A method and system for optimizing the configuration of UAV nest space for integrated inspection

Abstract

This invention discloses a method and system for optimizing the spatial configuration of UAV nests for comprehensive inspection, relating to the field of UAV nest site selection technology. The method includes: acquiring inspection targets; obtaining geographical data of the actual inspection range of the inspection targets; and performing point-based discretization processing on the geographical data according to inspection accuracy requirements; constructing a candidate set of take-off and landing points for comprehensive inspection; selecting a set of alternative take-off and landing points suitable for both construction environment and take-off conditions; based on each inspection demand point being visited at least once per day, using the lowest combined construction cost and flight cost of the UAV nest as the optimization objective; obtaining the constraints in the input objective optimization process; and performing a joint optimization solution for UAV nest site selection and path selection to obtain the location set of each UAV nest and the inspection intersection set. This method can obtain the location selection of UAV nests for multi-object joint inspection, providing a reference for the deployment and construction of comprehensive inspection UAVs.

Description

Technical Field

[0001] This invention relates to the field of UAV nest site selection technology, specifically to a method and system for optimizing the configuration of UAV nest space for comprehensive inspection. Background Technology

[0002] The construction of drone nesting sites requires supporting public facilities such as electricity, communication, and fire protection. Existing research has not considered the rational use of public resources, reducing the construction costs of supporting facilities such as electricity, and improving land use efficiency when considering drone nesting site selection. At the same time, most current drone nesting site selection studies only consider single inspection targets (such as forest areas, power grids, gas pipelines, etc.) for nesting site selection analysis, resulting in low drone utilization efficiency. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for optimizing the configuration of drone nest space for comprehensive inspection. It determines the location of drone nests and inspection intersections based on the rational use of public resources, providing a reference for the deployment and construction of comprehensive inspection drones.

[0004] This invention is achieved through the following technical solution:

[0005] In a first aspect, the first embodiment of the present invention provides a method for optimizing the configuration of UAV nest space for comprehensive inspection, comprising:

[0006] Obtain the inspection object, obtain the geographical data of the actual inspection range of the inspection object, and perform point-based discretization processing on the geographical data according to the inspection accuracy requirements, and use the point set to represent the inspection requirements.

[0007] Acquire public resource points of interest and construct a set of candidate locations for comprehensive inspection take-off and landing points;

[0008] Based on limiting factors, a set of alternative take-off and landing sites suitable for both construction environment and take-off conditions is selected from the candidate site set.

[0009] Based on the premise that each inspection point is visited at least once a day, the optimization goal is to minimize the combined cost of drone nest construction and flight.

[0010] Obtain the constraints in the target optimization process from the input;

[0011] A genetic algorithm is used to solve the joint optimization of UAV nest location and path selection, and the location set and inspection intersection set of each UAV nest are obtained.

[0012] Furthermore, the limiting factors include construction conditions, meteorological conditions, airspace conditions, and noise sensitivity.

[0013] Furthermore, the objective function formula for the optimization objective is:

[0014] ;

[0015] Where F represents the set of selected nest points, y i Represents a binary variable, y i =1 indicates that nest i is selected; otherwise, it is 0. 总 K represents the total length of all inspection paths, W represents the cost of building the aircraft nest, and W represents the cost of flight per unit distance.

[0016] Furthermore, the constraints include a coverage constraint, whereby each inspected object j must be covered by at least one unmanned aerial vehicle (UAV) nest i. The UAV nest coverage constraint is specifically expressed as follows:

[0017] ;

[0018] Where S represents the set of inspection objects, a ij Represents a binary parameter. ij =1 indicates that nest i can cover the inspected object j, otherwise it is 0;

[0019] Each inspection point s j It must be covered by at least one path, and the path coverage constraint is specifically expressed as:

[0020] ;

[0021] Among them, R i Let r represent the set of all inspection paths for nest i, r represent the path, and 𝕀 represent the indicator function.

[0022] Furthermore, the constraints also include path length constraints, which are specifically expressed as follows:

[0023] ;

[0024] ;

[0025] Where, d r E represents the distance a drone can fly in a single flight. max This indicates the maximum range of the drone.

