text), CR total is the global coverage rate, CR vital is the critical area coverage rate, Res is the performance of the sensing device, and V is the number of sensing devices.

[0069] In some embodiments of the present application, the global coverage rate is obtained by the following method: the deployment airspace within the range of the deployment airspace is divided into voxels according to a preset size to obtain a plurality of voxels; calculate the coverage index of each voxel among the plurality of voxels, where the coverage index represents the number of central points of each voxel covered by the UAV sensing device; based on the coverage index of each voxel, determine the global coverage rate, where the global coverage rate represents the ratio of voxels covered by at least one UAV sensing device.

[0070] For example, in some embodiments of the present application, in order to accurately estimate the detection range of the sensing device in three-dimensional space, it is necessary to divide the deployment airspace into spatial voxels. Specifically, after determining the height range (i.e., the deployment airspace range) H min and H max , the airspace between H min and H max above the deployment area is divided into cuboids (length × width × height) of a fixed size (as a specific example of the preset size), for example, the fixed size is 100m × 100m × 10m. By dividing the airspace into voxels, the position of the UAV (e.g., which height range it is located in) can be effectively determined during the sensing process. At the same time, the voxels will also be used in the calculation of the sensing coverage rate. Assume that after the deployment airspace is divided into voxels, q voxels (as a specific example of the plurality of voxels) are obtained, and C vox represents the set composed of the central points of all voxels: C vox ={P1, P2, ..., P q}.

[0071] After that, construct the coverage matrix A= a ij , for the central point j of a certain voxel, if there is a potential deployment point i such that the distance between i and j is less than R and there is no occlusion between i and j (calculated by the digital surface model DSM), then a ijIt is 1, otherwise it is 0. Further, the coverage index of each voxel j can be calculated. cov j , cov j represents the number of sensing devices covering the voxel j. cov j The calculation formula of is as follows:

[0072]

[0073] Then, the global coverage rate of the UAV sensing devices, that is, calculate the ratio of the voxels covered by at least one sensing device. CR total , as follows:

[0074]

[0075] In some embodiments of the present application, the key area coverage rate is obtained by the following method: mapping the key element information to multiple voxels to obtain key voxels; determining the key area coverage rate based on the coverage index of each voxel in the key voxels, where the key area coverage rate represents the ratio of the voxels covered by at least two UAV sensing devices.

[0076] For example, in some embodiments of the present application, after obtaining the POI data, the POI of the important area (a specific example of the key element information) can be mapped to the corresponding voxels. For example, since the POI points include important areas such as hospitals and factories, the voxels directly above these areas are also more important than other voxels and require key monitoring of UAV flight. Represent the important voxels with l For important voxels, their coverage index is at least 2 (for important voxels, sensing has redundancy), then the important area coverage rate can be calculated. CR vital (a specific example of the key area coverage rate), as follows:

[0077]

[0078] In some embodiments of the present application, the performance of the sensing device is obtained by the following method: randomly removing a preset number of the UAV sensing devices in each deployment plan to be arranged multiple times to obtain the global coverage rate each time; taking the mean value of the global coverage rates obtained multiple times as the performance of the sensing device.

[0079] For example, in some embodiments of the present application, since the sensing device may fail or encounter faults during operation, it is necessary to evaluate the sensing performance of the sensing system under extreme conditions (for example, some components fail). The performance of the sensing device is represented by sensing resilience.Res It is represented by. Removal tests are carried out by randomly removing 10% of the sensing devices (as a specific example of the preset number). After that, multiple (such as 50 times) removal tests will be carried out, and then take CR total the average value, and the calculation formula is:

[0080]

[0081] S230, using the multi-objective layout optimization algorithm, evaluate the multiple to-be-laid-out schemes, and determine the target layout scheme; the target layout scheme represents the layout points of the UAV sensing device layout.

[0082] For example, in some embodiments of the present application, through the above-mentioned constructed multi-objective layout optimization algorithm f select the optimal target layout scheme from multiple to-be-laid-out schemes. That is, the goal of the present application is to generate a set of target layout schemes for sensing devices, so that under the condition of ensuring as few sensing devices as possible, it has a relatively high global coverage rate, important area coverage rate, and sensing resilience. For this purpose, a genetic algorithm can be used to f solve it.

