Low-altitude air route intelligent planning method for unmanned aerial vehicle

Abstract

This invention discloses an intelligent low-altitude flight path planning method for unmanned aerial vehicles (UAVs), relating to the field of UAV technology. The method includes: receiving a flight mission and collecting static and dynamic environmental information, as well as UAV status information; dividing the flight area into a three-dimensional grid; calculating the terrain, wind field, and airspace accessibility indicators for each grid and converting them into risk indicators; weighting and fusing all risk indicators and correcting them with an environmental coupling coefficient to obtain a comprehensive risk value, thus constructing a three-dimensional dynamic risk field; generating a flight cost field based on the risk field and distance cost; searching for candidate flight paths; querying approved flight missions based on the estimated arrival time; statistically analyzing three indicators—conflict frequency, conflict intensity, and conflict distribution—and weighting them to obtain a traffic impact coefficient; correcting the cumulative flight cost and selecting the final flight path; resolving traffic conflicts on the final flight path and then issuing it for execution; continuously monitoring the environment during flight, predicting risks based on risk evolution sequences, and triggering dynamic replanning.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, specifically to a method and system for intelligent low-altitude flight path planning for UAVs. Background Technology

[0002] In recent years, with the rapid development of the low-altitude economy, drones have been widely used in logistics delivery, power line inspection, emergency rescue, urban management, surveying and mapping, and low-altitude transportation. In existing technologies, before a drone can perform a flight mission, it typically needs to plan its flight path based on the mission's starting point, target location, and flight restrictions to ensure the mission's safe completion. Specifically, this usually involves path searching based on electronic maps, no-fly zone information, and preset flight rules, generating a flight route by optimizing a single objective such as the shortest distance or minimum flight time. While some solutions can incorporate weather or obstacle information into route planning, these typically only provide localized corrections for a single environmental factor.

[0003] However, in actual low-altitude flight, the airspace environment in which UAVs operate exhibits strong dynamic changes. For example, changes in local wind fields, temporary airspace control, and low-altitude aircraft traffic all constantly change over time, collectively affecting the flight safety and flight path feasibility of UAVs. Current flight path planning still primarily relies on static environmental data. Given these dynamic characteristics, it is difficult to comprehensively reflect the real-time changes in the low-altitude airspace environment and fully utilize dynamic environmental information to form a basis for flight path planning. Consequently, the generated flight paths are ill-suited to the complex and dynamic low-altitude environment. Summary of the Invention

[0004] To address the problems in related technologies, this invention provides an intelligent low-altitude flight path planning method for unmanned aerial vehicles (UAVs) to overcome the aforementioned technical problems in existing related technologies.

[0005] To solve the aforementioned technical problem, the present invention is achieved through the following technical solution:

[0006] A method for intelligent low-altitude flight path planning for unmanned aerial vehicles (UAVs) includes:

[0007] The system acquires the flight mission information of the UAV and loads the corresponding 3D digital map of the flight area. It collects the static environmental information, dynamic environmental information, and UAV operating status information of the flight area and performs preprocessing. The flight area is divided into a 3D grid. For each grid, the system determines the terrain accessibility index, wind field impact index, and airspace access index. Based on this, the system comprehensively analyzes and determines the comprehensive risk value of the corresponding grid. Based on the comprehensive risk value of each grid, the system constructs a 3D dynamic risk field of the flight area.

[0008] A flight cost field is established based on the three-dimensional dynamic risk field of the environment. Several candidate flight routes are generated in the flight cost field. The degree of traffic impact is determined by combining the traffic conflict situation corresponding to the candidate flight routes. The candidate flight routes are screened according to the degree of traffic impact to determine the final flight route.

[0009] The final flight path is submitted to the low-altitude flight supervision platform for approval. After approval, it is sent to the UAV flight control system to control the UAV to execute the flight mission. During the flight, the UAV continuously acquires real-time environmental information of the flight area and updates the three-dimensional dynamic risk field of the environment. Based on the updated environmental risk status, it is determined whether the flight path replanning conditions are met. When the flight path replanning conditions are met, the flight path is regenerated, and after approval, the UAV is controlled to continue to execute the flight mission.

[0010] Preferably, the flight mission information includes at least the mission start point, target location, and permitted flight altitude; the static environment information includes at least building height, terrain elevation, and no-fly zone boundary information; the dynamic environment information includes at least wind speed, wind direction, low-altitude aircraft density, and temporary airspace restriction level; the UAV operating status information includes at least the current position, flight speed, flight heading angle, and the UAV's maximum permissible wind resistance speed; the preprocessing includes time synchronization processing of various types of data based on a unified time reference, and normalization processing of different types of environmental parameters to eliminate sampling time differences and parameter dimension differences between different data sources; specific implementation methods include:

[0011] Suppose any data source is at time... The collected data values ​​are If the unified processing time is t, then linear interpolation is used to complete time synchronization, eliminating data deviations caused by asynchronous sampling times from different data sources, so that all subsequent environmental parameters correspond to the low-altitude environmental state at the same time. The specific synchronization formula is as follows:

[0012]

[0013] To unify the data values ​​corresponding to the time axis, where p is a positive integer representing the sampling time sequence number, and These are two adjacent sampling times, where t is the unified time node, satisfying... ≤t≤ ;

[0014] After time synchronization is completed, the values ​​of each parameter are normalized to eliminate differences in the dimensions of different physical quantities. In this embodiment, the maximum-minimum normalization method is used, and the specific normalization formula is as follows:

[0015]

[0016] and These represent the maximum and minimum values ​​within the historical statistical range of this parameter, respectively.

[0017] Step two: Divide the 3D digital map corresponding to the flight area into a regular 3D grid. Assume the flight area is divided along the X, Y, and Z directions according to preset grid spacing ΔX, ΔY, and ΔZ, respectively. The specific values ​​for ΔX, ΔY, and ΔZ are set to 20–100m based on the UAV's flight accuracy. After division, the spatial grid corresponding to the i-th row, j-th column, and k-th layer is represented as follows: Each grid is treated as an independent environmental unit, and it stores both static and dynamic environmental information.

[0018] Preferably, the determination of terrain accessibility indicators includes:

[0019] The system calculates the building heights and terrain elevations within the current grid, determines the remaining passable altitude of the current grid based on the allowed flight altitude of the UAV, judges the flight status of the current grid based on the remaining passable altitude, and generates a corresponding terrain passability index based on the flight status. The flight status includes at least safe passability, restricted passability, and impassable passability, and the terrain passability index characterizes the spatial passability of the current grid for UAV flight. Specific implementation methods include:

[0020] The system calculates the maximum building height, maximum terrain elevation, and planned drone flight altitude within the current grid. The maximum value of the maximum building height and maximum terrain elevation within the current grid is subtracted from the planned drone flight altitude to obtain the remaining passable altitude of the current grid. If the remaining passable altitude is less than or equal to zero, the current grid is considered unflyable. Otherwise, the grid is categorized into three flight indicators based on the remaining passable altitude: safe passage, restricted passage, and impassable. Safe passage indicates that the drone has sufficient safety margin and can pass through the current grid normally; restricted passage indicates that the drone can still pass through the current grid, but the safety margin is small and it should be considered a low-priority flight area; impassable indicates that there are buildings or terrain obstacles intruding into the planned flight altitude, failing to meet the drone's safe flight conditions. A mapping relationship is established between flight indicators and terrain passability indicators. Corresponding terrain passability indicators are generated based on the flight indicators of the current grid. The terrain passability index ranges from 0 to 1. A larger value indicates stronger flight passability for the current grid and a higher safety margin for the UAV to pass through the current grid. A smaller value indicates weaker flight passability for the current grid and stronger constraints on UAV flight. When the current grid is safe to pass, a larger terrain passability index is assigned. When it is restricted to pass, the terrain passability index is reduced in order of degree of restriction. When it is impassable, the terrain passability index is set to 0, indicating that the grid cannot be used as a UAV flight path.

