Load distribution method and system for low-altitude unmanned aerial vehicle

Abstract

The invention provides a load distribution method and system for a low-altitude unmanned aerial vehicle, and relates to the technical field of unmanned aerial vehicles, and the method comprises the steps: inputting boundary longitude and latitude coordinates of a target flight region into a GIS platform, and obtaining region GIS data; performing complexity quantification on the target flight area to obtain a complexity thermodynamic diagram; after the standardized flight trajectory is drawn, trajectory space obstacle avoidance optimization is carried out; projecting a target flight trajectory to the complexity thermodynamic diagram, and segmenting to obtain a trajectory complexity sequence; traversing the load distribution rule base to obtain a load distribution sequence; and executing computing power distribution dynamic switching of the multi-task module according to the load distribution sequence. Through the method and the device, the technical problem that the flight safety is influenced due to the fact that a preset load distribution scheme is not applicable due to variability and uncertainty of a flight environment in the prior art is solved, and the flight safety and the task execution efficiency of the low-altitude unmanned aerial vehicle are improved through combination of the complexity thermodynamic diagram and the flight path.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle technology, and in particular to a payload allocation method and system for low-altitude unmanned aerial vehicles. Background Technology

[0002] With the rapid development of unmanned aerial vehicles (UAVs), low-altitude UAVs are widely used in various scenarios such as urban patrol and emergency rescue due to their advantages of flexible operation and low cost. Low-altitude flight areas typically feature complex terrain, diverse obstacles, and dynamically changing weather conditions, all of which significantly impact the aircraft's flight attitude, power consumption, and flight safety. Existing static payload allocation methods are mostly static configurations, making it difficult to perceive and respond to environmental changes in real time, leading to payload mismatches and affecting the aircraft's stability and mission completion. Furthermore, low-altitude UAVs often face frequent adjustments and switching of mission requirements in practical applications, such as expanding the monitoring area or temporarily adding emergency missions. Traditional payload allocation schemes lack dynamic adjustment mechanisms for mission changes, resulting in inflexible payload configurations that cannot meet the flexible switching requirements of multiple missions and scenarios, leading to low payload utilization efficiency and increased flight safety risks.

[0003] In summary, existing technologies suffer from the technical problem that the preset load distribution schemes are not applicable due to the variability and uncertainty of the flight environment, thus affecting flight safety. Summary of the Invention

[0004] The purpose of this application is to provide a payload allocation method and system for low-altitude unmanned aerial vehicles, in order to solve the technical problem that the preset payload allocation scheme is not applicable due to the variability and uncertainty of the flight environment, thereby affecting flight safety.

[0005] In view of the above problems, this application provides a payload allocation method and system for low-altitude unmanned aerial vehicles.

[0006] Firstly, this application provides a payload allocation method for low-altitude unmanned aerial vehicles (UAVs). This method is implemented through a payload allocation system for low-altitude UAVs. The method includes: inputting the boundary latitude and longitude coordinates of a target flight area into a GIS platform to obtain regional GIS data; quantifying the complexity of the target flight area based on the regional GIS data to obtain a complexity heatmap; drawing a standardized flight trajectory with the target flight area as a flight plane constraint, and then optimizing the trajectory space obstacle avoidance based on the regional GIS data to obtain the target flight trajectory; segmenting the target flight trajectory onto the complexity heatmap to obtain a trajectory complexity sequence; traversing a payload allocation rule base using the trajectory complexity sequence to obtain a payload allocation sequence; and dynamically switching the computing power allocation of multiple task modules based on the payload allocation sequence during the spatial motion control of the low-altitude UAV according to the target flight trajectory.

[0007] Optionally, an outer envelope rectangle is constructed based on the vertex latitude and longitude coordinate set of the target flight area; after extracting the physical size of the outer envelope rectangle, the 1 / M granularity of the physical size is used as the grid scale; the target flight area is divided into multiple equal-sized grids using the grid scale as a segmentation constraint; the complexity of the multiple equal-sized grids is quantified based on the regional GIS data to obtain the complexity heatmap.

[0008] Optionally, multiple grid GIS data of the multiple equal-sized grids are extracted from the regional GIS data; a first building density, a first terrain undulation, and a first obstacle height variance are extracted from the first grid GIS data based on a preset complexity correlation index; the first building density, the first terrain undulation, and the first obstacle height variance are loaded into a weighted model, and the first grid complexity is calculated and output; and so on, the multiple grid complexities of the multiple equal-sized grids are calculated; adjacent grids are merged according to the multiple grid complexities until the complexity heatmap is output.

[0009] Optionally, a preset grid spacing distance and a complexity deviation threshold are established; with the grid spacing distance as a constraint, W seed grids are randomly selected from the plurality of equal-sized grids; the W seed grids are used to traverse the 4-neighborhood grids, and adjacent grids that meet the complexity deviation threshold are selected for grid expansion and merging, and then W updated grid complexities are calculated; this process is repeated, and the grid merging iteration is performed until the number of grids is less than the preset segmentation region value, and multiple region boundaries are output; the multiple region boundaries are seamlessly stitched together in space to generate the complexity heatmap.

