Unmanned aerial vehicle route planning method for low-altitude surveying and mapping

Abstract

The invention discloses an unmanned aerial vehicle route planning method for low-altitude surveying and mapping, and relates to the technical field of unmanned aerial vehicles. The method comprises the following steps: regularizing and partitioning a surveying and mapping area to obtain surveying and mapping sub-areas, acquiring topographic feature data and demand data of the surveying and mapping sub-areas, and calculating priority values of the surveying and mapping sub-areas; unmanned aerial vehicle performance data and surveying and mapping sub-area performance demand data are collected, single-parameter performance matching is carried out to obtain feature matching values, and an unmanned aerial vehicle surveying and mapping sub-area adaptation matrix is constructed by combining the surveying and mapping sub-area priority values; based on the unmanned aerial vehicle surveying and mapping sub-area adaptation matrix, distributing unmanned aerial vehicles through a Hungary algorithm and generating a local initial route; constructing a flight path conflict analysis model through the terrain factors, and inputting the local initial flight path into the model to obtain an adjusted flight path of the unmanned aerial vehicle; when the unmanned aerial vehicle executes a task, the operation state and progress data are collected in real time, a risk value is calculated, and if the risk value exceeds the threshold value, early warning is performed and the unmanned aerial vehicle is exchanged to realize safe and efficient route planning.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles (UAVs), and in particular to a UAV route planning method for low-altitude mapping. Background Art

[0002] Low-altitude mapping drones, with their flexibility, efficiency, low cost, and ability to acquire high-resolution data, have been widely used in a variety of fields, including urban planning, resource surveys, disaster monitoring, and engineering construction. In these fields, efficient and secure route planning technology is crucial for ensuring the smooth completion of mapping missions and improving data acquisition efficiency and quality. This is especially true when faced with complex and changing geographical environments (such as undulating terrain and densely built-up areas) and dynamic mission requirements (such as emergency mapping response), placing even higher demands on the intelligence of drone path planning.

[0003] However, existing drone path planning methods have obvious shortcomings in dealing with multi-aircraft track conflicts in complex low-altitude environments. Existing technologies mostly use fixed safety thresholds (such as horizontal spacing and vertical height difference) for conflict detection, and lack adaptability to the dynamic terrain characteristics of the surveying area (such as building density, terrain undulation, and geographical environment complexity). This leads to a high risk of conflict in the planned routes, which in turn affects the efficiency and data quality of low-altitude mapping. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the present invention provides a UAV route planning method for low-altitude mapping to solve the problems existing in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A UAV route planning method for low-altitude mapping, comprising the following steps: Step S1: Regularly partitioning the surveying area to obtain a number of surveying sub-areas; collecting geographical features, topographic relief, and building density of the surveying sub-areas to obtain topographic feature data of the surveying sub-areas; Step S2: collecting the task urgency and historical data missing status of the surveying and mapping sub-areas to obtain the surveying and mapping sub-area demand data; and calculating the surveying and mapping sub-area priority value based on the surveying and mapping sub-area demand data and the surveying and mapping sub-area terrain feature data; Step S3: Collect data on the drone's flight time, sensor resolution, and route density to obtain drone performance data; collect data on drone performance requirements for the mapping sub-area to obtain mapping sub-area performance requirement data; perform single-parameter performance matching on the drone performance data and the mapping sub-area performance requirement data to obtain a performance matching value for each parameter in the drone performance data; and construct a drone mapping sub-area adaptation matrix by combining the performance matching value and the mapping sub-area priority value. Step S4: Based on the UAV mapping sub-area adaptation matrix, the UAVs are assigned to corresponding mapping sub-areas using the Hungarian algorithm, and a local initial route is generated for the mapping sub-area; Step S5: obtaining a terrain factor by calculating the terrain feature data of the surveying sub-area; constructing a track conflict analysis model based on a space-time grid based on the terrain factor; inputting the local initial route into the track conflict analysis model to obtain an adjusted UAV route, thereby realizing route planning for the UAV.

[0006] Preferably, the step of calculating the priority value of the surveying and mapping sub-area according to the surveying and mapping sub-area demand data and the surveying and mapping sub-area terrain feature data comprises the following specific steps: According to the surveying and mapping sub-area demand data and the surveying and mapping sub-area terrain feature data, the surveying and mapping sub-area priority value is calculated:

[0007] Among them, Q is the priority value of the surveying and mapping sub-area, G is the geographical feature, DX is the terrain undulation, B is the building density, U is the task urgency, and LS is the missing historical data.

[0008] Preferably, performing single parameter performance matching on the UAV performance data and the performance requirement data of the surveying and mapping sub-area to obtain a performance matching value of each parameter in the UAV performance data comprises the following steps: By performing single parameter performance matching on the UAV performance data and the performance requirement data of the surveying and mapping sub-area, a performance matching value of each parameter in the UAV performance data is obtained; By matching the UAV flight time with the UAV required flight time:

[0009] in, is the performance matching value of the battery life. For the drone's flight time, The required flight time for the drone is It is the penalty coefficient of battery life, the default value is 0.8; By matching the performance of the drone resolution with the required drone resolution:

[0010] in, is the performance matching value of the sensor resolution, The resolution required for the drone is is the drone resolution, is the penalty coefficient of resolution, the default value is 0.7; By matching the UAV route density with the UAV required route density:

[0011] in, is the performance matching value of route density, The density of flight routes required for drones is is the drone route density, It is the penalty coefficient for route density, and the default value is 0.5.

[0012] Preferably, the step of constructing the UAV mapping sub-area adaptation matrix by combining the performance matching value and the mapping sub-area priority value comprises the following steps: By combining the performance matching value and the mapping sub-area priority value, the UAV mapping sub-area adaptation matrix is ​​constructed:

[0013] in, represents the UAV mapping sub-area adaptation value of the kth UAV and the jth mapping sub-area in the UAV mapping sub-area adaptation matrix, is the weight of the i-th feature of the UAV performance data, represents the performance matching value of the i-th feature of the k-th UAV, is the mapping sub-area priority value of the jth mapping sub-area.

