Low-altitude air route operation feasibility determination method and system

Abstract

This application relates to a method and system for determining the feasibility of low-altitude flight path operations. The method includes obtaining the actual flight position of an aircraft at the current moment; projecting the actual flight position onto a preset flight path of the aircraft to obtain a current reference point; constructing a residual flight path based on the current reference point; generating redundant nodes in the neighborhood space of the residual flight path; constructing a residual flight path graph based on the redundant nodes and the master node; calculating the dynamic risk weight of each directed edge; obtaining an updated residual flight path graph based on the dynamic risk weight and a preset risk threshold; calculating a topological stability value based on the dynamic risk weight and the topological structure of the updated residual flight path graph; and determining the feasibility of the residual flight path when the topological stability value is greater than or equal to a preset stability threshold. This method can update the residual flight path graph based on dynamic risk weights, thereby determining the feasibility of the residual flight path and achieving real-time determination of the feasibility of low-altitude flight path operations.

Description

Technical Field

[0001] This application relates to the field of low-altitude airway planning and operation control technology, and in particular to methods and systems for determining the feasibility of low-altitude airway operation. Background Technology

[0002] With the rapid development of the low-altitude economy, low-altitude flight applications such as drone delivery, urban air traffic, and emergency rescue are increasing. Low-altitude airspace is characterized by significant complexity and dynamism: local weather conditions (such as gusts and rainfall) can change rapidly, and temporary airspace restrictions (such as sudden no-fly zones) can arise dynamically. Therefore, feasibility assessments of low-altitude routes are particularly important.

[0003] Existing technologies often employ machine learning algorithms to predict route feasibility before aircraft takeoff. However, these predictions are based solely on static pre-takeoff environmental data and historical samples, failing to adapt to dynamic risks during flight such as sudden weather changes, temporary airspace control, aircraft conflicts, and communication signal fluctuations. They also lack the ability to continuously track and predict the real-time evolution of route feasibility trends. Therefore, these technologies suffer from the inability to assess the feasibility of low-altitude routes in real time.

[0004] There is currently no effective solution to the problem that related technologies cannot make real-time feasibility assessments of low-altitude air route operations. Summary of the Invention

[0005] This embodiment provides a method and system for determining the feasibility of low-altitude air route operations, in order to solve the problem in related technologies that it is impossible to determine the feasibility of low-altitude air route operations in real time.

[0006] Firstly, this embodiment provides a method for determining the feasibility of low-altitude air route operations, including:

[0007] Obtain the actual flight position of the aircraft at the current moment, project the actual flight position onto the aircraft's preset flight path to obtain the current reference point corresponding to the actual flight position; construct the residual flight path based on the current reference point;

[0008] Redundant nodes are generated in the neighborhood space of the residual route; a residual operation graph is constructed based on the redundant nodes and the main nodes on the residual route; the residual operation graph includes several directed edges;

[0009] Calculate the dynamic risk weight of each directed edge in the residual running graph; update the edge set of the residual running graph according to the dynamic risk weight and the preset risk threshold to obtain the updated residual running graph;

[0010] Based on the dynamic risk weight of each directed edge in the updated residual running graph and the topology of the updated residual running graph, calculate the topological stability value of the updated residual running graph.

[0011] When the topology stability value is greater than or equal to a preset stability threshold, the residual route is determined to be feasible.

[0012] In some embodiments, projecting the actual flight position onto the aircraft's preset flight path to obtain a current reference point corresponding to the actual flight position includes:

[0013] The projection parameters of the actual flight position projected onto the preset flight path of the aircraft are obtained by using the minimum distance projection algorithm;

[0014] Based on the projection parameters, the current reference point corresponding to the actual flight position is obtained.

[0015] In some embodiments, generating redundant nodes in the neighborhood space of the residual route includes:

[0016] Determine the neighborhood space of the remaining route; within the neighborhood space, construct a cross-sectional plane;

[0017] Regular network sampling is performed within the cross-sectional plane to generate regular redundant nodes; an enhanced redundant node is generated using a conditional generative adversarial network with the current environmental conditions as input.

[0018] In some embodiments, constructing the residual flight graph based on the redundant nodes and the master nodes on the residual routes includes:

[0019] The redundant nodes and the main nodes on the residual route are merged and deduplicated to obtain the node set of the residual route.