[0010] Optionally, along the target flight trajectory, trajectory points are inserted at fixed intervals to obtain a trajectory point sequence; the spatial assignment of the trajectory point sequence is determined by traversing the complexity heatmap to obtain a spatial complexity assignment sequence; based on complexity consistency, adjacent trajectory points in the spatial complexity assignment sequence are merged to output the trajectory complexity sequence.

[0011] Optionally, flight sampling requirements are extracted locally, and a set of sample trajectory complexity and a set of sample load allocation strategies are invoked locally based on the flight sampling requirements; load features are aggregated on the set of sample trajectory complexity and the set of sample load allocation strategies to obtain multiple sample complexity intervals and multiple sample load module allocations; index rules for the multiple sample complexity intervals and multiple sample load module allocations are constructed to obtain the load allocation rule library; the load allocation sequence is obtained by traversing the load allocation rule library using the trajectory complexity sequence.

[0012] Optionally, when the low-altitude unmanned aerial vehicle enters the (N+1)th trajectory segment from the Nth trajectory segment of the trajectory complexity sequence, the (N+1)th payload module allocation is called from the payload allocation sequence; based on the (N+1)th payload module allocation, the computing power ratio of the perception module, decision module, and control module in the task scheduler is dynamically adjusted; and so on, the payload switching of each trajectory segment in the trajectory complexity sequence is performed until the flight mission of the target flight trajectory is terminated.

[0013] Secondly, this application also provides a payload allocation system for low-altitude unmanned aerial vehicles (UAVs), used to execute the payload allocation method for low-altitude UAVs as described in the first aspect. The payload allocation system for low-altitude UAVs includes: a data acquisition module for inputting the boundary latitude and longitude coordinates of a target flight area into a GIS platform to acquire regional GIS data; a complexity quantification module for performing complexity quantification on the target flight area based on the regional GIS data to obtain a complexity heatmap; a trajectory optimization module for drawing a standardized flight trajectory with the target flight area as a flight plane constraint, and then performing trajectory spatial obstacle avoidance optimization based on the regional GIS data to obtain a target flight trajectory; a sequence segmentation module for segmenting the target flight trajectory onto the complexity heatmap to obtain a trajectory complexity sequence; a rule traversal module for traversing a payload allocation rule library using the trajectory complexity sequence to obtain a payload allocation sequence; and a computing power allocation module for dynamically switching the computing power allocation of multiple task modules based on the payload allocation sequence during the spatial motion control of the low-altitude UAV based on the target flight trajectory.

[0014] One or more technical solutions provided in this application have at least the following beneficial effects:

[0015] By inputting the boundary latitude and longitude coordinates of the target flight area into a GIS platform, regional GIS data is obtained. Based on this regional GIS data, the complexity of the target flight area is quantified to obtain a complexity heatmap. After drawing a standardized flight trajectory using the target flight area as a flight plane constraint, spatial obstacle avoidance optimization is performed based on the regional GIS data to obtain the target flight trajectory. The target flight trajectory is projected onto the complexity heatmap to obtain a trajectory complexity sequence. This trajectory complexity sequence is then used to traverse a load allocation rule base to obtain a load allocation sequence. During the control of the low-altitude unmanned aerial vehicle's spatial movement based on the target flight trajectory, dynamic switching of computing power allocation among multiple task modules is performed according to the load allocation sequence. In other words, by quantifying the complexity of the target flight area using regional GIS data, generating a complexity heatmap, optimizing the standardized flight trajectory for spatial obstacle avoidance based on the regional GIS data, and projecting the trajectory onto the complexity heatmap to generate a load allocation sequence, dynamic switching of computing power allocation among multiple task modules is performed, thereby improving the flight safety and mission execution efficiency of the low-altitude unmanned aerial vehicle.

[0016] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

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

[0018] Figure 1 This is a flowchart illustrating the payload allocation method for low-altitude unmanned aerial vehicles proposed in this application.

[0019] Figure 2 This is a schematic diagram of the payload distribution system for low-altitude unmanned aerial vehicles in this application.

[0020] Figure labeling: Data acquisition module 11, complexity quantification module 12, trajectory optimization module 13, sequence segmentation module 14, rule traversal module 15, computing power allocation module 16. Detailed Implementation

[0021] This application provides a payload allocation method and system for low-altitude unmanned aerial vehicles (UAVs), solving the technical problem in existing technologies where the pre-set payload allocation scheme is inapplicable due to the variability and uncertainty of the flight environment, thus affecting flight safety. By using regional GIS data, the complexity of the target flight area is quantified, generating a complexity heatmap. Based on the regional GIS data, standardized flight trajectories are optimized for spatial obstacle avoidance, and this optimized trajectory is projected onto the complexity heatmap to generate a payload allocation sequence. Dynamic switching of computing power allocation across multiple task modules is then implemented, improving the flight safety and mission execution efficiency of low-altitude UAVs.

[0022] The technical solutions of this application will now be clearly and completely described 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. It should be understood that this application is not limited to the exemplary embodiments described herein. 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. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0023] Example 1, please refer to the appendix. Figure 1 This application provides a payload allocation method for low-altitude unmanned aerial vehicles (UAVs), wherein the payload allocation method for low-altitude UAVs is executed by a payload allocation system for low-altitude UAVs, and the payload allocation method for low-altitude UAVs specifically includes the following steps:

[0024] S100: Input the latitude and longitude coordinates of the target flight area boundary into the GIS platform to obtain regional GIS data.