[0026] Furthermore, the constraints also include path continuity constraints and spatial domain constraints.

[0027] Furthermore, the method of using a genetic algorithm for the joint optimization of UAV nest location and path selection specifically includes:

[0028] A greedy algorithm is used to solve for candidate nesting schemes covering all inspection points, which are then used as high-quality initial individuals;

[0029] The continuous geographic space is discretized into grids. The A* algorithm is used to plan feasible paths for UAVs on the grid map, avoiding no-fly zones. Inspection points are assigned to each UAV nest and the flight sequence is planned. The total inspection path length is calculated using the objective function of the total inspection path length to obtain candidate flight paths.

[0030] Genetic algorithms are used to optimize candidate flight paths and candidate drone nest location schemes to obtain the location set and inspection intersection set for each drone nest.

[0031] Furthermore, the specific method for discretizing continuous geographic space into a raster includes:

[0032] Determine the map boundaries based on the coordinates of all nests and inspection points;

[0033] Based on the accuracy requirements of the drone flight, the grid size is set to 100m×100m;

[0034] Mark raster attributes.

[0035] Furthermore, the specific steps of assigning inspection points to each nest and planning flight sequences include:

[0036] Based on the coverage area of ​​the nest, the inspection point is assigned to the nearest nest;

[0037] For each nest's inspection point set, the grid path length is calculated based on the A* algorithm. A distance matrix is ​​constructed based on the path length, and the shortest inspection path is solved using the TSP algorithm or VRP algorithm.

[0038] Secondly, another embodiment of the present invention provides a UAV nest space optimization configuration system for comprehensive inspection, used to implement the UAV nest space optimization configuration method for comprehensive inspection described in the above embodiment. The system includes: a first acquisition module, a construction module, a filtering module, an optimization target determination module, a second acquisition module, and an optimization module.

[0039] The first acquisition module is used to acquire the inspection object, acquire the geographical data of the actual inspection range of the inspection object, and perform point-based discretization processing on the geographical data according to the inspection accuracy requirements, and use the point set to represent the inspection requirements.

[0040] The construction module is used to obtain public resource points (POIs) and construct a set of candidate locations for comprehensive inspection take-off and landing points.

[0041] The filtering module is used to filter out a set of alternative take-off and landing sites from the candidate site set based on limiting factors, where both the construction environment and take-off conditions are suitable.

[0042] The optimization target determination module is used to take the lowest combined cost of drone nest construction and flight cost as the optimization target, based on the premise that each inspection demand point is visited at least once a day.

[0043] The second acquisition module is used to acquire the constraints in the input target optimization process;

[0044] The optimization module uses a genetic algorithm to solve the joint optimization of UAV nest location and path selection, and obtains the location set and inspection intersection set of each UAV nest.

[0045] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0046] This invention provides a method for optimizing the configuration of drone nest space for comprehensive inspection. By comprehensively considering the inspection targets and the rational utilization of public resources, a set of inspection take-off and landing points is determined. While ensuring that each inspection point is visited at least once daily, the optimization objective is to minimize the combined construction and flight costs of drone nests. A genetic algorithm is used to perform joint optimization of drone nest location and path selection, obtaining the location set of each drone nest and the set of inspection intersections. This method can obtain the location selection of drone nests for joint inspection of multiple targets such as forests, power grids, transportation, and urban areas, providing a reference for the deployment and construction of comprehensive inspection drones.

[0047] The UAV nest space optimization configuration system and the UAV nest space optimization configuration method for comprehensive inspection provided in this embodiment of the invention are based on the same inventive concept and have the same beneficial effects. Attached Figure Description

[0048] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0049] Figure 1 A flowchart of a method for optimizing the configuration of UAV nest space for comprehensive inspection, provided in the first embodiment of the present invention;

[0050] Figure 2 The diagram below shows a structural block diagram of a UAV nest space optimization configuration system for comprehensive inspection, provided as another embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0052] like Figure 1 As shown, the first embodiment of the present invention provides a method for optimizing the configuration of UAV nest space for comprehensive inspection, including:

[0053] Obtain the inspection object, obtain the geographical data of the actual inspection range of the inspection object, and perform point-based discretization processing on the geographical data according to the inspection accuracy requirements, and use the point set to represent the inspection requirements.

