Low-altitude risk distribution guided unmanned aerial vehicle multi-altitude course planning method and system

Abstract

The invention relates to the technical field of unmanned aerial vehicle route planning, in particular to a low-altitude risk distribution guided unmanned aerial vehicle multi-altitude route planning method and system. Obtaining an original risk influence range of the target unmanned aerial vehicle at a plurality of height layers; acquiring an unmanned aerial vehicle real-time scheduling scheme in the target area, and evaluating and acquiring airspace risk distribution information of the target area in combination with task real-time features; based on the high-precision geographic information data, acquiring ground function partition information of the target area, and correspondingly evaluating ground risk distribution information on a predetermined plane route of the target unmanned aerial vehicle; and based on the airspace risk distribution information, the ground risk distribution information and the original risk influence range, combined search is carried out on multiple height layers in combination with a path planning algorithm based on a greedy strategy, and a multi-height-layer space route of the target unmanned aerial vehicle is obtained. According to the invention, accurate risk management and control and efficient operation optimization of the multi-altitude course of the unmanned aerial vehicle can be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle route planning, and particularly relates to a low-altitude risk distribution oriented unmanned aerial vehicle multi-altitude layer route planning method and system. BACKGROUND

[0002] Current low-altitude unmanned aerial vehicle route planning methods are mostly based on single-altitude layer plane path planning, and the core relies on a static obstacle map. Some methods combine basic performance of unmanned aerial vehicles, but do not systematically integrate multi-dimensional risk factors. The existing technology has the problems of one-sided risk assessment, missing altitude layer planning, and single optimization target, which easily leads to airspace conflict or misjudgment of ground risk, increases energy consumption and operation time, and has poor adaptability to complex low-altitude environments. SUMMARY

[0003] The present application provides a low-altitude risk distribution oriented unmanned aerial vehicle multi-altitude layer route planning method and system to solve the problems of one-sided risk assessment, missing altitude layer planning, and single optimization target in the prior art.

[0004] The technical solution of the present application to solve the above technical problems is as follows: In a first aspect, the present application provides a low-altitude risk distribution oriented unmanned aerial vehicle multi-altitude layer route planning method, comprising: based on intrinsic static characteristics and task real-time characteristics of a target unmanned aerial vehicle, obtaining an original risk influence range of the target unmanned aerial vehicle at multiple altitude layers; obtaining a real-time scheduling scheme of unmanned aerial vehicles in a target region, and evaluating airspace risk distribution information of the target region in combination with the task real-time characteristics; based on high-precision geographic information data, obtaining ground function partition information of the target region, and corresponding to evaluating ground risk distribution information on a predetermined plane route of the target unmanned aerial vehicle; based on the airspace risk distribution information, the ground risk distribution information and the original risk influence range, combined with a path planning algorithm based on a greedy strategy, jointly searching at multiple altitude layers to obtain a multi-altitude layer space route of the target unmanned aerial vehicle.

[0005] Optionally, based on intrinsic static characteristics and task real-time characteristics of a target unmanned aerial vehicle, an original risk influence range of the target unmanned aerial vehicle at multiple altitude layers is obtained, comprising: obtaining flight performance parameters of the target unmanned aerial vehicle, and corresponding to extracting the intrinsic static characteristics, wherein the intrinsic static characteristics include the most unfavorable cruising speed and aerodynamic stability parameters; obtaining a predetermined plane route of the target unmanned aerial vehicle and a task load parameter, and outputting as the task real-time characteristics; based on the intrinsic static characteristics and the task real-time characteristics, combining a prior physical property model to construct a risk influence range model of the target unmanned aerial vehicle, wherein the risk influence range model takes an elevation parameter of an altitude layer as the only variable; inputting the elevation parameter of multiple altitude layers into the risk influence range model to obtain the original risk influence range of multiple altitude layers.

[0006] The method comprises: obtaining a real-time scheduling scheme of a UAV in a target region, combining real-time characteristics of the task, and evaluating airspace risk distribution information of the target region, wherein the real-time scheduling scheme of the UAV in the target region comprises intrinsic characteristics of the UAV and load characteristics of the UAV; extracting a most unfavorable original risk influence range from the original risk influence range, and combining a sliding window method to perform time sequence partitioning on the target region along a predetermined planar route of a target UAV, to obtain a plurality of path time sequence partitions, wherein the predetermined planar route is determined based on the real-time characteristics of the task; traversing the plurality of path time sequence partitions, and determining and analyzing time sequence labels based on the predetermined planar route; traversing the plurality of path time sequence partitions, and performing airspace risk assessment based on the intrinsic characteristics of the UAV and the load characteristics of the UAV and the analysis time sequence labels, to obtain the airspace risk distribution information, wherein the airspace risk distribution information comprises a plurality of partition risk distribution clusters corresponding to the plurality of path time sequence partitions, and each partition risk distribution cluster comprises risk distribution graphs of a plurality of airspace heights.

[0007] The method comprises: in the plurality of path time sequence partitions, randomly selecting an airspace height as a first analysis plane, and correspondingly extracting a first set of intrinsic characteristics of a UAV and a first set of load characteristics of the UAV; according to the first set of load characteristics of the UAV, matching and obtaining a first reference risk index; according to the first set of intrinsic characteristics of the UAV, combining a preset evaluation function, and evaluating and obtaining a first risk diffusion coefficient; based on a mapping relationship between the intrinsic characteristics of the UAV and the load characteristics of the UAV, correspondingly weighting the first reference risk index and the first risk diffusion coefficient, to obtain a first planar risk scatter plot; smoothing and fitting the first planar risk scatter plot, to obtain a first height risk distribution graph; traversing the plurality of airspace heights of the path time sequence partitions, to obtain a plurality of risk distribution graphs, and outputting as the partition risk distribution cluster; traversing the plurality of path time sequence partitions, to obtain a plurality of partition risk distribution clusters, and outputting as the airspace risk distribution information.

[0008] The ground function partition information of the target region is acquired based on high-precision geographic information data, and the ground risk distribution information on the predetermined planar route of the evaluation target unmanned aerial vehicle is correspondingly acquired, including: determining the ground risk boundary in combination with the predetermined planar route and the most unfavorable original risk influence range; extracting the ground function partition information and partition flow heat information of the target region by taking the ground risk boundary as a constraint; determining a reference weight based on the ground function partition information, and calculating and acquiring a regional risk index of each function partition by taking the normalized partition flow heat information as an adjustment coefficient; and generating the ground risk distribution information by spatial interpolation calculation based on the regional risk index.

