Unmanned aerial vehicle flight path planning method and device for rescue scene

Abstract

The invention discloses a rescue scene-oriented unmanned aerial vehicle flight path planning method and device, and relates to the technical field of low-altitude unmanned aerial vehicles. The method comprises the following steps: constructing an airspace risk index system, analyzing and processing each index in the airspace risk index system, and determining a key risk index; performing airspace risk quantitative evaluation based on the key risk indexes to obtain airspace risk factors; based on the airspace risk factors, determining an airspace model; constructing an unmanned aerial vehicle route planning model based on the airspace model; and performing flight path planning based on the unmanned aerial vehicle flight path planning model to obtain a flight path planning result. According to the invention, the trajectory planning efficiency and flight safety of the unmanned aerial vehicle are improved.

Description

Technical Field

[0001] This application relates to the field of low-altitude unmanned aerial vehicle (UAV) technology, and in particular to a UAV trajectory planning method and apparatus for rescue scenarios. Background Technology

[0002] Aviation emergency rescue, with its wide rescue range, strong mobility, and rapid response speed, has become an important means of responding to natural disasters and accidents. However, the environmental conditions in disaster-stricken areas pose significant challenges to aircraft operational safety, especially under the influence of factors such as fire spread, smoke obstruction, and complex terrain. Under these constraints, rapidly searching for one or more collision-free, highly safe, and low-risk routes from the starting point to the target location has become a critical requirement in forest and grassland aviation rescue scenarios.

[0003] However, current aerial rescue operations in forest and grassland areas face significant challenges in operational risk assessment. The lack of systematic and scientific assessment methods hinders effective support for flight safety in complex environments. Research on path planning primarily focuses on obstacle avoidance on the ground, failing to adequately integrate comprehensive consideration of airspace risk factors and neglecting the dynamic threats and airspace constraints inherent in aerial rescue missions. Therefore, current UAV trajectory planning is ineffective and results in low flight safety. Summary of the Invention

[0004] The embodiments of this application aim to at least partially solve one of the technical problems in the related art. Therefore, the first objective of the embodiments of this application is to provide a method, apparatus, device, medium, and program product for unmanned aerial vehicle (UAV) trajectory planning in rescue scenarios.

[0005] This application provides a method for UAV trajectory planning in rescue scenarios. The method includes: constructing an airspace risk indicator system, analyzing and processing each indicator in the airspace risk indicator system to determine key risk indicators; performing a quantitative assessment of airspace risk based on the key risk indicators to obtain airspace risk factors; determining an airspace model based on the airspace risk factors; constructing a UAV trajectory planning model based on the airspace model; and performing trajectory planning based on the UAV trajectory planning model to obtain the trajectory planning result.

[0006] In other implementations, the airspace risk indicator system includes multiple main indicators and multiple sub-indicators corresponding to each main indicator. The analysis and processing of each indicator in the airspace risk indicator system to determine key risk indicators includes: performing hierarchical analysis on the multiple main indicators to determine a first weight value for each main indicator; performing hierarchical analysis on the multiple sub-indicators corresponding to each main indicator to determine a second weight value for each sub-indicator; for each sub-indicator, determining a comprehensive weight based on the corresponding second weight value and the first weight value of the main indicator corresponding to the sub-indicator; and based on the comprehensive weight of each sub-indicator, determining at least one sub-indicator from the multiple sub-indicators as a key risk indicator.

[0007] In other implementations, airspace risk is quantitatively assessed based on key risk indicators to obtain airspace risk factors. This includes: dividing the airspace into grids based on preset rules to obtain airspace grids, wherein the airspace grids include multiple grids; determining the indicator values ​​of key risk indicators corresponding to each grid; inputting the indicator values ​​as input variables into the risk quantification assessment model, setting the output variables in the risk quantification assessment model to preset values, and minimizing the risk variables in the risk quantification assessment model to obtain the operational risk quantification values ​​corresponding to the risk variables as airspace risk factors.

[0008] In other implementations, the airspace model includes a set of static obstacles; based on the airspace model, a UAV trajectory planning model is constructed, including: determining the initial node and target node of the UAV flight in the airspace based on the airspace model; constructing the UAV trajectory based on the initial node and target node; determining the trajectory cost based on the UAV trajectory to obtain the UAV trajectory planning model, wherein the UAV trajectory planning model includes a trajectory cost model, a first constraint and a second constraint, the first constraint indicating that the UAV trajectory has no collision conflict with the set of static obstacles, and the second constraint indicating that the position of the target node is the position of the preset target endpoint.

[0009] In other implementations, the trajectory cost model includes operating costs and operating risks. Determining operating costs and operating risks includes: determining spatial distance costs based on the distance between the positions of adjacent trajectory nodes in the UAV trajectory in three-dimensional space; determining operating costs based on spatial distance costs; determining the minimum distance between any trajectory node in the UAV trajectory and a set of static obstacles; determining risk values ​​based on airspace risk factors of adjacent trajectory nodes in the UAV trajectory; and determining operating risks based on the minimum distance, risk values, operating risk attention adjustment coefficient, and operating risk sensitivity adjustment coefficient.

[0010] In other implementations, trajectory planning is performed based on the UAV trajectory planning model to obtain trajectory planning results, including: constructing a cost estimation model based on a heuristic function, wherein the cost estimation model is used to determine the estimated cost between the current node and the preset target endpoint in the trajectory, the estimated cost between the current node and the preset target endpoint includes a weighted average of estimated operating costs and estimated operating risks, and the heuristic function includes a weight factor corresponding to the estimated operating costs and a weight factor corresponding to the estimated operating risks; and performing search iterative calculations based on the UAV trajectory planning model and the cost estimation model to obtain trajectory planning results.

[0011] In other implementations, determining the estimated operating cost includes: determining the spatial distance between the current node's position and the position of a preset target endpoint; if an assigned track exists, determining a first adaptive adjustment factor based on the actual operating cost of the assigned track and the distance between the start and end points of the assigned track; if no assigned track exists, determining the first adaptive adjustment factor as a preset value; and determining the estimated operating cost based on the first adaptive adjustment factor and the spatial distance.

[0012] In other implementations, the airspace model includes a set of static obstacles; determining the estimated operational risk includes: determining a straight three-dimensional path from the current node to a preset target endpoint; determining first conflict data between the straight three-dimensional path and the set of static obstacles; if an assigned track exists, determining a second adaptive adjustment factor based on the actual operational risk of the assigned track and the second conflict data of the assigned track; if no assigned track exists, determining the second adaptive adjustment factor as a preset value; and determining the estimated operational cost based on the second adaptive adjustment factor and the first conflict data.

