A method and system for airspace adaptive gridding for unmanned aerial vehicle flight

Abstract

The application discloses an airspace adaptive gridding method and system for unmanned aerial vehicle flight, and relates to the technical field of airspace data processing.The method comprises the following steps: obtaining multiple sets of maneuverability parameters of multiple types of unmanned aerial vehicles, and configuring an initial grid size; evaluating the terrain complexity of a region to be gridded to obtain a grid size correction coefficient distribution map; correcting the initial grid size according to the grid size correction coefficient to obtain a regular grid size of the unmanned aerial vehicle; collecting historical obstacle avoidance data of the unmanned aerial vehicle in the region to be gridded within a historical time, and marking an obstacle avoidance hotspot region in the region to be gridded; if the region where the unmanned aerial vehicle is located is the obstacle avoidance hotspot region, the grid size is reduced to obtain a final grid size; if the region where the unmanned aerial vehicle is located is a non-obstacle avoidance hotspot region, the regular grid size is taken as the final grid size, and airspace adaptive gridding is performed.The application solves the technical problem of poor adaptability of the airspace gridding result in the prior art.

Description

Technical Field

[0001] This invention relates to the field of airspace data processing technology, specifically to an airspace adaptive gridding method and system for unmanned aerial vehicle (UAV) flights. Background Technology

[0002] Current UAV airspace data processing commonly employs a fixed-size static gridding method, dividing the airspace into uniform three-dimensional grid units to standardize flight paths and obstacle avoidance logic. However, this technology has significant drawbacks: First, the fixed grid size cannot adapt to the maneuverability differences of different UAV models, resulting in excessive restrictions on UAVs and wasted airspace resources; second, existing methods ignore the dynamic requirements of large amounts of data from different terrains for grid precision, making precise static division difficult to process, leading to poor adaptability and insufficient accuracy in airspace data processing. Summary of the Invention

[0003] This application provides an airspace adaptive gridding method and system for UAV flight, which is used to address the technical problem of poor adaptability of airspace gridding results in the prior art.

[0004] In view of the above problems, this application provides an airspace adaptive gridding method and system for UAV flight.

[0005] In a first aspect, this application provides an airspace adaptive gridding method for unmanned aerial vehicle (UAV) flight, the method comprising:

[0006] Multiple sets of maneuverability parameters for various types of UAVs are obtained, a mapping relationship between maneuverability parameters and initial grid size is established, and initial grid size is configured for each type of UAV based on the multiple sets of maneuverability parameters.

[0007] Obtain a high-precision map of the area to be gridded, assess the terrain complexity of the area based on the high-precision map, obtain a terrain complexity distribution map of the area to be gridded, and calculate the grid size correction coefficient distribution map based on the terrain complexity distribution map.

[0008] When a specific model of drone enters the area to be gridded, the initial grid size corresponding to the drone is obtained according to the model, the grid size correction coefficient distribution map of the area to be gridded is retrieved, the grid size correction coefficient is obtained according to the real-time position of the drone, and the initial grid size is corrected according to the grid size correction coefficient to obtain the conventional grid size of the drone.

[0009] Collect historical obstacle avoidance data of UAVs in the area to be gridded over a historical period, and mark obstacle avoidance hotspots in the area to be gridded based on the historical obstacle avoidance data of UAVs.

[0010] If the area where the drone is located is an obstacle avoidance hotspot, the grid size is reduced based on the regular grid size to obtain the final grid size; if the area where the drone is located is not an obstacle avoidance hotspot, the regular grid size is used as the final grid size, and airspace adaptive gridding is performed.

[0011] Secondly, this application provides an airspace adaptive gridding system for unmanned aerial vehicle (UAV) flight, including:

[0012] The initial grid size acquisition module is used to obtain multiple sets of maneuverability parameters for various types of UAVs, establish a mapping relationship between maneuverability parameters and initial grid size, and configure initial grid size for each type of UAV based on the multiple sets of maneuverability parameters.

[0013] The grid size correction module is used to obtain a high-precision map of the area to be gridded, evaluate the terrain complexity of the area to be gridded based on the high-precision map, obtain a terrain complexity distribution map of the area to be gridded, and calculate the grid size correction coefficient distribution map based on the terrain complexity distribution map.

[0014] The conventional grid size acquisition module is used to obtain the initial grid size corresponding to the drone model when a specific model of drone enters the area to be gridded, retrieve the grid size correction coefficient distribution map of the area to be gridded, obtain the grid size correction coefficient according to the real-time position of the drone, and correct the initial grid size according to the grid size correction coefficient to obtain the conventional grid size of the drone.

[0015] The hotspot area marking module is used to collect historical obstacle avoidance data of UAVs in the area to be gridded over a historical period, and mark obstacle avoidance hotspot areas in the area to be gridded based on the historical obstacle avoidance data of UAVs.

[0016] The adaptive gridding module is used to reduce the grid size from the standard grid size to obtain the final grid size if the area where the UAV is located is an obstacle avoidance hotspot area; if the area where the UAV is located is not an obstacle avoidance hotspot area, the standard grid size is used as the final grid size for airspace adaptive gridding.

[0017] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0018] This application proposes an adaptive airspace gridding method and system for UAV flight. By dynamically fusing UAV maneuverability parameters, terrain complexity, and historical obstacle avoidance hotspot data, it significantly improves the adaptive allocation of airspace resources and flight safety. Compared with traditional methods, the technical solution provided in this application significantly overcomes the rigidity of airspace utilization caused by static gridding, achieving the technical effect of adaptive gridding of airspace resources based on multi-source parameters. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the airspace adaptive lattice method for UAV flight provided in an embodiment of this application.

[0021] Figure 2 This is a schematic diagram of the airspace adaptive gridding system for UAV flight provided in an embodiment of this application.

[0022] The components represented by each number in the attached diagram are explained below:

[0023] The module includes: initial grid size acquisition module 100, grid size correction module 200, regular grid size acquisition module 300, hotspot area marking module 400, and adaptive gridding module 500. Detailed Implementation

[0024] This application provides an airspace adaptive lattice method and system for UAV flight, which addresses the technical problem of poor adaptability of airspace lattice results in the prior art.