[0021] Preferably, the indicators for determining wind field impact include:

[0022] The system acquires the wind speed, wind direction, and UAV flight path corresponding to the current grid; determines the wind field direction based on the angle between the wind direction and the UAV flight path, and determines the wind field impact index for the corresponding grid by combining the wind speed and the UAV's maximum allowable wind resistance speed; wherein, the wind field impact index is used to characterize the degree of influence of the wind field on the flight stability of the UAV, and is dynamically adjusted with changes in wind speed and wind field direction; the specific implementation includes:

[0023] Extract the current grid's wind speed, wind direction, and UAV heading. Convert the wind direction and UAV heading into angles to obtain the current grid's wind direction angle and UAV heading angle. Subtract the two and take the absolute value to obtain the angle between the wind direction and the UAV's flight direction. To avoid calculation errors caused by the periodicity of the heading angle, the angle is normalized. If the calculated angle is no greater than 180°, it is taken as the actual angle; if the calculated wind direction angle is greater than 180°, the difference between 360° and the angle is taken as the actual angle, ensuring that the obtained angle is the minimum angle between the wind direction and the UAV's flight direction. Calculate the wind field impact index using the following formula: wind speed, maximum allowable wind resistance speed of the UAV, and actual angle. The calculation formula is as follows:

[0024]

[0025] in The actual included angle is V, where V is the current wind speed on the grid, and Vamx is the maximum wind resistance speed allowed for the drone. When the wind is tailwind, =0, during crosswinds =1, when facing the wind =0, wind field impact index It can highlight the impact of crosswinds on the attitude stability of drones. The larger the field influence index, the more severe the impact of the current grid wind field.

[0026] Preferably, the airspace clearance indicators include:

[0027] Obtain the airspace restriction status corresponding to the airspace to which the current grid belongs; determine the airspace access index corresponding to the current grid based on the pre-established correspondence between airspace restriction status and airspace accessibility; wherein, the airspace restriction status includes at least open airspace, restricted airspace, approved airspace, and temporary no-fly airspace, and the airspace access index is used to characterize the flight permission level of the current grid, and the specific implementation includes:

[0028] Based on the airspace restriction level obtained from the flight monitoring platform, the specific airspace restrictions include, but are not limited to, open airspace, restricted airspace, approved airspace, and temporarily no-fly zone. Since different airspace restriction states correspond to different levels of flight permission, a mapping relationship between airspace restriction states and airspace access indicators is pre-established, and a corresponding airspace access indicator is generated based on the current airspace restriction state of the grid. The airspace access index ranges from 0 to 1. A larger value indicates less flight restriction and higher accessibility for the current grid, while a smaller value indicates stricter flight restriction. When the current grid is open airspace, a larger airspace access index is assigned. When it is restricted airspace or approved airspace, the airspace access index is reduced in order of degree of restriction. When it is a temporary no-fly zone, the airspace access index is set to 0, indicating that the grid cannot be used as a flight path for drones.

[0029] Preferably, determining the comprehensive risk value of the corresponding grid based on the terrain accessibility index, wind field impact index, and airspace accessibility index includes:

[0030] Based on the terrain accessibility index, wind field impact index, and airspace accessibility index, corresponding terrain risk index, wind field risk index, and airspace risk index are generated respectively. Based on the terrain risk index, wind field risk index, and airspace risk index, the degree of dispersion between each risk index is determined, and the environmental coupling coefficient is determined in combination with the overall risk level of each risk index to characterize the degree of synergistic influence between multiple environmental risk factors in the current grid.

[0031] The basic risk value is obtained by fusing various risk indicators according to preset weights, and the basic risk value is dynamically corrected using the environmental coupling coefficient to obtain the comprehensive risk value corresponding to the current grid. The comprehensive risk value corresponding to each grid is mapped to the corresponding three-dimensional grid to form a three-dimensional dynamic environmental risk field corresponding to the flight area. The specific implementation method includes:

[0032] Set the terrain accessibility index of the current grid. Wind field impact indicators Airspace passage indicators The terrain risk indicators are uniformly converted into corresponding risk indicators. Wind farm risk indicators and airspace risk indicators And respectively represented as:

[0033]

[0034] Terrain risk indicators Wind farm risk indicators and airspace risk indicators The average risk value is obtained by calculating the mean. Then, the standard deviation is used to calculate the dispersion of each risk indicator relative to the average risk value. The calculation formula is as follows:

[0035]

[0036] Then the dispersion and average risk value The environmental coupling coefficient K is calculated using the following formula:

[0037]

[0038] The terrain risk index, wind field risk index, and airspace risk index are weighted and fused according to preset weights to obtain the basic risk value of the current grid. The environmental coupling coefficient is used as a risk enhancement factor to dynamically correct the basic risk value, resulting in the comprehensive risk value of the current grid. The specific calculation formula is as follows:

[0039]

[0040] in, , and The weighting coefficients for terrain risk, wind field risk, and airspace risk are dynamically adjusted based on the UAV mission type or aircraft performance. λ is the environmental coupling enhancement coefficient, ranging from 0.2 to 0.5. A truncation function with an upper limit of one is then used to truncate the comprehensive risk value. The environmental coupling coefficient characterizes the degree of synergy between multiple environmental risks. When multiple environmental risks coexist and have strong coupling effects, the environmental coupling coefficient is large, thus enhancing the basic risk value. When environmental risks are mainly caused by a single risk factor, the environmental coupling coefficient is small, and its enhancement effect on the basic risk value is weakened. The above calculation and analysis method can reflect the importance of different environmental risks and highlight the impact of multiple risks on UAV flight safety, making the comprehensive risk value more consistent with the actual low-altitude flight environment. Thus, the comprehensive risk value for each grid can be obtained. They are then filled into the corresponding grid to construct a three-dimensional dynamic risk field of the flight area.