[0013] In other implementations, a search iterative calculation is performed based on the UAV trajectory planning model and the cost estimation model to obtain the trajectory planning result, including: determining the actual cost between the current node and its neighboring nodes based on the trajectory cost model in the UAV trajectory planning model; determining the estimated cost between the neighboring node and the target node based on the cost estimation model; calculating the target cost based on the actual cost and the estimated cost between the neighboring node and the target node; if the neighboring node is not in the open list, adding the neighboring node to the open list and determining the parent node of the neighboring node as the current node; if the neighboring node is in the open list, but the actual cost to reach the current node is lower than the actual cost to reach the neighboring node, updating the actual cost, the target cost, and the parent node corresponding to the current node; selecting the node with the lowest target cost from the open list as the current node for iterative calculation, and obtaining the trajectory planning result after satisfying the iteration conditions.

[0014] This application provides a drone trajectory planning device for rescue scenarios. The device includes: an analysis and processing module for constructing an airspace risk indicator system and analyzing and processing each indicator in the airspace risk indicator system to determine key risk indicators; an evaluation module for performing a quantitative evaluation of airspace risk based on the key risk indicators to obtain airspace risk factors; a determination module for determining an airspace model based on the airspace risk factors; a construction module for constructing a drone trajectory planning model based on the airspace model; and a planning module for performing trajectory planning based on the drone trajectory planning model to obtain trajectory planning results.

[0015] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any of the above embodiments.

[0016] This application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method of any of the above embodiments.

[0017] This application provides a computer program product that includes instructions that, when executed by a processor of a computer device, enable the computer device to perform the steps of the method described in any of the above embodiments. Attached Figure Description

[0018] Figure 1 This is a general schematic diagram of a drone trajectory planning method for rescue scenarios provided in the embodiments of this application.

[0019] Figure 2 This is a schematic diagram of an airspace height layer division provided for an embodiment of this application.

[0020] Figure 3 A flowchart illustrating a drone trajectory planning method for rescue scenarios, provided as an embodiment of this application.

[0021] Figure 4 This is a schematic diagram of an airspace risk index system provided for the implementation of this application.

[0022] Figure 5 This is a schematic diagram of a UAV trajectory planning result provided for an embodiment of this application.

[0023] Figure 6 A schematic diagram of a drone trajectory planning device for rescue scenarios provided in an embodiment of this application.

[0024] Figure 7 A block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0025] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0026] Some algorithms have been applied to the aircraft path planning problem, among which A The algorithm is a classic heuristic search algorithm, widely used because of its low computational cost and good path planning performance.

[0027] Based on A Improvements to the algorithm mainly focus on optimizing the heuristic function and improving search efficiency. Some techniques target A... To address the issues of excessive node traversal and large turning angles in the algorithm, an improved method, A, that adapts to the scene map is proposed. Algorithms. Some techniques are based on raster maps for A. The heuristic function weight coefficients of the algorithm are improved to adapt the algorithm to different scenarios. Some interpretations propose extending the A* algorithm: when multiple minimum cost values ​​exist for a path node, each node is expanded and its cost value is calculated until the node with the minimum path cost value is found, then that path is selected as the optimal path. Some techniques aim to plan the trajectory with minimum risk, considering both time and risk dimensions in the A* algorithm. The cost estimation function of the algorithm is reconstructed to propose a minimum-hazard path planning algorithm for urban ultra-low-altitude logistics scenarios. Some techniques address the complex low-altitude logistics UAV path planning problem, proposing an improved A... Algorithms, incorporating coefficients such as grid hazard factors, ensure UAV flight safety. Some technologies address the need to improve the safety and efficiency of general aviation aircraft in disaster relief flights, researching the trajectory planning problem for general aviation aircraft conducting disaster relief operations in complex low-altitude environments.

[0028] To address the issues of poor drone trajectory planning performance and low flight safety, this application provides the following drone trajectory planning method for rescue scenarios, including forest and grassland aerial rescue sites, as detailed below. Figure 1 As shown.

[0029] Figure 1 This is a general schematic diagram of a drone trajectory planning method for rescue scenarios provided in the embodiments of this application.

[0030] S1. Construct an airspace model for forest and grassland aerial rescue operations. Based on the concept of discrete grids, the airspace is modeled as a three-dimensional block. On the basis of the new grid architecture, targets with different characteristics in the airspace are digitally reconstructed, specifically including: a fixed three-dimensional volume model, an aircraft three-dimensional point model, and a trajectory three-dimensional line model.

[0031] S2, Airspace Altitude Layer Division. Considering factors such as on-site airspace density, safe distances between aircraft, and the requirements for aircraft collaborative missions, the airspace below 1000m is divided into three altitude layers: 0–400m, 400–700m, and 700–1000m. For example... Figure 2 As shown, considering factors such as on-site airspace density, safe distances between aircraft, and the requirements for aircraft collaborative missions, the airspace below 1000m for air rescue operations is divided into three altitude layers: 0–400m, 400–700m, and 700–1000m. The upper limit of each altitude layer is used as the obstacle projection surface, and aircraft are abstracted as point masses for two-dimensional path planning within this projection surface. The 0–400m low-altitude layer is suitable for helicopters, performing air rescue and material delivery missions. The 400–700m mid-altitude layer is suitable for medium-range reconnaissance by unmanned aerial vehicles (UAVs) and relaying command and communication nodes. The 700–1000m high-altitude layer is suitable for large aircraft or unmanned aerial platforms, performing strategic reconnaissance and airborne command and control.

[0032] S3. Construct an airspace risk indicator system for aviation rescue sites. The Analytic Hierarchy Process (AHP) method is used to systematically analyze the hazards of rescue airspace, constructing a three-dimensional main indicator system with meteorological factors, airspace conditions, and terrain factors as the criterion layers. Weights are then assigned to the relevant sub-indicators under the main indicators.

[0033] S4, a quantitative risk assessment model for airspace rescue in forest and grassland areas (FG-DEA). This model is based on the sub-indicator weights obtained from the aforementioned risk indicator system. Four key sub-indicators were selected as input variables. The output indicator was fixed to a single value, and the FG-DEA model was used to assess airspace risk. Under this improved DEA model framework, the assessment results directly map to the airspace's risk value.

[0034] S5. Construct an autonomous path planning model based on airspace risk. A mathematical model for a general graph-based path planning problem is established, and a cost function incorporating operational costs and risks is presented for the RA-A* planning method based on an aviation rescue airspace risk map model.

[0035] S6, Improved Adaptive Heuristic Function. RA-A* planning introduces spatial operational risk factors into the 3D rasterized search. As the current node gets closer to the target node, its spatial distance cost decreases, but the operational risk cost should increase accordingly. The improved heuristic function considers both operational cost and operational risk, and adjusts their magnitude through an adaptive factor.