[0014] Preferably, allocating the drones to corresponding mapping sub-areas by using the Hungarian algorithm based on the drone mapping sub-area adaptation matrix comprises the following steps: According to the number of UAVs and the number of surveying and mapping sub-areas, the UAV surveying and mapping sub-area adaptation matrix is ​​adjusted to obtain the adjusted UAV surveying and mapping sub-area adaptation matrix ; Construct the cost matrix C, where: =max( )- , is the value of the k-th UAV and the j-th mapping sub-area in the cost matrix, represents the value of the kth UAV and the jth mapping sub-area in the adjusted UAV mapping sub-area adaptation matrix, where max( ) represents the maximum value in the adjusted UAV mapping sub-area adaptation matrix; Perform row reduction on the cost matrix, subtract the minimum value of each row from the value in the row, and then perform column reduction on the cost matrix, subtract the minimum value of each column from the value in the column. Finally, the number of straight lines that can cover all zero elements in the cost matrix is ​​obtained, that is, the number of independent zero elements, and the independent zero elements are found; The cost matrix after row and column reduction is C'. Find the independent zero elements in C' and construct the drone allocation matrix FP:

[0015] in, The value of the kth drone and the jth mapping sub-area in the drone allocation matrix is ​​assigned to the kth drone. A value of 1 indicates that the kth drone is assigned to the jth mapping sub-area, and a value of 0 indicates that the kth drone is not assigned to the jth mapping sub-area.

[0016] Preferably, generating a local initial route for the surveying sub-area comprises the following specific steps: The UAVs are allocated to each mapping sub-area through the UAV allocation matrix. A "Z"-shaped scanning route is used for the mapping sub-area with regular shapes; Delaunay triangulation is used for the mapping sub-area with complex polygons and then cruise along the boundary. The local initial route of the k-th UAV is , ={( , , , ),( , , , ),...,( , , , ),...,( , , , )},in, represents the local initial route of the k-th UAV, , , , represents the mth route coordinate of the local initial route of the kth UAV, represents the time when the kth UAV is at the mth route coordinate, and M represents the total number of route coordinates.

[0017] Preferably, the obtaining of the terrain factor by calculating the terrain feature data of the surveying sub-area comprises the following specific steps: The terrain factor of the conflict area is calculated by using the terrain feature data of the surveyed sub-area of ​​the conflict area of ​​the UAV:

[0018] Among them, TCI is the terrain factor, G is the geographical feature, DX is the terrain undulation, and B is the building density.

[0019] Preferably, the constructing of a track conflict analysis model based on a time-space grid based on the terrain factors comprises the following specific steps: The grid resolution of the conflict area is modified by the terrain factor: = *(1- ), = *(1- ), = *(1- ),in, is the grid size in the horizontal X direction, is the grid size in the horizontal Y direction, z is the grid size in the vertical Z direction, is the attenuation coefficient, which is used to control the attenuation degree of the grid size with the terrain factor. , , is the base grid size, = = , vertical resolution = , is the minimum safety threshold in the horizontal direction, is the minimum safety threshold in the vertical direction, and the time step is t, maps the local initial route to a discrete grid; the spatiotemporal grid unit is represented by: =[ , , , ],in, represents the spatiotemporal grid cell of the k-th UAV, , , The x, y, and z coordinates of the mth route coordinates of the local initial route of the kth UAV are mapped to the space-time grid unit. It represents the time when the k-th UAV is mapped to the time grid unit at the m-th route coordinate. , , , ; Conflict detection module:

[0020] Where F is the conflict detection function, , , is the coordinate of the mth route coordinate of the local initial route of the k1th UAV mapped to the coordinate of the space-time grid unit, , , The mth route coordinates of the local initial route of the k2th UAV are mapped to the x, y, and z coordinates of the space-time grid cell. It represents the time when the k1th UAV is mapped to the time grid unit at the mth route coordinate. It represents the time when the k2th UAV is mapped to the time grid unit at the mth route coordinate. t represents the time step, is the time tolerance coefficient, when When it is 1, it means that the time difference is less than ,when When it is 2, it means that the allowed time difference is less than 2 ; Route adjustment module: The route adjustment module includes time adjustment, altitude adjustment, and level adjustment of the UAV's local initial route; Time adjustment:

[0021] in, ' is the time of the k2th UAV at the mth route coordinate after adjustment mapped to the time-space grid unit, is the time step, py is the number of offset steps, and the timing rationality must be met; Height Adjustment:

[0022] in, ' is the z coordinate of the adjusted k2th UAV in the mth route coordinate mapping in the space-time grid unit, The mth route coordinate of the local initial route of the k2th UAV is mapped to the z coordinate of the space-time grid cell, py is the number of offset steps to avoid collision with the terrain, is the grid size in the vertical Z direction; Horizontal adjustment:

[0023] in, The x-coordinate and y-coordinate of the adjusted k2th UAV in the mth route coordinate are mapped to the space-time grid unit. The x-coordinate and y-coordinate of the k2th UAV in the mth route coordinate are mapped to the space-time grid unit, is the offset angle, is the horizontal distance between the k1th UAV and the k2th UAV.

[0024] Preferably, the inputting of the local initial route into the route adjustment module to obtain the adjusted UAV route and implement the route planning of the UAV includes the following specific steps: By inputting the local initial route of the UAV into the track conflict analysis model, the adjusted UAV route is obtained. ', '=[ , , , ].

[0025] Preferably, the method further includes step S6, specifically: Step S6: When the drone is performing low-altitude mapping along the adjusted route, the drone's operating status data and drone mapping progress data are collected in real time. By combining the drone's operating status data and drone mapping progress data, a drone operation risk value is calculated. If the drone operation risk value is greater than a preset threshold, an early warning is issued and other drones are replaced to continue the mapping task. By combining the UAV's operating status data and the UAV mapping progress data, the UAV operation risk value is calculated:

[0026] Among them, FX is the UAV operation risk value, DL is the remaining battery power, ZT is the attitude stability, XT is the flight control response delay, and QU is the area of ​​the sub-area that has been completed; If the drone operation risk value is greater than the preset threshold, an early warning will be issued and other drones will be replaced to continue the surveying and mapping mission.

[0027] Beneficial effects: The present invention provides a UAV route planning method for low-altitude mapping, involving image sensor technology, which has the following beneficial effects: (1) The adaptation matrix is ​​constructed by combining the performance matching degree of drones with the priority weights of surveying and mapping sub-areas, breaking through the limitations of traditional static task allocation. Instead of considering distance or simple priority alone, it realizes the dynamic quantitative integration of two key dimensions: the degree to which the drone's own capabilities (endurance, resolution, route density) meet the actual needs of the sub-area; and the priority of the sub-area itself due to the urgency and importance of the task. Combining the two can achieve more accurate task matching, ensuring that high-performance drones give priority to serving complex and critical sub-areas with high demand and high priority, avoiding resource mismatch (such as high-performance drones being used in simple areas, or drones with insufficient capabilities being unable to handle high-demand areas); significantly improving the overall fleet utilization efficiency and the completion effect of surveying and mapping tasks.