[0083] In some embodiments of the present application, S230 may include: repeatedly execute the following operations until it is confirmed that the fitness value of any to-be-laid-out scheme in the multiple to-be-laid-out schemes remains unchanged at the maximum value, then use the any to-be-laid-out scheme as the target layout scheme: calculate the fitness value of each to-be-laid-out scheme in the layout scheme set by using the multi-objective layout optimization algorithm; classify and randomly recombine the multiple to-be-laid-out schemes based on the fitness values of each to-be-laid-out scheme to obtain multiple new layout schemes; add the multiple new layout schemes to the layout scheme set.

[0084] For example, in some embodiments of the present application, calculate the global coverage rate, important area coverage rate, sensing resilience and the number of sensing devices used corresponding to each to-be-laid-out scheme by using the above calculation formula, and finally use the multi-objective layout optimization algorithm f calculate the fitness value of each to-be-laid-out scheme.

[0085] After that, the fitness values of each layout scheme to be arranged are sorted in reverse order. Then, crossover is randomly performed among the layout schemes with relatively high rankings (e.g., the top 20%), for example, the first half of one scheme is combined with the second half of another scheme to form a new scheme, and mutation is randomly performed on the schemes with relatively low rankings (e.g., the last 80%). For example, the values of a few digits in a certain scheme are swapped, such as 0 becoming 1 or 1 becoming 0 (as a specific example of classification and random recombination). At the same time, multiple new layout schemes generated by crossover and mutation are also added to the layout scheme set.

[0086] After such repeated iterations as described above (i.e., continuously performing random crossover and mutation while expanding the candidate layout schemes), if it is found that the best scheme has not changed after multiple rounds (e.g., 10 rounds) (i.e., the fitness value of any layout scheme to be arranged is always the maximum), then the iteration is stopped and the best scheme is output (as a specific example of the target layout scheme).

[0087] The following Figure 3 exemplarily elaborates the specific process of the layout of the UAV sensing device provided by some embodiments of the present application.

[0088] Please refer to the Figure 3 , Figure 3 which is a flowchart of a method for arranging a UAV sensing device provided by some embodiments of the present application.

[0089] The following exemplarily elaborates the above process.

[0090] The first step is input data preparation.

[0091] For example, input the digital surface model DSM of the layout area, points of interest POI, communication parameters of the sensing devices (i.e., the communication distance between the sensing devices), etc.

[0092] The second step is the generation of potential layout points (as a specific example of multiple layout points):

[0093] For example, first select high points in the city, then select ordinary points in the city, and finally merge the high points and ordinary points in the city to obtain potential layout points.

[0094] The third step is spatial voxel preprocessing:

[0095] For example, the airspace of the layout area is divided into multiple voxels, and important voxels are marked based on the important voxel marking of the POI.

[0096] The fourth step is multi-objective layout optimization based on the genetic algorithm:

[0097] For example, multiple sets of candidate solutions (or referred to as populations, as a specific example of multiple proposed layout solutions) are generated based on the above-mentioned potential layout point docking;

[0098] Calculate indicators for each set of solutions, that is CR total 、 CR vital 、 Res and N (N is the number of sensing devices). Using the multi-objective layout optimization algorithm, calculate the fitness value f for each set of solutions and perform reverse sorting. Determine whether the best solution has remained unchanged for multiple iterations. If not, perform random crossover on the top-ranked solutions in the current population, perform random mutation on the bottom-ranked solutions in the current population, update the population, recalculate the indicators, and perform iterative loops again.

[0099] If the best solution has remained unchanged for multiple iterations, at this time, output the optimal solution in the population as the final sensing device layout solution (as a specific example of the target layout solution).

[0100] It can be understood that the specific implementation process above can refer to the method embodiments provided above. To avoid repetition, the detailed description is appropriately omitted here.