[0025] Specifically, this involves obtaining the boundary coordinates of the target flight area, specifically the longitude and latitude data of the four corners (or multiple points) of the flight area. Longitude refers to the east-west angle on the Earth's surface, while latitude is the north-south angle. Using this longitude and latitude data, the location of the flight area can be precisely defined. The boundary coordinates define a polygonal region through their precise location, thus indicating the boundary of the flight area.

[0026] A GIS platform is a Geographic Information System, a system used to acquire, manage, analyze, and display spatial data. GIS platforms can process different types of geographic data, such as remote sensing images, terrain data, and road networks. By inputting the boundary latitude and longitude coordinates into the GIS platform, and utilizing its spatial analysis functions, detailed spatial data (regional GIS data) of the target flight area can be obtained, including terrain relief, building distribution, road grids, meteorological data, and potential no-fly zones.

[0027] The GIS platform integrates with satellite imagery and geographic databases to automatically generate detailed spatial data for the target area. Regional GIS data refers to geospatial data related to the target flight area, including topographic data, building distribution, roads, water bodies, no-fly zones, and other information. For example, suppose the boundary coordinates of the target flight area are from 34.0522°N, 118.2437°W to 34.0525°N, 118.2432°W. After inputting these coordinates into the GIS platform, the elevation data (100-150 meters), land cover type (e.g., city, vegetation, water bodies), and transportation network (e.g., roads, railways) for the area are obtained. Acquiring regional GIS data helps identify topographic features, potential obstacles, and key points of interest in the flight area, thereby improving the safety and efficiency of flight missions.

[0028] S200: Based on the regional GIS data, the complexity of the target flight area is quantified to obtain a complexity heatmap.

[0029] Furthermore, this application S200 includes:

[0030] An outer envelope rectangle is constructed based on the vertex latitude and longitude coordinates of the target flight area; after extracting the physical size of the outer envelope rectangle, the 1 / M granularity of the physical size is used as the grid scale; the target flight area is divided into multiple equal-sized grids using the grid scale as the segmentation constraint; the complexity of the multiple equal-sized grids is quantified based on the regional GIS data to obtain the complexity heatmap.

[0031] Extract multiple grid GIS data from the multiple equal-sized grids from the regional GIS data; extract the first building density, first terrain undulation, and first obstacle height variance from the first grid GIS data based on a preset complexity correlation index; load the first building density, first terrain undulation, and first obstacle height variance into a weighted model, calculate and output the first grid complexity; and so on, calculate the multiple grid complexities of the multiple equal-sized grids; perform adjacent merging on the multiple equal-sized grids according to the multiple grid complexities until the complexity heatmap is output.

[0032] A preset grid spacing distance and complexity deviation threshold are established. Using the grid spacing distance as a constraint, W seed grids are randomly selected from the plurality of equal-sized grids. The W seed grids are used to traverse 4-neighborhood grids, and adjacent grids that meet the complexity deviation threshold are selected for grid expansion and merging. Then, W updated grid complexities are calculated. This process is repeated until the number of grids is less than a preset segmentation region value, and multiple region boundaries are output. The multiple region boundaries are then seamlessly stitched together to generate the complexity heatmap.

[0033] Specifically, the process involves obtaining the vertex latitude and longitude coordinates set of the target flight area, which is the set of latitude and longitude coordinates describing multiple points within the target flight area. These points typically form a polygonal region in a certain order. An outer envelope rectangle, the smallest rectangle that completely encompasses the target flight area, is then constructed based on the vertex latitude and longitude coordinate set. Geometric calculations are used to construct an outer envelope rectangle that encloses the target area using this smallest rectangle. The outer envelope rectangle is the smallest rectangle in two-dimensional space that can enclose the entire target area, simplifying the analysis and rasterization of complex-shaped regions.

[0034] Extract the physical dimensions of the outer envelope rectangle, i.e., its width and height in actual geographic space, rather than pixels or coordinate units. Set a granularity control parameter M (e.g., M=100), using 1 / M of the physical dimensions as the raster scale. That is, divide the target flight area into M cells on each side; if M=100, each grid represents 1 / 100 of the width and height of the entire area. Divide the entire outer envelope rectangle into multiple equal-sized grids according to the raster scale; all equal-sized grids have the same dimensions and shape.

[0035] This process extracts multiple grid GIS data points from regional GIS data, each with an equal-sized grid. These grid data points include building distribution, terrain elevation, and obstacle information for each grid. Each grid cell contains a subset of GIS data, typically including all buildings, terrain elevation points, and obstacles within that grid. A first grid is randomly selected from these grid data points. From this first grid, the following parameters are extracted: first building density, first terrain undulation, and first obstacle height variance. The first building density is the proportion of buildings within a unit area of ​​the first grid, calculated as the ratio of the total projected area of ​​buildings to the total area of ​​the grid. The first terrain undulation is the degree of elevation variation within the first grid area, usually measured by the maximum elevation difference or standard deviation. The first obstacle height variance is the variance of obstacle heights within the first grid area, measuring the unevenness of height distribution; a larger variance indicates a more complex obstacle distribution. For example, the first grid might have a building density of 30% (coverage), a terrain undulation of 50 meters (elevation difference), and an obstacle height variance of 10 meters. 2 (Degree of fluctuation)

[0036] After normalizing the variances of the first building density, first terrain undulation, and first obstacle height, these values ​​are loaded into a weighted model to calculate the first grid complexity of the first grid. The weighted model assigns different weights to each indicator based on its relative importance, such as C. i =w1*d i +w2*r i +w3*v i , where Ci The complexity is the i-th grid, d i It is the building density of the i-th grid, r i It is the undulation of the i-th grid, v i is the height variance of the i-th grid, and w1, w2, and w3 are the weights.