[0028] (2) Based on the terrain complexity factor (calculated based on geographical features, terrain undulation, and building density), a spatiotemporal grid track conflict analysis model is constructed, which achieves adaptive and refined conflict detection and resolution, and can dynamically perceive the actual complexity of the environment. The terrain complexity factor quantifies the terrain challenges of the area (such as high-rise building areas or steep mountains). The model automatically adjusts the fineness of the spatiotemporal grid used for conflict detection based on this, using finer grids in complex areas to improve detection sensitivity, and using looser grids in open and flat areas to avoid unnecessary conservative avoidance. This greatly improves the safety of multi-aircraft coordinated flight at low altitudes, especially in complex environments.

[0029] (3) The initial local routes generated after allocation are input into the above-mentioned conflict analysis model. After testing and adjustment, the final executable optimized route is output, ensuring the safe, reliable and efficient execution of multi-machine collaborative operations. The model accurately identifies the potential conflict points in time and space of different drone routes (such as the possibility of meeting at the same location or getting too close), and uses strategies such as time fine-tuning (staggering arrival times), altitude adjustment (changing flight altitude layers) or horizontal avoidance (lateral deviation of routes) to completely eliminate the risk of collision. This is crucial for safe flight in low-altitude complex environments (such as urban buildings). BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0031] Figure 1 This is a flowchart of the steps of a UAV route planning method for low-altitude mapping proposed by the present invention; Figure 2 A hierarchical diagram of the steps of a UAV route planning method for low-altitude mapping proposed in the present invention; DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] See also Figure 1-2 ,The present invention provides a technical solution: a UAV route planning method for low-altitude mapping.

[0034] Step S1: Regularly partition the surveying and mapping area to obtain several surveying and mapping sub-areas; collect the geographical features, terrain undulations, and building density of the surveying and mapping sub-areas to obtain terrain feature data of the surveying and mapping sub-areas.

[0035] By partitioning the survey area into several sub-areas using regularized rules, the zoning rules are determined based on the survey objectives and regional characteristics. For example, natural geographic boundaries (rivers, ridgelines, etc.), artificial geographic boundaries (roads, street redlines), or task priorities (core operating areas, peripheral auxiliary areas) can be used as the zoning basis. Taking geographic boundaries as an example, a GIS platform can be used to import remote sensing imagery and vector maps of the survey area, extract boundary features such as rivers and main roads, and use spatial analysis tools to divide the survey area into multiple polygonal sub-areas along the boundary lines, ensuring that each sub-area boundary aligns with actual geographic features. For example, an urban survey area can be divided into independent street sub-areas based on main roads and street boundaries, or the survey area can be divided into left-bank and right-bank sub-areas along the direction of the river. After the zoning is completed, terrain feature data collection work is carried out for each surveying and mapping sub-area. High-precision DEM data is obtained through drone-mounted lidar to calculate terrain undulation parameters (such as slope and altitude difference). Remote sensing image interpretation technology is used to count the building area and distribution density in the sub-area. Combined with field surveys, information such as building height and type is recorded. At the same time, other geographical features such as vegetation coverage and water distribution are collected. Finally, a terrain feature dataset of the surveying and mapping sub-area is formed, which includes terrain undulation indicators, building density data and other related attributes, providing accurate basic data support for subsequent regional priority weight assessment and drone task allocation.

[0036] Step S2: collecting the task urgency and historical data missing of the surveying and mapping sub-area to obtain the surveying and mapping sub-area demand data; and calculating the surveying and mapping sub-area priority value based on the surveying and mapping sub-area demand data and the surveying and mapping sub-area terrain feature data.

[0037] Collect the task urgency and historical data missing of the surveying and mapping sub-areas to obtain the surveying and mapping sub-area demand data. Standardize the surveying and mapping sub-area terrain feature data and surveying and mapping sub-area demand data.

[0038] It should be noted that the terrain feature data of the surveyed sub-area include: geographical features (G), terrain relief (DX), and building density (B). Geographical features (G) are quantified as 0-1 according to the landform type (mountainous, plain, water, etc.) (e.g., mountainous = 1, water = 0.6, plain = 0.3). The complex terrain of the mountain makes route planning difficult and the risk of signal obstruction is high (frequent altitude adjustment is required). The mapping accuracy requirement is high (large terrain changes), so it is necessary to give priority to mapping and give it the highest priority. When the landform type is mountainous, the landform type value is 1. The water surface reflection in the water interferes with the sensor (lidar / camera), requiring a special flight altitude (to prevent wave splashing). If there is no terrain obstruction but signal interference exists, then When the landform type is water, the value of the landform type can be 0.6. Although the equipment loss in water operations is higher (the high humidity environment makes the annual maintenance cost reach 8,000 US dollars, which is 8 times that of plains), its key role in flood control decision-making and waterway management is irreplaceable. Therefore, its priority is still significantly higher than that of plains. The flat terrain of the plain is easy to plan straight routes, and the lack of obstructions reduces flight risks. Processing can be postponed without affecting the overall progress. Therefore, when the landform type is plain, the value of the landform type can be 0.3; Terrain relief (DX) is determined by the value of the relative relief of the surveying sub-area. , The relative relief reveals the intrinsic relationship between the horizontal expansion and vertical relief of the terrain. When the altitude change is the same, the smaller the area, the more drastic the terrain change, the greater the challenge to the drone route planning, the greater the relative relief, and the larger the DX value. DX processes the value range of the relative relief, limiting the value range of the relative relief to between 0 and 1, with the maximum DX value being 1; the building density (B) represents the proportion of the building site area to the area of ​​the surveying sub-area. The higher the building density, the greater the degree of obstruction of the buildings in the surveying sub-area to drone mapping. The building density is standardized to 0-1 (for example, if the building density is 100%, it is standardized to 1, and if the building density is 70%, it is standardized to 0.7).

[0039] It should be noted that the surveying and mapping sub-area demand data includes: mission urgency (U) and historical data missingness (LS). Mission urgency (U) is set according to the type of operation (e.g., if the disaster relief area is set to level 10, the mission urgency value is 10; if the conventional surveying and mapping area is set to level 5, the mission urgency value is 5); historical data missingness (LS) is determined by evaluating the time span and resolution of the existing data in the sub-area (the historical data missingness value is 0-10, if the data is more than 5 years old or the resolution is less than 1m, it is considered high missingness, and the historical data missingness value is 10; if the data is more than 4 years old or the resolution is less than 2m, it is considered very missing, and the historical data missingness value is 8).

[0040] According to the surveying and mapping sub-area demand data and the surveying and mapping sub-area terrain feature data, the surveying and mapping sub-area priority value is calculated:

[0041] Among them, Q is the priority value of the surveying and mapping sub-area, G is the geographical feature, DX is the terrain undulation, B is the building density, U is the task urgency, and LS is the missing historical data.