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0026] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0027] Example 1, as Figure 1 As shown, this application provides an airspace adaptive gridding method for UAV flight, wherein the method includes:

[0028] S10: Obtain multiple sets of maneuverability parameters for various types of UAVs, establish a mapping relationship between maneuverability parameters and initial grid size, and configure initial grid size for each type of UAV based on the multiple sets of maneuverability parameters.

[0029] Existing static gridding methods divide airspace with uniform dimensions, which cannot adapt to the different maneuverability differences of different UAV models. For example, less maneuverable models cannot avoid obstacles in time in an excessively large grid due to insufficient capabilities, increasing the risk of collision; while more maneuverable models are forced to frequently adjust their paths in an excessively dense grid, resulting in wasted airspace resources and reduced flight efficiency.

[0030] Step S10 in the method provided in this application embodiment includes:

[0031] Multiple sets of maneuverability parameters for various UAV models were obtained, including minimum turning radius, maximum rate of climb, and maximum rate of descent.

[0032] Based on the minimum turning radius, determine the horizontal dimension of the initial grille size for a specific model of UAV;

[0033] The height dimension of the initial grille size for a specific model of UAV is determined based on the maximum climb rate and maximum descent rate.

[0034] By combining the horizontal and vertical dimensions of the initial grille size for a specific drone model, the initial grille size for that drone model can be obtained.

[0035] Based on the aforementioned multiple sets of mobility parameters, initial grid sizes are configured for each type of UAV.

[0036] In this embodiment of the application, multiple sets of maneuverability parameters of various UAV models are obtained based on the design parameters of various UAV models. One set of maneuverability parameters includes the minimum turning radius in meters; the maximum rate of climb in meters per second; and the maximum rate of descent in meters per second.

[0037] Based on the minimum turning radius, determine the horizontal dimension of the initial grille size for a specific drone model. For example, the horizontal dimension of the initial grille size = the minimum turning radius of the specific drone model × a safety factor. The safety factor is a constant set to prevent accidents caused by an excessively small grille; for example, the safety factor can be set to 1.5.

[0038] Based on the maximum climb rate and maximum descent rate, the initial grid height dimension of a specific UAV model is determined. Specifically, the maximum climb rate and maximum descent rate of a specific UAV model are compared, and the larger of the two rates is selected for the height dimension calculation. For example, if the descent rate is larger than the maximum climb rate, then the initial grid height dimension = descent rate × descent unit time. The descent unit time can be set to 2 seconds to obtain the initial grid height dimension.

[0039] By combining the horizontal and vertical dimensions of the initial grille size for a specific drone model, the initial grille size for that drone model is obtained, where the initial grille size is calculated as horizontal dimension × horizontal dimension × vertical dimension, in meters.

[0040] Based on multiple sets of mobility parameters, initial grid sizes were configured for various types of UAVs.

[0041] By establishing a mapping relationship between maneuverability parameters and initial grid size, initial grid allocation based on aircraft capabilities is achieved. The horizontal grid size is determined based on the minimum turning radius, and the vertical dimension is set in conjunction with the climb and fall rate, ensuring that the grid size accurately matches the UAV's performance and fundamentally avoiding conflicts between maneuverability and grid constraints.

[0042] S20: Obtain a high-precision map of the area to be gridded, evaluate the terrain complexity of the area to be gridded based on the high-precision map, obtain a terrain complexity distribution map of the area to be gridded, and calculate the grid size correction coefficient distribution map based on the terrain complexity distribution map.

[0043] Traditional methods neglect the dynamic requirements of complex parameters such as terrain and environment on grid precision. For example, in densely built-up urban areas, the distribution of obstacles is complex, requiring smaller grid sizes to improve obstacle avoidance accuracy; while using the same grid in open suburban areas may lead to redundant computing power.

[0044] Step S20 in the method provided in this application embodiment includes:

[0045] The area to be gridded is pre-divided into multiple basic grids, and the terrain complexity is calculated in units of the basic grids.

[0046] Calculate the ratio of the total floor area of ​​all buildings in each basic grid to the area of ​​the basic grid to obtain the building area density of each basic grid;

[0047] Calculate the ratio of the total number of buildings in each basic grid to the area of ​​the basic grid to obtain the building density of each basic grid;

[0048] Calculate the standard deviation of the height of all buildings within each base grid to obtain the building height difference of each base grid;

[0049] The building area density, building number density, and building height difference values ​​corresponding to each basic grid are normalized and then weighted and summed to obtain the terrain complexity corresponding to each basic grid. The terrain complexity value ranges from greater than 0 to less than or equal to 1.

[0050] Within the high-precision map of the area to be gridded, the range of each basic grid and its corresponding terrain complexity are marked to obtain the terrain complexity distribution map;

[0051] Extract the terrain complexity corresponding to each basic grid in the terrain complexity distribution map, obtain the distribution range of multiple terrain complexity values, and obtain the terrain complexity distribution interval;

[0052] A preset range of grid size correction coefficients is established, and a mapping relationship is created between the terrain complexity distribution range and the range of grid size correction coefficients.

[0053] Based on the terrain complexity corresponding to each basic grid, the grid size correction coefficient corresponding to each basic grid is obtained by mapping.

[0054] In this embodiment, a publicly available high-precision map, such as OpenStreetMap, is obtained, and the area to be gridded is pre-divided into multiple basic grids. The terrain complexity is calculated using the basic grids as units. For example, the area to be gridded can be divided into an average of 30m × 30m basic grids.

[0055] Calculate the ratio of the total floor area of ​​all buildings in each basic grid to the area of ​​the basic grid to obtain the building area density of each basic grid. The building area density of the basic grid = total floor area of ​​buildings in the grid ÷ grid area.

[0056] Calculate the ratio of the total number of buildings in each basic grid to the area of ​​the basic grid to obtain the building density of each basic grid. The building density of the basic grid = total number of buildings in the grid ÷ grid area, with the unit being buildings / square meter.