[0101] From some embodiments of the present application above, it can be seen that the embodiments of the present application simplify the three-dimensional voxel division and the regularized point generation method (such as uniform grid based on DSM and high-point extraction), and combine with the multi-objective genetic algorithm for optimization, avoiding the high computational cost of complex three-dimensional urban modeling (such as BIM, CityGML) and electromagnetic simulation in traditional methods. The present application is convenient for calculation, can significantly improve the calculation efficiency, and is convenient for rapid deployment in an environment with limited resources. Moreover, the embodiments of the present application do not rely on the historical data of the UAV flight path, but directly derive the layout points using urban basic data (such as DSM and POI). This avoids the problem of prediction failure when the historical data is missing or incomplete, making the solution applicable to newly developed areas or scenarios where the UAV activity pattern is unknown, enhancing the generality and implementation flexibility of the technology.

[0102] Please refer to Figure 4 , Figure 4 which shows the block diagram of the composition of the device for UAV sensing device layout provided by some embodiments of the present application. It should be understood that this device for UAV sensing device layout corresponds to the above method embodiments and can execute each step involved in the above method embodiments. The specific functions of this device for UAV sensing device layout can be seen in the description above. To avoid repetition, the detailed description is appropriately omitted here.

[0103] Figure 4The apparatus for deploying drone sensing equipment includes at least one software function module that can be stored in a memory in the form of software or firmware or solidified in the apparatus for deploying drone sensing equipment. The apparatus for deploying drone sensing equipment includes: a point determination module 410, which is used to determine multiple deployment points based on the deployment parameters of the drone sensing equipment; wherein the deployment parameters include: deployment area size, deployment airspace range, spatial location information of geographical elements in the deployment area, key element information and communication distance between the sensing devices; a solution generation module 420, which is used to randomly generate a set of deployment plans based on the multiple deployment points, wherein the set of deployment plans includes multiple plans to be deployed; an optimal solution evaluation module 430, which is used to evaluate the multiple plans to be deployed using a multi-objective deployment optimization algorithm to determine a target deployment plan, wherein the multi-objective deployment optimization algorithm is related to the global coverage rate, key area coverage rate, sensing device performance and number of sensing devices of each plan to be deployed; the target deployment plan represents the deployment points of the drone sensing equipment deployment.

[0104] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method, and will not be described in detail here.

[0105] Some embodiments of the present application further provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the operations of the method corresponding to any of the above methods provided in the above embodiments.

[0106] Some embodiments of the present application further provide a computer program product, which includes a computer program, wherein when the computer program is executed by a processor, it can implement the operations corresponding to any of the above methods provided in the above embodiments.

[0107] like Figure 5 As shown, some embodiments of the present application provide an electronic device 500, which includes: a memory 510, a processor 520, and a computer program stored in the memory 510 and executable on the processor 520, wherein the processor 520 can implement a method as described in any of the above embodiments when reading the program from the memory 510 through a bus 530 and executing the program.

[0108] Processor 520 can process digital signals and can include various computing structures, such as complex instruction set computer structure, reduced instruction set computer structure, or a structure that implements a combination of multiple instruction sets. In some examples, processor 520 can be a microprocessor.

[0109] The memory 510 can be used to store instructions executed by the processor 520 or data related to the instruction execution process. These instructions and / or data can include code for implementing some or all of the functions of one or more modules described in the embodiments of the present application. The processor 520 of the embodiments of the present disclosure can be used to execute the instructions in the memory 510 to implement the method shown above. The memory 510 includes dynamic random access memory, static random access memory, flash memory, optical memory, or other memories well known to those skilled in the art.

[0110] The above are only the embodiments of the present application and are not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0111] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application and should be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0112] It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the presence of another identical element in the process, method, article or device including the element.