[0054] Acquire public resource POI locations and construct a comprehensive inspection take-off and landing point candidate location set. POI (Point of Interest) refers to data used in geographic information systems to describe a specific location, which usually includes information such as location coordinates, name, address and category. POI is widely used in navigation, map services, business site selection, urban planning and tourism, helping users quickly find surrounding facilities and providing businesses with market analysis basis.

[0055] Based on limiting factors, a set of alternative take-off and landing sites suitable for both construction environment and take-off conditions is selected from the candidate site set.

[0056] Based on the premise that each inspection point is visited at least once a day, the optimization goal is to minimize the combined cost of drone nest construction and flight.

[0057] Obtain the constraints in the target optimization process from the input;

[0058] A genetic algorithm is used to solve the joint optimization of UAV nest location and path selection, and the location set and inspection intersection set of each UAV nest are obtained.

[0059] Considering that inspection objects such as rivers and power grids mainly use linear inspection paths, while inspection objects such as forests and farmland mainly use gridded inspection paths, and that different inspection objects may have spatial overlaps, such as rivers passing through forests and farmland, making the inspection paths complex, transforming continuous facility site selection into discrete facility site selection can reduce the difficulty of solving the problem and improve computational efficiency.

[0060] Public resource points of interest (POIs) include government agencies, village committees, etc. By acquiring public resource POIs, a comprehensive inspection set of candidate take-off and landing points is constructed. Based on four types of limiting factors—construction conditions, meteorological conditions, airspace conditions, and noise sensitivity—a set of alternative take-off and landing points with suitable construction environment and take-off conditions is selected.

[0061] Construction conditions: Considering that some public institutions are located inside residential, commercial or office buildings and do not have independent space for setting up drone nests, the POI locations are overlaid with land use planning data to screen out public resource locations that are not located on independent government or institutional land or service facility land.

[0062] Meteorological conditions: Stable meteorological conditions are required for drone takeoff and landing. Drone nests should be located far from densely packed high-rise buildings to avoid interference from rapid wind changes such as angular currents, eddies, and funneling effects, which could disrupt flight safety. Based on existing urban wind environment research, the core areas affected by wind speed changes are the 0-120 meter radius around high-rise buildings, narrow passages between high-rise buildings, and the 50-200 meter downstream area of ​​their exits. By overlaying POI points with 3D building vector data, public resource points with buildings taller than 60 meters within a 200-meter radius are excluded.

[0063] Airspace conditions: Public resource sites located in no-fly zones, dangerous zones, and various restricted airspace areas will be excluded.

[0064] Noise Sensitivity: In accordance with the noise control requirements of the "Environmental Noise Quality Standard (GB3096-2008)" and considering the noise attenuation characteristics of aircraft, noise-sensitive land use such as rehabilitation and convalescent facilities, residential areas, medical and health facilities, and educational and research facilities should be avoided within approximately 50 meters of the UAV take-off and landing facilities. By overlaying POI locations with land use planning data, public resource locations within a 50-meter radius of noise-sensitive land use are filtered out.

[0065] While ensuring that each inspection point is visited at least once daily, the overall construction and flight costs of the unmanned aerial vehicle (UAV) nest are minimized. Construction cost equals the fixed construction cost of the UAV nest and its associated communication equipment, converted to K yuan / day. Flight cost includes daily charging, maintenance, and inspection expenses, converted to the UAV nest inspection path length multiplied by the flight cost per unit distance. The objective function for optimization is specifically expressed as:

[0066] ;

[0067] Where F represents the set of selected nest points, y i Represents a binary variable, y i =1 indicates that nest i is selected; otherwise, it is 0. 总 K represents the total length of all inspection paths, W represents the cost of building the aircraft nest, and W represents the cost of flight per unit distance.