[0042] It should be noted that the sub-area priority values ​​can be used to construct a sub-area priority map, providing importance grading and spatial mapping. The sub-area priority values ​​are categorized into a three-tier system: high priority (0.8-1), medium priority (0.5-0.79), and low priority (<0.5). Using a GIS platform, each sub-area's score is mapped to a color gradient (e.g., red represents high priority, blue represents medium priority, and yellow represents low priority), generating a sub-area priority map. A dynamic correction mechanism is established. If the operational environment changes (e.g., sudden strong winds) or the task is adjusted (e.g., new emergency requirements), the environmental complexity or task urgency index is updated in real time, the sub-area priority value is recalculated, and the sub-area priority map is updated to ensure that priority rankings are synchronized with actual operational needs. For example, if a sudden disaster occurs in a sub-area, the task urgency index increases from level 5 to level 10, and the sub-area priority value increases accordingly. The sub-area's color on the sub-area priority map changes from yellow to red, automatically adjusting it to a high-priority operational area.

[0043] Step S3: Collect data on the drone's flight time, sensor resolution, and route density to obtain drone performance data; collect data on drone performance requirements in the mapping sub-area to obtain mapping sub-area performance requirement data; perform single-parameter performance matching on the drone performance data and the mapping sub-area performance requirement data to obtain a performance matching value for each parameter in the drone performance data; and construct a drone mapping sub-area adaptation matrix by combining the performance matching value and the mapping sub-area priority value.

[0044] Collect data on drone flight time, sensor resolution, and route density to obtain drone performance data. Collect data on drone performance requirements for the mapping sub-area to obtain mapping sub-area performance requirement data, including required drone flight time, required drone resolution, and required drone route density.

[0045] By performing single-parameter performance matching on the UAV performance data and the performance requirement data of the surveying and mapping sub-area, a performance matching value of each parameter in the UAV performance data is obtained.

[0046] By matching the UAV flight time with the UAV required flight time:

[0047] in, is the performance matching value of the battery life. For the drone's flight time, The required flight time for the drone is It is the penalty coefficient of battery life, the default value is 0.8.

[0048] It should be noted that when the drone's flight time Meet the UAV flight time requirements of the surveying and mapping sub-area , the matching value is directly assigned 1 point (complete adaptation); if the drone's flight time is insufficient, it will be converted proportionally and multiplied by the penalty coefficient , reflecting the principle of "points will be deducted for capability gaps". For example: the drone in the surveying and mapping sub-area requires a flight time of 100 minutes, and the drone's flight time is 80 minutes. =0.8, then the performance matching value of the battery life is 0.64.

[0049] By matching the performance of the drone resolution with the required drone resolution:

[0050] in, is the performance matching value of the sensor resolution, The resolution required for the drone is is the drone resolution, It is the penalty coefficient of resolution, and the default value is 0.7.

[0051] The smaller the sensor resolution value (the higher the accuracy), the stronger the capability. ≤Sub-area demand When the resolution is greater than 5cm (e.g. 2cm≤5cm), the matching value is 1 point (capability is better than demand); if the resolution value is larger (insufficient accuracy, e.g. 6cm>5cm), the ratio of demand to capability is converted and multiplied by the penalty coefficient of resolution. For example: the required resolution of the drone in the surveying sub-area is ≤5cm, and the resolution of the drone is 6cm. =0.7, the performance matching value of the sensor resolution is 0.58.

[0052] By matching the UAV route density with the UAV required route density:

[0053] in, is the performance matching value of route density, The density of flight routes required for drones is is the drone route density, It is the penalty coefficient for route density, and the default value is 0.5.

[0054] It should be noted that the exponential function is used to describe the characteristic that "the closer the route density is to the demand, the higher the matching value is." - | represents the difference between the flight route density required by the drone and the mapping sub-area, multiplied by the sensitivity coefficient , through the form of index, the smaller the gap, the closer the matching value is to 1. For example: the human-machine demand route density in the surveying and mapping sub-area is 8 lines / km, and the drone route density is 10 lines / km, If is 0.5, the performance matching value of route density is 0.37, which means that the matching value is low due to the large density difference (the route needs to be adjusted).

[0055] It is important to note that flight time (threshold matching): Flight time is a fundamental requirement for a drone to complete a mission. If the capability does not meet the requirements (for example, a drone with a 100-minute flight time performs a 120-minute mission), the mission will be aborted. Threshold matching, using the logic of "full score if meeting the requirements, proportional penalties if not," ensures a minimum minimum capability and prevents mission failure due to insufficient flight time. Sensor resolution (inverse matching): Resolution is inversely proportional to accuracy (for example, a 0.5cm resolution is better than a 1cm resolution). If a drone's resolution exceeds the required accuracy (for example, a 1cm capability and a 0.5cm requirement), it will not meet high-precision mapping requirements. Inverse matching ensures that data collection accuracy meets the requirements by achieving full scores when the capability value is less than or equal to the required accuracy, and penalties when it exceeds the required accuracy. Route density (progressive matching): Excessive route density can lead to redundant flights (wasting flight time), while too low a density can result in missed mapping areas. Progressive matching uses an exponential function to drive the matching value closer to 1 as the gap narrows. This allows density to fluctuate within a reasonable range, balancing mission efficiency and coverage completeness.

[0056] By combining the performance matching value and the mapping sub-area priority value, the UAV mapping sub-area adaptation matrix is ​​constructed:

[0057] in, represents the UAV mapping sub-area adaptation value of the kth UAV and the jth mapping sub-area in the UAV mapping sub-area adaptation matrix, is the weight of the i-th feature of the UAV performance data, represents the performance matching value of the i-th feature of the k-th UAV, is the mapping sub-area priority value of the jth mapping sub-area.

[0058] Step S4: Based on the UAV mapping sub-area adaptation matrix, the UAVs are assigned to corresponding mapping sub-areas using the Hungarian algorithm, and a local initial route is generated for the mapping sub-area.

[0059] Based on the UAV mapping sub-area adaptation matrix, the UAVs are assigned to corresponding mapping sub-areas through the Hungarian algorithm, and a local initial route is generated for the mapping sub-area.

[0060] According to the number of UAVs and the number of surveying and mapping sub-areas, the UAV surveying and mapping sub-area adaptation matrix is ​​adjusted to obtain the adjusted UAV surveying and mapping sub-area adaptation matrix .