[0036] In S7, the UAV obtains the optimal flight path by comprehensively considering operational risks, costs, and spatial distance. By continuously searching and iterating the flight path points with the lowest overall cost in S4, the spatiotemporally optimal flight path is obtained.

[0037] Next, in conjunction with the following Figures 3 to 5 This paper specifically describes the UAV trajectory planning method for rescue scenarios proposed in this application.

[0038] Figure 3 A flowchart illustrating a drone trajectory planning method for rescue scenarios, provided as an embodiment of this application.

[0039] like Figure 3 As shown, the UAV trajectory planning method 100 for rescue scenarios provided in this application includes steps S310-S350.

[0040] Step S310: Construct an airspace risk indicator system, analyze and process each indicator in the airspace risk indicator system, and determine key risk indicators.

[0041] Step S320: Conduct a quantitative assessment of airspace risk based on key risk indicators to obtain airspace risk factors.

[0042] Step S330: Determine the airspace model based on airspace risk factors.

[0043] Step S340: Based on the airspace model, construct the UAV trajectory planning model.

[0044] Step S350: Perform trajectory planning based on the UAV trajectory planning model to obtain the trajectory planning result.

[0045] In the embodiments of this application, in order to effectively ensure the safety of airspace for forest and grassland aerial rescue, the embodiments of this application conduct an in-depth analysis of various risk factors within the airspace and construct an airspace risk indicator system for forest fire fighting and rescue airspace. The Analytic Hierarchy Process (AHP) is then used to analyze and process each indicator in the airspace risk indicator system to determine key risk indicators.

[0046] An airspace risk indicator system may include multiple layers of indicators. If the multiple layers consist of two layers, the first layer is the primary indicator and the second layer is the sub-indicator. If the multiple layers consist of three layers, the first layer is the primary indicator and the second layer is the sub-indicator, or the second layer is the primary indicator and the third layer is the sub-indicator.

[0047] The embodiments of this application take an airspace risk indicator system comprising two layers of indicators as an example. The first layer of indicators are the main indicators, and the second layer of indicators are the sub-indicators, such as... Figure 4 As shown, the main indicators include three dimensions of indicators in the criterion layer: meteorological factors, airspace conditions, and topographic factors. Each main indicator corresponds to multiple sub-indicators. For example, the sub-indicators corresponding to meteorological factors include wind direction and speed, temperature, air pressure, and humidity; the sub-indicators corresponding to airspace conditions include airspace visibility, airspace density, flight altitude restrictions, and airspace management; and the sub-indicators corresponding to topographic factors include topographic relief, distribution of combustible resources, and distribution of ultra-high obstacles.

[0048] The sub-indicators are divided into qualitative and quantitative indicators. Qualitative indicators include airspace management, terrain undulation, distribution of combustibles, distribution of super-high obstacles, and distribution of combustible resources, while quantitative indicators include humidity, wind direction and speed, temperature, air pressure, airspace density, and airspace visibility.

[0049] After identifying key risk indicators, airspace risk can be quantitatively assessed based on these indicators to obtain airspace risk factors. And based on airspace risk factors Deterministic spatial model Spatial domain model The construction process is shown in formulas (1) to (3).

[0050] Spatial domain model Modeled as a three-dimensional array block. 3D voxel world (spatial model) It is by Three-dimensional voxels The set that constitutes it. Voxel unit. It can be represented as: (1) In the formula unit voxel coordinate, For spatial coordinates, This is a risk factor for airspace. Therefore, the airspace model... The structure is as follows: (2) In the formula Distribution represents spatial domain of The complete set of coordinates. Represents airspace The number of voxels.

[0051] Spatial domain model Includes a collection of static obstacles For example, fixed, time-invariant spaces or objects in an airspace scene (such as buildings, no-fly zones, etc.) can be represented using a fixed volume model. In this case, the fixed, time-invariant spaces or objects in the airspace scene can be represented as an unordered set of meshes: (3) After obtaining the spatial domain model, we can base our work on the spatial domain model. A UAV trajectory planning model is constructed, and then trajectory planning is performed based on the UAV trajectory planning model to obtain the trajectory planning results.

[0052] The embodiments of this application construct an airspace risk indicator system, then conduct key risk indicator assessments and quantitative airspace risk assessments based on this system to obtain airspace risk factors. Based on these airspace risk factors, an airspace model is determined for trajectory planning. Therefore, the embodiments of this application fully consider airspace risks during trajectory planning, improving the efficiency and safety of UAV collaborative planning.

[0053] In one embodiment of this application, step S310 involves analyzing and processing each indicator in the airspace risk indicator system to determine key risk indicators. This includes: the airspace risk indicator system includes multiple main indicators and multiple sub-indicators corresponding to each main indicator; performing hierarchical analysis on the multiple main indicators to determine a first weight value for each main indicator; performing hierarchical analysis on the multiple sub-indicators corresponding to each main indicator to determine a second weight value for each sub-indicator; and for each sub-indicator, determining a comprehensive weight based on the corresponding second weight value and the first weight value of the main indicator corresponding to the sub-indicator. Based on the comprehensive weight of each sub-indicator At least one sub-indicator is selected from multiple sub-indicators as the key risk indicator.

[0054] Specifically, because the indicators have varying degrees of impact on airspace safety risks for aviation rescue, the Analytic Hierarchy Process (AHP) is used to assign weights to the indicators in the indicator system to ensure the accuracy of the assessment results. The first weight value of the main indicator is calculated based on the indicator scaling value, and the determination steps are as follows: Step 1: Construct the judgment matrix . It can be any one of the three main indicators. It can also be any one of the three main indicators, using express right Importance values, among which, The scale values ​​are obtained from Table 1.

[0055] Table 1. Index Scaling Table Based on AHP

[0056] As shown in Table 1, in the Analytic Hierarchy Process (AHP), the basic scale values ​​are 1, 3, 5, 7, and 9, representing equal importance, slight importance, significant importance, strong importance, and extreme importance, respectively. The median of adjacent judgments refers to the value between two adjacent integer scales. For example, if a decision-maker considers two indicators to be between two integer scales—slightly more important than "slightly important (scale value 3)" but not quite "significantly important (scale value 5)"—then the median value of 4 can be used as the scale value.

[0057] Step 2: Calculate the first weight value for each main indicator. See the eigenvalue vector in equation (4). The value in the table is the first weight value of each main indicator. To determine the matrix The largest eigenvalue.

[0058] (4) Step 3: Consistency check of the judgment matrix. Since the importance of indicators is subjective, a consistency check of the judgment matrix is ​​necessary.

[0059] Calculate the consistency index : (5) Calculate test indicators : (6) in, It is the Random Index.