[0068] Obtain the constraints in the input target optimization process: The constraints include coverage constraints, path length constraints, path continuity constraints, and spatial condition constraints.

[0069] Coverage constraint: Each inspected object j must be covered by at least one drone nest i. To ensure signal connection between the drone and the drone nest, the maximum inspection range of the drone is within a 10km radius of the drone nest. The drone nest coverage constraint is specifically expressed as follows:

[0070] ;

[0071] Where S represents the set of inspection objects, a ij Represents a binary parameter. ij =1 indicates that nest i can cover the inspected object j (distance ≤ 10 km), otherwise it is 0;

[0072] Each inspection point s j It must be covered by at least one path, and the path coverage constraint is specifically expressed as:

[0073] ;

[0074] Among them, R i Let r represent the set of all inspection paths for nest i, r represent the path, and 𝕀 represent the indicator function.

[0075] Path length constraint: The distance of a single drone flight cannot exceed 80% of its maximum range; to ensure inspection efficiency and reduce drone wear and tear, each drone will not take off more than 10 times per day. It is assumed that the drone maintains a constant altitude and speed throughout the inspection process, and the flight distance during takeoff and landing is ignored. The path length constraint is specifically expressed as follows:

[0076] ;

[0077] ;

[0078] Where, d r E represents the distance a drone can fly in a single flight. max This indicates the maximum range of the drone.

[0079] Path continuity constraint: Each path starts from and returns to the nest; each inspection object can only be assigned to one path.

[0080] Airspace constraints: The drone inspection route cannot pass through no-fly zones, dangerous zones, or any type of restricted airspace.

[0081] The specific steps for solving the UAV nest location-path joint optimization problem using genetic algorithms include:

[0082] (1) For nest selection, in order to improve the solution efficiency, heuristic generation is adopted in the initial population generation, and greedy algorithm is used to solve the candidate nest selection scheme covering all inspection points as high-quality initial individuals.

[0083] (2) For inspection path planning, a combination of rasterization and the A* algorithm is used to solve the problem. The continuous geographic space is discretized into a raster, and the feasible path of the UAV is planned on the raster map using the A* algorithm to avoid no-fly zones. Inspection points are assigned to each UAV nest and the flight sequence is planned. The total inspection path length is calculated using the objective function of the total inspection path length to obtain candidate flight paths. Comprehensive inspections generally cover a wide range and involve many inspection objects. The airspace no-fly zones, airspace danger zones, and various airspace restricted zones constitute a complex structure. Spatial rasterization can effectively reduce the amount of computation and improve computational efficiency.

[0084] Specific methods for discretizing continuous geographic space into rasters include:

[0085] Determine the map boundaries based on the coordinates of all nests and inspection points;

[0086] Based on the accuracy requirements of the drone flight, the grid size is set to 100m×100m;

[0087] Mark grid attributes, including: free grid (flyable area), no-fly grid (grid that intersects with no-fly zone vector data), and nest / checkpoint grid (grid location of nest and checkpoint).

[0088] The A* algorithm is used to find the shortest feasible path from the starting point (nest) to the target point (inspection point) on the grid map, avoiding the no-fly zone.

[0089] Assign inspection points to each nacelle and plan flight sequences to meet range constraints, specifically including:

[0090] Assign inspection points to nests: Based on the nest's coverage area (10km radius), assign inspection points to the nearest nest;

[0091] Single-nest path planning: For each nest's inspection point set, the grid path length is calculated based on the A* algorithm. A distance matrix is ​​constructed based on the path lengths, and the shortest inspection path is found using either the TSP algorithm (used when a single flight distance is within 80% of the drone's maximum range and can cover all inspection points) or the VRP algorithm (used when a single flight cannot cover all inspection points due to range limitations and needs to be split into multiple flights). The Traveling Salesman Problem (TSP) involves a salesman starting from one of n cities, visiting the remaining n-1 cities without repetition, and returning to his starting point. The goal is to find the shortest possible path. The Vehicle Routing Problem (VRP) involves optimizing vehicle routes to achieve the shortest distance, lowest cost, or least time while meeting customer needs. VRP is a combinatorial optimization problem and a hot topic in operations research and combinatorial optimization.