[0041] Preferably, establishing the flight cost field based on the three-dimensional environmental dynamic risk field includes:

[0042] Based on the comprehensive risk value corresponding to each grid and the distance information from each grid to the target location, the flight cost value corresponding to each grid is determined; a flight cost field is established based on the flight cost values ​​corresponding to all grids; taking the grid corresponding to the mission starting point as the starting node and the grid corresponding to the target location as the target node, a continuous flight path with a smaller cumulative flight cost is searched in the flight cost field to generate several candidate flight routes; wherein, each candidate flight route is sorted in ascending order of cumulative flight cost, and a preset number of candidate flight routes are selected; the specific implementation method includes:

[0043] For any grid within the flight area Calculate the Euclidean distance from the grid to the target location as follows: and combine it with the grid's overall risk value. The current grid flight cost is calculated. The calculation formula is as follows:

[0044]

[0045] Where α is the risk cost weight and β is the distance cost weight, satisfying α + β = 1. The maximum flight distance within the flight area is represented by the flight cost value. The higher the flight cost value, the higher the flight risk and cost that the UAV needs to bear when passing through the current grid. The lower the flight cost value, the more suitable the current grid is as the area through which the flight path passes. After calculating the flight cost value of all grids, a flight cost field is established for the flight area. The grid corresponding to the mission starting point is used as the starting node, and the grid corresponding to the target position is used as the target node. Continuous flight grids are searched within the flight cost field according to the principle of minimizing the cumulative flight cost, generating several initial flight paths for the UAV. Each initial flight path is sorted in ascending order of cumulative flight cost, and the top few initial flight paths are selected as candidate flight paths. In this embodiment, 6 candidate flight paths are usually selected.

[0046] Preferably, determining the degree of traffic impact by combining the traffic conflict situation corresponding to the candidate flight routes includes:

[0047] The time nodes when the UAV is expected to pass through each grid are determined based on the planned flight speed corresponding to the candidate flight route; based on the expected arrival time corresponding to each grid, the approved flight missions within the corresponding time period are obtained from the low-altitude flight monitoring platform; it is determined whether the candidate flight route and the approved flight mission pass through the same grid within the preset time range, and the grids that meet the conditions are identified as conflict points.

[0048] Based on the number of conflict points corresponding to candidate flight routes, the scale of each conflict point, and the spatial distribution of traffic conflict points along the candidate flight routes, conflict frequency, conflict intensity, and conflict distribution indices are determined respectively; based on the conflict frequency, conflict intensity, and conflict distribution indices, the traffic impact coefficient corresponding to the candidate flight routes is determined; the specific implementation method includes:

[0049] Choose any candidate flight path, and calculate the estimated time points when the UAV will pass through each grid within the candidate flight path based on the UAV's planned flight speed. Let the estimated time for the UAV to enter the nth grid be denoted as . Let n represent the index of the grid within the candidate flight path. Then, the expected flight time series corresponding to the candidate flight path is represented as: , Indicates the starting node. The target node is represented by N, a positive integer, which represents the total number of grids within the initial flight path. Based on the estimated arrival time of each grid, approved flight missions are retrieved from the low-altitude flight monitoring platform, and all reported flight paths are queried. The preset time tolerance is... The system counts events where at least two flight missions need to pass through the same grid within the time tolerance, and treats these events as traffic conflict events. The number of flight missions passing through the grid in a traffic conflict event is taken as the conflict number, and the corresponding grid point is taken as the conflict point of the traffic conflict event. It should be noted that the conflict number is greater than or equal to 2. The system counts the number of conflict points where traffic conflicts occur within the candidate flight route. Each conflict point is counted only once, and the number is divided by the total number of grids on the candidate flight route to obtain the conflict frequency. The conflict frequency reflects the overall density of conflicts encountered by the candidate flight route in the spatiotemporal dimension.

[0050] Quantitative analysis of traffic conflicts along candidate flight routes yields conflict frequency, conflict intensity, and conflict distribution indices to characterize the traffic operation status of these routes. Specifically, for each conflict point on a candidate flight route, the number of conflicts at each point is extracted. The conflict numbers at all conflict points are summed and then divided by the total number of conflict points; the average value is the conflict intensity index. This index is not less than 2. A higher value indicates more simultaneous or near-simultaneous flight missions at a single conflict point, higher air traffic congestion at that point, and greater flight conflict risk and coordination difficulty. Conversely, a lower value indicates smaller conflict scales at each conflict point, making coordination and resolution relatively easier. The conflict intensity index characterizes the congestion level or conflict severity at each conflict point on the candidate flight route. The flight direction of the flight path is determined by measuring the cumulative flight arc length of each conflict point from the starting point of the route. The average arc length of all conflict points relative to the starting point of the route is calculated, and then the standard deviation of the arc length of each conflict node relative to this average is calculated. The ratio obtained by dividing the standard deviation by the average is the conflict distribution index. The conflict distribution index is used to characterize the dispersion or clustering of each conflict node on the candidate flight path in the direction of the route extension. The conflict frequency index, conflict intensity index, and conflict distribution index corresponding to the candidate flight path are normalized to eliminate the differences in the dimensions and value ranges between different indicators, so that each indicator has a unified evaluation scale. The three normalized indicators are fused according to the preset weights to obtain the traffic impact coefficient corresponding to the candidate flight path. The traffic impact coefficient is used to comprehensively characterize the overall degree of impact of traffic conflicts on the candidate flight path.

[0051] Preferably, the step of screening candidate flight routes based on the degree of traffic impact includes:

[0052] The conflict frequency index, conflict intensity index, and conflict distribution index are uniformly processed to eliminate the dimensional differences between different evaluation indicators; based on the pre-set weights, the traffic impact coefficients corresponding to the candidate flight routes are obtained.

[0053] The cumulative flight cost corresponding to the candidate flight routes is corrected based on the traffic impact coefficient to obtain the comprehensive execution cost corresponding to each candidate flight route; the comprehensive execution costs corresponding to each candidate flight route are compared, and the candidate flight route with the smallest comprehensive execution cost is selected as the final flight route.

[0054] Preferably, the step of determining whether the flight path replanning conditions are met based on the updated environmental risk status during the drone's flight includes:

[0055] The system continuously acquires real-time environmental information and UAV operational status information for the flight area, and updates the three-dimensional dynamic risk field corresponding to the flight area based on the acquired real-time environmental information. It constructs a risk evolution sequence based on the comprehensive risk values ​​of each grid within multiple consecutive sampling periods, and predicts the environmental risk state corresponding to the expected arrival of the UAV at each grid based on the risk evolution sequence. Combining the current comprehensive risk state, predicted comprehensive risk state, changes in airspace restriction status, and UAV flight performance, it determines whether the route replanning conditions are met. When the route replanning conditions are met, it identifies the risk impact area and selects either local or global route replanning based on the impact range of the risk impact area on the current flight route.

[0056] The local flight path replanning uses the grid corresponding to the current position of the UAV as the new starting node and the safety grid after the risk impact area as the new target node to regenerate the local flight segment and replace the original flight path segment.

[0057] The global flight path replanning uses the current position of the UAV as the new mission starting point and the original mission target position as the new target node to regenerate candidate flight paths and redetermine the final flight path. The regenerated flight path is submitted to the low-altitude flight monitoring platform for approval. After approval, the UAV is controlled to continue to perform the flight mission according to the regenerated flight path.