[0037] For other grids of multiple equal-sized grids, repeat the above steps to obtain multiple grid complexities for these equal-sized grids. Based on these multiple grid complexities, merge adjacent grids of the same size, combining grid cells with similar complexity values ​​and spatial proximity into larger regions to reduce color jumps in the heatmap and improve visualization continuity. Specifically, a preset grid spacing distance is established, which is the minimum distance that must be maintained between seed grids when selecting seed grids (initial cluster centers) to prevent seeds from clustering in local areas and affecting the merging effect. A preset complexity deviation threshold is the maximum allowable difference in complexity values ​​between two grids. If the difference between the complexity of an adjacent grid and the complexity of the current seed grid is less than this threshold, they are considered similar and can be merged.

[0038] In multiple uniformly sized grids, the grid spacing is used as a constraint. W seed grids are randomly selected, where the distance between the center points of any two seed grids is less than the grid spacing. The W seed grids are used to traverse the 4-neighborhood grids, searching for adjacent grids in the four directions above, below, left, and right. Neighboring grids that meet the complexity deviation threshold are selected for grid expansion and merging. In other words, if the difference between the complexity of the Xianlin grid and the current grid complexity is less than the complexity deviation threshold, grid expansion and merging are performed, and W updated grid complexities are calculated. These updated grid complexities are recalculated based on the grid complexities before merging.

[0039] This mesh merging iteration process is repeated until the number of meshes is less than the preset segmentation region value. Expansion stops, resulting in multiple merged regions. The boundaries of these merged regions are then output. The preset segmentation region value is the upper limit of the number of clustering targets, representing the maximum acceptable number of complex regions in the final output. If some regions are merged, their boundaries are frozen to avoid overlap. These multiple region boundaries serve as the dividing lines between merged regions of different complexities. The multiple region boundaries are spatially seamlessly stitched together to generate the final complexity heatmap, ensuring its continuity and integrity. Each region is filled with a color corresponding to its complexity value, forming a color gradient layer. Spatially seamless stitching refers to combining the boundaries of multiple merged mesh regions into a complete and continuous region map in space, avoiding overlapping or blank stitching gaps.

[0040] By constructing an outer envelope rectangle, determining the grid scale, dividing the grid into equal-sized meshes, and quantifying the complexity of the meshes, the flight difficulty of the target flight area is assessed. Complexity heatmaps provide an intuitive way to visualize the flight complexity of different areas, helping UAVs avoid highly complex areas when planning flight paths, thereby improving flight safety and efficiency.

[0041] S300: After drawing a standardized flight trajectory with the target flight area as the flight plane constraint, the trajectory space obstacle avoidance optimization is performed based on the regional GIS data to obtain the target flight trajectory.

[0042] Specifically, a standardized flight trajectory is drawn based on the flight plane constraints of the target flight area. Flight plane constraints mean that the flight trajectory must be strictly confined within the target area, and the trajectory design cannot exceed these boundaries. The standardized flight trajectory is an initially planned, regular trajectory that does not consider obstacles. Within the defined target area, using the flight plane as a two-dimensional constraint, a standardized path construction method (such as grid scanning or covering curve method) is used to generate a set of trajectory points covering the entire area, resulting in the standardized flight trajectory.

[0043] Regional GIS data is introduced for spatial obstacle avoidance analysis. Building, obstacle, and terrain change data are extracted from each standard trajectory point and its surrounding radius R (e.g., 10m). If a point falls into an area marked as high complexity on a complexity heatmap (e.g., greater than 0.7), or is less than a preset safe distance from an obstacle (e.g., 5m), it is considered a danger point. Potential collision points on the identified flight trajectory are then optimized. The optimization process may include adjusting flight altitude, changing flight path, or adjusting flight speed to ensure the UAV can safely avoid all obstacles. The adjusted path must maintain uniformity in coverage of the target area. Obstacle avoidance principles include minimum flight length, maximum safety margin, and minimum energy consumption.

[0044] The output is an optimized target flight trajectory, the final generated and executable flight path. It possesses obstacle avoidance capabilities and meets mission coverage requirements, taking into account constraints and obstacles within the flight area to ensure flight safety. By drawing a standardized flight trajectory with the target flight area as constraints and optimizing the trajectory space for obstacle avoidance based on regional GIS data, a safe and effective target flight trajectory is generated, ensuring that the UAV avoids collisions during mission execution and improving flight safety. Optimizing the flight path can reduce flight time and improve mission efficiency.

[0045] S400: By projecting the target flight trajectory onto the complexity heatmap, a trajectory complexity sequence is obtained.