[0061] It should be noted that the UAV mapping sub-area adaptation matrix is ​​adjusted according to the number of UAVs and the number of mapping sub-areas: the Hungarian algorithm requires the UAV mapping sub-area adaptation matrix to be a square matrix. When the number of UAVs is greater than the number of mapping sub-areas, the UAV mapping sub-area adaptation matrix is ​​expanded. The newly added Kj columns in the expanded UAV mapping sub-area adaptation matrix represent virtual sub-areas. The UAV mapping sub-area adaptation values ​​of the virtual sub-areas are all 0. K represents the total number of UAVs, and j represents the total number of mapping sub-areas. For example, the original UAV mapping sub-area adaptation matrix is , the adjusted UAV mapping sub-area adaptation matrix is ; When the number of drones is less than the number of surveying and mapping sub-areas, the drone surveying and mapping sub-area adaptation matrix is ​​expanded, and the newly added jK rows in the expanded drone surveying and mapping sub-area adaptation matrix represent virtual sub-areas; when the number of drones is equal to the number of surveying and mapping sub-areas, the drone surveying and mapping sub-area adaptation matrix remains unchanged.

[0062] Construct the cost matrix C in the Hungarian algorithm. The cost matrix is ​​the core input of the Hungarian algorithm and quantifies the cost of each task allocation combination. The row index represents the drone and the column index represents the mapping sub-area, where: =max( )- , is the value of the k-th UAV and the j-th mapping sub-area in the cost matrix, represents the value of the kth UAV and the jth mapping sub-area in the adjusted UAV mapping sub-area adaptation matrix, where max( ) represents the maximum value in the adjusted UAV mapping sub-area adaptation matrix.

[0063] Perform row reduction on the cost matrix, subtract the minimum value of each row from the value in the row, and then perform column reduction on the cost matrix, subtract the minimum value of each column from the value in the column. Finally, the number of straight lines that can cover all zero elements in the cost matrix is ​​obtained, that is, the number of independent zero elements, and the independent zero elements are found.

[0064] It should be noted that independent zero elements are a set of zero elements located in different rows and columns, that is, there are no two independent zero elements in the same row or column in the matrix. Each independent zero element corresponds to a valid allocation pair: row index → ​​drone number, column index → ​​mapping sub-area number It should be noted that the minimum number of horizontal and vertical lines that can cover all zero elements in the cost matrix C is found. 1. Mark all rows that are not assigned independent zero elements, 2. For each zero element in the marked row (regardless of whether it is an independent zero element), mark its column, 3. For each zero element in the marked column (regardless of whether it is an independent zero element), mark its row, repeat steps 2-3 until it is no longer possible to mark, and cover all unmarked rows and marked columns with straight lines. If the number of covered straight lines is less than the matrix order, perform the following adjustment: find the minimum value among the elements not covered by any straight line , the uncovered elements are subtracted , the elements covered by the two lines plus , the elements covered by only one line remain unchanged; if the number of covered lines is equal to the matrix order, the currently assigned independent zeros is the optimal solution.

[0065] The cost matrix after row and column reduction is C'. Find the independent zero elements in C' and construct the drone allocation matrix FP:

[0066] in, The value of the kth drone and the jth mapping sub-area in the drone allocation matrix is ​​assigned to the kth drone. A value of 1 indicates that the kth drone is assigned to the jth mapping sub-area, and a value of 0 indicates that the kth drone is not assigned to the jth mapping sub-area.

[0067] The drone allocation matrix is ​​used to allocate drones to the mapping sub-areas (when the number of drones is greater than the number of mapping sub-areas, some drones will remain; when the number of drones is less than the number of mapping sub-areas, some mapping sub-areas will not have drones allocated, and will enter the waiting sequence. When all drones have completed their tasks, the drone allocation matrix of all drones and mapping sub-areas in the waiting sequence will be recalculated, and the drones will be allocated to the drones in the waiting sequence). For mapping sub-areas with regular shapes (such as rectangles), a "Z"-shaped scanning route is adopted; for mapping sub-areas with complex polygons, Delaunay triangulation is used and then cruises along the boundaries. The local initial route of the kth drone is , ={( , , , ),( , , , ),...,( , , , ),...,( , , , )},in, represents the local initial route of the k-th UAV, , , , represents the mth route coordinate of the local initial route of the kth UAV, represents the time when the kth UAV is at the mth route coordinate, and M represents the total number of route coordinates.

[0068] Step S5: obtaining a terrain factor by calculating the terrain feature data of the surveying sub-area; constructing a track conflict analysis model based on a space-time grid based on the terrain factor; inputting the local initial route into the track conflict analysis model to obtain an adjusted UAV route, thereby realizing route planning for the UAV.

[0069] A track conflict analysis model based on space-time grid is constructed to make local adjustments to track segments with potential conflicts and realize decoupled flight route planning.

[0070] The track conflict analysis model includes: a time grid division module, a conflict detection module, and a route adjustment module.

[0071] Time grid division module: The terrain factor of the conflict area is calculated by using the terrain feature data of the surveyed sub-area of ​​the conflict area of ​​the UAV:

[0072] Among them, TCI is the terrain factor, G is the geographical feature, DX is the terrain undulation, and B is the building density.

[0073] It should be noted that the conflict area of ​​the drone is the boundary area of ​​the surveying and mapping sub-areas. The boundary area is often an area with sudden changes in geographical features (such as the transition zone between plains and mountains, and the edge of a building complex), which leads to increased climbing or turning maneuvers of the drone and reduced route stability.

[0074] Conflicts are meshed, and the grid resolution is modified using the terrain factor. In low-altitude mapping, terrain complexity directly determines the effective sensor accuracy required. The terrain factor (TCI) quantifies the combined effects of geographic features, topographic relief, and building density. When the TCI approaches 1 (e.g., in steep mountainous terrain or urban areas with densely populated buildings), abrupt changes in surface geometry can cause significant distortion in standard-resolution scan data. This distortion stems from multiple reflections in three-dimensional space: in densely built-up areas, lidar pulses can generate multiple echoes between adjacent buildings, resulting in ghost outlines in the point cloud data. In complex mountainous terrain, where the surface inclination exceeds 15°, fixed-resolution scans can lose key terrain features due to projection distortion. Dynamically attenuating resolution using the terrain factor establishes an inversely proportional adaptation mechanism between resolution and terrain complexity, avoiding wasted flight time due to oversampling in plain areas while ensuring mapping quality in complex areas.

[0075] The grid resolution of the conflict area is modified by the terrain factor: = *(1- ), = *(1- ), = *(1- ),in, is the grid size in the horizontal X direction, is the grid size in the horizontal Y direction, z is the grid size in the vertical Z direction, is the attenuation coefficient, which is used to control the attenuation degree of the grid size with the terrain factor. , , is the base grid size, = = , vertical resolution = , is the minimum safety threshold in the horizontal direction, is the minimum safety threshold in the vertical direction, and the time step is t, the local initial route is mapped to a discrete grid, and the spatiotemporal grid unit is represented as: =[ , , , ],in, represents the spatiotemporal grid cell of the k-th UAV, , , The x, y, and z coordinates of the mth route coordinates of the local initial route of the kth UAV are mapped to the space-time grid unit. It represents the time when the k-th UAV is mapped to the time grid unit at the m-th route coordinate. , , , .