[0057] Calculate the standard deviation of the height of all buildings within each base grid to obtain the building height variation of each base grid.

[0058] For each basic grid, the building area density, building number density, and building height difference values ​​are normalized and then weighted and summed to obtain the terrain complexity for each basic grid. The terrain complexity value ranges from greater than 0 to less than or equal to 1. Specifically, the average building area density, average building number density, and average building height difference for all basic grids are calculated. Normalized building area density = basic grid building density ÷ average building area density; normalized building number density = basic grid building number density ÷ average building number density; normalized building height difference = basic grid height difference ÷ average height difference. The normalized results are then weighted and summed to obtain the terrain complexity. Preferably, the terrain complexity = 0.35 × normalized building area density + 0.35 × normalized building number density + 0.3 × normalized building height difference.

[0059] Within a high-precision map of the area to be gridded, the extent of each basic grid and its corresponding terrain complexity are marked to obtain a terrain complexity distribution map.

[0060] Extract the terrain complexity corresponding to each basic grid in the terrain complexity distribution map, obtain the distribution range of multiple terrain complexity values, and obtain the terrain complexity distribution interval.

[0061] A preset range of grid size correction coefficients is established to map the distribution range of terrain complexity to the range of grid size correction coefficients. For example, the range of grid size correction coefficients is set to -1 to 1, which means that the range of grid correction is from shrinking by half to enlarging by half.

[0062] Based on the terrain complexity corresponding to each basic grid, the grid size correction coefficient corresponding to each basic grid is obtained by mapping. For example, if the terrain complexity is 0.2, then 0.2 is mapped to the preset grid size correction coefficient range, and the obtained grid size correction coefficient = 0.2 ÷ 1 × [1 - (-1)] = 0.4.

[0063] Within a high-precision map of the area to be gridded, the range of each basic grid and its corresponding grid size correction coefficient are marked to obtain a grid size correction coefficient distribution map.

[0064] By analyzing high-precision maps, a terrain complexity distribution map is generated and converted into grid correction coefficients, enabling environmentally adaptive dynamic scaling of the grid. Based on parameters such as building density and height differences, terrain complexity is accurately quantified. In areas with complex buildings, the grid size is automatically reduced to improve obstacle recognition accuracy, while in flat areas, the grid size is increased to optimize computational efficiency.

[0065] S30: When a specific model of UAV enters the area to be gridded, obtain the initial grid size corresponding to the UAV based on the model, retrieve the grid size correction coefficient distribution map of the area to be gridded, obtain the grid size correction coefficient based on the real-time position of the UAV, and correct the initial grid size based on the grid size correction coefficient to obtain the conventional grid size of the UAV.

[0066] In this embodiment, when a specific model of UAV enters the area to be gridded, the initial grid size corresponding to the UAV is obtained according to the model, and the grid size correction coefficient distribution map of the area to be gridded is retrieved. Based on the UAV's real-time position, the grid size correction coefficient is obtained, and the initial grid size is corrected according to the grid size correction coefficient to obtain the UAV's conventional grid size. When the grid size correction coefficient is less than 0, the conventional grid size = initial grid size × (1 + 0.5 × grid size correction coefficient); when the grid size correction coefficient is equal to 0, the conventional grid size = initial grid size; when the grid size correction coefficient is greater than 0, the conventional grid size = initial grid size × (1 + grid size correction coefficient). For example, if an initial grid size is 30 meters × 30 meters and the grid size correction coefficient is -0.4, then the conventional grid size = 30 × [1 + 0.5 × (-0.4)] = 24, that is, the corrected conventional grid size is 24 meters × 24 meters.

[0067] By fusing the initial grid size with correction coefficients based on the real-time position, a regular grid size that evolves with the terrain in real time is generated. When the UAV enters the area to be gridded, the correction coefficients corresponding to the current position are immediately retrieved, and the initial size is dynamically scaled to improve the adaptability of the gridding process.

[0068] S40: Collect historical obstacle avoidance data of UAVs in the area to be gridded over a historical period, and mark obstacle avoidance hotspots in the area to be gridded based on the historical obstacle avoidance data of UAVs.

[0069] Historical obstacle avoidance data contains accident patterns. For example, high-frequency accident points such as densely built-up areas and blind spots are not marked, leading to the recurrence of similar obstacle avoidance events.

[0070] Step S40 in the method provided in this application embodiment includes:

[0071] Based on the flight logs of the UAV flying in the area to be gridded within a historical period, historical obstacle avoidance events are extracted. The historical obstacle avoidance events include the coordinates of the obstacle avoidance position when the obstacle avoidance occurs and N historical obstacle avoidance features of N obstacle avoidance attempts at the corresponding obstacle avoidance position. The historical obstacle avoidance features include the maneuverability parameter features, flight parameter features, obstacle avoidance time features, and obstacle avoidance weather features of the corresponding obstacle avoidance UAV. Among them, the maneuverability parameter features include the minimum turning radius, the maximum climb rate, and the maximum descent rate.

[0072] The maneuverability parameters, flight parameters, current time and weather characteristics of a specific UAV model are obtained as features to be compared.

[0073] Clustering algorithms are used to cluster the feature to be compared with N historical obstacle avoidance features. Multiple similar historical obstacle avoidance features that are similar to the feature to be compared are extracted from the N historical obstacle avoidance features. Multiple obstacle avoidance position coordinates that correspond one-to-one with the multiple similar historical obstacle avoidance features are obtained as multiple obstacle avoidance hotspot coordinates.

[0074] Specifically, a clustering algorithm is used to cluster the feature to be compared with N historical obstacle avoidance features, and multiple similar historical obstacle avoidance features that are similar to the feature to be compared are extracted from the N historical obstacle avoidance features, including:

[0075] The features to be compared and N historical obstacle avoidance features are grouped into the same set to obtain the sample obstacle avoidance dataset.

[0076] The K-means clustering algorithm is used to divide the obstacle avoidance dataset into K clusters based on Euclidean distance, where K is a positive integer greater than or equal to 2.