[0009] Optionally, based on the airspace risk distribution information, the ground risk distribution information and the original risk influence range, a joint search is performed on multiple height layers in combination with a path planning algorithm based on a greedy strategy to acquire a multi-height layer space route of the target unmanned aerial vehicle, including: constructing a greedy target function, wherein the greedy target function includes: a ground direct risk factor determined based on the ground risk distribution information and the original risk influence range corresponding to the height layer; and an airspace conflict risk factor determined based on the airspace risk distribution information and the original risk influence range corresponding to the height layer; performing a joint search on multiple height layers based on the greedy strategy to acquire the multi-height layer space route, with the optimization target of minimizing the greedy target function and the constraint of the predetermined planar route.

[0010] Optionally, based on the airspace risk distribution information, the ground risk distribution information and the original risk influence range, a joint search is performed on multiple height layers in combination with a path planning algorithm based on a greedy strategy to acquire a multi-height layer space route of the target unmanned aerial vehicle, and further including: acquiring low-altitude environment historical wind field data of the target airspace; determining a most unfavorable wind condition parameter according to the low-altitude environment historical wind field data, and correspondingly calculating a wind field correction coefficient for shape correction of the original risk influence range, wherein the wind field correction coefficient is anisotropic; and performing axial correction and update on the original risk influence range in combination with the wind field correction coefficient.

[0011] In a second aspect, the present application provides a low-altitude risk distribution oriented unmanned aerial vehicle multi-height layer route planning system, including: An original risk influence range acquisition module is configured to acquire original risk influence ranges of a target unmanned aerial vehicle at multiple height layers based on intrinsic static characteristics and task real-time characteristics of the target unmanned aerial vehicle; An airspace risk distribution information evaluation module is configured to acquire a real-time scheduling scheme of an unmanned aerial vehicle in a target region, and evaluate airspace risk distribution information of the target region in combination with the task real-time characteristics; The ground risk distribution information evaluation module is configured to acquire ground function partition information of a target region based on high-precision geographic information data, and correspondingly evaluate ground risk distribution information on a predetermined planar flight route of the target UAV. The multi-height-layer space flight route acquisition module is configured to jointly search on multiple height layers based on the airspace risk distribution information, the ground risk distribution information, and the original risk influence range, in combination with a path planning algorithm based on a greedy strategy, to acquire a multi-height-layer space flight route of the target UAV.

[0012] By implementing the present application, the original risk influence range of the target UAV on multiple height layers can be acquired based on intrinsic static characteristics and task real-time characteristics of the target UAV, thereby providing a "UAV self-risk benchmark" for subsequent risk assessment and flight route planning, avoiding risk misjudgment due to neglect of UAV performance differences, and ensuring the pertinence and accuracy of risk analysis. By implementing the present application, a real-time scheduling scheme of a UAV in a target region can be acquired, in combination with the task real-time characteristics, to evaluate airspace risk distribution information of the target region, dynamically capture real-time running states of other UAVs in the airspace, explicitly determine airspace conflict risks at different height layers and different time periods, avoid time and space overlap of flight route planning and other UAVs, and improve airspace running safety. By implementing the present application, ground function partition information of a target region can be acquired based on high-precision geographic information data, and ground risk distribution information on a predetermined planar flight route of the target UAV can be correspondingly evaluated, thereby accurately identifying ground high-risk areas, avoiding UAV risk range covering ground sensitive areas, reducing ground personnel and property damage risks, and improving ground safety of the flight route. By implementing the present application, a multi-height-layer space flight route of the target UAV can be acquired based on the airspace risk distribution information, the ground risk distribution information, and the original risk influence range, in combination with a path planning algorithm based on a greedy strategy, to jointly search on multiple height layers, thereby ensuring low risk, reducing height change frequency through the greedy strategy, reducing UAV energy consumption, balancing safety and running efficiency, and realizing risk-controllable and energy-efficient flight route optimization.

[0013] In summary, by implementing the present application, accurate risk control and efficient running optimization of a multi-height-layer flight route of a UAV can be realized, avoiding airspace conflict and ground safety hazards, reducing energy consumption and height changes, and significantly improving safety, pertinence, and economy of low-altitude UAV flight route planning. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 A flowchart of a low-altitude risk distribution oriented multi-height-layer flight route planning method of a UAV according to the present application is shown. Figure 2 This is a schematic diagram of the structure of a low-altitude risk distribution-oriented UAV multi-altitude flight path planning system provided by the present invention.

[0015] In the attached diagram, the components represented by each number are as follows: The module includes: Module 11 for obtaining the scope of original risk impact, Module 12 for evaluating airspace risk distribution information, Module 13 for evaluating ground risk distribution information, and Module 14 for obtaining multi-altitude spatial routes. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0018] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0019] Example 1, as Figure 1 As shown, this invention provides a method and system for planning multi-altitude flight paths of unmanned aerial vehicles (UAVs) guided by low-altitude risk distribution, including: S100: Based on the intrinsic static characteristics and real-time characteristics of the target UAV, obtain the original risk impact range of the target UAV at multiple altitude levels; S200: Obtain a real-time scheduling scheme of a target region for a UAV, and evaluate airspace risk distribution information of the target region in combination with real-time characteristics of a task; S300: Obtain ground function partition information of a target region based on high-precision geographic information data, and correspondingly evaluate ground risk distribution information on a predetermined planar flight path of a target UAV; S400: Based on the airspace risk distribution information, the ground risk distribution information, and the original risk influence range, a joint search is performed on multiple height layers in combination with a path planning algorithm based on a greedy strategy to obtain a multi-height-layer space flight path of the target UAV.

[0020] In step S100 of the embodiment, based on intrinsic static characteristics and real-time characteristics of a task of a target UAV, an original risk influence range of the target UAV on multiple height layers is obtained, including: Flight performance parameters of the target UAV are obtained, and the intrinsic static characteristics are correspondingly extracted, wherein the intrinsic static characteristics include a most unfavorable cruising speed and an aerodynamic stability parameter; A predetermined planar flight path and a task load parameter of the target UAV are obtained, and the task real-time characteristics are output; Based on the intrinsic static characteristics and the task real-time characteristics, a risk influence range model of the target UAV is constructed in combination with a prior physical property model, wherein the risk influence range model takes an elevation parameter of a height layer as a unique variable; The elevation parameter of the multiple height layers is input into the risk influence range model to obtain the original risk influence range of the multiple height layers.