[0058] The present invention has the following beneficial effects:

[0059] First, this invention acquires static environmental information, dynamic environmental information, and UAV operating status information within the flight area simultaneously before the UAV performs a flight mission, and constructs a unified three-dimensional digital grid. For each grid, terrain accessibility indicators, wind field impact indicators, and airspace accessibility indicators are calculated, and environmental coupling analysis and quantification are embedded in the grid to comprehensively evaluate the synergistic effects between multiple types of environmental risks. This constructs a three-dimensional dynamic environmental risk field that can reflect the real-time changes in the low-altitude environment, expanding the basis for route planning from traditional static map information to dynamic risk information, effectively improving the adaptability of route planning to complex low-altitude environmental changes.

[0060] Secondly, this invention constructs a flight cost field based on a three-dimensional dynamic risk field of the environment, and combines the expected arrival time of the UAV at each grid with the spatiotemporal correlation analysis of the approved flight missions in the low-altitude flight supervision platform. By establishing multi-dimensional traffic evaluation indicators such as traffic conflict frequency, conflict intensity and conflict distribution, the traffic operation status of candidate flight routes is quantitatively analyzed, and the flight cost of the route is dynamically corrected using the traffic impact coefficient. This achieves synergistic optimization of flight environment risk and airspace traffic conflict, so that the final flight route selected not only has a lower environmental risk, but also effectively reduces the probability of flight conflict, and improves the overall feasibility of the route and the efficiency of airspace coordination.

[0061] Finally, this invention continuously updates the environmental information of the flight area during the flight of the UAV, constructs a risk evolution sequence by combining the comprehensive risk values ​​of each grid within a continuous sampling period, predicts the environmental risks on the future flight path of the UAV, and intelligently determines whether to perform local replanning or global replanning based on the prediction results and real-time environmental changes. This allows the route adjustment to be completed in advance before the environmental risks affect flight safety, avoids the UAV from entering the high-risk area that is about to be formed, improves the foresight and continuity of the route adjustment, and reduces frequent route switching and flight interruption caused by sudden environmental changes.

[0062] In summary, compared to existing technologies that primarily rely on static environmental information for route planning, this invention comprehensively utilizes dynamic environmental risks, airspace traffic operation status, and risk evolution trends to participate in route generation and flight control decisions. This transforms UAV route planning from static planning to intelligent planning oriented towards dynamic environments, improving the safety, environmental adaptability, flight stability, and mission execution efficiency of route planning in complex low-altitude environments. It is particularly suitable for UAV application scenarios with frequently changing dynamic environments, such as low-altitude logistics delivery, urban inspection, emergency rescue, power line inspection, and low-altitude airspace collaborative flight.

[0063] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0064] To more clearly illustrate the technical solutions of the embodiments of the invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For those skilled in the art, the drawings can be obtained from these drawings without creative effort.

[0065] Figure 1 The present invention provides a flowchart of a method for intelligent low-altitude flight path planning for unmanned aerial vehicles. Detailed Implementation

[0066] 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.

[0067] Application Scenarios: In actual low-altitude flight, the airspace environment in which UAVs operate exhibits strong dynamic changes. For example, changes in local wind fields, temporary airspace control, and low-altitude aircraft traffic all continuously change over time, collectively affecting the flight safety and flight path feasibility of UAVs. Current flight path planning primarily relies on static environmental data, which, given the aforementioned dynamic characteristics, struggles to comprehensively reflect real-time changes in the low-altitude airspace environment and fails to fully utilize dynamic environmental information to form a basis for flight path planning. This results in generated flight paths that are ill-suited to complex and dynamic low-altitude environments. To address these technical problems, this invention provides an intelligent low-altitude flight path planning method for UAVs. By integrating static environmental information, dynamic environmental information, and UAV operational status information, a three-dimensional dynamic risk field is constructed. Combined with traffic conflict analysis and dynamic replanning during flight, this method achieves intelligent generation and real-time optimization of flight paths in complex low-altitude environments, improving flight path safety, feasibility, and environmental adaptability. It is applicable to low-altitude logistics delivery, urban inspection, power line inspection, emergency rescue, fire fighting, police patrol, surveying, and other UAV low-altitude flight application scenarios.

[0068] like Figure 1 As shown, this embodiment of the invention provides a method for intelligent low-altitude flight path planning for unmanned aerial vehicles (UAVs), specifically including:

[0069] Step one: After receiving the flight mission, the UAV acquires the mission origin, target location, and permitted flight altitude, and loads the corresponding 3D digital map based on the flight area. Simultaneously, it collects static environmental information, dynamic environmental information, and the UAV's operational status that affect flight path planning within the flight area. Static environmental information includes building height, terrain elevation, and no-fly zone boundary information. Building height and terrain elevation describe the 3D terrain features of the low-altitude flight space, while no-fly zone boundaries determine areas the UAV cannot enter. These three elements together constitute the basic environmental data for the flight area. Dynamic environmental information includes wind speed, wind direction, low-altitude aircraft density, and temporary airspace restriction levels within the current flight area. Wind speed and wind direction reflect the impact of airflow on the UAV's flight stability. Density reflects the airspace congestion level of the target area. It is calculated by obtaining the real-time positions of surrounding aircraft in the flight area through a low-altitude flight monitoring platform and calculating the aircraft density per unit airspace volume. Temporary airspace restriction level indicates the degree of flight restriction in different areas. UAV operating status includes current position, flight speed, current heading angle, and maximum wind resistance speed. Current position is used to determine the UAV's real-time position on the 3D map. Flight speed and heading angle characterize the current flight status. Maximum wind resistance speed is used as a performance parameter of the UAV itself. Since the above data comes from electronic maps, meteorological monitoring systems, low-altitude flight monitoring platforms, and UAV onboard flight control systems, the sampling time and data format of various data types differ. All collected data undergoes unified preprocessing: assuming any data source is at time... The collected data values ​​are If the unified processing time is t, then linear interpolation is used to complete time synchronization, eliminating data deviations caused by asynchronous sampling times from different data sources, so that all subsequent environmental parameters correspond to the low-altitude environmental state at the same time. The specific synchronization formula is as follows:

[0070]

[0071] To unify the data values ​​corresponding to the time axis, where p is a positive integer representing the sampling time sequence number, and These are two adjacent sampling times, where t is the unified time node, satisfying... ≤t≤ ;

[0072] After time synchronization is completed, the values ​​of each parameter are normalized to eliminate differences in the dimensions of different physical quantities. In this embodiment, the maximum-minimum normalization method is used, and the specific normalization formula is as follows:

[0073]

[0074] and These represent the maximum and minimum values ​​within the historical statistical range of this parameter, respectively.