[0060] As shown in Table 2, a third-order judgment matrix was constructed for meteorological factors, airspace conditions, and topographic factors in the criterion layer (main indicators) to conduct an AHP hierarchical analysis. The resulting eigenvectors are: The weight values ​​corresponding to the three risk factors are as follows: By combining the eigenvectors, the largest eigenvalue, 3.013, can be calculated. The consistency index is then calculated using this largest eigenvalue. The value is 0.006, and the random consistency index is obtained from a third-order table lookup. The value is 0.520, and the calculated test index is... Value This indicates that the judgment matrix satisfies the consistency test, and the first weight values ​​of each main index calculated are consistent.

[0061] Table 2. AHP Hierarchy Process Table for Criterion Level (Main Indicators)

[0062] Similarly, by grouping multiple sub-indicators corresponding to each main indicator into a set, the second weight values ​​of the sub-criteria layer (sub-indicators) are obtained through the above steps, as shown in Tables 3, 4, and 5. The meteorological conditions sub-criteria layer indicators include wind direction and speed, temperature, air pressure, and humidity. Wind direction and speed and temperature share the same feature vector of 1.650 and a weight value of 41.252%, while air pressure and humidity have weight values ​​of 12.750% and 4.745%, respectively. The airspace conditions sub-criteria layer indicators include airspace visibility, airspace density, flight altitude restrictions, and airspace management. Airspace visibility has the highest feature vector of 1.715 and a weight value of 42.876%, followed by airspace density with a feature vector of 1.624 and a weight value of 40.603%. Flight altitude restrictions and airspace management have lower weight values ​​of 12.212% and 4.309%, respectively. The topographic factor sub-criteria layer includes sub-indicators such as topographic relief, distribution of super-high obstacles, and distribution of combustible resources. The eigenvector of super-high obstacle distribution is the highest at 1.900, with a weight of 63.335%. The weights of topographic relief and combustible resource distribution are 26.050% and 10.616%, respectively. The maximum eigenvalues ​​of all sub-criteria layers are 4.219, 4.182, and 3.039, respectively. Furthermore, the validation indicators... The values ​​are all less than 0.1, the judgment matrix satisfies the consistency test, and the calculated second weight values ​​are consistent.

[0063] Table 3. AHP Hierarchical Analysis Table of Sub-Criterion Layers (Meteorological Sub-Indicators)

[0064] For each sub-indicator, the overall weight is determined based on the corresponding second weight value and the first weight value of the corresponding main indicator. Based on the comprehensive weight of each sub-indicator, at least one sub-indicator is selected as the key risk indicator from multiple sub-indicators. For example, for the sub-indicator of ultra-high obstacle distribution, the first weight is 7.817% of the weight corresponding to the main indicator of terrain factors, and the second weight is 63.335%. Therefore, the comprehensive weight... =4.95% is the first weight value of 7.817% multiplied by the second weight value of 63.335%.

[0065] Based on the comprehensive weight of each sub-indicator Select comprehensive weight The most significant sub-indicators are selected as key risk indicators, for example, four. These four key risk indicators include: wind direction and speed, temperature, airspace visibility, and airspace density. These key risk indicators are used as inputs for the subsequent operation of the airspace risk assessment model to ensure the accuracy and reliability of the assessment results.

[0066] According to one embodiment of this application, after screening out key risk indicators, step S320 above performs a quantitative assessment of airspace risk based on the key risk indicators to obtain an airspace risk factor. This includes the following content.

[0067] The spatial domain is divided into meshes based on preset rules, including the meshing method and mesh size. The spatial domain mesh comprises multiple meshes, which can form a spatial domain model. The process of mesh generation can be found in formulas (1), (2) and (3) above.

[0068] Determine the key risk indicators for each grid. For example, for a certain grid, key risk indicators include wind direction and speed, temperature, airspace visibility, and airspace density. Collect the specific values ​​of each key risk indicator in that grid as indicator values. Indicator values ​​include, for example, specific values ​​of wind direction and speed, temperature, airspace visibility, and airspace density.

[0069] Use the indicator value as an input variable Input the variables into the risk quantification assessment model and output the variables from the risk quantification assessment model. Set to preset values ​​and apply them to the risk variables in the risk quantification assessment model. Minimize the variables to obtain the risk variables. The corresponding operational risk quantification value serves as the airspace risk factor. The operational risk quantification value can be obtained based on the Forest and Grassland Aviation Rescue Airspace Risk Quantification Assessment Model (FG-DEA), and the specific process is as follows.

[0070] Input indicators (input variables) Output indicators (output variables) represent the resources consumed in the production or service process, used to measure the amount of resources invested by a decision-making unit (such as an enterprise, organization, or project) in its activities; while output indicators (output variables) represent the resources consumed in the production or service process. Input and output variables represent the results or benefits obtained during the production or service process, measuring the value or outcome created by the decision-making unit after using the invested resources. Based on this approach, this application uses the DEA model to assess the risk value of high-risk factors and redefines input and output variables. In emergency rescue airspace safety risk assessment, input indicators should represent risk factors within the airspace, while output indicators should represent the occurrence of undesirable events (such as accidents) caused by these risk factors.

[0071] Four key risk indicators will be selected as input variables, and output indicators (output variables) will be selected as output variables. The risk of the airspace is fixed at a specific preset value, and a super-efficient model is used to assess the risk. Within this FG-DEA model framework, the assessment results directly reflect the risk value of the airspace. The model can be expressed as: (7) in, and Let i represent the input (any one of the four key risk indicators) and r represent the output of the j-th sample (a unit voxel or grid in the spatial domain is a sample), respectively. and This indicates the input and output corresponding to the sample o currently being evaluated. These are the weighting coefficients for each sample. In the field of risk assessment, the corresponding inputs are potential risk factors, while the output indicators represent the undesirable events caused by these risk factors, which are strictly controlled and limited. To reflect this reality, we fix the output indicators; specifically, we will... Abstracted to fixed constants, the ultra-efficient DEA model can therefore be further expressed as: (8) in, Represents the j-th sample (one unit voxel or grid in the spatial domain is one sample) i Item input (any one of the four key risk indicators), This indicates the sample currently being evaluated. o The corresponding investment. The operational risk quantification value output by the super-efficient DEA model is used as the airspace risk factor in formula (1). .

[0072] In formulas (7) and (8) and Essentially, they have the same meaning, both representing the weight coefficients of the sample. The difference is that formula (8) is for risk assessment, an improved model after fixing the output. Therefore, for ease of distinction, its weight coefficients are represented by... Indicates weighting coefficients. and These are not fixed parameters preset by humans, but rather unknown parameters of the model, obtained through linear programming. The model automatically finds the optimal reference combination for each decision unit based on data characteristics. The value represents the contribution of the j-th decision-making unit (i.e., the j-th sample (a unit voxel or grid in the spatial domain is considered a sample)) in constructing this virtual optimal frontier (reference surface). In the hyper-efficient DEA model, the virtual optimal frontier (also called the efficient frontier or reference surface) refers to an efficiency frontier formed by the most efficient decision-making units (DMUs). This frontier represents the optimal relationship between inputs and outputs that can be achieved under current technological conditions.