[0021] In the embodiment, the purpose of step S100 is to accurately determine the original risk influence range of the target UAV on different height layers due to events such as crashes and failures, to provide a UAV self-risk benchmark for subsequent airspace risk, ground risk evaluation, and multi-height-layer flight path planning, to avoid risk misjudgments caused by neglecting the differences between UAV performance and tasks, and to ensure the pertinence and accuracy of subsequent risk analysis.

[0022] First, flight performance parameters of the target UAV are obtained, and the intrinsic static characteristics are correspondingly extracted.

[0023] Specifically, flight performance parameters of the target UAV are obtained first, and then the most unfavorable cruising speed and the aerodynamic stability parameter, which are two core intrinsic static characteristics, are extracted.

[0024] The most unfavorable cruising speed is defined as "the maximum speed under the task endurance requirement", and under this speed, if the UAV fails, the initial kinetic energy is larger and the risk influence range is wider.

[0025] The aerodynamic stability parameter can be measured by aerodynamic derivative. When the UAV can automatically restore to the original equilibrium state after being disturbed, it is defined as static stability, at which time the aerodynamic derivative satisfies the static stability criterion, and the aerodynamic derivative is quantified as "1". If the aerodynamic derivative does not satisfy the static stability criterion, a "number greater than 1" is assigned according to the degree of deviation of the aerodynamic derivative from the stability criterion, as the aerodynamic stability parameter. The larger the aerodynamic stability parameter, the worse the stability, the larger the trajectory deviation of the UAV after losing control, and the larger the risk impact range.

[0026] Second step, the predetermined planar route and task load parameters of the target UAV need to be obtained, and the output is the task real-time feature. The predetermined planar route is the planar position of the flight path, and the task load parameters include load weight and type, which affect the inertia and impact force of the UAV in falling. The two kinds of data are integrated into the task real-time feature.

[0027] Third step, the risk impact range model of the target UAV needs to be constructed based on the intrinsic static feature and the task real-time feature, combined with the prior physical model, wherein the risk impact range model takes the elevation parameter of the height layer as the only variable.

[0028] The prior physical model mainly includes two core sub-models: free fall model and aerodynamics model.

[0029] The free fall model is based on the balance principle of gravity and air resistance in classical mechanics. After the UAV fails, it enters a powerless falling state. By inputting the UAV weight, shape size (affecting the aerodynamic stability parameter), initial falling speed (i.e. the most unfavorable cruising speed), etc., the falling time, vertical acceleration and final pre-landing motion trajectory range of the UAV from different height layers are calculated, which is the basic framework model of the risk impact range.

[0030] The aerodynamics model further refines the influence of environmental airflow on the falling trajectory of the UAV, and quantifies the aerodynamic characteristics of the UAV by introducing the aerodynamic derivative. For example, when the aerodynamic derivative of the UAV is greater than 1 (statically unstable state), the aerodynamics model will calculate the lateral deviation caused by airflow disturbance during falling, and correct the "circular basic range" output by the free fall model to a more realistic "elliptical or irregular range". If a special airflow environment such as a high-rise "air port" is encountered, the aerodynamics model can also adjust the air resistance coefficient combined with historical wind field data to make the risk range prediction more accurate.

[0031] The prior physical model only takes the elevation parameter of the height layer as the only variable, ensuring that the risk range of different height layers can be calculated separately to meet the multi-height layer planning requirements.

[0032] In the fourth step, the elevation parameters of multiple height layers are input into the risk impact range model to obtain the original risk impact range of the multiple height layers. That is, the elevation parameters of multiple height layers, such as 100 meters, 150 meters, 200 meters, etc., are input into the constructed risk impact range model one by one, and the original risk impact range corresponding to each height layer is obtained through model calculation, that is, the spatial boundary that may be affected by the unmanned aerial vehicle after failure at this height.

[0033] In step S200 of the embodiment of the present application, the real-time scheduling scheme of the unmanned aerial vehicle in the target area is obtained, and the airspace risk distribution information of the target area is evaluated in combination with the real-time characteristics of the task, including: The real-time scheduling scheme of the unmanned aerial vehicle in each height layer of the target area is obtained, wherein the real-time scheduling scheme of the unmanned aerial vehicle includes intrinsic characteristics of the unmanned aerial vehicle and load characteristics of the unmanned aerial vehicle; The most unfavorable original risk impact range in the original risk impact range is extracted, and the target area is time-sequentially partitioned along the predetermined planar route of the target unmanned aerial vehicle in combination with the sliding window method to obtain multiple path time-sequential partitions, wherein the predetermined planar route is determined based on the real-time characteristics of the task; The analysis time sequence label is determined in combination with the predetermined planar route by traversing multiple path time-sequential partitions; The airspace risk distribution information is obtained by performing airspace risk assessment in combination with the real-time scheduling scheme of the unmanned aerial vehicle and the analysis time sequence label with multiple airspace heights as analysis planes by traversing multiple path time-sequential partitions; The airspace risk distribution information includes multiple partition risk distribution clusters corresponding to multiple path time-sequential partitions, and each partition risk distribution cluster includes risk distribution maps of multiple airspace heights.

[0034] In the embodiment of the present application, the purpose of step S200 is to dynamically capture the airspace conflict risk of different height layers and different time periods in the target area, generate accurate airspace risk distribution information, provide an airspace dynamic risk map for subsequent multi-height layer route planning, avoid risk overlap between the target unmanned aerial vehicle route and other unmanned aerial vehicles in the time-space+height dimension, and ensure airspace operation safety.

[0035] First, the real-time scheduling scheme of the unmanned aerial vehicle in each height layer of the target area needs to be obtained. That is, the real-time scheduling scheme of other unmanned aerial vehicles in all height layers in the target area is collected, including intrinsic characteristics of the unmanned aerial vehicle and load characteristics of the unmanned aerial vehicle. The intrinsic characteristics of the unmanned aerial vehicle, that is, the most unfavorable cruising speed and aerodynamic stability parameters of other unmanned aerial vehicles, determine the risk range of the unmanned aerial vehicle itself; the load characteristics of the unmanned aerial vehicle, that is, the load weight and type, affect the risk level.

[0036] Next, it is necessary to extract the most unfavorable original risk impact range from the original risk impact range, and combine it with the sliding window method to divide the target area into temporal partitions along the predetermined plane route of the target UAV to obtain multiple path temporal partitions, wherein the predetermined plane route is determined based on the real-time characteristics of the task.