[0075] Step two: Divide the 3D digital map corresponding to the flight area into a regular 3D grid. Assume the flight area is divided along the X, Y, and Z directions according to preset grid spacing ΔX, ΔY, and ΔZ, respectively. The specific values ​​for ΔX, ΔY, and ΔZ are set to 20–100m based on the UAV's flight accuracy. After division, the spatial grid corresponding to the i-th row, j-th column, and k-th layer is represented as follows: Each grid cell serves as an independent environmental unit, storing both static and dynamic environmental information. Environmental risk quantification and overlay are performed on each grid cell to construct a dynamic risk field, specifically including:

[0076] The system calculates the maximum building height, maximum terrain elevation, and planned drone flight altitude within the current grid. The maximum value of the maximum building height and maximum terrain elevation within the current grid is subtracted from the planned drone flight altitude to obtain the remaining passable altitude of the current grid. If the remaining passable altitude is less than or equal to zero, the current grid is considered unflyable. Otherwise, the grid is categorized into three flight indicators based on the remaining passable altitude: safe passage, restricted passage, and impassable. Safe passage indicates that the drone has sufficient safety margin and can pass through the current grid normally; restricted passage indicates that the drone can still pass through the current grid, but the safety margin is small and it should be considered a low-priority flight area; impassable indicates that there are buildings or terrain obstacles intruding into the planned flight altitude, failing to meet the drone's safe flight conditions. A mapping relationship is established between flight indicators and terrain passability indicators. Corresponding terrain passability indicators are generated based on the flight indicators of the current grid. The terrain passability index ranges from 0 to 1. A larger value indicates stronger flight passability for the current grid and a higher safety margin for the UAV to pass through the current grid; a smaller value indicates weaker flight passability for the current grid and stronger constraints on UAV flight. When the current grid is safe to pass, a larger terrain passability index is assigned; when it is restricted to pass, the terrain passability index is reduced sequentially according to the degree of restriction; when it is impassable, the terrain passability index is set to 0, indicating that the grid cannot be used as a UAV flight path.

[0077] Extract the current grid's wind speed, wind direction, and UAV heading. Convert the wind direction and UAV heading into angles to obtain the current grid's wind direction angle and UAV heading angle. Subtract the two and take the absolute value to obtain the angle between the wind direction and the UAV's flight direction. To avoid calculation errors caused by the periodicity of the heading angle, the angle is normalized. If the calculated angle is no greater than 180°, it is taken as the actual angle; if the calculated wind direction angle is greater than 180°, the difference between 360° and the angle is taken as the actual angle, ensuring that the obtained angle is the minimum angle between the wind direction and the UAV's flight direction. Calculate the wind field impact index using the following formula: wind speed, maximum allowable wind resistance speed of the UAV, and actual angle. The calculation formula is as follows:

[0078]

[0079] in The actual included angle is V, where V is the current wind speed on the grid, and Vamx is the maximum wind resistance speed allowed for the drone. When the wind is tailwind, =0, during crosswinds =1, when facing the wind =0, wind field impact index It can highlight the impact of crosswinds on the attitude stability of UAVs. The larger the field influence index, the more severe the impact of the current grid wind field.

[0080] Based on the airspace restriction level obtained from the flight monitoring platform, the specific airspace restrictions include, but are not limited to, open airspace, restricted airspace, approved airspace, and temporarily no-fly zone. Since different airspace restriction states correspond to different levels of flight permission, a mapping relationship between airspace restriction states and airspace access indicators is pre-established, and a corresponding airspace access indicator is generated based on the current airspace restriction state of the grid. The airspace access index ranges from 0 to 1. A larger value indicates less restriction and higher accessibility for the current grid, while a smaller value indicates stricter restrictions. When the current grid is open airspace, a larger airspace access index is assigned. When it is restricted or approved airspace, the airspace access index is reduced according to the degree of restriction. When it is a temporary no-fly zone, the airspace access index is set to 0, indicating that the grid cannot be used as a flight path for drones.

[0081] Set the terrain accessibility index of the current grid. Wind field impact indicators Airspace passage indicators The terrain risk indicators are uniformly converted into corresponding risk indicators. Wind farm risk indicators and airspace risk indicators And respectively represented as:

[0082]

[0083] Terrain risk indicators Wind farm risk indicators and airspace risk indicators The average risk value is obtained by calculating the mean. Then, the standard deviation is used to calculate the dispersion of each risk indicator relative to the average risk value. The calculation formula is as follows:

[0084]

[0085] Then the dispersion and average risk value The environmental coupling coefficient K is calculated using the following formula:

[0086]

[0087] The environmental coupling coefficient K ranges from 0 to 1. When the three risk indicators are generally high and their values ​​are similar, the average risk value is large and the standard deviation is small. Therefore, the environmental coupling coefficient is close to 1, indicating that multiple risk factors act on the current grid simultaneously, and environmental risks have a strong coupling effect. When only a few risk indicators are high while the rest are low, the standard deviation increases and the environmental coupling coefficient decreases, indicating that the current grid is mainly affected by local risks, and there is no obvious coupling effect between multiple risks. The environmental coupling coefficient can comprehensively reflect the degree of synergistic influence between multiple types of environmental risks, providing a dynamic correction basis for the subsequent calculation of the grid's comprehensive risk value, rather than relying solely on the magnitude of a single environmental factor.

[0088] The terrain risk index, wind field risk index, and airspace risk index are weighted and fused according to preset weights to obtain the basic risk value of the current grid. The environmental coupling coefficient is used as a risk enhancement factor to dynamically correct the basic risk value, resulting in the comprehensive risk value of the current grid. The specific calculation formula is as follows:

[0089]

[0090] in, , and The weighting coefficients for terrain risk, wind field risk, and airspace risk are dynamically adjusted based on the UAV mission type or aircraft performance. λ is the environmental coupling enhancement coefficient, ranging from 0.2 to 0.5. A truncation function with an upper limit of one is then used to truncate the comprehensive risk value. The environmental coupling coefficient characterizes the degree of synergy between multiple environmental risks. When multiple environmental risks coexist and have strong coupling effects, the environmental coupling coefficient is large, thus enhancing the basic risk value. When environmental risks are mainly caused by a single risk factor, the environmental coupling coefficient is small, and its enhancement effect on the basic risk value is weakened. The above calculation and analysis method can reflect the importance of different environmental risks and highlight the impact of multiple risks on UAV flight safety, making the comprehensive risk value more consistent with the actual low-altitude flight environment. Thus, the comprehensive risk value for each grid can be obtained. And fill it into the corresponding grid to construct a three-dimensional dynamic risk field of the flight area;

[0091] Step 3: Convert the dynamic risk field of the 3D environment into a flight cost field, and use the flight cost as the basis for route search to generate an initial flight path. Specifically, for any grid within the flight area... Calculate the Euclidean distance from the grid to the target location as follows: and combine it with the grid's overall risk value. The current grid flight cost is calculated. The calculation formula is as follows:

[0092]

[0093] Where α is the risk cost weight and β is the distance cost weight, satisfying α + β = 1. The maximum flight distance within the flight area is represented by the flight cost value. The higher the flight cost value, the higher the flight risk and cost that the UAV needs to bear when passing through the current grid. The lower the flight cost value, the more suitable the current grid is as the area through which the flight path passes. After calculating the flight cost value of all grids, a flight cost field is established for the flight area. The grid corresponding to the mission start point is used as the starting node, and the grid corresponding to the target position is used as the target node. Continuous flight grids are searched within the flight cost field according to the principle of minimizing the cumulative flight cost, generating several initial flight paths for the UAV. Each initial flight path is sorted in ascending order of cumulative flight cost, and the top several initial flight paths are selected as candidate flight paths. In this embodiment, 6 candidate flight paths are usually selected.