[0076] Conflict detection module:

[0077] Where F is the conflict detection function, , , is the coordinate of the mth route coordinate of the local initial route of the k1th UAV mapped to the coordinate of the space-time grid unit, , , The mth route coordinates of the local initial route of the k2th UAV are mapped to the x, y, and z coordinates of the space-time grid cell. It represents the time when the k1th UAV is mapped to the time grid unit at the mth route coordinate. It represents the time when the k2th UAV is mapped to the time grid unit at the mth route coordinate. t represents the time step, is the time tolerance coefficient, when When it is 1, it means that the time difference is less than ,when When it is 2, it means that the allowed time difference is less than 2 .

[0078] It should be noted that the UAV conflict monitoring module is the core component of the trajectory conflict analysis model. It determines the conflict through two conditions: when two UAVs are at the same spatial grid coordinates (i.e., three-dimensional position overlap) and the time difference is less than the set fault tolerance threshold When the collision risk is detected, for physical collision avoidance, the spatial coordinates must be completely consistent and the time difference must be close to zero (η≈0.1). For surveying and mapping task conflicts (such as repeated scanning of the same area), the time condition is relaxed (η=1.5).

[0079] Route adjustment module: The route adjustment module includes time adjustment, altitude adjustment, and level adjustment of the UAV's local initial route.

[0080] When the conflict monitoring module identifies a route conflict, it determines the mapping sub-area priority values ​​of the mapping sub-areas corresponding to the two UAVs. For example, when adjusting the k2th UAV, the mapping sub-area priority value of the mapping sub-area corresponding to the k2th UAV should be smaller than the mapping sub-area priority value corresponding to the k1th UAV.

[0081] Time adjustment: , in, ' is the time of the k2th UAV at the mth route coordinate after adjustment mapped to the time-space grid unit, is the time step, py is the number of offset steps, and timing rationality must be met.

[0082] Height Adjustment:

[0083] in, The z coordinate of the adjusted k2th UAV in the mth route coordinate is mapped to the space-time grid unit, is the mth route coordinate of the local initial route of the k2th UAV mapped to the z coordinate of the space-time grid unit, py is the number of offset steps, is the grid size in the vertical Z direction.

[0084] Horizontal adjustment:

[0085] in, The x-coordinate and y-coordinate of the adjusted k2th UAV in the mth route coordinate are mapped to the space-time grid unit. The x-coordinate and y-coordinate of the k2th UAV in the mth route coordinate are mapped to the space-time grid unit, is the offset angle, is the horizontal distance between the k1th UAV and the k2th UAV.

[0086] It should be noted that when two drones are at the same spatial grid coordinates (i.e., three-dimensional positions overlap) and the time difference is less than the set fault tolerance threshold When there is a collision risk, the mapping sub-area priority values ​​of the mapping sub-areas corresponding to the two drones are determined. For example, the mapping sub-area priority value of the mapping sub-area corresponding to the k2th drone is adjusted to be smaller than the mapping sub-area priority value of the mapping sub-area corresponding to the k1th drone. Time is used to judge the situation of the two drones and make corresponding adjustments. Horizontal adjustment: before the two drones collide, At time t, when the actual horizontal distance between the k1th UAV and the k2th UAV is less than the horizontal safety threshold, the horizontal offset is calculated; for example, when the horizontal distance between the k1th UAV and the k2th UAV is 150m, the minimum horizontal safety threshold is 200m, and a horizontal offset is required to make the horizontal distance between the two UAVs reach the safety threshold. The required offset distance is 200-150=50m, and the offset direction is determined by the angle Determine that, by decomposing the offset into the x-axis and y-axis through trigonometric functions, the coordinates of the adjusted k2th UAV on the mth route are mapped to the space-time grid unit as follows: . Altitude Adjustment: Before the two drones collide When the vertical distance between the k1th UAV and the k2th UAV is less than the safety threshold, the height adjustment is performed; Time adjustment: when the allowed time tolerance If it is too small, it may cause rear-end collision or there will be obstacles when adjusting the horizontal offset distance, so time offset adjustment is performed.

[0087] By inputting the local initial route of the UAV into the route adjustment module, the adjusted UAV route is obtained. ', '=[ , , , ].

[0088] Step S6: When the UAV is performing low-altitude mapping via the adjusted route, the UAV's operating status data and the UAV's mapping progress data are collected in real time. By combining the UAV's operating status data and the UAV's mapping progress data, the UAV's operating risk value is calculated. If the UAV's operating risk value is greater than a preset threshold, an early warning is issued and other UAVs are replaced to continue the mapping task.

[0089] As the drone conducts low-altitude mapping along the adjusted route, real-time operating status data and mapping progress data are collected and normalized to [0, 1]. The operating status data includes the remaining battery life, attitude stability, and flight control response delay. The mapping progress data includes the percentage of sub-areas that have been mapped.

[0090] By combining the UAV's operating status data and the UAV mapping progress data, the UAV operation risk value is calculated:

[0091] Among them, FX is the UAV operation risk value, DL is the remaining battery power, ZT is the attitude stability, XT is the flight control response delay, and QU is the area ratio of the sub-area that has been completed.

[0092] If the drone operation risk value is greater than the preset threshold, an early warning will be issued and other drones will be replaced to continue the surveying and mapping mission.

[0093] It should be noted that when a drone's operational risk exceeds a preset threshold (e.g., 0.7), an alert is activated, triggering a drone replacement mechanism. With the risk drone as the center, a search is conducted within a set radius (e.g., 1000 meters) for nearby drones that meet the battery requirements (remaining battery power at least 10% above the minimum safety value). These candidate drones must have a lower priority value for their mapping sub-area than the sub-area corresponding to the drone whose operational risk exceeds the threshold. Next, the system prioritizes the candidate drone based on its performance match for that sub-area, prioritizing it with a higher performance match and more sufficient battery power to minimize mission interruption. After determining the optimal candidate drone, the system sends the coordinates of the original drone's unfinished mapping mission to the candidate drone and, using a track conflict analysis model, reroutes the candidate drone. The original risk drone then executes a return or standby command based on its status. The entire replacement process is completed through real-time data exchange, ensuring seamless mapping mission continuity and reducing the risk of mission interruption due to single-machine failures.