[0077] Extract the cluster where the feature to be compared is located, and obtain multiple historical obstacle avoidance features that are classified into the same cluster as the feature to be compared, as multiple similar historical obstacle avoidance features;

[0078] Mark the base grid containing the coordinates of multiple obstacle avoidance hotspots as multiple obstacle avoidance hotspot regions;

[0079] This involves marking the base grid containing multiple obstacle avoidance hotspot coordinates as multiple obstacle avoidance hotspot regions, including:

[0080] Count the number of obstacle avoidance hotspot coordinates contained in each basic grid;

[0081] Based on the number of obstacle avoidance hotspot coordinates contained in each basic grid and a preset threshold for the number of obstacle avoidance hotspot coordinates, grids containing obstacle avoidance hotspot coordinates are labeled with obstacle avoidance heat indicators. The obstacle avoidance heat indicators include multiple obstacle avoidance heat levels, and multiple basic grids with obstacle avoidance heat indicators are regarded as multiple obstacle avoidance hotspot regions.

[0082] In this embodiment, historical obstacle avoidance events are extracted based on the flight logs of the UAV flying in the area to be gridded within a historical time period. These historical obstacle avoidance events include the coordinates of the obstacle avoidance location when the obstacle avoidance occurred, and N historical obstacle avoidance features representing N obstacle avoidance attempts at the corresponding location. The historical obstacle avoidance features include the maneuverability parameters of the corresponding obstacle avoidance UAV, flight parameters such as flight speed and altitude, obstacle avoidance time features (e.g., 10:00 AM, as different times can affect visibility or interference signal strength), obstacle avoidance weather features (e.g., sunny or rainy), and maneuverability parameters including minimum turning radius, maximum climb rate, and maximum descent rate.

[0083] The maneuverability parameters, flight parameters, current time and weather characteristics of a specific UAV model are obtained as features to be compared.

[0084] The features to be compared are grouped together with N historical obstacle avoidance features into the same set to obtain the sample obstacle avoidance dataset.

[0085] The K-means clustering algorithm is used to divide the obstacle avoidance dataset into K clusters based on Euclidean distance, where K is a positive integer greater than or equal to 2.

[0086] Extract the cluster to which the feature to be compared belongs, and obtain multiple historical obstacle avoidance features that are classified into the same cluster as the feature to be compared, as multiple similar historical obstacle avoidance features.

[0087] The number of obstacle avoidance hotspot coordinates contained in each basic grid is counted. These hotspot coordinates are defined as the coordinates of drones with similar obstacle avoidance characteristics to a specific drone model that have repeatedly triggered obstacle avoidance within the gridded area within a preset historical time window. For example, obstacle avoidance hotspot coordinates can be defined as the coordinates of drones with similar obstacle avoidance characteristics to a specific drone model that have triggered obstacle avoidance more than 10 times within the gridded area in the past 60 days.

[0088] Based on the number of obstacle avoidance hotspot coordinates in each basic grid and a preset threshold for the number of obstacle avoidance hotspot coordinates, grids containing obstacle avoidance hotspot coordinates are labeled with obstacle avoidance heat indicators. The preset threshold for the number of obstacle avoidance hotspot coordinates refers to the threshold for the number of hotspot coordinates used to classify obstacle avoidance heat levels. The obstacle avoidance heat indicator includes multiple obstacle avoidance heat levels. For example, obstacle avoidance heat can be represented as high-level, medium-level, and low-level. The threshold for the number of obstacle avoidance hotspot coordinates for high-level can be set to 6, meaning basic grids containing 6 or more obstacle avoidance hotspot coordinates are labeled as high-level obstacle avoidance heat. The threshold for the number of obstacle avoidance hotspot coordinates for medium-level can be set to 3-5, meaning basic grids containing 3 or more but less than or equal to 5 obstacle avoidance hotspot coordinates are labeled as medium-level obstacle avoidance heat. The threshold for the number of obstacle avoidance hotspot coordinates for low-level can be set to 2, meaning basic grids containing 2 or fewer obstacle avoidance hotspot coordinates are labeled as low-level obstacle avoidance heat. Multiple basic grids with obstacle avoidance heat indicators are considered as multiple obstacle avoidance hotspot regions.

[0089] By analyzing the spatiotemporal characteristics of historical events, a predictive model for obstacle avoidance hotspot areas is constructed. Obstacle avoidance location coordinates and associated aircraft parameters, flight status, and other features are extracted, and hotspot areas are clustered and labeled on a map, providing data support for the proactive optimization of airspace grids.

[0090] S50: If the area where the UAV is located is an obstacle avoidance hotspot, the grid size is reduced based on the regular grid size to obtain the final grid size; if the area where the UAV is located is not an obstacle avoidance hotspot, the regular grid size is used as the final grid size for airspace adaptive gridding.

[0091] Existing methods often use conventional grids in hotspot areas, which cannot take into account the correlation between aircraft models and accidents in hotspot areas. This results in grid optimization lacking specificity and the optimized grid lacking adaptability.

[0092] Step S50 in the method provided in this application embodiment includes:

[0093] Retrieve the maneuverability parameters of a specific model of UAV;

[0094] Retrieve multiple maneuvering parameters of multiple obstacle avoidance drones corresponding to multiple obstacle avoidance hotspot coordinates within the basic grid of a specific drone model, and use them as a set of historical obstacle avoidance maneuvering parameters;

[0095] Based on predefined comparison rules, the proportion of multiple maneuverability parameters in the historical obstacle avoidance maneuverability parameter set that are superior to those of a specific UAV model is calculated, and this proportion is used as the maneuverability risk level.

[0096] Based on the obstacle avoidance heat index, set the initial grid size reduction ratio for each type of obstacle avoidance heat index;

[0097] Obtain the obstacle avoidance heat map of the obstacle avoidance hotspot area where a specific drone model is located, and the corresponding initial grid size reduction ratio;

[0098] Based on the maneuverability risk level of a specific UAV model in obstacle avoidance hotspot areas, the initial grille size reduction ratio is adjusted to obtain the grille size reduction ratio.