[0020] Based on the set of nodes of the remaining routes, a set of candidate edges for the remaining routes is obtained; based on the set of candidate edges for the remaining routes, the remaining route graph is constructed.

[0021] In some embodiments, calculating the dynamic risk weight of each directed edge in the residual running graph includes:

[0022] The continuous risk field of the residual route is estimated using a trained deep neural network;

[0023] Based on the continuous risk field of the residual route, calculate the dynamic risk weight of each directed edge in the residual flight graph.

[0024] In some embodiments, updating the edge set of the residual running graph based on the dynamic risk weight and a preset risk threshold to obtain the updated residual running graph includes:

[0025] When the dynamic risk weight is greater than the risk threshold, the directed edge corresponding to the dynamic risk weight is deleted from the edge set of the residual running graph to obtain the edge set of the updated residual running graph.

[0026] The updated residual running graph is obtained based on the edge set of the updated residual running graph.

[0027] In some embodiments, the low-altitude airway operation feasibility determination method further includes:

[0028] When the topological stability value is less than the stability threshold, the key failure region subgraph of the residual operation graph is located according to the dynamic risk weight.

[0029] Using graph neural networks and reinforcement learning algorithms, local alternative paths are generated within the critical failure region subgraph; these local alternative paths are then concatenated with the unfailed paths in the residual route to obtain the updated residual route.

[0030] In some embodiments, locating the critical failure region subgraph of the residual running graph based on the dynamic risk weight when the topological stability value is less than the stability threshold includes:

[0031] When the topological stability value is less than the stability threshold, the set of critical failure edges of the residual running graph is located according to the dynamic risk weight.

[0032] Based on the set of critical failure edges in the residual running graph, locate the critical failure region subgraph of the residual running graph.

[0033] Secondly, this embodiment provides a low-altitude route operation feasibility determination system, including a residual route construction module, a residual flight map construction module, a residual flight map update module, a topology stability value calculation module, and a feasibility determination module, wherein:

[0034] The residual route construction module is used to obtain the actual flight position of the aircraft at the current moment, project the actual flight position onto the preset route of the aircraft to obtain the current reference point corresponding to the actual flight position, and construct the residual route based on the current reference point;

[0035] The residual operation graph construction module is used to generate redundant nodes in the neighborhood space of the residual route; and to construct a residual operation graph based on the redundant nodes and the main nodes on the residual route; the residual operation graph includes several directed edges.

[0036] The residual running graph update module is used to calculate the dynamic risk weight of each directed edge in the residual running graph; and update the edge set of the residual running graph according to the dynamic risk weight and a preset risk threshold to obtain the updated residual running graph.

[0037] The topology stability value calculation module is used to calculate the topology stability value of the updated residual running graph based on the dynamic risk weight of each directed edge in the updated residual running graph and the topology of the updated residual running graph.

[0038] The feasibility determination module is used to determine the feasibility of the residual route when the topology stability value is greater than or equal to a preset stability threshold.

[0039] In some embodiments, the low-altitude route operation feasibility determination system further includes a critical failure sub-map location module and a residual route update module, wherein:

[0040] The critical failure subgraph localization module is used to locate the critical failure region subgraph of the residual operation graph according to the dynamic risk weight when the topological stability value is less than the stability threshold.

[0041] The residual route update module is used to generate local alternative paths within the critical failure region subgraph using graph neural networks and reinforcement learning algorithms; and to concatenate the local alternative paths with the unfailed paths in the residual routes to obtain the updated residual routes.

[0042] Compared with related technologies, the low-altitude route operation feasibility determination method provided in this application obtains the actual flight position of the aircraft at the current moment, projects the actual flight position onto the aircraft's preset route to obtain the current reference point corresponding to the actual flight position; constructs a residual route based on the current reference point; generates redundant nodes in the neighborhood space of the residual route; constructs a residual operation graph based on the redundant nodes and the main nodes on the residual route; the residual operation graph includes several directed edges; calculates the dynamic risk weight of each directed edge in the residual operation graph; updates the edge set of the residual operation graph based on the dynamic risk weight and a preset risk threshold to obtain an updated residual operation graph; calculates the topological stability value of the updated residual operation graph based on the dynamic risk weight of each directed edge in the updated residual operation graph and the topological structure of the updated residual operation graph; when the topological stability value is greater than or equal to a preset stability threshold, the residual route is determined to be feasible. It can update the residual operation graph in real time based on the dynamic risk weight of the directed edges and determine the feasibility of the residual route based on the topological stability value of the updated residual operation graph, thereby achieving real-time determination of the feasibility of low-altitude route operation.