[0094] This paper proposes an intelligent multi-UAV collaborative route planning method for low-altitude mapping scenarios. The method first divides the mapping area into regular subareas. For each subarea, terrain data, such as geographic features, topographic relief, and building density, is collected, along with demand data such as mission urgency and missing historical data. Based on this data, a priority weight is calculated for each subarea. Furthermore, the performance parameters of available UAVs (such as flight time, sensor resolution, and route density) are collected and evaluated. Next, the UAVs' performance is matched to the subarea requirements. An adaptation matrix is ​​constructed based on the subarea priorities. An optimization algorithm (such as the Hungarian algorithm) is then used to assign the optimal mapping subarea to the UAVs and generate an initial route. Subsequently, a track conflict analysis model is dynamically constructed, taking into account the complexity of the subarea terrain, to safely adjust the initial route. Finally, the UAVs' status and progress are monitored in real time during the mission, operational risks are assessed, and UAVs are dynamically replaced when risks become excessive, ensuring efficient, safe, and reliable completion of the mapping mission.

[0095] Combining drone performance matching with the priority weights of mapping sub-areas to construct an adaptation matrix breaks through the limitations of traditional static task allocation. Instead of simply considering distance or simple priority, it achieves a dynamic, quantitative fusion of two key dimensions: the degree to which the drone's capabilities (endurance, resolution, and route density) meet the actual needs of the sub-area; and the sub-area's own priority, reflected by the urgency and importance of the task. Combining these two allows for more precise task matching, ensuring that high-performance drones prioritize complex, high-priority sub-areas with high demand, avoiding resource mismatches (such as high-performance drones being used in simple areas or less capable drones being unable to handle demanding areas). This significantly improves overall fleet utilization and the effectiveness of mapping tasks.

[0096] Based on the terrain complexity factor (calculated based on geographical features, terrain relief, and building density), a space-time grid track conflict analysis model is constructed, enabling adaptive and refined conflict detection and resolution, dynamically sensing the actual complexity of the environment. The terrain complexity factor quantifies the challenging terrain of an area (such as a densely populated area or steep mountainous terrain). The model automatically adjusts the fineness of the space-time grid used for conflict detection, using a finer grid in complex areas to improve detection sensitivity and a looser grid in open, flat areas to avoid unnecessary conservative avoidance. This significantly improves the safety of multi-aircraft coordinated flight at low altitudes, especially in complex environments.

[0097] The initial local routes generated after allocation are input into the aforementioned conflict analysis model. After testing and adjustment, the final, executable optimized routes are output, ensuring the safe, reliable, and efficient execution of multi-drone collaborative operations. The model accurately identifies potential conflict points in time and space between different drone routes (such as the possibility of encountering at the same location or approaching too close). Strategies such as time adjustments (staggering arrival times), altitude adjustments (changing flight levels), or horizontal avoidance (laterally shifting routes) are then employed to completely eliminate collision risks. This is crucial for safe flight in complex low-altitude environments, such as urban complexes.

[0098] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include," "comprise," or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations. The phrase "includes an element defined by..." does not exclude the presence of other identical elements in the process, method, article, or device that includes the element.

[0099] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A UAV route planning method for low-altitude mapping, characterized by: The following steps are involved: Step S1: Regularly partitioning the surveying area to obtain a number of surveying sub-areas; collecting geographical features, topographic relief, and building density of the surveying sub-areas to obtain topographic feature data of the surveying sub-areas; Step S2: collecting the task urgency and historical data missing status of the surveying and mapping sub-areas to obtain the surveying and mapping sub-area demand data; and calculating the surveying and mapping sub-area priority value based on the surveying and mapping sub-area demand data and the surveying and mapping sub-area terrain feature data; Step S3: Collect data on the drone's flight time, sensor resolution, and route density to obtain drone performance data; collect data on drone performance requirements for the mapping sub-area to obtain mapping sub-area performance requirement data; perform single-parameter performance matching on the drone performance data and the mapping sub-area performance requirement data to obtain a performance matching value for each parameter in the drone performance data; and construct a drone mapping sub-area adaptation matrix by combining the performance matching value and the mapping sub-area priority value. Step S4: Based on the UAV mapping sub-area adaptation matrix, the UAVs are assigned to corresponding mapping sub-areas using the Hungarian algorithm, and a local initial route is generated for the mapping sub-area; Step S5: obtaining a terrain factor by calculating the terrain feature data of the surveying sub-area; constructing a track conflict analysis model based on a space-time grid based on the terrain factor; inputting the local initial route into the track conflict analysis model to obtain an adjusted UAV route, thereby realizing route planning for the UAV.

2. The method for UAV route planning for low-altitude mapping according to claim 1, characterized in that: The step of calculating the priority value of the surveying and mapping sub-area according to the surveying and mapping sub-area demand data and the surveying and mapping sub-area terrain feature data comprises the following specific steps: According to the surveying and mapping sub-area demand data and the surveying and mapping sub-area terrain feature data, the surveying and mapping sub-area priority value is calculated: ； Among them, Q is the priority value of the surveying and mapping sub-area, G is the geographical feature, DX is the terrain undulation, B is the building density, U is the task urgency, and LS is the missing historical data.

3. The method for UAV route planning for low-altitude mapping according to claim 2, characterized in that: The method of performing single parameter performance matching on the UAV performance data and the performance requirement data of the surveying and mapping sub-area to obtain a performance matching value of each parameter in the UAV performance data includes the following steps: By performing single parameter performance matching on the UAV performance data and the performance requirement data of the surveying and mapping sub-area, a performance matching value of each parameter in the UAV performance data is obtained; By matching the UAV flight time with the UAV required flight time: ； in, is the performance matching value of the battery life. For the drone's flight time, The required flight time for the drone is It is the penalty coefficient of battery life, the default value is 0.8; By matching the performance of the drone resolution with the required drone resolution: ； in, is the performance matching value of the sensor resolution, The resolution required for the drone is is the drone resolution, is the penalty coefficient of resolution, the default value is 0.7; By matching the UAV route density with the UAV required route density: ； in, is the performance matching value of route density, The density of flight routes required for drones is is the drone route density, It is the penalty coefficient for route density, and the default value is 0.

5.

4. The method for UAV route planning for low-altitude mapping according to claim 3, characterized in that: The method of constructing a UAV mapping sub-area adaptation matrix by combining the performance matching value and the mapping sub-area priority value includes the following steps: By combining the performance matching value and the mapping sub-area priority value, the UAV mapping sub-area adaptation matrix is ​​constructed: ； in, represents the UAV mapping sub-area adaptation value of the kth UAV and the jth mapping sub-area in the UAV mapping sub-area adaptation matrix, is the weight of the i-th feature of the UAV performance data, represents the performance matching value of the i-th feature of the k-th UAV, is the mapping sub-area priority value of the jth mapping sub-area.