[0099] Based on the reduction ratio of the grille size of a specific model of UAV entering the obstacle avoidance hotspot area, the regular grille size is reduced to obtain the final grille size.

[0100] In this embodiment of the application, the maneuverability parameters of a specific model of UAV are retrieved, including the minimum turning radius, maximum climb rate, and maximum descent rate.

[0101] Retrieve multiple maneuvering parameters of multiple obstacle avoidance drones corresponding to multiple obstacle avoidance hotspot coordinates within the basic grid of a specific drone model, and use them as a historical obstacle avoidance maneuvering parameter set.

[0102] Based on predefined comparison rules, the proportion of multiple maneuverability parameters in the historical obstacle avoidance maneuverability parameter set that are superior to those of a specific UAV model is calculated as the maneuverability hazard level. For example, a weighted sum is performed on the proportion of maneuverability parameters superior to those of a specific UAV model. First, a smaller turning radius, a higher maximum climb rate, and a higher maximum descent rate are defined as superior maneuverability parameters. If a specific model of UAV has a turning radius of 2 meters, a maximum climb rate of 3 meters per second, and a maximum descent rate of 3 meters per second in its maneuverability parameters, and its historical obstacle avoidance maneuverability parameters include a turning radius of 1 meter, a maximum climb rate of 4 meters per second, and a maximum descent rate of 4 meters per second, then the following ratios are considered excellent: Turning radius of specific model UAV - Turning radius of historical obstacle avoidance UAVs ÷ Turning radius of specific model UAV = (2-1) ÷ 2 = 0.5; Maximum climb rate of excellent ratio = (Maximum climb rate of historical obstacle avoidance UAVs - Maximum climb rate of specific model UAV) ÷ Maximum climb rate of specific model UAV = (4-3) ÷ 3 = 0.33; Maximum descent rate of excellent ratio = (Maximum descent rate of historical obstacle avoidance UAVs - Maximum descent rate of specific model UAV) ÷ Maximum descent rate of specific model UAV = (4-3) ÷ 3 = 0.33. The three ratios are then weighted and summed to obtain the overall excellent maneuverability parameters. For example, the excellent maneuverability comprehensive parameter = 0.3 × excellent turning radius ratio + 0.35 × excellent maximum climb rate ratio + 0.35 × excellent maximum descent rate ratio = 0.3 × 0.5 + 0.35 × 0.33 + 0.35 × 0.33 = 0.38. A threshold for the excellent maneuverability parameter can be set, such as defining an excellent maneuverability comprehensive parameter greater than 0.2 as superior to that of a specific UAV model. The maneuverability risk level = number of historical obstacle avoidance maneuverability parameters with an excellent maneuverability comprehensive parameter greater than 0.2 ÷ total number of historical obstacle avoidance maneuverability parameters. For example, if 600 out of 1000 historical obstacle avoidance maneuverability parameters have an excellent maneuverability comprehensive parameter greater than 0.2, then the maneuverability risk level = 600 ÷ 1000 = 0.6. A higher maneuverability risk level indicates a higher maneuverability requirement for obstacle avoidance within these obstacle avoidance hotspot coordinates.

[0103] Based on the obstacle avoidance heat index, set the initial grid size reduction ratio for each obstacle avoidance heat index. For example, the reduction ratio for high level can be set to 0.8, which means the grid size is reduced to 80% of the initial size; the reduction ratio for medium level can be set to 0.5; and the reduction ratio for low level can be set to 0.3, thus obtaining the initial grid size reduction ratio for each obstacle avoidance heat index.

[0104] Obtain the obstacle avoidance heat index of the obstacle avoidance hotspot area where a specific model of drone is located, as well as the corresponding initial grid size reduction ratio.

[0105] If the area where the drone is located is an obstacle avoidance hotspot, the grille size is reduced from the standard grille size to obtain the final grille size. Specifically, based on the maneuverability hazard level corresponding to a specific drone model in the obstacle avoidance hotspot area, the initial grille size reduction ratio is adjusted to obtain the grille size reduction ratio. The grille size reduction ratio = initial grille size reduction ratio × (1 - maneuverability hazard level). For example, if an obstacle avoidance hotspot area is marked as high-level, the initial grille size reduction ratio is 0.8, and the maneuverability hazard level is 0.6, then the grille size reduction ratio = 0.8 × (1 - 0.6) = 0.32, which means the grille size is reduced to 32% of the initial size.

[0106] Based on the reduction ratio of the grille size of a specific model of drone entering the obstacle avoidance hotspot area, the regular grille size is reduced to obtain the final grille size. For example, if the regular grille size is 24 meters × 24 meters, the reduced grille size = 24 × 0.32 = 7.68, and the final grille size is 7.68 meters × 7.68 meters.

[0107] If the area where the drone is located is not an obstacle avoidance hotspot, the standard grid size will be used as the final grid size for airspace adaptive gridding.

[0108] Based on obstacle avoidance hotspot markers and aircraft characteristics, risk-adaptive grid optimization is achieved. When a UAV enters a hotspot area, the size of the regular grid is reduced to enhance path planning precision. Furthermore, historical data is used to dynamically adjust the reduction ratio. By linking hotspot areas to aircraft risk, proactive optimization in hotspot areas significantly improves grid adaptability in complex airspace.

[0109] Example 2, as Figure 2 As shown, based on the same inventive concept as the airspace adaptive gridding method for UAV flight provided in Embodiment 1, this embodiment of the invention also provides an airspace adaptive gridding system for UAV flight, including:

[0110] The initial grid size acquisition module 100 is used to obtain multiple sets of maneuverability parameters for various types of UAVs, establish a mapping relationship between maneuverability parameters and initial grid size, and configure initial grid size for each type of UAV according to the multiple sets of maneuverability parameters.

[0111] The grid size correction module 200 is used to obtain a high-precision map of the area to be gridded, evaluate the terrain complexity of the area to be gridded based on the high-precision map, obtain a terrain complexity distribution map of the area to be gridded, and calculate the grid size correction coefficient distribution map based on the terrain complexity distribution map.