[0043] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0044] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0045] Figure 1 This is a hardware structure block diagram of the terminal of the low-altitude air route operation feasibility determination method in this embodiment;

[0046] Figure 2 This is a flowchart of the low-altitude air route operation feasibility determination method in this embodiment;

[0047] Figure 3 This is a preferred flowchart of the low-altitude air route operation feasibility determination method in this embodiment;

[0048] Figure 4 This is a structural block diagram of the low-altitude air route operation feasibility assessment system in this embodiment;

[0049] Figure 5 This is a preferred structural block diagram of the low-altitude air route operation feasibility determination system in this embodiment. Detailed Implementation

[0050] To better understand the purpose, technical solution, and advantages of this application, the application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0051] Unless otherwise defined, the technical or scientific terms used in this application shall have the general meaning understood by one of ordinary skill in the art to which this application pertains. Words such as “a,” “an,” “an,” “the,” “the,” and “these” used in this application do not indicate quantitative limitation and may be singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or modules (units) is not limited to the listed steps or modules (units) but may include steps or modules (units) not listed, or may include other steps or modules (units) inherent to these processes, methods, products, or devices. Words such as “connected,” “linked,” and “coupled” used in this application are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. “Multiple” used in this application refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. Normally, the character " / " indicates that the objects before and after it are in an "or" relationship. The terms "first," "second," "third," etc., used in this application are merely to distinguish similar objects and do not represent a specific order of objects.

[0052] The method embodiments provided in this example can be executed on a terminal, computer, or similar computing device. For example, it can run on a terminal. Figure 1 This is a hardware structure block diagram of the terminal for the low-altitude airway operation feasibility determination method in this embodiment. (See diagram for example.) Figure 1 As shown, a terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 and a memory 104 for storing data are also included. The processor 102 may be, but is not limited to, a microprocessor (MCU) or a programmable logic device (FPGA). The terminal may also include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that… Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the terminal described above. For example, the terminal may also include components that are larger than... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown are illustrated.

[0053] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the low-altitude airway operation feasibility determination method in this embodiment. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, thereby implementing the aforementioned method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0054] The transmission device 106 is used to receive or send data via a network. This network includes a wireless network provided by the terminal's communication provider. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 can be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0055] This embodiment provides a method for determining the feasibility of low-altitude air route operations. Figure 2 This is a flowchart of the low-altitude airway operation feasibility determination method in this embodiment, such as... Figure 2 As shown, the process includes the following steps:

[0056] Step S201: Obtain the actual flight position of the aircraft at the current moment, project the actual flight position onto the preset flight path of the aircraft to obtain the current reference point corresponding to the actual flight position; construct the residual flight path based on the current reference point.

[0057] Specifically, during flight, the aircraft's current actual flight position is acquired at fixed intervals. A minimum distance projection algorithm is then used to project this current actual flight position onto a pre-planned continuous flight path curve. This means finding a point on the preset flight path with the minimum Euclidean distance to the actual flight position; this point becomes the current reference point corresponding to the actual flight position. Using this current reference point as the starting position and the end point of the original preset flight path as the ending position, a residual flight path is constructed.

[0058] Step S202: Generate redundant nodes in the neighborhood space of the residual route; construct the residual operation graph based on the redundant nodes and the main nodes on the residual route.