[0092] Constraint handling: If a single flight path exceeds 80% of the drone's maximum range, it will be split into multiple flights; all paths are generated on a grid map using A* to avoid no-fly zones.

[0093] Calculating the total inspection path length: The objective function for calculating the total inspection path length is specifically expressed as:

[0094] ;

[0095] ;

[0096] Among them, L ir D represents the length of the r-th path of nest i. i Indicates the position of nest i, s r1 , ..., s rn This represents the inspection points visited sequentially by path r, and d(.,.) represents the actual flight distance between the two points.

[0097] (3) Use genetic algorithm to find the best candidate flight path and candidate nest location scheme to obtain the location set and inspection intersection set for each UAV nest.

[0098] The first embodiment of this invention provides a method for optimizing the configuration of UAV nest space for comprehensive inspection. By comprehensively considering the inspection targets and the rational utilization of public resources, a set of inspection take-off and landing points is determined. While ensuring that each inspection point is visited at least once daily, the optimization objective is to minimize the combined construction and flight costs of UAV nests. A genetic algorithm is used to perform joint optimization of UAV nest location and path selection, obtaining the location set of each UAV nest and the set of inspection intersections. This method can obtain the location selection of UAV nests for joint inspection of multiple targets such as forests, power grids, transportation, and urban areas, providing a reference for the deployment and construction of comprehensive inspection UAVs.

[0099] like Figure 2 As shown, another embodiment of the present invention provides a UAV nest space optimization configuration system for comprehensive inspection, which is used to implement the UAV nest space optimization configuration method for comprehensive inspection described in the first embodiment above. The system includes: a first acquisition module, a construction module, a filtering module, an optimization target determination module, a second acquisition module, and an optimization module.

[0100] The first acquisition module is used to acquire the inspection object, acquire the geographical data of the actual inspection range of the inspection object, and perform point-based discretization processing on the geographical data according to the inspection accuracy requirements, and use the point set to represent the inspection requirements.

[0101] The construction module is used to obtain public resource points (POIs) and construct a set of candidate locations for comprehensive inspection take-off and landing points.

[0102] The filtering module is used to filter out a set of alternative take-off and landing sites from the candidate site set based on limiting factors, where both the construction environment and take-off conditions are suitable.

[0103] The optimization target determination module is used to take the lowest combined cost of drone nest construction and flight cost as the optimization target, based on the premise that each inspection demand point is visited at least once a day.

[0104] The second acquisition module is used to acquire the constraints in the input target optimization process;

[0105] The optimization module uses a genetic algorithm to solve the joint optimization of UAV nest location and path selection, and obtains the location set and inspection intersection set of each UAV nest.

[0106] This invention provides a UAV nest space optimization configuration system for comprehensive inspection. By comprehensively considering the inspection targets and the rational utilization of public resources, it determines the set of inspection take-off and landing points. While ensuring that each inspection point is visited at least once daily, it uses the lowest combined cost of UAV nest construction and flight as the optimization objective. A genetic algorithm is used to perform joint optimization of UAV nest location and path selection, obtaining the location set of each UAV nest and the set of inspection intersections. This system can obtain the location selection of UAV nests for joint inspection of multiple targets such as forests, power grids, transportation, and urban areas, providing a reference for the deployment and construction of comprehensive inspection UAVs.