[0073] According to embodiments of this application, based on airspace risk factors... Constructing the spatial domain model Subsequently, in step S340 above, based on the spatial domain model... Construct a drone trajectory planning model, including the following:

[0074] Specifically, the airspace model Includes a collection of static obstacles Based on spatial domain model Determine the initial node for the drone's flight in airspace. and target node Based on the initial node and target node Constructing drone trajectories And based on the drone trajectory Determine the trajectory cost to obtain the UAV trajectory planning model.

[0075] Because the airspace information has been digitally reconstructed at multiple scales, and the reconstructed grid is unique, any UAV appearing in any location can establish a logical mapping with the airspace grid, completing the UAV data reconstruction. A UAV in the airspace can be represented as: (9) in, This represents the spatiotemporal grid airspace occupied by a drone that appears at any location at any given time.

[0076] Given the initial and target positions of the UAV, this application employs an efficient search algorithm to plan a safe and effective path. According to the planning method described above, the UAV's trajectory can be discretized into a series of heterogeneous grids. At this point, the UAV's trajectory can be represented by a path derived from... An ordered set of individual prime points To indicate: (10) The starting point (initial node) of the path is... The endpoint (target node) is .

[0077] 3D spatial model The block (voxel, mesh) occupancy state in the array is shown in Equation (11), and can be represented as a three-dimensional array. Each element Voxel representation Is it occupied? 1 indicates occupied, 0 indicates unoccupied. Blocks overlapping with static obstacles such as buildings should be identified as occupied.

[0078] (11) Next, based on the drone trajectory By determining the trajectory cost, a UAV trajectory planning model is obtained. This UAV trajectory planning model includes a trajectory cost model. First constraint and second constraint, where the first constraint represents the drone trajectory. With static obstacle set No collisions or conflicts; the second constraint represents the target node. Location To set a target endpoint Location .

[0079] Specifically, this application proposes a mathematical model for the graph-based path planning problem and provides a mathematical model of the local functions used in the RA-A* planning method based on a spatiotemporal unified map model.

[0080] Based on formula (2), let's set it as a spatial model. The three-dimensional state space of the mesh, using and Let represent the initial point (initial node) and the target point (target node) in the flight plan, respectively. Given a static set of obstacles, the flight path planning problem for a UAV can be viewed as finding a path from its current position. The first voxel Begin, reach the predetermined target voxel This makes the entire flight path cost The minimum, collision-free continuous three-dimensional trajectory . The spatial coordinates are fixed, for However, the risks are uncertain. Let's assume this flight path... As shown in equation (10) If we represent the problem as an ordered set of individual points, then the trajectory planning problem can be expressed as follows: (12) in, For the trajectory cost model, in the first constraint express and In the spatial grid spatial domain The intersection of the lines, where the constraint represents the track. With occupied voxels (Including obstacles and assigned paths) No collisions. The second constraint represents the track. The last three-dimensional voxel The spatial location is the spatial location of the preset target endpoint. .

[0081] According to embodiments of this application, the trajectory cost model Including operating costs and operational risks Determine operating costs and operational risks The process is as follows.

[0082] Regarding operating costs Based on drone trajectory The distance between adjacent trajectory nodes in three-dimensional space Determine the spatial distance cost, and then determine the operating cost based on the spatial distance cost; for example, use the spatial distance cost as the operating cost. The operating cost is shown in equation (13).

[0083] (13) In the formula express Euclidean distance in three-dimensional space.

[0084] Regarding operational risks Determine the drone's trajectory Any trajectory node Minimum distance between the static obstacle set Based on drone trajectory Determining risk values ​​by spatial risk factors of adjacent trajectory nodes And based on minimum distance Risk Value Operational risk attention adjustment coefficient and operational risk sensitivity adjustment coefficient The operational risks are determined. The operational risks are shown in equation (14).

[0085] (14) in, express and The minimum Euclidean distance of all voxels in three-dimensional space. This indicates the occupied voxels at this minimum distance. A positive real number, representing the risk awareness adjustment coefficient. When The larger the value, the more attention will be paid to operational risks during path planning; conversely, the smaller the value, the more attention will be paid to operational costs. is a positive real number, representing the risk sensitivity adjustment coefficient. When The larger the value, the more likely the farther away the data is to be occupied. It will also significantly increase operational risks; conversely, it will only be considered as Only when they are close enough will the operational risks increase significantly.

[0086] According to an embodiment of this application, in step S350 above, trajectory planning is performed based on the UAV trajectory planning model to obtain trajectory planning results, including the following:

[0087] A cost estimation model is constructed based on heuristic functions. Among them, the cost estimation model Used to confirm the current node in the trajectory. and the preset target endpoint The estimated cost between the current node and the preset target endpoint The estimated costs between them include estimated operating costs. and estimating operational risks The weighted, heuristic function includes estimating operating costs. Corresponding weighting factors And estimate operational risks Corresponding weighting factors .

[0088] Regarding the improved adaptive heuristic function, RA-A* programming introduces operational risk into a three-dimensional rasterized search. With the current node... The closer to the target node The spatial distance cost will decrease over time. An improved heuristic function considers both operational cost and risk, and adjusts its size using an adaptive factor. Current node Distance to target node The estimated cost is as shown in equation (15).

[0089] (15) in, As a weighting factor, it plays a crucial role in comprehensively considering spatial distance (operating costs) and operational risk costs. The weighting factor is adjusted according to the specific application scenario to balance the importance of space and time. This application adopts... All values ​​are set to 1 as an example.

[0090] For example, determining the estimated operating costs The specific process is as follows.

[0091] Determine the current node Location and the preset target endpoint Location Spatial distance between If an assigned track exists Based on the assigned track Actual operating cost and assigned tracks Distance between the starting point and the ending point Determine the first adaptive adjustment factor If no assigned track exists Determine the first adaptive adjustment factor The preset value (e.g., 1) is used; based on the first adaptive adjustment factor. and spatial distance Determine estimated operating costs See formulas (16) and (17) for details.

[0092] In formula (15) Indicates that the current node To the target location The estimated operating cost of a flight path is generally proportional to its spatial distance, specifically: (16) In the formula, As an adaptive adjustment factor, it can be adjusted based on the actual operating costs of the allocated tracks in the environment. We obtain the set of assigned tracks. ,but It can be obtained from the following expression: (17) In the formula, express The Euclidean distance between the starting point and the ending point in space. If no path is assigned in the initial state, take... .