[0037] This involves extracting the most unfavorable original risk impact range with the largest coverage area and highest risk level from the multi-altitude layer original risk impact range obtained from S100, such as the risk range of a high-payload, low-stability UAV at a certain altitude. Then, using the target UAV's predetermined planar flight path as a benchmark, a sliding window method is employed to divide the entire flight path and surrounding airspace along the flight direction into multiple continuous "path-time partitions," such as one partition corresponding to one minute of flight distance. This decomposes large-scale, long-term airspace risks into small-scale, short-term partition risks, reducing the computational complexity of subsequent risk assessments while ensuring that the risk analysis accurately matches the flight timeline of the target UAV.

[0038] Then, it is necessary to traverse multiple path time-series partitions and determine the analysis time-series labels in conjunction with the predetermined plane flight path. That is, traverse all path time-series partitions, and calculate the time when the target UAV enters each partition by combining the flight speed of the target UAV on the predetermined plane flight path and the distance length of each partition. For example, partition 1 corresponds to the flight time of minute 0-1, partition 2 corresponds to minute 1-2, and each partition is labeled with a unique analysis time-series label, such as "T0-T1" or "T1-T2".

[0039] Furthermore, it is necessary to traverse multiple path time-series partitions, using multiple airspace heights as analysis planes, and combine the UAV real-time scheduling scheme with the analysis time-series labels to conduct airspace risk assessment and obtain the airspace risk distribution information.

[0040] In step S200 of this embodiment, multiple path time-series partitions are traversed, and multiple airspace altitudes are used as analysis planes. Airspace risk assessment is performed by combining the UAV real-time scheduling scheme with the analysis time-series labels to obtain airspace risk distribution information, including: In the multiple path time-series partitions, a spatial altitude is randomly selected as the first analysis plane, and the first UAV intrinsic feature set and the first UAV payload feature set are extracted accordingly. Based on the characteristics of the first UAV payload, a first benchmark risk index is obtained; Based on the intrinsic characteristics of the first UAV and combined with the preset evaluation function, the first risk diffusion coefficient is evaluated and obtained. Based on the mapping relationship between the intrinsic characteristics of UAVs and the characteristics of UAV payloads, the first plane risk scatter plot is obtained by weighting the first benchmark risk index and the first risk diffusion coefficient accordingly. Smoothly fit the first plane risk scatter plot to obtain the first altitude risk distribution map; Traverse multiple spatial domain heights of the path time-series partitions to obtain multiple risk distribution maps, and output them as the risk distribution clusters of the partitions; Traverse multiple path time-series partitions to obtain multiple partition risk distribution clusters, and output the spatial risk distribution information.

[0041] In this embodiment of the application, the purpose of the above-mentioned subdivision step S200 is to perform refined airspace risk calculation for different altitude layers of each path time-series partition, transform the abstract UAV intrinsic characteristics and UAV payload characteristics into a visualized and quantifiable risk distribution map, and finally form airspace risk distribution information covering "time-series partition - altitude layer", providing an accurate airspace risk heat map for subsequent route planning, and ensuring that airspace conflict risks at different altitudes and at different times can be accurately avoided.

[0042] The first step is to randomly select an airspace altitude from the multiple path time-series partitions as the first analysis plane, and extract the first UAV intrinsic feature set and the first UAV payload feature set accordingly.

[0043] In other words, among the multiple path time series partitions that have been divided, an airspace altitude is randomly selected as the "first analysis plane", such as 120 meters altitude of a certain path time series partition. From the UAV real-time scheduling scheme of the UAV in the path time series partition, the "first UAV intrinsic feature set" of all other UAVs at this altitude is extracted, that is, the most unfavorable cruise speed and aerodynamic stability parameters of each UAV; and the "first UAV payload feature set", that is, the payload weight and type of each UAV.

[0044] The second step is to obtain a first baseline risk index based on the payload characteristics of the first UAV. Specifically, based on the extracted payload characteristics of the first UAV, such as payload weight, a corresponding "first baseline risk index" can be matched for each other UAV at that altitude by referring to a preset "payload-risk index mapping table". For example, in the payload-risk index mapping table: payload < 5kg corresponds to baseline risk index 1, 5kg ≤ payload < 10kg corresponds to risk index 2, and payload ≥ 10kg corresponds to risk index 3.

[0045] The third step is to evaluate and obtain the first risk diffusion coefficient based on the intrinsic characteristics of the first UAV and in conjunction with the preset evaluation function.

[0046] Specifically, it is necessary to use aerodynamic stability parameters, namely aerodynamic derivatives greater than or equal to 1 as defined in the previous steps, to substitute these parameters into a preset evaluation function to calculate the "first risk diffusion coefficient" for each other UAV. For example, the risk diffusion coefficient is calculated as follows: Risk diffusion coefficient = 0.5 × aerodynamic derivative + 0.5 × (most unfavorable cruise speed / actual cruise speed).

[0047] It is easy to see from the above formula that the larger the aerodynamic derivative (the more unstable the drone), the higher the most unfavorable cruise speed, the wider the trajectory deviation after failure, the greater the risk diffusion coefficient, and the greater the correction range of the risk impact.

[0048] The fourth step is to obtain a first planar risk scatter plot by weighting the first benchmark risk index and the first risk diffusion coefficient based on the mapping relationship between the intrinsic characteristics of the UAV and the UAV payload characteristics.

[0049] Specifically, based on the preset mapping relationship between the intrinsic characteristics and payload characteristics of the UAV, the first benchmark risk index and the first risk diffusion coefficient of each other UAV need to be calculated by weighting. For example, based on historical experience, the weight of the benchmark risk index can be set to 0.6 and the weight of the risk diffusion coefficient can be set to 0.4.

[0050] The final risk value is calculated as: benchmark risk index × 0.6 + risk diffusion coefficient × 0.4. The calculated "final risk value" is then mapped to the real-time position (latitude and longitude) of the UAV on the first analysis plane to generate a "first plane risk scatter plot". Each point in the plot represents the position of a UAV and its corresponding risk value.

[0051] The fifth step is to smoothly fit the first planar risk scatter plot to obtain the first altitude risk distribution map.

[0052] Specifically, interpolation algorithms such as Kriging interpolation can be used to smoothly fit the discrete risk points in the first plane risk scatter plot, transforming isolated risk points into a "first-level risk distribution map" covering the entire first analysis plane. For example, red areas represent high-risk areas with a final risk value ≥ 3, yellow areas represent medium-risk areas with a final risk value ≤ 3, and green areas represent low-risk areas with a final risk value < 2.

[0053] The sixth step involves traversing multiple spatial heights of the path time-series partitions to obtain multiple risk distribution maps, which are then output as the partition risk distribution clusters. This process also involves traversing multiple path time-series partitions to obtain multiple partition risk distribution clusters, which are then output as the spatial risk distribution information.