[0046] Furthermore, this application S400 includes:

[0047] Along the target flight trajectory, trajectory points are inserted at fixed intervals to obtain a trajectory point sequence; the spatial assignment of the trajectory point sequence is determined by traversing the complexity heatmap to obtain a spatial complexity assignment sequence; based on complexity consistency, adjacent trajectory points in the spatial complexity assignment sequence are merged to output the trajectory complexity sequence.

[0048] Specifically, trajectory points are inserted at fixed intervals along the target flight path to form a sequence of trajectory points. The fixed interval insertion of trajectory points represents discretizing the flight path into multiple position points at a set physical distance (e.g., every 10 meters). A fixed interval, such as 10 meters, is used to insert trajectory points along the target flight path, one point every 10 meters, resulting in a sequence of trajectory points. This sequence is an ordered set of points formed by the inserted points, each point carrying latitude and longitude information, representing the spacecraft's spatial position at a given moment.

[0049] Traverse the sequence of trajectory points in the complexity heatmap, set the coordinates of all trajectory points in the complexity heatmap, locate the grid cell to which each trajectory point belongs based on its latitude and longitude, query the complexity value of that grid cell, and assign it to the trajectory point. The spatial complexity assignment sequence is a sequence composed of the corresponding complexity values ​​of each trajectory point in the complexity heatmap, for example, [0.2,0.2,0.3,0.8,0.85,0.9,0.2].

[0050] Based on complexity consistency, adjacent trajectory points in the spatial complexity attribution sequence are merged. If adjacent trajectory points have similar complexity levels, they can be merged into a single complexity segment. Typically, a complexity consistency threshold is set; if the complexity difference between adjacent trajectory points is less than or equal to the threshold, they are determined to be in the same complexity segment, forming a new trajectory segment structure. The merged trajectory complexity sequence is output, representing the complexity level of different segments on the flight trajectory. This helps the UAV make corresponding adjustments during flight to adapt to different flight environments.

[0051] By inserting trajectory points along the target flight path, determining the spatial complexity attribution, and merging adjacent trajectory points based on complexity consistency, a sequence representing the complexity of different segments on the flight path is generated. This helps the UAV better adapt to environmental changes during flight, improving flight safety and efficiency.

[0052] S500: Use the trajectory complexity sequence to traverse the load allocation rule base to obtain the load allocation sequence.

[0053] Furthermore, this application S500 includes:

[0054] The flight sampling requirements are extracted locally, and the sample trajectory complexity set and sample load allocation strategy set are invoked locally based on the flight sampling requirements; the load features of the sample trajectory complexity set and sample load allocation strategy set are aggregated to obtain multiple sample complexity intervals and multiple sample load module allocations; index rules for the multiple sample complexity intervals and multiple sample load module allocations are constructed to obtain the load allocation rule library; the load allocation sequence is obtained by traversing the load allocation rule library using the trajectory complexity sequence.

[0055] Specifically, the flight sampling requirements are extracted locally, i.e., the specific requirements of the current flight mission for sampling (such as images, temperature, gas concentration, etc.), including accuracy, frequency, and payload type. Based on the flight sampling requirements, the sample trajectory complexity set and sample payload allocation strategy set are retrieved locally. The sample trajectory complexity set is a set of predefined trajectory complexity samples used to represent the complexity characteristics of different flight environments; the sample payload allocation strategy set is a set of predefined payload allocation strategy samples used to optimize payload allocation under different flight environments. The sample payload allocation strategy set contains payload usage records corresponding to the sample trajectory complexity set.

[0056] Load feature aggregation is performed on the sample trajectory complexity set and sample load allocation strategy set. Similar complexity features and load allocation strategies are combined to form a more specific load allocation scheme. Load feature aggregation is performed on the sample trajectory complexity set and sample load allocation strategy set to obtain multiple sample complexity intervals and multiple sample load module allocations. For each categorized sample complexity interval, a corresponding load module allocation strategy is formulated. Based on the complexity characteristics and task requirements within that interval, different load module allocation schemes are determined. For example, in intervals with higher complexity, more load resources need to be allocated for obstacle avoidance and navigation, while in intervals with lower complexity, more load resources are allocated for data acquisition and transmission.

[0057] The system outputs the results of multiple sample complexity ranges and multiple sample payload module allocations, establishes index rules for these allocations, and obtains a payload allocation rule library. The index rules, formulated based on the sample complexity ranges and sample payload module allocations, are used to construct the payload allocation rule library. The payload allocation rule library is a collection of strategies containing multiple sample complexity ranges and multiple sample payload module allocations, used to guide payload allocation for UAVs.

[0058] Based on the current trajectory complexity sequence, the payload allocation rule base is traversed to obtain a payload allocation sequence. This sequence, generated from the trajectory complexity sequence and the payload allocation rule base, guides the UAV's payload allocation during flight. For example, in low-complexity areas, a payload allocation of 30% for the perception module, 40% for the decision-making module, and 30% for the control module is suitable for cruise mode and data transmission. In medium-complexity areas, a payload allocation of 50% for the perception module, 30% for the decision-making module, and 20% for the control module is suitable for obstacle avoidance. In high-complexity areas, a payload allocation of 70% for the perception module, 20% for the decision-making module, and 10% for the control module is suitable for obstacle avoidance in dense obstacles. The payload allocation sequence guides the UAV on how to allocate payload resources during flight to optimize mission performance.