[0112] The conventional grid size acquisition module 300 is used to obtain the initial grid size corresponding to the drone model when a specific model of drone enters the area to be gridded, retrieve the grid size correction coefficient distribution map of the area to be gridded, obtain the grid size correction coefficient according to the real-time position of the drone, and correct the initial grid size according to the grid size correction coefficient to obtain the conventional grid size of the drone.

[0113] The hotspot area marking module 400 is used to collect historical obstacle avoidance data of UAVs in the area to be gridded over a historical period, and mark obstacle avoidance hotspot areas in the area to be gridded based on the historical obstacle avoidance data of UAVs.

[0114] The adaptive gridding module 500 is used to reduce the grid size based on the regular grid size to obtain the final grid size if the area where the UAV is located is an obstacle avoidance hotspot area; if the area where the UAV is located is not an obstacle avoidance hotspot area, the regular grid size is used as the final grid size for airspace adaptive gridding.

[0115] In one embodiment, the initial grid size acquisition module 100 is further configured to:

[0116] Multiple sets of maneuverability parameters for various UAV models were obtained, including minimum turning radius, maximum rate of climb, and maximum rate of descent.

[0117] Based on the minimum turning radius, determine the horizontal dimension of the initial grille size for a specific model of UAV;

[0118] The height dimension of the initial grille size for a specific model of UAV is determined based on the maximum climb rate and maximum descent rate.

[0119] By combining the horizontal and vertical dimensions of the initial grille size for a specific drone model, the initial grille size for that drone model can be obtained.

[0120] Based on the aforementioned multiple sets of mobility parameters, initial grid sizes are configured for each type of UAV.

[0121] In one embodiment, the grille size correction module 200 is further configured to:

[0122] The area to be gridded is pre-divided into multiple basic grids, and the terrain complexity is calculated in units of the basic grids.

[0123] Calculate the ratio of the total floor area of ​​all buildings in each basic grid to the area of ​​the basic grid to obtain the building area density of each basic grid;

[0124] Calculate the ratio of the total number of buildings in each basic grid to the area of ​​the basic grid to obtain the building density of each basic grid;

[0125] Calculate the standard deviation of the height of all buildings within each base grid to obtain the building height difference of each base grid;

[0126] The building area density, building number density, and building height difference values ​​corresponding to each basic grid are normalized and then weighted and summed to obtain the terrain complexity corresponding to each basic grid. The terrain complexity value ranges from greater than 0 to less than or equal to 1.

[0127] Within the high-precision map of the area to be gridded, the range of each basic grid and its corresponding terrain complexity are marked to obtain the terrain complexity distribution map;

[0128] Extract the terrain complexity corresponding to each basic grid in the terrain complexity distribution map, obtain the distribution range of multiple terrain complexity values, and obtain the terrain complexity distribution interval;

[0129] A preset range of grid size correction coefficients is established, and a mapping relationship is created between the terrain complexity distribution range and the range of grid size correction coefficients.

[0130] Based on the terrain complexity corresponding to each basic grid, the grid size correction coefficient corresponding to each basic grid is obtained by mapping.

[0131] In one embodiment, the hotspot area marking module 400 is further used for:

[0132] Based on the flight logs of the UAV flying in the area to be gridded within a historical period, historical obstacle avoidance events are extracted. The historical obstacle avoidance events include the coordinates of the obstacle avoidance position when the obstacle avoidance occurs and N historical obstacle avoidance features of N obstacle avoidance attempts at the corresponding obstacle avoidance position. The historical obstacle avoidance features include the maneuverability parameter features, flight parameter features, obstacle avoidance time features, and obstacle avoidance weather features of the corresponding obstacle avoidance UAV. Among them, the maneuverability parameter features include the minimum turning radius, the maximum climb rate, and the maximum descent rate.

[0133] The maneuverability parameters, flight parameters, current time and weather characteristics of a specific UAV model are obtained as features to be compared.

[0134] Clustering algorithms are used to cluster the feature to be compared with N historical obstacle avoidance features. Multiple similar historical obstacle avoidance features that are similar to the feature to be compared are extracted from the N historical obstacle avoidance features. Multiple obstacle avoidance position coordinates that correspond one-to-one with the multiple similar historical obstacle avoidance features are obtained as multiple obstacle avoidance hotspot coordinates.

[0135] Specifically, a clustering algorithm is used to cluster the feature to be compared with N historical obstacle avoidance features, and multiple similar historical obstacle avoidance features that are similar to the feature to be compared are extracted from the N historical obstacle avoidance features, including:

[0136] The features to be compared and N historical obstacle avoidance features are grouped into the same set to obtain the sample obstacle avoidance dataset.

[0137] The K-means clustering algorithm is used to divide the obstacle avoidance dataset into K clusters based on Euclidean distance, where K is a positive integer greater than or equal to 2.

[0138] Extract the cluster where the feature to be compared is located, and obtain multiple historical obstacle avoidance features that are classified into the same cluster as the feature to be compared, as multiple similar historical obstacle avoidance features;

[0139] Mark the base grid containing the coordinates of multiple obstacle avoidance hotspots as multiple obstacle avoidance hotspot regions;

[0140] This involves marking the base grid containing multiple obstacle avoidance hotspot coordinates as multiple obstacle avoidance hotspot regions, including:

[0141] Count the number of obstacle avoidance hotspot coordinates contained in each basic grid;

[0142] Based on the number of obstacle avoidance hotspot coordinates contained in each basic grid and a preset threshold for the number of obstacle avoidance hotspot coordinates, grids containing obstacle avoidance hotspot coordinates are labeled with obstacle avoidance heat indicators. The obstacle avoidance heat indicators include multiple obstacle avoidance heat levels, and multiple basic grids with obstacle avoidance heat indicators are regarded as multiple obstacle avoidance hotspot regions.

[0143] In one embodiment, the adaptive gridding module 500 is further configured to:

[0144] Retrieve the maneuverability parameters of a specific model of UAV;

[0145] Retrieve multiple maneuvering parameters of multiple obstacle avoidance drones corresponding to multiple obstacle avoidance hotspot coordinates within the basic grid of a specific drone model, and use them as a set of historical obstacle avoidance maneuvering parameters;

[0146] Based on predefined comparison rules, the proportion of multiple maneuverability parameters in the historical obstacle avoidance maneuverability parameter set that are superior to those of a specific UAV model is calculated, and this proportion is used as the maneuverability risk level.