[0054] That is, for all spatial heights of the current path time series partition, such as 100 meters, 110 meters, 120 meters...200 meters, repeat the above five steps to generate a risk distribution map for each height, and integrate these distribution maps into a "partition risk distribution cluster" for that partition.

[0055] Then, the above operations are performed on all path time-series partitions one by one to obtain the partition risk distribution cluster of each partition. Finally, the risk distribution clusters of all partitions are integrated to output "airspace risk distribution information" covering "full time-series partitions - full altitude layers".

[0056] In step S300 of this embodiment, based on high-precision geographic information data, ground functional zoning information of the target area is obtained, and ground risk distribution information on the predetermined plane flight path of the target UAV is evaluated accordingly, including: By combining the predetermined planar flight path with the range of impact of the most unfavorable original risk, the ground risk boundary is determined; Using ground risk boundaries as constraints, extract ground functional zoning information and zoning traffic heat information for the target area; Based on the ground functional zoning information, a baseline weight is determined, and the regional risk index of each functional zoning is calculated using the normalized zoning traffic heat information as an adjustment coefficient. Spatial interpolation is performed based on the regional risk index to generate the ground risk distribution information.

[0057] In this embodiment of the application, the purpose of step S300 is to accurately identify the risk distribution of the ground area below and around the predetermined flight path of the target UAV, identify high-risk ground areas such as densely populated areas and important facility areas, provide a ground risk map for subsequent multi-altitude flight path planning, avoid the risk range of the UAV failure covering sensitive ground areas, and reduce the risk of damage to ground personnel and property.

[0058] First, the ground risk boundary needs to be determined by combining the predetermined plane route with the range of the most unfavorable original risk impact.

[0059] That is, by combining the "target UAV's planned plane path" and the "most unfavorable original risk impact range" obtained in S100, the planned plane path is used as the central reference, and the maximum radius of the most unfavorable original risk impact range is used as the extension distance to delineate the "ground risk boundary" that needs to be assessed in key areas on the ground, that is, the outline of the ground area that may be affected after the UAV fails.

[0060] Then, ground functional zoning information and zoning traffic heat information of the target area need to be extracted, using ground risk boundaries as constraints.

[0061] Specifically, the ground risk boundary determined by the aforementioned steps is a spatial constraint, and two types of key information are extracted from high-precision geographic information data such as urban GIS maps.

[0062] First, there is information on ground functional zoning, which clarifies the classification of ground areas within the boundaries, such as residential areas, schools, hospitals, industrial parks, open green spaces, etc. Different functional zones have different risk sensitivities. Second, there is the information on traffic intensity in different functional zones, which involves obtaining dynamic activity intensity data for each zone, such as population density in residential areas, pedestrian traffic during school hours, and vehicle traffic in industrial parks. The higher the traffic intensity in a zone, the more severe the risk and consequences.

[0063] Furthermore, it is necessary to determine the baseline weights based on the aforementioned ground functional zoning information, and to calculate the regional risk index for each functional zoning using the normalized zoning traffic heat information as the adjustment coefficient.

[0064] First, the "benchmark weight" needs to be determined based on the ground functional zoning information. Specifically, the weight value needs to be preset according to the risk sensitivity of different zones. For example, the benchmark weight is set to 0.8 for densely populated and vulnerable areas such as schools and hospitals, 0.5 for residential areas, and 0.1 for open green spaces.

[0065] Then, the traffic heat information of the partition is "normalized" to convert the original traffic data, such as 1000 people / hour and 500 vehicles / hour, into a standardized coefficient between 0 and 1, which is used as an "adjustment coefficient". Next, the specific risk index for each functional zone is calculated using the formula "Regional Risk Index = Baseline Weight × Adjustment Coefficient". For example, for a densely populated area of ​​a city, the adjustment coefficient during peak hours is 1, and the regional risk index is 0.8 × 1 = 0.8; the adjustment coefficient during off-peak hours is 0.2, and the regional risk index is 0.8 × 0.2 = 0.16.

[0066] Finally, spatial interpolation calculations are needed based on the regional risk index to generate the ground risk distribution information.

[0067] This involves using a spatial interpolation algorithm to transform the "regional risk index" calculated in the previous steps into continuous risk values ​​covering the entire ground risk boundary, thereby generating "ground risk distribution information," typically presented in the form of a heatmap. For example, red represents high-risk areas, yellow represents medium-risk areas, and green represents low-risk areas. This provides a visual representation of the spatial gradient distribution of ground risks.

[0068] The information, along with the original risk impact range, is combined with a path planning algorithm based on a greedy strategy to perform a joint search at multiple altitude levels to obtain the multi-altitude spatial flight path of the target UAV. Prior to this, it also includes: Acquire historical low-altitude wind field data for the target airspace; The most unfavorable wind condition parameters are determined based on the historical wind field data of the low-altitude environment, and wind field correction coefficients are calculated accordingly to correct the shape of the original risk impact range, wherein the wind field correction coefficients are anisotropic. Based on the wind field correction coefficient, the original risk impact range is updated with axial correction.

[0069] In step S400 of this embodiment, based on the airspace risk distribution information, the ground risk distribution information, and the original risk impact range, a joint search is performed at multiple altitude levels using a path planning algorithm based on a greedy strategy to obtain the multi-altitude spatial flight path of the target UAV, including: Construct a greedy objective function, wherein the greedy objective function includes: The ground direct risk factor is determined based on the ground risk distribution information and the original risk impact range corresponding to the height layer; Airspace conflict risk factor determined based on the airspace risk distribution information and the original risk impact range corresponding to the altitude layer; With minimizing the greedy objective function as the optimization objective and the predetermined planar route as the constraint, a joint search is performed at multiple altitude levels based on a greedy strategy to obtain the multi-altitude spatial route.

[0070] In step S400 of this embodiment, based on the airspace risk distribution information and the ground risk distribution... In this embodiment of the application, the purpose of step S400 is to first correct the original risk impact range of the UAV to fit the actual wind conditions, then construct a greedy objective function that takes into account both ground and airspace risks, jointly search for the optimal route at multiple altitudes, and finally achieve a balance between risk minimization and operational efficiency, so as to ensure that the UAV flies safely and with low energy consumption in complex low-altitude environments.