[0059] By extracting flight sampling requirements locally, calling sample trajectory complexity sets and sample payload allocation strategy sets, performing payload feature aggregation, constructing a payload allocation rule base, and using trajectory complexity sequence to traverse the payload allocation rule base to generate a payload allocation sequence for the current flight mission, the payload allocation scheme is adjusted in real time according to the flight environment and mission requirements, thereby improving the adaptability and flexibility of the UAV.

[0060] S600: During the process of controlling the spatial motion of the low-altitude unmanned aerial vehicle based on the target flight trajectory, the computing power allocation of the multi-task module is dynamically switched according to the payload allocation sequence.

[0061] Furthermore, this application S600 includes:

[0062] When the low-altitude unmanned aerial vehicle enters the (N+1)th trajectory segment from the Nth trajectory segment of the trajectory complexity sequence, the (N+1)th payload module is mapped and allocated from the payload allocation sequence. Based on the (N+1)th payload module allocation, the computing power ratio of the perception module, decision module, and control module in the task scheduler is dynamically adjusted. This process is repeated to perform payload switching for each trajectory segment in the trajectory complexity sequence until the flight mission of the target flight trajectory is terminated.

[0063] Specifically, based on the determined target flight trajectory, the low-altitude unmanned aerial vehicle (UAV) is controlled, and the UAV moves spatially according to the target flight trajectory. The system continuously determines the trajectory segment corresponding to the UAV's current position (e.g., segment N) and calls the corresponding payload module configuration for that segment. When the UAV moves from segment N to segment N+1 (determined by latitude and longitude), the system automatically detects the trajectory segment switch. In other words, when the UAV flies from one trajectory segment to another, the system finds the appropriate payload module allocation scheme from the payload allocation sequence based on the new complexity characteristics.

[0064] Based on the allocation of the N+1th payload module, the computing power ratio of the perception module, decision-making module, and control module in the task scheduler is dynamically adjusted to meet new task requirements. The task scheduler is a control component used to manage the allocation of computing resources within the aircraft, enabling on-demand allocation and switching of computing power between modules, and includes at least a perception module, a decision-making module, and a control module.

[0065] The above process is repeated continuously until the trajectory flight mission is completed (i.e., the final segment of the trajectory flight is completed). During flight, the UAV's position and trajectory complexity are continuously monitored, and payload switching is performed as needed to ensure that the UAV always adopts the optimal payload allocation scheme. Payload switching is continuously performed throughout the entire flight until the UAV completes all tasks and returns to base. By dynamically switching the computing power allocation of multi-task modules according to the payload allocation sequence, the UAV can adjust the payload allocation scheme in real time according to the flight environment and mission requirements, improving flight safety and efficiency. For example, in areas with high complexity, the UAV can automatically switch to a more conservative payload allocation strategy to reduce the risk of collision; while in areas with low complexity, a more aggressive payload allocation strategy can be adopted to improve mission efficiency.

[0066] In summary, the payload allocation method for low-altitude unmanned aerial vehicles provided in this application has the following beneficial effects:

[0067] By inputting the boundary latitude and longitude coordinates of the target flight area into a GIS platform, regional GIS data is obtained. Based on this regional GIS data, the complexity of the target flight area is quantified to obtain a complexity heatmap. After drawing a standardized flight trajectory using the target flight area as a flight plane constraint, spatial obstacle avoidance optimization is performed based on the regional GIS data to obtain the target flight trajectory. The target flight trajectory is projected onto the complexity heatmap to obtain a trajectory complexity sequence. This trajectory complexity sequence is then used to traverse a load allocation rule base to obtain a load allocation sequence. During the control of the low-altitude unmanned aerial vehicle's spatial movement based on the target flight trajectory, dynamic switching of computing power allocation among multiple task modules is performed according to the load allocation sequence. In other words, by quantifying the complexity of the target flight area using regional GIS data, generating a complexity heatmap, optimizing the standardized flight trajectory for spatial obstacle avoidance based on the regional GIS data, and projecting the trajectory onto the complexity heatmap to generate a load allocation sequence, dynamic switching of computing power allocation among multiple task modules is performed, thereby improving the flight safety and mission execution efficiency of the low-altitude unmanned aerial vehicle.

[0068] Example 2: Based on the same inventive concept as the payload allocation method for low-altitude unmanned aerial vehicles in Example 1, this application also provides a payload allocation system for low-altitude unmanned aerial vehicles. Please refer to the appendix. Figure 2The payload distribution system for low-altitude unmanned aerial vehicles includes:

[0069] The data acquisition module 11 is used to input the boundary latitude and longitude coordinates of the target flight area into the GIS platform to acquire regional GIS data; the complexity quantification module 12 is used to perform complexity quantification on the target flight area based on the regional GIS data to obtain a complexity heatmap; the trajectory optimization module 13 is used to draw a standardized flight trajectory with the target flight area as the flight plane constraint, and then perform trajectory spatial obstacle avoidance optimization based on the regional GIS data to obtain the target flight trajectory; the sequence segmentation module 14 is used to segment the target flight trajectory onto the complexity heatmap to obtain a trajectory complexity sequence; the rule traversal module 15 is used to traverse the load allocation rule library using the trajectory complexity sequence to obtain a load allocation sequence; and the computing power allocation module 16 is used to dynamically switch the computing power allocation of the multi-task module according to the load allocation sequence during the control of the low-altitude unmanned aerial vehicle's spatial motion based on the target flight trajectory.