[0094] Choose any candidate flight path, and calculate the estimated time points when the UAV will pass through each grid within the candidate flight path based on the UAV's planned flight speed. Let the estimated time for the UAV to enter the nth grid be denoted as . Let n represent the index of the grid within the candidate flight path. Then, the expected flight time series corresponding to the candidate flight path is represented as: , Indicates the starting node. The target node is represented by N, a positive integer, which represents the total number of grids within the initial flight path. Based on the estimated arrival time of each grid, approved flight missions are retrieved from the low-altitude flight monitoring platform, and all reported flight paths are queried. The preset time tolerance is... The system counts events where at least two flight missions need to pass through the same grid within the time tolerance, and treats these events as traffic conflict events. The number of flight missions passing through the grid in a traffic conflict event is taken as the conflict number, and the corresponding grid point is taken as the conflict point of the traffic conflict event. It should be noted that the conflict number is greater than or equal to 2. The system counts the number of conflict points where traffic conflicts occur within the candidate flight route. Each conflict point is counted only once, and the number is divided by the total number of grids on the candidate flight route to obtain the conflict frequency. The conflict frequency reflects the overall density of conflicts encountered by the candidate flight route in the spatiotemporal dimension.

[0095] Quantitative analysis of traffic conflicts on candidate flight routes yields conflict frequency, conflict intensity, and conflict distribution indices to characterize the traffic operation status of these routes. Specifically, for each conflict point on a candidate flight route, the number of conflicts at each point is extracted. The conflict numbers at all conflict points are summed, and the sum is divided by the total number of conflict points. The average value obtained is the conflict intensity index. This index has a value no less than 2. A higher value indicates more simultaneous or near-simultaneous flight missions at a single conflict point, higher air traffic congestion at that point, and greater flight conflict risk and coordination difficulty. Conversely, a lower value indicates smaller conflict scales at each conflict point, and relatively lower difficulty in coordination and resolution. The conflict intensity index is used to characterize the congestion or severity of conflict nodes on the candidate flight route. Along the flight direction of the candidate flight route, the cumulative flight arc length of each conflict point from the start of the route is measured, the average arc length of all conflict point positions relative to the start of the route is calculated, and the standard deviation of the arc length of each conflict node relative to this average value is calculated. The ratio obtained by dividing the standard deviation by the average value is the conflict distribution index. The sudden distribution index is used to characterize the degree of dispersion or clustering of each conflict node on the candidate flight route in the direction of route extension. This index is a dimensionless parameter. The larger the value, the more dispersed the distribution of conflict nodes on the route, that is, the conflict points are scattered in various sections of the route, and the scope of route adjustment or avoidance is wide and the implementation is difficult.

[0096] The conflict frequency, conflict intensity, and conflict distribution indices corresponding to candidate flight routes are normalized to eliminate differences in dimensions and value ranges among different indices, ensuring a unified evaluation scale. The three normalized indices are then fused according to preset weights to obtain the traffic impact coefficient for the candidate flight routes. The traffic impact coefficient comprehensively characterizes the overall degree of impact of traffic conflicts on the candidate flight routes, with a value ranging from 0 to 1. A higher value indicates that the candidate flight route not only experiences frequent traffic conflict events but also has a large number of flight tasks requiring coordination at individual conflict nodes, and a wide distribution of conflict nodes along the route, indicating significant overall traffic pressure and higher difficulty in route adjustment and flight coordination. The traffic impact coefficient is used as a traffic correction factor to dynamically adjust the cumulative flight cost of the candidate flight routes, yielding a comprehensive execution cost. The specific calculation formula is as follows:

[0097]

[0098] in The cumulative flight cost for candidate flight routes. The traffic impact coefficient corresponding to the candidate flight routes. The traffic impact enhancement coefficient is used to adjust the degree of impact of traffic conflicts on the cumulative flight cost, and its value ranges from 0.2 to 0.6. When the traffic impact coefficient of the candidate flight route is small, it indicates that although there may be local traffic conflicts on the candidate flight route, the overall traffic operation is relatively smooth, so the comprehensive execution cost is basically consistent with the cumulative flight cost. When the traffic impact coefficient of the candidate flight route gradually increases, it indicates that the number of traffic conflict events on the candidate flight route is gradually increasing, and the complexity of conflict coordination is constantly increasing, which increases the comprehensive execution cost accordingly, thereby reducing the priority of the candidate flight route in the final selection process. Among all candidate flight routes, the candidate flight route with the smallest comprehensive execution cost is selected as the final flight route. At the same time, the final flight route is reported to the low-altitude flight supervision platform for approval. Once approved, the final flight route is issued to the UAV flight control system, and the UAV is controlled to complete the flight mission according to the grid sequence corresponding to the final flight route.

[0099] Step four: After the final flight route is approved, to avoid local traffic conflicts arising from time overlaps between approved flight missions during actual execution, traffic conflict resolution is performed on the final flight route. Specifically: Based on the planned flight speed and grid of the UAV corresponding to the final flight route, the estimated arrival time of the UAV to each grid of the final flight route is calculated. Combined with the approved flight missions returned in real time by the low-altitude flight monitoring platform, the spatiotemporal occupancy relationship corresponding to the final flight route is re-established. For each grid on the final flight route, the estimated time of the UAV entering the grid and the estimated time of the approved flight mission occupying the grid are compared to see if they meet the preset time tolerance. When the time difference is less than or equal to the preset time tolerance and the corresponding flight altitude meets the preset altitude tolerance, it is determined that there is still a traffic conflict in the current grid, and the corresponding grid is identified as a grid to be resolved. Several consecutive grids before and after the grid to be resolved form a local conflict segment, and the estimated entry time window corresponding to the local conflict segment is calculated. A local timing adjustment permit is applied for from the low-altitude flight monitoring platform. If the low-altitude flight monitoring platform returns permission to proceed, the process is completed. For inter-altitude flight coordination, the final flight path remains unchanged. Based on the coordination time returned by the monitoring platform, the drone's flight speed is locally adjusted to allow the drone to enter the conflict-to-be-resolved grid earlier or later, thus avoiding airspace occupation time corresponding to other flight missions. Local speed adjustments prioritize drone flight performance constraints, ensuring the adjusted speed remains within the drone's permissible range while maintaining stable flight attitude. If the low-altitude flight monitoring platform indicates the current conflict cannot be resolved through time coordination, the overall final flight path remains unchanged. A neighboring grid with the lowest overall risk value that meets flight conditions is searched around the conflict-to-be-resolved grid. This neighboring grid replaces the conflict-to-be-resolved grid, forming a local detour segment. The estimated flight time for this local detour segment is recalculated, and approval is submitted to the low-altitude flight monitoring platform again. Upon approval, the local detour segment is moved to the corresponding position on the final flight path, thus resolving the traffic conflict. After all conflict-to-be-resolved grids are processed, the corresponding flight control commands in the drone's flight control system are updated simultaneously, controlling the drone to complete the flight mission according to the resolved final execution path.