[0071] To achieve the above objectives, the first step is to obtain historical wind field data of the low-altitude environment of the target airspace. This involves collecting historical wind field data accumulated over a long period of time in the target airspace, including parameters such as wind speed, wind direction, and wind frequency at different altitude levels. For example, data such as the annual average wind speed distribution at altitudes of 100–300 meters, the location of seasonal "wind gaps," and airflow intensity can be collected.

[0072] The second step is to determine the most unfavorable wind condition parameters based on the historical wind field data of the low-altitude environment, and to calculate the corresponding wind field correction coefficient for shape correction of the original risk impact range.

[0073] The most unfavorable wind condition parameter refers to the wind condition characteristic parameter selected from historical wind field data of the low-altitude environment of the target airspace that has the greatest "expansion" and "most significant trajectory deviation" in terms of the risk impact range after UAV failure. For example, the most unfavorable wind condition parameter can be the maximum instantaneous wind speed in the historical records of a certain altitude layer, such as 20 m / s.

[0074] Specifically, it is necessary to select the "most unfavorable wind condition parameter" that has the greatest impact on the risk range of the drone from historical wind field data; and calculate the "wind field correction coefficient" based on this most unfavorable wind condition parameter. Since different wind directions have different effects on the offset direction, the wind field correction coefficient needs to be differentiated by direction and has anisotropy. During the calculation, the correction base coefficient is first determined based on the most unfavorable wind speed.

[0075] For example, the wind speed data from historical wind field data at a certain altitude in the target airspace can be divided into multiple intervals, with the maximum instantaneous wind speed in the historical records at that altitude as the upper limit of the interval, such as 0-5 m / s, 5-10 m / s, 10-15 m / s, 15-20 m / s, etc. A corresponding "baseline correction factor" can be preset for each interval: 1.0 for 0-5 m / s, 1.2 for 5-10 m / s, 1.5 for 10-15 m / s, and 1.8 for 15-20 m / s.

[0076] Then, the direction is distinguished. For example, when the wind direction is the direction the drone is traveling, the wind field correction coefficient is 1.5 times the correction base coefficient. When the wind direction is at a 45° angle to the direction the drone is traveling, the wind field correction coefficient is 1.3 times the correction base coefficient. When the wind direction is opposite to the direction the drone is traveling, the wind field correction coefficient is 0.9 times the correction base coefficient. Following this logic, the wind field correction coefficients for different directions are finally obtained, which are used to correct the original circular risk range into an ellipse or irregular shape that fits the actual wind conditions.

[0077] The third step is to update the original risk impact range by axial correction using the wind field correction coefficient. This involves substituting the wind field correction coefficient into the original risk impact range model obtained in S100 and adjusting the spatial coordinates in the model axially, such as multiplying the risk radius by 1.5 along the wind direction axis, to obtain the corrected original risk impact range.

[0078] The fourth step is to construct the greedy objective function.

[0079] First, it is necessary to determine the direct ground risk factor based on the ground risk distribution information and the original risk impact range corresponding to the altitude layer.

[0080] Obtaining the ground direct risk factor involves overlaying the corrected original risk impact range with ground risk distribution information to calculate the percentage of the original risk impact range that covers the high-risk area. The higher the percentage, the larger the factor value. For example, covering 50% of the high-risk area corresponds to a ground direct risk factor of 0.8, and covering 10% of the high-risk area corresponds to a ground direct risk factor of 0.2.

[0081] Then, an airspace conflict risk factor needs to be determined based on the airspace risk distribution information and the original risk impact range corresponding to the altitude layer.

[0082] Obtaining the airspace conflict risk factor allows matching the corrected original risk impact range with airspace risk distribution information, i.e., the zonal risk distribution clusters, and calculating the overlap probability between the original risk impact range and the original risk impact range of other UAVs. The higher the probability, the larger the airspace conflict risk factor value. For example, an overlap probability of 80% corresponds to an airspace conflict risk factor of 0.9, 0% corresponds to an airspace conflict risk factor of 0.1, and so on.

[0083] Furthermore, it is necessary to optimize by minimizing the greedy objective function and constrain the predetermined planar route, and to perform a joint search at multiple altitude levels based on a greedy strategy to obtain the multi-altitude spatial route.

[0084] Specifically, the greedy objective function formula can be simplified to: Cost = α × Ground Direct Risk Factor + β × Airspace Conflict Risk Factor + γ × Climb Energy Consumption. Climb energy consumption is positively correlated with the actual energy consumption of the UAV reaching a certain altitude, used to balance the UAV's overall energy consumption. This climb energy consumption is a mapped value, of the same order of magnitude as the two risk factors, and dimensionless. For example, for a cargo UAV, a climb from 100m to 120m maps to a climb energy consumption of 0.15, a climb from 120m to 150m maps to a climb energy consumption of 0.25, and so on.

[0085] α, β, and γ are weighting coefficients, and their sum is 1. The specific values ​​can be set according to the task priority. For example, α is 0.6 for large drones and 0.4 for small drones. The value of γ can be appropriately increased for cargo drones and decreased for patrol drones.

[0086] Then, based on the "predetermined horizontal flight path" in S100, it is necessary to ensure that the searched flight path does not deviate from the preset path in the horizontal direction. Then, the entire predetermined horizontal flight path is divided into multiple segments according to the flight sequence, such as every 5 kilometers. For each segment, all selectable altitude layers are traversed, such as traversing multiple altitude layers between 100 and 200 meters, with each layer spaced 10 meters apart. The "total cost value" of selecting each altitude layer is calculated, and the altitude with the minimum total cost value of the segment is selected as the "local optimal altitude".

[0087] Next, the local optimal altitudes of each road segment are connected to form a complete multi-altitude layer flight path; if the difference in optimal altitude between adjacent road segments exceeds a preset threshold, such as 30 meters, the intermediate altitude is finely adjusted, such as gradually transitioning from 120 meters to 150 meters, to reduce the number of altitude changes.

[0088] Finally, the route coordinates obtained from the joint search, including longitude, latitude, and altitude, are organized into structured data and output as the final flight path of the target UAV, i.e., a multi-altitude spatial route, which can be directly used for flight control.

[0089] By implementing the low-altitude risk distribution-oriented UAV multi-altitude flight path planning method provided in the embodiments of this application, at least the following can be achieved: 1. Precise risk management and efficient operation of UAV multi-altitude flight paths.

[0090] 2. Accurately determine the original risk range of drones at multiple altitudes to avoid misjudgment of risks.

[0091] 3. Provides multi-altitude joint search for optimal flight routes, balancing risk and energy consumption, and improving the flight safety and operational efficiency of UAVs.