[0070] Furthermore, the complex quantification module 12 in the payload allocation system for low-altitude unmanned aerial vehicles is also used for:

[0071] An outer envelope rectangle is constructed based on the vertex latitude and longitude coordinates of the target flight area; after extracting the physical size of the outer envelope rectangle, the 1 / M granularity of the physical size is used as the grid scale; the target flight area is divided into multiple equal-sized grids using the grid scale as the segmentation constraint; the complexity of the multiple equal-sized grids is quantified based on the regional GIS data to obtain the complexity heatmap.

[0072] Furthermore, the complex quantification module 12 in the payload allocation system for low-altitude unmanned aerial vehicles is also used for:

[0073] Extract multiple grid GIS data from the multiple equal-sized grids from the regional GIS data; extract the first building density, first terrain undulation, and first obstacle height variance from the first grid GIS data based on a preset complexity correlation index; load the first building density, first terrain undulation, and first obstacle height variance into a weighted model, calculate and output the first grid complexity; and so on, calculate the multiple grid complexities of the multiple equal-sized grids; perform adjacent merging on the multiple equal-sized grids according to the multiple grid complexities until the complexity heatmap is output.

[0074] Furthermore, the complex quantification module 12 in the payload allocation system for low-altitude unmanned aerial vehicles is also used for:

[0075] A preset grid spacing distance and complexity deviation threshold are established. Using the grid spacing distance as a constraint, W seed grids are randomly selected from the plurality of equal-sized grids. The W seed grids are used to traverse 4-neighborhood grids, and adjacent grids that meet the complexity deviation threshold are selected for grid expansion and merging. Then, W updated grid complexities are calculated. This process is repeated until the number of grids is less than a preset segmentation region value, and multiple region boundaries are output. The multiple region boundaries are then seamlessly stitched together to generate the complexity heatmap.

[0076] Furthermore, the sequence segmentation module 14 in the payload distribution system for low-altitude unmanned aerial vehicles is also used for:

[0077] Along the target flight trajectory, trajectory points are inserted at fixed intervals to obtain a trajectory point sequence; the spatial assignment of the trajectory point sequence is determined by traversing the complexity heatmap to obtain a spatial complexity assignment sequence; based on complexity consistency, adjacent trajectory points in the spatial complexity assignment sequence are merged to output the trajectory complexity sequence.

[0078] Furthermore, the rule traversal module 15 in the payload allocation system for low-altitude unmanned aerial vehicles is also used for:

[0079] The flight sampling requirements are extracted locally, and the sample trajectory complexity set and sample load allocation strategy set are invoked locally based on the flight sampling requirements; the load features of the sample trajectory complexity set and sample load allocation strategy set are aggregated to obtain multiple sample complexity intervals and multiple sample load module allocations; index rules for the multiple sample complexity intervals and multiple sample load module allocations are constructed to obtain the load allocation rule library; the load allocation sequence is obtained by traversing the load allocation rule library using the trajectory complexity sequence.

[0080] Furthermore, the computing power allocation module 16 in the payload allocation system for low-altitude unmanned aerial vehicles is also used for:

[0081] When the low-altitude unmanned aerial vehicle enters the (N+1)th trajectory segment from the Nth trajectory segment of the trajectory complexity sequence, the (N+1)th payload module is mapped and allocated from the payload allocation sequence. Based on the (N+1)th payload module allocation, the computing power ratio of the perception module, decision module, and control module in the task scheduler is dynamically adjusted. This process is repeated to perform payload switching for each trajectory segment in the trajectory complexity sequence until the flight mission of the target flight trajectory is terminated.

[0082] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Figure 1The payload allocation method and specific examples for low-altitude unmanned aerial vehicles in Embodiment 1 are also applicable to the payload allocation system for low-altitude unmanned aerial vehicles in this embodiment. Through the foregoing detailed description of the payload allocation method for low-altitude unmanned aerial vehicles, those skilled in the art can clearly understand the payload allocation system for low-altitude unmanned aerial vehicles in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0083] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0084] Obviously, those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A payload allocation method for low-altitude unmanned aerial vehicles, characterized in that, include: Input the latitude and longitude coordinates of the target flight area into the GIS platform to obtain regional GIS data; Based on the regional GIS data, the complexity of the target flight area is quantified to obtain a complexity heatmap; After drawing a standardized flight trajectory with the target flight area as the flight plane constraint, the trajectory space obstacle avoidance optimization is performed based on the GIS data of the area to obtain the target flight trajectory. By projecting the target flight trajectory onto the complexity heatmap, a trajectory complexity sequence is obtained by segmentation. The load allocation sequence is obtained by traversing the load allocation rule base using the trajectory complexity sequence. During the process of controlling the spatial motion of the low-altitude unmanned aerial vehicle based on the target flight trajectory, the computing power allocation of the multi-task module is dynamically switched according to the payload allocation sequence.