[0100] Step 5: During the UAV's flight along the final execution route, the environmental conditions of the flight area are continuously monitored. When a change in the environmental conditions is detected that causes the current flight route to no longer meet the safe flight conditions, or when it is predicted that the environmental risk will exceed the safety threshold during subsequent flights, dynamic route replanning is initiated. Specifically:

[0101] The drone continuously acquires real-time wind speed, wind direction, temporary airspace restriction levels, temporary no-fly zones issued by the flight monitoring platform, and the drone's current location according to a preset sampling period. Simultaneously, it acquires the drone's current flight speed, flight heading, and remaining flight range. Centered on the drone's current location, it recalculates the terrain risk indicators, wind field risk indicators, and airspace risk indicators for the corresponding grid within a preset monitoring range, and recalculates the comprehensive risk value for each grid. This updates the three-dimensional environmental risk field corresponding to the flight area in real time, continuously recording the comprehensive risk value of each grid over multiple sampling periods, forming a risk evolution sequence for the corresponding grid. Because local wind field changes, temporary airspace restrictions, and flight activity changes in the low-altitude environment typically exhibit a certain degree of continuous evolution... Therefore, this study utilizes risk evolution sequence analysis to predict the environmental risks along the subsequent flight path of the UAV by analyzing the changing trend of the comprehensive risk value of each grid over time. Specifically, a risk evolution sequence is established, consisting of the comprehensive risk values ​​corresponding to m consecutive sampling periods of a certain grid. The rate of change of the comprehensive risk is calculated based on the risk evolution sequence, where m is a positive integer representing the length of the risk prediction window. The predicted comprehensive risk value of the UAV at the expected arrival time of the corresponding grid is predicted based on the rate of change of the comprehensive risk. The predicted comprehensive risk value is compared with a preset risk threshold, and a comprehensive judgment is made in conjunction with the latest temporary no-fly information released by the flight monitoring platform, changes in airspace restriction levels, and whether the real-time wind speed exceeds the maximum allowable wind resistance speed of the UAV. Dynamic route replanning is triggered when any of the following conditions are met:

[0102] (1) The current overall risk value exceeds the preset risk threshold;

[0103] (2) The predicted comprehensive risk value exceeds the preset risk threshold;

[0104] (3) The flight monitoring platform issues new temporary no-fly zones or the airspace restriction level changes;

[0105] (4) The real-time wind speed exceeds the maximum wind resistance speed allowed for the drone;

[0106] In this embodiment, the risk threshold is set to 0.75. When the comprehensive risk value or the predicted comprehensive risk value reaches 0.75 or above, it indicates that the current grid is simultaneously affected by multiple environmental risk factors, which has a significant impact on the flight safety of the UAV. Dynamic route replanning should be initiated to prevent the UAV from entering high-risk areas, increase the lead time for route adjustments, and improve flight continuity.

[0107] The comprehensive risk value of each grid obtained in real time is compared with a preset risk threshold. Grids with a comprehensive risk value greater than or equal to the risk threshold are marked as high-risk grids. Then, according to the spatial adjacency relationship between grids, the spatially continuous high-risk grids are clustered. Each group of continuous high-risk grids is considered as a risk impact region. The risk impact region represents a continuous high-risk space that can continuously affect the safe flight of the UAV. The risk impact region is mapped to the final execution flight path, and the number of grids corresponding to the final execution flight path covered by the risk impact region is counted. The coverage rate of the risk impact region is calculated. The coverage rate of the risk impact region is the ratio between the number of flight path grids covered by the risk impact region and the total number of grids in the final execution flight path, which is used to characterize the degree of influence of the risk region on the current flight path. In this embodiment, when the coverage rate is less than 30%, it is considered that the risk impact region only affects a local flight segment of the final execution flight path; when the coverage rate is greater than or equal to 30%, it is considered that the risk impact region has covered the remaining main flight segment, and the remaining flight path needs to be replanned.

[0108] If the risk-affected area covers the current final execution route and the coverage rate of the risk-affected area is less than 30%, then local dynamic replanning is performed. Specifically, the grid corresponding to the current position of the UAV is used as the new starting node, and the search continues along the flight direction of the final execution route. When three consecutive grids are not located within the risk-affected area, the first consecutive safe grid is determined as the local replanning endpoint. In the updated flight cost field, the local flight path between the starting node and the local replanning endpoint is searched again, and the flight cost calculation process in step three is re-executed to generate a new local flight segment. The new local flight segment is then used to replace the local segment corresponding to the final execution route, thereby forming a locally replanned route.

[0109] If the risk impact area coverage is greater than or equal to 30%, or if local dynamic replanning cannot find a continuous flight path that meets the flight conditions, then global dynamic replanning is executed. Specifically, using the UAV's current position as the new starting node and the original mission target position as the target node, the candidate flight path generation process in step three and the flight path screening process in step four are re-executed within the updated flight cost field to regenerate a new final execution route. After the new final execution route is generated, it is submitted again to the low-altitude flight monitoring platform for approval. After approval, the new flight control commands are sent to the UAV flight control system to control the UAV to continue following the new final execution route. Continue to complete the flight mission; if the approval is not approved, further determine whether the remaining flight segments of the current final execution route still meet the flight safety conditions; if the remaining flight segments will not enter the risk impact area within the preset safe flight time, maintain the current flight status and continue to fly, and continuously update the three-dimensional environmental risk field and risk prediction results; if it is expected that the UAV will enter the risk impact area within the preset safe flight time, control the UAV to fly to the safe waiting area closest to the current position and meet the flight conditions to perform hovering waiting, while continuously acquiring real-time environmental information and re-performing dynamic route replanning until a new final execution route is generated and approved before continuing to execute the flight mission.

[0110] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0111] The preferred embodiments of the invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A method for intelligent low-altitude flight path planning for unmanned aerial vehicles (UAVs), characterized in that, include: The system acquires the flight mission information of the UAV and loads the corresponding 3D digital map of the flight area. It collects the static environmental information, dynamic environmental information, and UAV operating status information of the flight area and performs preprocessing. The flight area is divided into a 3D grid. For each grid, the system determines the terrain accessibility index, wind field impact index, and airspace access index. Based on this, the system comprehensively analyzes and determines the comprehensive risk value of the corresponding grid. Based on the comprehensive risk value of each grid, the system constructs a 3D dynamic risk field of the flight area. A flight cost field is established based on the three-dimensional dynamic risk field of the environment. Several candidate flight routes are generated in the flight cost field. The degree of traffic impact is determined by combining the traffic conflict situation corresponding to the candidate flight routes. The candidate flight routes are screened according to the degree of traffic impact to determine the final flight route. The final flight path is submitted to the low-altitude flight supervision platform for approval. After approval, it is sent to the UAV flight control system to control the UAV to execute the flight mission. During the flight, the UAV continuously acquires real-time environmental information of the flight area and updates the three-dimensional dynamic risk field of the environment. Based on the updated environmental risk status, it is determined whether the flight path replanning conditions are met. When the flight path replanning conditions are met, the flight path is regenerated, and after approval, the UAV is controlled to continue to execute the flight mission.