[0147] Based on the obstacle avoidance heat index, set the initial grid size reduction ratio for each type of obstacle avoidance heat index;

[0148] Obtain the obstacle avoidance heat map of the obstacle avoidance hotspot area where a specific drone model is located, and the corresponding initial grid size reduction ratio;

[0149] Based on the maneuverability risk level of a specific UAV model in obstacle avoidance hotspot areas, the initial grille size reduction ratio is adjusted to obtain the grille size reduction ratio.

[0150] Based on the reduction ratio of the grille size of a specific model of UAV entering the obstacle avoidance hotspot area, the regular grille size is reduced to obtain the final grille size.

[0151] In summary, the embodiments of this application have at least the following technical effects:

[0152] This application proposes an adaptive gridding method and system for UAV flight. By dynamically fusing UAV maneuverability parameters, terrain complexity, and historical obstacle avoidance hotspot data, it significantly improves the adaptive allocation of airspace resources and flight safety. First, based on the initial grid configuration considering the maneuverability differences among various UAV models, it fundamentally adapts to the different flight capabilities of each model, preventing high-maneuverability models from being constrained by overly dense grids and wasting airspace resources. Second, relying on the dynamic terrain complexity correction mechanism of high-precision maps, it automatically reduces the grid size in high-risk environments such as densely built-up areas and complex terrain to improve obstacle avoidance accuracy, while appropriately increasing the grid size in open areas to optimize path planning efficiency, achieving a balance between safety and airspace utilization. Finally, by mining historical obstacle avoidance event characteristics to identify hotspot areas, it finely adjusts the grid in these areas to prevent the recurrence of historical obstacle avoidance risks. Compared to traditional methods, the technical solution provided in this application significantly overcomes the rigidity of airspace utilization caused by static gridding, achieving the technical effect of adaptive gridding of airspace resources based on multi-source parameters.

[0153] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0154] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0155] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. An airspace adaptive gridding method for UAV flight, characterized in that, The method includes: This process involves obtaining multiple sets of maneuverability parameters for various UAV models, establishing a mapping relationship between these parameters and initial grid sizes, and configuring initial grid sizes for each UAV model based on these parameters. The process includes: obtaining multiple sets of maneuverability parameters for various UAV models, where each set includes minimum turning radius, maximum climb rate, and maximum descent rate; determining the horizontal dimension of the initial grid size for a specific UAV model based on the minimum turning radius; determining the vertical dimension of the initial grid size for a specific UAV model based on the maximum climb rate and maximum descent rate; fusing the horizontal and vertical dimensions of the initial grid size for a specific UAV model to obtain the initial grid size for that specific model; and configuring initial grid sizes for each UAV model based on these multiple sets of maneuverability parameters. Obtain a high-precision map of the area to be gridded, assess the terrain complexity of the area based on the high-precision map, obtain a terrain complexity distribution map of the area to be gridded, and calculate the grid size correction coefficient distribution map based on the terrain complexity distribution map. When a specific model of drone enters the area to be gridded, the initial grid size corresponding to the drone is obtained according to the model, the grid size correction coefficient distribution map of the area to be gridded is retrieved, the grid size correction coefficient is obtained according to the real-time position of the drone, and the initial grid size is corrected according to the grid size correction coefficient to obtain the conventional grid size of the drone. Collect historical obstacle avoidance data of UAVs in the area to be gridded over a historical period, and mark obstacle avoidance hotspots in the area to be gridded based on the historical obstacle avoidance data of UAVs. If the area where the drone is located is an obstacle avoidance hotspot, the grid size is reduced based on the regular grid size to obtain the final grid size; if the area where the drone is located is not an obstacle avoidance hotspot, the regular grid size is used as the final grid size, and airspace adaptive gridding is performed.

2. The airspace adaptive gridding method for UAV flight according to claim 1, characterized in that, Based on a high-precision map, the terrain complexity of the area to be gridded is assessed, and a terrain complexity distribution map of the area to be gridded is obtained, including: The area to be gridded is pre-divided into multiple basic grids, and the terrain complexity is calculated in units of the basic grids. Calculate the ratio of the total floor area of ​​all buildings in each basic grid to the area of ​​the basic grid to obtain the building area density of each basic grid; Calculate the ratio of the total number of buildings in each basic grid to the area of ​​the basic grid to obtain the building density of each basic grid; Calculate the standard deviation of the height of all buildings within each base grid to obtain the building height difference of each base grid; The building area density, building number density, and building height difference values ​​corresponding to each basic grid are normalized and then weighted and summed to obtain the terrain complexity corresponding to each basic grid. The terrain complexity value ranges from greater than 0 to less than or equal to 1. Within the high-precision map of the area to be gridded, the range of each basic grid and its corresponding terrain complexity are marked to obtain the terrain complexity distribution map.

3. The airspace adaptive gridding method for UAV flight according to claim 2, characterized in that, Based on the terrain complexity distribution map, a grid size correction factor distribution map was obtained, including: Extract the terrain complexity corresponding to each basic grid in the terrain complexity distribution map, obtain the distribution range of multiple terrain complexity values, and obtain the terrain complexity distribution interval; A preset range of grid size correction coefficients is established, and a mapping relationship is created between the terrain complexity distribution range and the range of grid size correction coefficients. Based on the terrain complexity corresponding to each basic grid, the grid size correction coefficient corresponding to each basic grid is obtained by mapping.