2. The payload allocation method for low-altitude unmanned aerial vehicles as described in claim 1, characterized in that, The complexity of the target flight area is quantified based on the regional GIS data to obtain a complexity heatmap, including: Construct an outer envelope rectangle based on the latitude and longitude coordinates of the vertices of the target flight area; After extracting the physical dimensions of the outer envelope rectangle, the 1 / M granularity of the physical dimensions is used as the grid scale; Using the grid scale as a segmentation constraint, the target flight area is divided into multiple grids of equal size; The complexity heatmap is obtained by quantifying the multiple equal-sized grids based on the regional GIS data.

3. The payload allocation method for low-altitude unmanned aerial vehicles as described in claim 2, characterized in that, Based on the regional GIS data, the complexity of the multiple equal-sized grids is quantified to obtain the complexity heatmap, including: Extract multiple grid GIS data from the multiple equal-sized grids from the regional GIS data; Based on preset complexity correlation indicators, the first building density, the first terrain undulation, and the first obstacle height variance are extracted from the first grid GIS data. The first building density, the first terrain undulation, and the first obstacle height variance are loaded into the weighted model, and the first grid complexity is calculated and output. Similarly, calculate the mesh complexity of the multiple equal-sized meshes; Based on the multiple grid complexities, adjacent grids of equal size are merged until the complexity heatmap is output.

4. The payload allocation method for low-altitude unmanned aerial vehicles as described in claim 1, characterized in that, The load allocation sequence is obtained by traversing the load allocation rule base using the trajectory complexity sequence, including: Locally extract flight sampling requirements, and locally invoke the sample trajectory complexity set and sample payload allocation strategy set based on the flight sampling requirements; The load features of the sample trajectory complexity set and the sample load allocation strategy set are aggregated to obtain multiple sample complexity intervals and multiple sample load module allocations; Construct index rules for allocating the multiple sample complexity ranges and multiple sample load modules to obtain the load allocation rule library; The load allocation sequence is obtained by traversing the load allocation rule base using the trajectory complexity sequence.

5. The payload allocation method for low-altitude unmanned aerial vehicles as described in claim 1, characterized in that, During the process of controlling the spatial motion of the low-altitude unmanned aerial vehicle based on the target flight trajectory, the dynamic switching of computing power allocation of the multi-task module is performed according to the payload allocation sequence, including: When the low-altitude unmanned aerial vehicle enters the (N+1)th trajectory segment from the Nth trajectory segment of the trajectory complexity sequence, the (N+1)th payload module is called from the payload allocation sequence mapping for allocation. Based on the N+1th load module allocation, the computing power ratio of the perception module, decision module, and control module in the task scheduler is dynamically adjusted; This process continues, performing load switching on each trajectory segment in the trajectory complexity sequence until the flight mission of the target flight trajectory is completed.

6. The payload allocation method for low-altitude unmanned aerial vehicles as described in claim 3, characterized in that, Based on the multiple mesh complexities, adjacent meshes of equal size are merged until the complexity heatmap is output, including: Preset grid spacing and complexity deviation threshold; Using the grid spacing distance as a constraint, W seed grids are randomly selected from the plurality of equal-sized grids; The W seed grids are used to traverse the 4-neighborhood grids. After selecting neighboring grids that meet the complexity deviation threshold, the grids are expanded and merged, and the complexity of the W updated grids is calculated. This process continues until the number of grids is less than the preset segmentation region value, at which point multiple region boundaries are output. The spatial seamless stitching of the boundaries of the multiple regions generates the complexity heatmap.

7. The payload allocation method for low-altitude unmanned aerial vehicles as described in claim 6, characterized in that, By projecting the target flight trajectory onto the complexity heatmap, a trajectory complexity sequence is obtained, including: Along the target flight trajectory, trajectory points are inserted at fixed intervals to obtain a sequence of trajectory points; Spatial attribution is determined by traversing the sequence of trajectory points in the complexity heatmap to obtain the spatial complexity attribution sequence; Based on complexity consistency, adjacent trajectory points of the spatial complexity attribution sequence are merged to output the trajectory complexity sequence.

8. A payload distribution system for low-altitude unmanned aerial vehicles, characterized in that, The step of implementing the payload allocation method for low-altitude unmanned aerial vehicles according to any one of claims 1 to 7, wherein the payload allocation system for low-altitude unmanned aerial vehicles comprises: The data acquisition module is used to input the latitude and longitude coordinates of the target flight area into the GIS platform to obtain regional GIS data; The complexity quantification module is used to perform complexity quantification on the target flight area based on the regional GIS data to obtain a complexity heatmap. The trajectory optimization module is used to draw a standardized flight trajectory with the target flight area as the flight plane constraint, and then perform trajectory spatial obstacle avoidance optimization based on the regional GIS data to obtain the target flight trajectory. The sequence segmentation module is used to segment the trajectory complexity sequence by projecting the target flight trajectory onto the complexity heatmap; The rule traversal module is used to traverse the load allocation rule library using the trajectory complexity sequence to obtain the load allocation sequence. The computing power allocation module is used to dynamically switch the computing power allocation of the multi-task module according to the payload allocation sequence during the process of controlling the spatial movement of the low-altitude unmanned aerial vehicle based on the target flight trajectory.

Images