[0093] For example, spatial domain model Includes a collection of static obstacles Determine estimated operational risks The specific process is as follows.

[0094] Determine the current node and the preset target endpoint straight three-dimensional path Determine the three-dimensional path of the straight line. With static obstacle set First conflict data between If an assigned track exists Based on the assigned track Actual operational risks and assigned tracks Second conflict data Determine the second adaptive adjustment factor If no assigned track exists Determine the second adaptive adjustment factor The preset value (e.g., 1) is used; based on the second adaptive adjustment factor. Conflict data with the first Determine estimated operating costs See formulas (18) and (19) for details.

[0095] In formula (15) Indicates that the current node To the target location The estimated operational risks are set. For the current node The three-dimensional straight path leading to the target endpoint is expressed in the following form to estimate operational risk: (18) In the formula, This indicates how to count the number of elements in a set. This is an adaptive adjustment factor, which can be obtained based on the actual operational risks of the assigned tracks in the environment. Let the set of assigned tracks be... ,but It can be obtained from the following expression: (18) If no path is assigned in the initial state, take .

[0096] According to the embodiments of this application, the UAV trajectory planning model of equation (12) and the cost estimation model of equation (15) are obtained. Subsequently, based on the UAV trajectory planning model and cost estimation model... The search and iterative calculations are performed to obtain the trajectory planning results.

[0097] Specifically, based on UAV trajectory planning model and cost estimation model The search and iterative calculations are performed to obtain the trajectory planning results, including the following process: The trajectory cost model in UAV trajectory planning Determine the current node With adjacent nodes The actual cost between ; Based on cost estimation model Determine adjacent nodes With the target node Estimated costs between ; Based on actual cost Adjacent nodes With the target node Estimated costs between Calculate the target cost ; If adjacent nodes If the node is not in the open list, then the adjacent node will be... Add to the open list and determine adjacent nodes. The parent node is used as the current node; If adjacent nodes If the actual cost to reach the current node is lower than the actual cost to reach the adjacent node in the open list, update the actual cost, target cost, and parent node of the current node. The node with the lowest objective cost is selected from the open list as the current node for iterative calculation. The trajectory planning result is obtained after the iteration conditions are met. The iteration conditions include, for example, finding the objective node or the open list being empty.

[0098] Let's take the following specific example to explain.

[0099] S51. Initialization: Set the starting point Add to the open list and set the current actual cost. The value is 0. Indicates the voxel point To the starting point The actual flight path. Value is a voxel point Location to the target node The estimated cost, calculation value: .

[0100] S52. Check if the open list is empty: If the open list is empty, the algorithm terminates because no path was found.

[0101] Otherwise, select from the open list. The node with the smallest value As the current node, move it to the closed list.

[0102] S53. Determine the current node Target location reached : If so, backtrack the path from the target node to the starting point, and the algorithm ends.

[0103] If not, proceed to the next step.

[0104] S54. For the current node Each adjacent node Perform the following operations: 1. If adjacent nodes This node is already in the closed list; skip it.

[0105] 2. Calculate or update adjacent nodes of Value (the actual cost from the starting point to this node). 3. Calculate adjacent nodes of Value (estimated cost from this node to the preset target node).

[0106] 4. Calculate adjacent nodes of value

[0107] 5. If adjacent nodes If it's not in the open list, add it to the open list and set its parent node. This refers to the current node. 6. If adjacent nodes A path already on the open list but reached through the current node is better (i.e., the new path). If the value is lower, then update it. value, Value and parent node information.

[0108] S55. Return to S52 and continue iterating until the target node is found or the open list is empty.

[0109] By continuously searching and iterating to find the path point that minimizes the sum of expected operating costs and expected operating risks, the optimal path with the best operating costs and operating risks is obtained.

[0110] Figure 5 The UAV trajectory planning results of this application embodiment are shown in In the planar space display, the black grid area 510 represents the voxels occupied by static obstacles, and the shades of the red grid 520 represent the risk level. The green grid 530 and yellow grid 540 represent the voxels occupied by the initial node and the target endpoint, respectively, and the blue grid 550 represents the flight path planned by the UAV.

[0111] This application, from the perspective of airspace flight safety in aviation rescue, constructs an airspace risk assessment index system for forest and grassland aviation rescue sites. The Analytic Hierarchy Process (AHP) is applied to screen the airspace assessment system for forest and grassland aviation rescue, identifying key assessment indicators such as wind direction and speed, temperature, airspace visibility, and airspace density. Using these key assessment indicators as input variables and fixing the output indicator values, the FG-DEA risk assessment model is employed to quantify the risk of multiple points within the airspace, obtaining the relative risk values ​​of each point in a specific firefighting scenario. Furthermore, with the goal of minimizing the planned trajectory risk, the cost estimation function is reconstructed from two dimensions: path distance and airspace risk, proposing a minimum-risk path planning algorithm for forest and grassland aviation rescue scenarios.

[0112] The AHP method was applied to screen the airspace assessment system for forest and grassland aerial rescue, identifying key assessment indicators such as wind direction and speed, temperature, airspace visibility, and airspace density. These key indicators were used as input variables, and output indicator values ​​were fixed. A super-efficient DEA model was employed to quantify the risk of multiple points within the airspace, obtaining the relative risk values ​​of each point in a specific firefighting scenario. For example... Figure 5 As shown, the minimum risk path planning algorithm proposed in this application has a lower average risk value for the planned path compared with the traditional path planning algorithm under the above spatial conditions.

[0113] Figure 6 A schematic diagram of a drone trajectory planning device for rescue scenarios provided in an embodiment of this application.

[0114] This application provides a drone trajectory planning device 600 for rescue scenarios, including: The analysis and processing module 610 is used to construct an airspace risk indicator system and analyze and process each indicator in the airspace risk indicator system to determine key risk indicators.

[0115] The assessment module 620 is used to conduct a quantitative assessment of airspace risk based on the key risk indicators to obtain airspace risk factors.

[0116] The determination module 630 is used to determine the airspace model based on the airspace risk factor.

[0117] Module 640 is used to construct a UAV trajectory planning model based on the airspace model.

[0118] The planning module 650 is used to perform trajectory planning based on the UAV trajectory planning model and obtain trajectory planning results.

[0119] In other implementations, the airspace risk indicator system includes multiple main indicators and multiple sub-indicators corresponding to each main indicator; the analysis and processing module 610 is further configured to: perform hierarchical analysis and processing on the multiple main indicators to determine a first weight value for each main indicator; perform hierarchical analysis and processing on the multiple sub-indicators corresponding to each main indicator to determine a second weight value for each sub-indicator; for each sub-indicator, determine a comprehensive weight based on the corresponding second weight value and the first weight value of the main indicator corresponding to the sub-indicator; and based on the comprehensive weight of each sub-indicator, determine at least one sub-indicator from the multiple sub-indicators as a key risk indicator.