4. The airspace adaptive gridding method for UAV flight according to claim 1, characterized in that, Collect historical obstacle avoidance data of UAVs within a historical period in the area to be gridded. Based on the historical obstacle avoidance data, mark obstacle avoidance hotspot areas within the area to be gridded, including: Based on the flight logs of the UAV flying in the area to be gridded within a historical period, historical obstacle avoidance events are extracted. The historical obstacle avoidance events include the obstacle avoidance position coordinates when the obstacle avoidance occurs and N historical obstacle avoidance features of N obstacle avoidance attempts at the corresponding obstacle avoidance position. The historical obstacle avoidance features include the maneuverability parameter features, flight parameter features, obstacle avoidance time features, and obstacle avoidance weather features of the corresponding obstacle avoidance UAV. Among them, the maneuverability parameter features include the minimum turning radius, maximum climb rate, and maximum descent rate. The maneuverability parameters, flight parameters, current time and weather characteristics of a specific UAV model are obtained as features to be compared. Clustering algorithms are used to cluster the feature to be compared with N historical obstacle avoidance features. Multiple similar historical obstacle avoidance features that are similar to the feature to be compared are extracted from the N historical obstacle avoidance features. Multiple obstacle avoidance position coordinates that correspond one-to-one with the multiple similar historical obstacle avoidance features are obtained as multiple obstacle avoidance hotspot coordinates. The base grid containing the coordinates of multiple obstacle avoidance hotspots is marked as multiple obstacle avoidance hotspot regions.

5. The airspace adaptive gridding method for UAV flight according to claim 4, characterized in that, Clustering algorithms are used to cluster the feature to be compared with N historical obstacle avoidance features. Multiple similar historical obstacle avoidance features, which are similar to the feature to be compared, are extracted from the N historical obstacle avoidance features, including: The features to be compared and N historical obstacle avoidance features are grouped into the same set to obtain the sample obstacle avoidance dataset. The K-means clustering algorithm is used to divide the obstacle avoidance dataset into K clusters based on Euclidean distance, where K is a positive integer greater than or equal to 2. Extract the cluster to which the feature to be compared belongs, and obtain multiple historical obstacle avoidance features that are classified into the same cluster as the feature to be compared, as multiple similar historical obstacle avoidance features.

6. The airspace adaptive gridding method for UAV flight according to claim 5, characterized in that, The base grid containing the coordinates of multiple obstacle avoidance hotspots is marked as multiple obstacle avoidance hotspot regions, including: Count the number of obstacle avoidance hotspot coordinates contained in each basic grid; Based on the number of obstacle avoidance hotspot coordinates contained in each basic grid and a preset threshold for the number of obstacle avoidance hotspot coordinates, grids containing obstacle avoidance hotspot coordinates are labeled with obstacle avoidance heat indicators. The obstacle avoidance heat indicators include multiple obstacle avoidance heat levels, and multiple basic grids with obstacle avoidance heat indicators are regarded as multiple obstacle avoidance hotspot regions.

7. The airspace adaptive gridding method for UAV flight according to claim 6, characterized in that, If the area where the drone is located is an obstacle avoidance hotspot, the grille size is reduced from the standard grille size to obtain the final grille size, including: Retrieve the maneuverability parameters of a specific drone model; Retrieve multiple maneuvering parameters of multiple obstacle avoidance drones corresponding to multiple obstacle avoidance hotspot coordinates within the basic grid of a specific drone model, and use them as a set of historical obstacle avoidance maneuvering parameters; Based on predefined comparison rules, the proportion of multiple maneuverability parameters in the historical obstacle avoidance maneuverability parameter set that are superior to those of a specific UAV model is calculated, and this proportion is used as the maneuverability risk level. Based on the obstacle avoidance heat index, set the initial grid size reduction ratio for each type of obstacle avoidance heat index; Obtain the obstacle avoidance heat map of the obstacle avoidance hotspot area where a specific drone model is located, and the corresponding initial grid size reduction ratio; Based on the maneuverability risk level of a specific UAV model in obstacle avoidance hotspot areas, the initial grille size reduction ratio is adjusted to obtain the grille size reduction ratio. Based on the reduction ratio of the grille size of a specific model of UAV entering the obstacle avoidance hotspot area, the regular grille size is reduced to obtain the final grille size.

8. An airspace adaptive gridding system for unmanned aerial vehicle (UAV) flight, characterized in that, For implementing the airspace adaptive gridding method for UAV flight according to any one of claims 1 to 7, the system comprises: An initial grid size acquisition module is used to obtain multiple sets of maneuverability parameters for various UAV models, establish a mapping relationship between maneuverability parameters and initial grid sizes, and configure initial grid sizes for each UAV model based on the multiple sets of maneuverability parameters. This includes: obtaining multiple sets of maneuverability parameters for various UAV models, where one set of maneuverability parameters includes minimum turning radius, maximum climb rate, and maximum descent rate; determining the horizontal dimension of the initial grid size for a specific UAV model based on the minimum turning radius; determining the vertical dimension of the initial grid size for a specific UAV model based on the maximum climb rate and maximum descent rate; fusing the horizontal and vertical dimensions of the initial grid size for a specific UAV model to obtain the initial grid size for that specific UAV model; and configuring initial grid sizes for each UAV model based on the multiple sets of maneuverability parameters. The grid size correction module is used to obtain a high-precision map of the area to be gridded, evaluate the terrain complexity of the area to be gridded based on the high-precision map, obtain a terrain complexity distribution map of the area to be gridded, and calculate the grid size correction coefficient distribution map based on the terrain complexity distribution map. The conventional grid size acquisition module is used to obtain the initial grid size corresponding to the drone model when a specific model of drone enters the area to be gridded, retrieve the grid size correction coefficient distribution map of the area to be gridded, obtain the grid size correction coefficient according to the real-time position of the drone, and correct the initial grid size according to the grid size correction coefficient to obtain the conventional grid size of the drone. The hotspot area marking module is used to collect historical obstacle avoidance data of UAVs in the area to be gridded over a historical period, and mark obstacle avoidance hotspot areas in the area to be gridded based on the historical obstacle avoidance data of UAVs; the adaptive gridding module is used to reduce the grid size based on the regular grid size to obtain the final grid size if the area where the UAV is located is an obstacle avoidance hotspot area, and to use the regular grid size as the final grid size for airspace adaptive gridding if the area where the UAV is located is not an obstacle avoidance hotspot area.

Images