[0092] Example 2, as Figure 2 As shown, based on the same inventive concept as the low-altitude risk distribution-oriented UAV multi-altitude flight path planning method provided in Embodiment 1, this embodiment of the invention also provides a low-altitude risk distribution-oriented UAV multi-altitude flight path planning system, including: The original risk impact range acquisition module 11 is used to acquire the original risk impact range of the target UAV at multiple altitude layers based on the intrinsic static characteristics and real-time characteristics of the mission. The airspace risk distribution information evaluation module 12 is used to obtain the real-time scheduling scheme of UAVs in the target area and evaluate the airspace risk distribution information of the target area in combination with the real-time characteristics of the task. The ground risk distribution information evaluation module 13 is used to obtain ground functional zoning information of the target area based on high-precision geographic information data, and to evaluate the ground risk distribution information on the predetermined plane flight path of the target UAV accordingly. The multi-altitude spatial route acquisition module 14 is used to perform joint search at multiple altitude levels based on the airspace risk distribution information, the ground risk distribution information and the original risk impact range, combined with a path planning algorithm based on a greedy strategy, to obtain the multi-altitude spatial route of the target UAV.

[0093] Furthermore, the original risk impact scope acquisition module 11 includes the following execution steps: The flight performance parameters of the target UAV are obtained, and the intrinsic static features are extracted accordingly, wherein the intrinsic static features include the most unfavorable cruise speed and aerodynamic stability parameters. Obtain the target UAV's predetermined planar flight path and mission payload parameters, and output them as the real-time characteristics of the mission; Based on the intrinsic static features and the real-time features of the mission, a risk impact range model for the target UAV is constructed by combining the prior physical property model, wherein the risk impact range model uses the elevation parameter of the altitude layer as the only variable. Input the elevation parameters of multiple height layers into the risk impact range model to obtain the original risk impact range of multiple height layers.

[0094] Furthermore, the airspace risk distribution information evaluation module 12 includes the following execution steps: Obtain real-time scheduling schemes for UAVs at various altitude levels within the target area, wherein the real-time scheduling schemes for UAVs include UAV intrinsic characteristics and UAV payload characteristics; Extract the most unfavorable original risk impact range from the original risk impact range, and combine it with the sliding window method to divide the target area into temporal partitions along the predetermined plane route of the target UAV to obtain multiple path temporal partitions. The predetermined plane route is determined based on the real-time characteristics of the task. Traverse multiple path time-series partitions and determine the analysis time-series labels in conjunction with the predetermined plane route; By traversing multiple path time-series partitions and using multiple airspace heights as analysis planes, airspace risk assessment is performed in conjunction with the UAV real-time scheduling scheme and the analysis time-series labels to obtain the airspace risk distribution information. The airspace risk distribution information includes multiple partition risk distribution clusters corresponding to multiple path time-series partitions, and each partition risk distribution cluster includes multiple airspace height risk distribution maps.

[0095] The method involves acquiring real-time scheduling schemes for UAVs at various altitudes within the target area, wherein the real-time scheduling schemes for UAVs include UAV intrinsic characteristics and UAV payload characteristics. Extract the most unfavorable original risk impact range from the original risk impact range, and combine it with the sliding window method to divide the target area into temporal partitions along the predetermined plane route of the target UAV to obtain multiple path temporal partitions. The predetermined plane route is determined based on the real-time characteristics of the task. Traverse multiple path time-series partitions and determine the analysis time-series labels in conjunction with the predetermined plane route; By traversing multiple path time-series partitions and using multiple airspace heights as analysis planes, airspace risk assessment is performed in conjunction with the UAV real-time scheduling scheme and the analysis time-series labels to obtain the airspace risk distribution information. The airspace risk distribution information includes multiple partition risk distribution clusters corresponding to multiple path time-series partitions, and each partition risk distribution cluster includes multiple airspace height risk distribution maps.

[0096] Furthermore, the ground risk distribution information assessment module 13 includes the following execution steps: By combining the predetermined planar flight path with the range of impact of the most unfavorable original risk, the ground risk boundary is determined; Using ground risk boundaries as constraints, extract ground functional zoning information and zoning traffic heat information for the target area; Based on the ground functional zoning information, a baseline weight is determined, and the regional risk index of each functional zoning is calculated using the normalized zoning traffic heat information as an adjustment coefficient. Spatial interpolation is performed based on the regional risk index to generate the ground risk distribution information.

[0097] Furthermore, the multi-altitude spatial route acquisition module 14 includes the following execution steps: Acquire historical low-altitude wind field data for the target airspace; The most unfavorable wind condition parameters are determined based on the historical wind field data of the low-altitude environment, and wind field correction coefficients are calculated accordingly to correct the shape of the original risk impact range, wherein the wind field correction coefficients are anisotropic. Based on the wind field correction coefficient, the original risk impact range is updated with axial correction.

[0098] Construct a greedy objective function, wherein the greedy objective function includes: The ground direct risk factor is determined based on the ground risk distribution information and the original risk impact range corresponding to the height layer; Airspace conflict risk factor determined based on the airspace risk distribution information and the original risk impact range corresponding to the altitude layer; With minimizing the greedy objective function as the optimization objective and the predetermined planar route as the constraint, a joint search is performed at multiple altitude levels based on a greedy strategy to obtain the multi-altitude spatial route.

[0099] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0100] Those skilled in the art will understand that embodiments of the present invention can provide methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0101] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0102] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0103] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0104] Although preferred embodiments of the invention have been described, those skilled in the art, once they have learned the basic inventive concept, can make other changes and modifications to these embodiments.

[0105] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A low-altitude risk distribution-oriented multi-altitude flight path planning method for unmanned aerial vehicles (UAVs), characterized in that, include: Based on the intrinsic static characteristics and real-time characteristics of the target UAV, the original risk impact range of the target UAV at multiple altitude levels is obtained. Obtain a real-time drone scheduling plan within the target area, and evaluate the airspace risk distribution information of the target area based on the real-time characteristics of the task. Based on high-precision geographic information data, the ground functional zoning information of the target area is obtained, and the ground risk distribution information on the predetermined plane flight path of the target UAV is evaluated accordingly. Based on the airspace risk distribution information, the ground risk distribution information, and the original risk impact range, a joint search is performed at multiple altitude levels using a path planning algorithm based on a greedy strategy to obtain the multi-altitude spatial flight path of the target UAV.