Claims

1. A method for arranging an unmanned aerial vehicle sensing device, characterized in that, Including: Based on the layout parameters of the UAV sensing devices, determine multiple layout points; wherein, the layout parameters include: the size of the layout area, the layout airspace range, the spatial position information of geographical elements in the layout area, key element information, and the communication distance between the sensing devices; Based on the multiple layout points, randomly generate a set of layout plans, wherein the set of layout plans includes multiple to-be-layout plans; Use a multi-objective layout optimization algorithm to evaluate the multiple to-be-layout plans and determine a target layout plan, wherein the multi-objective layout optimization algorithm is related to the global coverage rate, key area coverage rate, sensing device performance, and the number of sensing devices of each to-be-layout plan; the target layout plan represents the layout points of the UAV sensing device layout; The global coverage rate is obtained by the following method: divide the layout airspace between the layout airspace ranges into voxels according to a preset size to obtain multiple voxels; calculate the coverage index of each voxel among the multiple voxels, wherein the coverage index represents the number of the central points of the voxels covered by the UAV sensing device; based on the coverage index of each voxel, determine the global coverage rate, wherein the global coverage rate represents the ratio of the voxels covered by at least one UAV sensing device; The key area coverage rate is obtained by the following method: map the key element information to multiple pairs of multiple voxels to obtain key voxels; based on the coverage index of each voxel in the key voxels, determine the key area coverage rate, wherein the key area coverage rate represents the ratio of the voxels covered by at least two UAV sensing devices; The sensing device performance is obtained by the following method: randomly remove a preset number of the UAV sensing devices in each to-be-layout plan multiple times to obtain the global coverage rate each time; take the mean value of the global coverage rates obtained multiple times as the sensing device performance.

2. The method according to claim 1, wherein The determining of multiple layout points based on the layout parameters of the UAV sensing devices includes: Select multiple spatial points from the geographical elements to form a set of spatial points, wherein the distance between adjacent two spatial points is not less than the point distance threshold; Divide the plane space corresponding to the layout area into multiple grids; Take the center of each grid in the multiple grids as a plane point to form a set of plane points; Obtain the multiple layout points by analyzing the distances between the plane points in the set of plane points and each spatial point in the multiple spatial points.

3. The method according to claim 2, characterized in that The obtaining of the multiple layout points by analyzing the distances between the plane points in the set of plane points and each spatial point in the multiple spatial points includes: Select the target plane point that is the farthest from each spatial point in the set of plane points; If it is confirmed that the distance between the target plane point and the adjacent spatial point is not less than the point distance threshold, then move the target plane point into the set of spatial points to obtain a set of layout points; wherein the set of layout points includes the multiple layout points.

4. The method according to any one of claims 1 to 3, characterized in that, Evaluating the multiple to-be-arranged schemes by using the multi-objective layout optimization algorithm to determine the target layout scheme, including: Repeatedly execute the following operations until it is confirmed that the fitness value of any to-be-arranged scheme in the multiple to-be-arranged schemes remains unchanged at the maximum value, and then use the any to-be-arranged scheme as the target layout scheme: Calculate the fitness value of each to-be-arranged scheme in the layout scheme set by using the multi-objective layout optimization algorithm; Classify and randomly recombine the multiple to-be-arranged schemes based on the fitness values of each to-be-arranged scheme to obtain multiple new layout schemes; Add the multiple new layout schemes to the layout scheme set.

5. The method according to any one of claims 1 to 3, characterized in that, The formula of the multi-objective layout optimization algorithm is as follows: Wherein, f is the fitness value, , , , are preset coefficients respectively, CR total is the global coverage rate, CR vital is the critical area coverage rate, Res is the performance of the sensing device, and V is the number of sensing devices.

6. A device for arranging an unmanned aerial vehicle sensing device, characterized in that, The device is used to execute the method as described in claim 1, including: A point position determination module, configured to determine multiple layout point positions based on the layout parameters of the UAV sensing device; wherein, the layout parameters include: the size of the layout area, the layout airspace range, the spatial position information of the geographical elements in the layout area, the key element information, and the communication distance between the sensing devices; A scheme generation module, configured to randomly generate a layout scheme set based on the multiple layout point positions, wherein the layout scheme set includes multiple to-be-arranged schemes; An optimal scheme evaluation module, configured to evaluate the multiple to-be-arranged schemes by using the multi-objective layout optimization algorithm to determine the target layout scheme, wherein the multi-objective layout optimization algorithm is related to the global coverage rate, the key area coverage rate, the performance of the sensing device, and the number of sensing devices of each to-be-arranged scheme; the target layout scheme represents the layout point positions of the UAV sensing device layout.

7. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, wherein the computer program, when run by a processor, executes the method as described in any one of claims 1-5.

8. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored on the memory and running on the processor, wherein the computer program, when run by the processor, executes the method as described in any one of claims 1-5.

Images