[0120] In other embodiments, the evaluation module 620 is further configured to: divide the airspace into grids based on preset rules to obtain an airspace grid, wherein the airspace grid includes multiple grids; determine the index value of the key risk indicator corresponding to each grid; input the index value as an input variable into the risk quantification assessment model, set the output variable in the risk quantification assessment model to a preset value, and minimize the risk variable in the risk quantification assessment model to obtain the operational risk quantification value corresponding to the risk variable as an airspace risk factor.

[0121] In other embodiments, the airspace model includes a set of static obstacles; the construction module 640 is further configured to: determine the initial node and target node of the UAV flying in the airspace based on the airspace model; construct the UAV trajectory based on the initial node and target node; determine the trajectory cost based on the UAV trajectory to obtain the UAV trajectory planning model, wherein the UAV trajectory planning model includes a trajectory cost model, a first constraint and a second constraint, the first constraint indicating that the UAV trajectory has no collision conflict with the set of static obstacles, and the second constraint indicating that the position of the target node is the position of the preset target endpoint.

[0122] In other implementations, the trajectory cost model includes operating costs and operating risks. Determining operating costs and operating risks includes: determining spatial distance costs based on the distance between the positions of adjacent trajectory nodes in the UAV trajectory in three-dimensional space; determining operating costs based on spatial distance costs; determining the minimum distance between any trajectory node in the UAV trajectory and a set of static obstacles; determining risk values ​​based on airspace risk factors of adjacent trajectory nodes in the UAV trajectory; and determining operating risks based on the minimum distance, risk values, operating risk attention adjustment coefficient, and operating risk sensitivity adjustment coefficient.

[0123] In other embodiments, the planning module 650 is further configured to: construct a cost estimation model based on a heuristic function, wherein the cost estimation model is used to determine the estimated cost between the current node and the preset target endpoint in the trajectory, the estimated cost between the current node and the preset target endpoint includes a weighted average of the estimated operating cost and the estimated operating risk, and the heuristic function includes a weight factor corresponding to the estimated operating cost and a weight factor corresponding to the estimated operating risk; and perform search iterative calculations based on the UAV trajectory planning model and the cost estimation model to obtain the trajectory planning result.

[0124] In other implementations, determining the estimated operating cost includes: determining the spatial distance between the current node's position and the position of a preset target endpoint; if an assigned track exists, determining a first adaptive adjustment factor based on the actual operating cost of the assigned track and the distance between the start and end points of the assigned track; if no assigned track exists, determining the first adaptive adjustment factor as a preset value; and determining the estimated operating cost based on the first adaptive adjustment factor and the spatial distance.

[0125] In other implementations, the airspace model includes a set of static obstacles; determining the estimated operational risk includes: determining a straight three-dimensional path from the current node to a preset target endpoint; determining first conflict data between the straight three-dimensional path and the set of static obstacles; if an assigned track exists, determining a second adaptive adjustment factor based on the actual operational risk of the assigned track and the second conflict data of the assigned track; if no assigned track exists, determining the second adaptive adjustment factor as a preset value; and determining the estimated operational cost based on the second adaptive adjustment factor and the first conflict data.

[0126] In other implementations, a search iterative calculation is performed based on the UAV trajectory planning model and the cost estimation model to obtain the trajectory planning result, including: determining the actual cost between the current node and its neighboring nodes based on the trajectory cost model in the UAV trajectory planning model; determining the estimated cost between the neighboring node and the target node based on the cost estimation model; calculating the target cost based on the actual cost and the estimated cost between the neighboring node and the target node; if the neighboring node is not in the open list, adding the neighboring node to the open list and determining the parent node of the neighboring node as the current node; if the neighboring node is in the open list, but the actual cost to reach the current node is lower than the actual cost to reach the neighboring node, updating the actual cost, the target cost, and the parent node corresponding to the current node; selecting the node with the lowest target cost from the open list as the current node for iterative calculation, and obtaining the trajectory planning result after satisfying the iteration conditions.

[0127] It is understandable that a detailed description of the UAV trajectory planning device 600 for rescue scenarios can be found in the description of the UAV trajectory planning method for rescue scenarios above, and will not be repeated here.

[0128] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method in any of the above embodiments.

[0129] One embodiment of this application provides a computer program product including instructions that, when executed by a processor of a computer device, enable the computer device to perform the steps of the method described in any of the above embodiments.

[0130] Figure 7 A block diagram of an electronic device provided in an embodiment of this application.

[0131] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method in any of the above embodiments.

[0132] like Figure 7 As shown, for ease of understanding, an embodiment of this application illustrates a specific electronic device 700.

[0133] Electronic device 700 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 700 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present application described and / or claimed herein.

[0134] like Figure 7 As shown, device 700 includes a computing unit 701, which can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) 702 or a computer program loaded from storage unit 708 into random access memory (RAM) 703. RAM 703 may also store various programs and data required for the operation of electronic device 700. The computing unit 701, ROM 702, and RAM 703 are interconnected via bus 704. Input / output (I / O) interface 705 is also connected to bus 704.

[0135] Multiple components in electronic device 700 are connected to I / O interface 705. These components include: input unit 706, such as a keyboard or mouse; output unit 707, such as various types of displays or speakers; storage unit 708, such as a disk or optical disk; and communication unit 709, such as a network interface card (NIC), modem, or wireless transceiver. Communication unit 709 allows electronic device 700 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0136] The computing unit 701 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 performs the various methods described above. For example, in some embodiments, any one or more of the methods described above can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 700 via ROM 702 and / or communication unit 709. When the computer program is loaded into RAM 703 and executed by the computing unit 701, one or more steps of any one or more of the methods described above can be performed. Alternatively, in other embodiments, the computing unit 701 can be configured to perform any one or more of the methods described above by any other suitable means (e.g., by means of firmware).

[0137] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this application, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0138] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0139] In the description of this application, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," 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 this application. In this application, 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.

[0140] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc., indicating the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0141] Furthermore, the terms "first," "second," etc., used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying relative importance, or implicitly specifying the number of technical features indicated in this embodiment. Therefore, features defined with terms such as "first" and "second" in the embodiments of this application can explicitly or implicitly indicate that the embodiment includes at least one of those features. In the description of this application, the word "multiple" means at least two or more, such as two, three, four, etc., unless otherwise explicitly and specifically defined in the embodiments.

[0142] In this application, unless otherwise explicitly specified or limited in the embodiments, the terms "installation," "connection," "joining," and "fixing" appearing in the embodiments should be interpreted broadly. For example, a connection can be a fixed connection, a detachable connection, or an integral part; it can also be a mechanical connection, an electrical connection, etc. Of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication between two components, or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific implementation.