2. The low-altitude risk distribution-oriented UAV multi-altitude flight path planning method as described in claim 1, characterized in that, Based on the intrinsic static characteristics and real-time characteristics of the target UAV, the original risk impact range of the target UAV at multiple altitude layers is obtained, including: The flight performance parameters of the target UAV are obtained, and the intrinsic static features are extracted accordingly, wherein the intrinsic static features include the most unfavorable cruise speed and aerodynamic stability parameters. Obtain the target UAV's predetermined planar flight path and mission payload parameters, and output them as the real-time characteristics of the mission; Based on the intrinsic static features and the real-time features of the mission, a risk impact range model for the target UAV is constructed by combining the prior physical property model, wherein the risk impact range model uses the elevation parameter of the altitude layer as the only variable. Input the elevation parameters of multiple height layers into the risk impact range model to obtain the original risk impact range of multiple height layers.

3. The low-altitude risk distribution-oriented UAV multi-altitude flight path planning method as described in claim 2, characterized in that, Obtain a real-time drone scheduling plan within the target area, and evaluate the airspace risk distribution information of the target area based on the real-time characteristics of the task, including: Obtain real-time scheduling schemes for UAVs at various altitude levels within the target area, wherein the real-time scheduling schemes for UAVs include UAV intrinsic characteristics and UAV payload characteristics; Extract the most unfavorable original risk impact range from the original risk impact range, and combine it with the sliding window method to divide the target area into temporal partitions along the predetermined plane route of the target UAV to obtain multiple path temporal partitions. The predetermined plane route is determined based on the real-time characteristics of the task. Traverse multiple path time-series partitions and determine the analysis time-series labels in conjunction with the predetermined plane route; By traversing multiple path time-series partitions and using multiple airspace heights as analysis planes, airspace risk assessment is performed in conjunction with the UAV real-time scheduling scheme and the analysis time-series labels to obtain the airspace risk distribution information. The airspace risk distribution information includes multiple partition risk distribution clusters corresponding to multiple path time-series partitions, and each partition risk distribution cluster includes multiple airspace height risk distribution maps.

4. The low-altitude risk distribution-oriented UAV multi-altitude flight path planning method as described in claim 3, characterized in that, By traversing multiple path time-series partitions and using multiple airspace altitudes as analysis planes, airspace risk assessment is performed in conjunction with the UAV real-time scheduling scheme and the analysis time-series labels to obtain the airspace risk distribution information, including: In the multiple path time-series partitions, a spatial altitude is randomly selected as the first analysis plane, and the first UAV intrinsic feature set and the first UAV payload feature set are extracted accordingly. Based on the characteristics of the first UAV payload, a first benchmark risk index is obtained; Based on the intrinsic characteristics of the first UAV and combined with the preset evaluation function, the first risk diffusion coefficient is evaluated and obtained. Based on the mapping relationship between the intrinsic characteristics of UAVs and the characteristics of UAV payloads, the first plane risk scatter plot is obtained by weighting the first benchmark risk index and the first risk diffusion coefficient accordingly. Smoothly fit the first plane risk scatter plot to obtain the first altitude risk distribution map; Traverse multiple spatial domain heights of the path time-series partitions to obtain multiple risk distribution maps, and output them as the partition risk distribution clusters; Traverse multiple path time-series partitions to obtain multiple partition risk distribution clusters, and output the spatial risk distribution information.

5. The low-altitude risk distribution-oriented UAV multi-altitude flight path planning method as described in claim 3, characterized in that, Based on high-precision geographic information data, ground functional zoning information of the target area is obtained, and ground risk distribution information on the predetermined plane flight path of the target UAV is evaluated accordingly, including: By combining the predetermined horizontal flight path with the range of impact of the most unfavorable original risk, the ground risk boundary is determined; Using ground risk boundaries as constraints, extract ground functional zoning information and zoning traffic heat information for the target area; Based on the ground functional zoning information, a baseline weight is determined, and the regional risk index of each functional zoning is calculated using the normalized zoning traffic heat information as an adjustment coefficient. Spatial interpolation is performed based on the regional risk index to generate the ground risk distribution information.

6. The low-altitude risk distribution-oriented UAV multi-altitude flight path planning method as described in claim 1, characterized in that, Based on the airspace risk distribution information, the ground risk distribution information, and the original risk impact range, a joint search is performed at multiple altitude levels using a path planning algorithm based on a greedy strategy to obtain the multi-altitude spatial flight path of the target UAV, including: Construct a greedy objective function, wherein the greedy objective function includes: The ground direct risk factor is determined based on the ground risk distribution information and the original risk impact range corresponding to the height layer; Airspace conflict risk factor determined based on the airspace risk distribution information and the original risk impact range corresponding to the altitude layer; With minimizing the greedy objective function as the optimization objective and the predetermined planar route as the constraint, a joint search is performed at multiple altitude levels based on a greedy strategy to obtain the multi-altitude spatial route.

7. The low-altitude risk distribution-oriented UAV multi-altitude flight path planning method as described in claim 1, characterized in that, Based on the airspace risk distribution information, the ground risk distribution information, and the original risk impact range, a joint search is performed at multiple altitude levels using a path planning algorithm based on a greedy strategy to obtain the multi-altitude spatial flight path of the target UAV. Prior to this, the following steps are also included: Acquire historical low-altitude wind field data for the target airspace; The most unfavorable wind condition parameters are determined based on the historical wind field data of the low-altitude environment, and wind field correction coefficients are calculated accordingly to correct the shape of the original risk impact range, wherein the wind field correction coefficients are anisotropic. Based on the wind field correction coefficient, the original risk impact range is updated with axial correction.

8. A low-altitude risk distribution-oriented UAV multi-altitude flight path planning system, characterized in that, A method for implementing a low-altitude risk distribution-oriented UAV multi-altitude flight path planning method as described in any one of claims 1-7 includes: The original risk impact range acquisition module is used to acquire the original risk impact range of the target UAV at multiple altitude layers based on the intrinsic static characteristics and real-time characteristics of the mission. The airspace risk distribution information evaluation module is used to obtain real-time scheduling schemes for UAVs within the target area and evaluate the airspace risk distribution information of the target area in conjunction with the real-time characteristics of the task. The ground risk distribution information evaluation module is used to obtain ground functional zoning information of the target area based on high-precision geographic information data, and to evaluate the ground risk distribution information on the predetermined plane flight path of the target UAV accordingly. The multi-altitude spatial flight path acquisition module is used to obtain the multi-altitude spatial flight path of the target UAV by jointly searching at multiple altitude layers based on the airspace risk distribution information, the ground risk distribution information and the original risk impact range, combined with a path planning algorithm based on a greedy strategy.

Images