5. The method for UAV route planning for low-altitude mapping according to claim 4, characterized in that: The method of allocating the drones to corresponding mapping sub-areas based on the drone mapping sub-area adaptation matrix by using the Hungarian algorithm includes the following steps: According to the number of UAVs and the number of surveying and mapping sub-areas, the UAV surveying and mapping sub-area adaptation matrix is ​​adjusted to obtain the adjusted UAV surveying and mapping sub-area adaptation matrix ; Construct the cost matrix C, where: =max( )- , is the value of the k-th UAV and the j-th mapping sub-area in the cost matrix, represents the value of the kth UAV and the jth mapping sub-area in the adjusted UAV mapping sub-area adaptation matrix, where max( ) represents the maximum value in the adjusted UAV mapping sub-area adaptation matrix; Perform row reduction on the cost matrix, subtract the minimum value of each row from the value in the row, and then perform column reduction on the cost matrix, subtract the minimum value of each column from the value in the column. Finally, the number of straight lines that can cover all zero elements in the cost matrix is ​​obtained, that is, the number of independent zero elements, and the independent zero elements are found; The cost matrix after row and column reduction is C'. Find the independent zero elements in C' and construct the drone allocation matrix FP: ； in, The value of the kth drone and the jth mapping sub-area in the drone allocation matrix is ​​assigned to the kth drone. A value of 1 indicates that the kth drone is assigned to the jth mapping sub-area, and a value of 0 indicates that the kth drone is not assigned to the jth mapping sub-area.

6. The method for UAV route planning for low-altitude mapping according to claim 5, characterized in that: The generating of a local initial route for the surveying sub-area comprises the following specific steps: The drones are allocated to each mapping sub-area through the drone allocation matrix. A "Z"-shaped scanning route is adopted for mapping sub-areas with regular shapes; Delaunay triangulation is adopted for mapping sub-areas with complex polygons, and the drones are cruised along the boundaries. , ={( , , , ),( , , , ),...,( , , , ),...,( , , , )},in, represents the local initial route of the k-th UAV, , , , represents the mth route coordinate of the local initial route of the kth UAV, represents the time when the kth UAV is at the mth route coordinate, and M represents the total number of route coordinates.

7. The method for UAV route planning for low-altitude mapping according to claim 6, characterized in that: The method of obtaining the terrain factor by calculating the terrain feature data of the surveying sub-area includes the following specific steps: The terrain factor of the conflict area is calculated by using the terrain feature data of the surveyed sub-area of ​​the conflict area of ​​the UAV: ； Among them, TCI is the terrain factor, G is the geographical feature, DX is the terrain undulation, and B is the building density.

8. The method for UAV route planning for low-altitude mapping according to claim 7, characterized in that: The method of constructing a track conflict analysis model based on a time-space grid based on the terrain factors includes the following specific steps: The grid resolution of the conflict area is modified by the terrain factor: = *(1- ), = *(1- ), = *(1- ),in, is the grid size in the horizontal X direction, is the grid size in the horizontal Y direction, z is the grid size in the vertical Z direction, is the attenuation coefficient, which is used to control the attenuation degree of the grid size with the terrain factor. , , is the base grid size, = = , vertical resolution = , is the minimum safety threshold in the horizontal direction, is the minimum safety threshold in the vertical direction, and the time step is t, maps the local initial route to a discrete grid; the spatiotemporal grid unit is represented by: =[ , , , ],in, represents the spatiotemporal grid cell of the k-th UAV, , , The x, y, and z coordinates of the mth route coordinates of the local initial route of the kth UAV are mapped to the space-time grid unit. It represents the time when the k-th UAV is mapped to the time grid unit at the m-th route coordinate. , , , ; Conflict detection module: ； Where F is the conflict detection function, , , is the coordinate of the mth route coordinate of the local initial route of the k1th UAV mapped to the coordinate of the space-time grid unit, , , The mth route coordinates of the local initial route of the k2th UAV are mapped to the x, y, and z coordinates of the space-time grid cell. It represents the time when the k1th UAV is mapped to the time grid unit at the mth route coordinate. It represents the time when the k2th UAV is mapped to the time grid unit at the mth route coordinate. t represents the time step, is the time tolerance coefficient, when When it is 1, it means that the time difference is less than ,when When it is 2, it means that the allowed time difference is less than 2 ; Route adjustment module: The route adjustment module includes time adjustment, altitude adjustment, and level adjustment of the UAV's local initial route; Time adjustment: ； in, ' is the time of the k2th UAV at the mth route coordinate after adjustment mapped to the time-space grid unit, is the time step, py is the number of offset steps, and the timing rationality must be met; Height Adjustment: ； in, ' is the z coordinate of the adjusted k2th UAV in the mth route coordinate mapping in the space-time grid unit, is the mth route coordinate of the local initial route of the k2th UAV mapped to the z coordinate of the space-time grid unit, py is the number of offset steps, is the grid size in the vertical Z direction; Horizontal adjustment: ； in, The x-coordinate and y-coordinate of the adjusted k2th UAV in the mth route coordinate are mapped to the space-time grid unit. The x-coordinate and y-coordinate of the k2th UAV in the mth route coordinate are mapped to the space-time grid unit, is the offset angle, is the horizontal distance between the k1th UAV and the k2th UAV.

9. The method for UAV route planning for low-altitude mapping according to claim 8, characterized in that: Inputting the local initial route into the track conflict analysis model to obtain the adjusted UAV route and implement route planning for the UAV includes the following specific steps: By inputting the local initial route of the UAV into the route adjustment module, the adjusted UAV route is obtained. ', '=[ , , , ].

10. The UAV route planning method for low-altitude mapping according to claim 9, characterized in that: The step S6 is further included, specifically: Step S6: When the drone is performing low-altitude mapping along the adjusted route, the drone's operating status data and drone mapping progress data are collected in real time. By combining the drone's operating status data and drone mapping progress data, a drone operation risk value is calculated. If the drone operation risk value is greater than a preset threshold, an early warning is issued and other drones are replaced to continue the mapping task. By combining the UAV's operating status data and the UAV mapping progress data, the UAV operation risk value is calculated: ； Among them, FX is the UAV operation risk value, DL is the remaining battery power, ZT is the attitude stability, XT is the flight control response delay, and QU is the area of ​​the sub-area that has been completed; If the drone operation risk value is greater than the preset threshold, an early warning will be issued and other drones will be replaced to continue the surveying and mapping mission.

Images