[0107] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing configuration of a UAV nest space for comprehensive inspection, characterized in that, The method comprises the following steps: acquiring an inspection object, acquiring geographical data of an actual inspection range of the inspection object according to the inspection object, performing point-based discretization processing on the geographical data according to an inspection accuracy requirement, and representing an inspection requirement by using a point set; acquiring public resource interest points and constructing a comprehensive inspection take-off and landing point candidate set; filtering out a set of selected take-off and landing points suitable for construction environment and take-off conditions from the candidate point set according to limiting factors, wherein the limiting factors include construction conditions, meteorological conditions, airspace conditions and noise sensitivity; on the basis that each inspection requirement point is visited at least once per day, comprehensively minimizing a UAV nest construction cost and a flight cost as an optimization target; acquiring input constraint conditions in a target optimization process; performing UAV nest site selection-path joint optimization solving by using a genetic algorithm to obtain a position set of each UAV nest and an inspection intersection set; the UAV nest site selection-path joint optimization solving by using the genetic algorithm specifically comprises: solving a candidate nest site selection scheme covering all inspection points by using a greedy algorithm as a high-quality initial individual; discretizing continuous geographical space into grids, planning a feasible path of the UAV on the grid map by using an A* algorithm, avoiding a no-fly zone, assigning inspection points to each nest and planning a flight sequence, calculating a total inspection path length by using a total path length objective function, and obtaining a candidate flight path; optimizing the candidate flight path and the candidate nest site selection scheme by using a genetic algorithm to obtain the position set of each UAV nest and the inspection intersection set; the method of discretizing the continuous geographical space into grids specifically comprises: determining a boundary of the map according to coordinates of all nests and inspection points; setting a grid size as 100m*100m according to a flight accuracy requirement of the UAV; labeling grid attributes; the method of assigning the inspection points to each nest and planning the flight sequence specifically comprises: assigning the inspection points to the nearest nest according to a coverage range of the nest; calculating a grid path length based on the A* algorithm for the inspection point set of each nest, constructing a distance matrix according to the path length, and solving a shortest inspection path by using a TSP algorithm or a VRP algorithm; the constraint conditions further include a path length constraint, and a take-off frequency of each UAV per day is not more than 10 times, assuming that the UAV keeps the same height and uniform speed during the whole inspection process, and ignoring a flight mileage of a take-off and landing process of the UAV, and the path length constraint condition is specifically expressed as: ； ； Wherein, d r represents the single flight distance of the UAV, E max represents the maximum endurance mileage of the UAV. 2.The method of claim 1, wherein, a target function formula of the optimization target is: ； where F represents the selected nest point set, y i is a binary variable, y i =1 represents selecting nest i, otherwise 0, L 总 represents the total length of all inspection paths, K represents the nest construction cost, and W represents the flight cost per unit distance. 3.The method of claim 2, wherein, the constraint conditions include a coverage constraint, and each inspection object j must be covered by at least one UAV nest i, and a UAV nest coverage constraint is specifically expressed as: ； where S denotes a set of inspection objects, a ij denotes a binary parameter, a ij =1 indicates that nest i can cover inspection object j, otherwise 0. Each inspection point s j At least covered by one path, the path coverage constraint is specifically represented as: ； wherein R i represents all the inspection path sets of the machine nest i, and r represents a path. 4.The method of claim 1, wherein, the constraint conditions further include a path continuity constraint and an airspace condition constraint.

5. A comprehensive inspection-oriented unmanned aerial vehicle nest space optimization configuration system, characterized in that, The system for implementing the comprehensive inspection-oriented UAV nest spatial optimization configuration method according to any one of claims 1-4 comprises a first acquisition module, a construction module, a filtering module, an optimization target determination module, a second acquisition module and an optimization module. The first obtaining module is configured to obtain a patrol object, obtain actual patrol range geographic data of the patrol object according to the patrol object, perform point-based discretization processing on the geographic data according to a patrol precision requirement, and represent the patrol requirement by using a point set; The constructing module is configured to obtain a public resource interest point, and construct a comprehensive patrol take-off and landing point candidate point set; The screening module is configured to screen, from the candidate point set, a set of selected take-off and landing points suitable for construction environment and take-off conditions according to a limiting factor; The optimization target determining module is configured to take, as an optimization target, a comprehensive minimum of a UAV nest construction cost and a flight cost on the basis that each patrol requirement point is visited at least once per day; The second obtaining module is configured to obtain an input constraint condition in a target optimization process; The optimization module adopts a genetic algorithm to perform joint optimization and solving of UAV nest site selection and a patrol path, and obtains a position set of each UAV nest and a patrol intersection set.

Images