[0143] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

Claims

1. A method for planning the flight path of a drone in a rescue scenario, characterized in that, The method includes: An airspace risk indicator system is constructed, and each indicator in the airspace risk indicator system is analyzed and processed to determine key risk indicators. Based on the aforementioned key risk indicators, a quantitative assessment of airspace risk is conducted to obtain an airspace risk factor. This process includes: dividing the airspace into grids based on preset rules to obtain an airspace grid, wherein the airspace grid comprises multiple grids; determining the indicator value of the key risk indicator corresponding to each grid; inputting the indicator value as an input variable into a risk quantification assessment model, setting the output variable in the risk quantification assessment model to a preset value, and minimizing the risk variable in the risk quantification assessment model to obtain the operational risk quantification value corresponding to the risk variable as the airspace risk factor. Based on the aforementioned airspace risk factors, an airspace model is determined; Based on the aforementioned airspace model, a UAV trajectory planning model is constructed; Based on the aforementioned UAV trajectory planning model, trajectory planning is performed to obtain the trajectory planning results.

2. The method according to claim 1, characterized in that, The airspace risk indicator system includes multiple main indicators and multiple sub-indicators corresponding to each main indicator; the analysis and processing of each indicator in the airspace risk indicator system to determine key risk indicators includes: Hierarchical analysis is performed on the multiple main indicators to determine the first weight value of each main indicator; A hierarchical analysis process is performed on the multiple sub-indicators corresponding to each main indicator to determine the second weight value of each sub-indicator. For each sub-indicator, the comprehensive weight of the sub-indicator is determined based on the corresponding second weight value and the first weight value of the main indicator corresponding to the sub-indicator. Based on the comprehensive weight of each of the sub-indicators, at least one sub-indicator is determined from the plurality of sub-indicators as the key risk indicator.

3. The method according to claim 1, characterized in that, The airspace model includes a set of static obstacles; the construction of a UAV trajectory planning model based on the airspace model includes: Based on the aforementioned airspace model, the initial node and target node for the UAV to fly in the airspace are determined; Based on the initial node and the target node, construct the UAV trajectory; Based on the UAV trajectory, the trajectory cost is determined, and the UAV trajectory planning model is obtained. The UAV trajectory planning model includes a trajectory cost model, a first constraint, and a second constraint. The first constraint indicates that the UAV trajectory has no collision with the set of static obstacles, and the second constraint indicates that the position of the target node is the position of the preset target endpoint.

4. The method according to claim 3, characterized in that, The trajectory cost model includes operating costs and operating risks. Determining the operating costs and operating risks includes: The spatial distance cost is determined based on the distance between the positions of adjacent trajectory nodes in the UAV trajectory in three-dimensional space, and the operating cost is determined based on the spatial distance cost. The minimum distance between any trajectory node in the UAV trajectory and the set of static obstacles is determined. The risk value is determined based on the airspace risk factor of adjacent trajectory nodes in the UAV trajectory. The operational risk is determined based on the minimum distance, the risk value, the operational risk attention adjustment coefficient, and the operational risk sensitivity adjustment coefficient.

5. The method according to claim 3, characterized in that, The trajectory planning based on the UAV trajectory planning model, to obtain the trajectory planning result, includes: A cost estimation model is constructed based on a heuristic function, wherein the cost estimation model is used to determine the estimated cost between the current node and the preset target endpoint in the trajectory. The estimated cost between the current node and the preset target endpoint includes a weighted average of the estimated operating cost and the estimated operating risk. The heuristic function includes a weight factor corresponding to the estimated operating cost and a weight factor corresponding to the estimated operating risk. The trajectory planning result is obtained by performing search and iterative calculations based on the UAV trajectory planning model and the cost estimation model.

6. The method according to claim 5, characterized in that, Determining the estimated operating cost includes: Determine the spatial distance between the current node's position and the preset target endpoint's position; If an assigned track exists, a first adaptive adjustment factor is determined based on the actual operating cost of the assigned track and the distance between the start and end points of the assigned track; if no assigned track exists, the first adaptive adjustment factor is determined to be a preset value. The estimated operating cost is determined based on the first adaptive adjustment factor and the spatial distance.

7. The method according to claim 5, characterized in that, The spatial model includes a set of static obstacles; Determining the estimated operational risk includes: Determine the straight-line three-dimensional path between the current node and the preset target endpoint; Determine the first conflict data between the straight three-dimensional path and the set of static obstacles; If an assigned track exists, a second adaptive adjustment factor is determined based on the actual operational risk of the assigned track and the second conflict data of the assigned track; if no assigned track exists, the second adaptive adjustment factor is determined to be a preset value. The estimated operating cost is determined based on the second adaptive adjustment factor and the first conflict data.

8. The method according to claim 5, characterized in that, The search and iterative calculation based on the UAV trajectory planning model and the cost estimation model to obtain the trajectory planning result includes: Based on the trajectory cost model in the UAV trajectory planning model, the actual cost between the current node and its neighboring nodes is determined. Based on the cost estimation model, the estimated cost between the adjacent node and the target node is determined; Calculate the target cost based on the actual cost and the estimated cost between the adjacent nodes and the target node; If the adjacent node is not in the open list, add the adjacent node to the open list and determine the parent node of the adjacent node as the current node; If the adjacent node is in the open list, but the actual cost to reach the current node is lower than the actual cost to reach the adjacent node, update the actual cost, the target cost, and the parent node corresponding to the current node; The node with the lowest objective cost is selected from the open list as the current node for iterative calculation. After the iteration conditions are met, the trajectory planning result is obtained.

9. A drone trajectory planning device for rescue scenarios, characterized in that, The device includes: The analysis and processing module is used to construct an airspace risk indicator system and analyze and process each indicator in the airspace risk indicator system to determine key risk indicators. An assessment module is used to perform airspace risk quantification assessment based on the key risk indicators to obtain airspace risk factors. The airspace risk quantification assessment based on the key risk indicators to obtain airspace risk factors includes: dividing the airspace into grids based on preset rules to obtain airspace grids, wherein the airspace grids include multiple grids; determining the indicator value of the key risk indicator corresponding to each grid; inputting the indicator value as an input variable into a risk quantification assessment model, setting the output variable in the risk quantification assessment model to a preset value, and minimizing the risk variable in the risk quantification assessment model to obtain the operational risk quantification value corresponding to the risk variable as the airspace risk factor. The determination module is used to determine the airspace model based on the airspace risk factors; A construction module is used to construct a UAV trajectory planning model based on the airspace model. The planning module is used to perform trajectory planning based on the UAV trajectory planning model and obtain the trajectory planning results.

Images