2. The intelligent low-altitude flight path planning method for unmanned aerial vehicles according to claim 1, characterized in that: The flight mission information includes at least the mission start point, target location, and permitted flight altitude; the static environment information includes at least building height, terrain elevation, and no-fly zone boundary information; the dynamic environment information includes at least wind speed, wind direction, low-altitude aircraft density, and temporary airspace restriction level; the UAV operating status information includes at least the current location, flight speed, flight heading angle, and the UAV's maximum permissible wind resistance speed; the preprocessing includes time synchronization processing of various types of data based on a unified time reference, and normalization processing of different types of environmental parameters.

3. The intelligent low-altitude flight path planning method for unmanned aerial vehicles according to claim 1, characterized in that, The indicators for determining terrain accessibility include: The system calculates the building height and terrain elevation within the current grid, determines the remaining passable height of the current grid based on the allowed flight altitude of the UAV, judges the flight status of the current grid based on the remaining passable height, and generates a corresponding terrain passability index based on the flight status. The flight status includes at least safe passability, restricted passability, and impassable passability, and the terrain passability index is used to characterize the spatial passability of the current grid for UAV flight.

4. The intelligent low-altitude flight path planning method for unmanned aerial vehicles according to claim 1, characterized in that, The indicators for determining wind field impact include: Obtain the wind speed, wind direction, and UAV flight heading corresponding to the current grid; determine the wind field direction based on the angle between the wind direction and the UAV flight heading, and determine the wind field impact index for the corresponding grid by combining the wind speed and the maximum allowable wind resistance speed of the UAV; wherein, the wind field impact index is used to characterize the degree of influence of the wind field on the flight stability of the UAV, and is dynamically adjusted with changes in wind speed and wind field direction.

5. The intelligent low-altitude flight path planning method for unmanned aerial vehicles according to claim 1, characterized in that, The criteria for determining airspace clearance include: Obtain the airspace restriction status corresponding to the airspace to which the current grid belongs; determine the airspace access index corresponding to the current grid based on the pre-established correspondence between airspace restriction status and airspace access capacity; wherein, the airspace restriction status includes at least open airspace, restricted airspace, approved airspace and temporary no-fly airspace, and the airspace access index is used to characterize the flight permission level of the current grid.

6. The intelligent low-altitude flight path planning method for unmanned aerial vehicles according to claim 1, characterized in that, The determination of the comprehensive risk value for the corresponding grid based on the terrain accessibility index, wind field impact index, and airspace accessibility index includes: Based on the terrain accessibility index, wind field impact index, and airspace accessibility index, corresponding terrain risk index, wind field risk index, and airspace risk index are generated respectively. Based on the terrain risk index, wind field risk index, and airspace risk index, the degree of dispersion between each risk index is determined, and the environmental coupling coefficient is determined in combination with the overall risk level of each risk index to characterize the degree of synergistic influence between multiple environmental risk factors in the current grid. The basic risk value is obtained by fusing various risk indicators according to preset weights, and the basic risk value is dynamically corrected by the environmental coupling coefficient to obtain the comprehensive risk value corresponding to the current grid. The comprehensive risk value corresponding to each grid is mapped to the corresponding three-dimensional grid to form a three-dimensional environmental dynamic risk field corresponding to the flight area.

7. The intelligent low-altitude flight path planning method for unmanned aerial vehicles according to claim 1, characterized in that, The step of establishing the flight cost field based on the three-dimensional environmental dynamic risk field includes: Based on the comprehensive risk value corresponding to each grid and the distance information from each grid to the target location, the flight cost value corresponding to each grid is determined; a flight cost field is established based on the flight cost values ​​corresponding to all grids; taking the grid corresponding to the mission starting point as the starting node and the grid corresponding to the target location as the target node, a continuous flight path with a smaller cumulative flight cost is searched in the flight cost field to generate several candidate flight routes; among them, each candidate flight route is sorted in ascending order of cumulative flight cost, and a preset number of candidate flight routes are selected.

8. The intelligent low-altitude flight path planning method for unmanned aerial vehicles according to claim 1, characterized in that, The determination of the degree of traffic impact by combining the traffic conflict situation corresponding to the candidate flight routes includes: The time nodes when the UAV is expected to pass through each grid are determined based on the planned flight speed corresponding to the candidate flight route; based on the expected arrival time corresponding to each grid, the approved flight missions within the corresponding time period are obtained from the low-altitude flight monitoring platform; it is determined whether the candidate flight route and the approved flight mission pass through the same grid within the preset time range, and the grids that meet the conditions are identified as conflict points. Based on the number of conflict points corresponding to the candidate flight routes, the scale of each conflict point, and the spatial distribution of traffic conflict points along the candidate flight routes, conflict frequency index, conflict intensity index, and conflict distribution index are determined respectively; based on the conflict frequency index, conflict intensity index, and conflict distribution index, the traffic impact coefficient corresponding to the candidate flight routes is determined.

9. The intelligent low-altitude flight path planning method for unmanned aerial vehicles according to claim 8, characterized in that, The process of screening candidate flight routes based on the degree of traffic impact includes: The conflict frequency index, conflict intensity index, and conflict distribution index are uniformly processed; based on the pre-set weighted fusion processing of each evaluation index, the traffic impact coefficient corresponding to the candidate flight route is obtained. The cumulative flight cost corresponding to the candidate flight routes is corrected based on the traffic impact coefficient to obtain the comprehensive execution cost corresponding to each candidate flight route; the comprehensive execution costs corresponding to each candidate flight route are compared, and the candidate flight route with the smallest comprehensive execution cost is selected as the final flight route.

10. The intelligent low-altitude flight path planning method for unmanned aerial vehicles according to claim 9, characterized in that, The process of determining whether the flight path replanning conditions are met based on the updated environmental risk status during UAV flight includes: The system continuously acquires real-time environmental information and UAV operational status information for the flight area, and updates the three-dimensional dynamic risk field corresponding to the flight area based on the acquired real-time environmental information. It constructs a risk evolution sequence based on the comprehensive risk values ​​of each grid within multiple consecutive sampling periods, and predicts the environmental risk state corresponding to the expected arrival of the UAV at each grid based on the risk evolution sequence. Combining the current comprehensive risk state, predicted comprehensive risk state, changes in airspace restriction status, and UAV flight performance, it determines whether the route replanning conditions are met. When the route replanning conditions are met, it identifies the risk impact area and selects either local or global route replanning based on the impact range of the risk impact area on the current flight route. The local flight path replanning uses the grid corresponding to the current position of the UAV as the new starting node and the safety grid after the risk impact area as the new target node to regenerate the local flight segment and replace the original flight path segment. The global flight path replanning uses the current position of the UAV as the new mission starting point and the original mission target position as the new target node to regenerate candidate flight paths and redetermine the final flight path. The regenerated flight path is submitted to the low-altitude flight monitoring platform for approval. After approval, the UAV is controlled to continue to perform the flight mission according to the regenerated flight path.

Images