[0059] Specifically, within the neighborhood of the residual flight path, a cross-sectional plane is constructed along the main nodes of the residual flight path. Regular mesh sampling is performed within this cross-sectional plane to generate regular redundant nodes. Using a conditional generative adversarial network (GAN), with the current environmental conditions as input, enhanced redundant nodes are generated. The regular redundant nodes and enhanced redundant nodes together constitute a set of candidate positions within the neighborhood of the residual flight path. The regular redundant nodes, enhanced redundant nodes, and the main nodes on the residual flight path are then fused and deduplicated to obtain the node set of the residual flight path. Based on spatial distance constraints and the flight direction, a set of candidate edges for the residual flight path is obtained from this set of nodes. A weighted residual operation graph is constructed based on this set of candidate edges. This weighted residual operation graph includes several directed edges.

[0060] Step S203: Calculate the dynamic risk weight of each directed edge in the residual running graph; based on the dynamic risk weight and the preset risk threshold, update the edge set of the residual running graph to obtain the updated residual running graph.

[0061] Specifically, a continuous risk field for the residual flight path is estimated using a trained deep neural network. This continuous risk field comprehensively considers dynamic risks from multiple sources, including weather, static obstacles, airspace conflicts, regulatory restrictions, and communication quality. Based on the continuous risk field of the residual flight path, the dynamic risk weight of each directed edge in the residual flight path graph is calculated. Then, based on the dynamic risk weight of each directed edge in the residual flight path graph and a preset risk threshold, the edge set of the residual flight path graph is updated, resulting in the updated residual flight path graph.

[0062] Step S204: Calculate the topological stability value of the updated residual graph based on the dynamic risk weight of each directed edge in the updated residual graph and the topological structure of the updated residual graph.

[0063] Specifically, the formula for calculating the topological stability value of the updated residual running graph is expressed as follows:

[0064] ;

[0065] in This represents the minimum risk cost required to disconnect the current reference point from the endpoint, i.e., the minimum cut cost. The formula for calculating the minimum cut cost is:

[0066] ;

[0067] in This represents the dynamic risk weight of directed edges in the updated residual graph. The larger the minimum cut cost, the more stable the residual graph; the smaller the minimum cut cost, the less stable the residual graph. This represents the number of feasible paths from the current reference point to the destination, i.e., the path redundancy index. This value can be estimated using the k-shortest path algorithm or the approximately disjoint path counting method. The larger the value, the stronger the system robustness. Let represent the set of bridge edges. A bridge edge is defined as an edge that, if deleted, would cause the current reference point and the endpoint to lose connectivity. The mean of the positive risk gradient is expressed as:

[0068] ;

[0069] in This represents the updated set of edges. This indicates the number of elements in the set. This represents the edge risk evolution rate, which is calculated using the following formula:

[0070] ;

[0071] in This represents the dynamic risk weight of directed edges in the updated residual graph. An edge risk evolution rate greater than 0 indicates an increase in risk, while an edge risk evolution rate less than 0 indicates a decrease in risk. The minimum cut value under the initial or calibration reference. For path redundancy under initial or calibration baseline, Let be the gradient normalization constant. These are the weighting coefficients. These are the weighting coefficients. These are the weighting coefficients. These are the weighting coefficients.

[0072] Step S205: When the topology stability value is greater than or equal to the preset stability threshold, the remaining route is determined to be feasible.

[0073] Specifically, when the topological stability value of the residual route is greater than or equal to a pre-set stability threshold, it indicates that the current residual route meets the requirements for safe operation in four dimensions: minimum cutting cost, path redundancy, bridge-side ratio, and risk evolution trend. That is, the residual route not only has a connecting path from the current reference point to the destination, but the path structure also has sufficient resistance to disruption and alternative detour space, while the overall risk does not show a rapid deterioration trend. Therefore, the system determines that the current residual route is feasible and path reconstruction is not required.

[0074] In this embodiment, by obtaining the actual flight position of the aircraft at the current moment, the actual flight position is projected onto the aircraft's preset flight path to obtain the current reference point corresponding to the actual flight position; a residual flight path is constructed based on the current reference point; redundant nodes are generated in the neighborhood space of the residual flight path; a residual operation graph is constructed based on the redundant nodes and the main nodes on the residual flight path; the residual operation graph includes several directed edges; the dynamic risk weight of each directed edge in the residual operation graph is calculated; the edge set of the residual operation graph is updated based on the dynamic risk weight and a preset risk threshold to obtain the updated residual operation graph; the topological stability value of the updated residual operation graph is calculated based on the dynamic risk weight of each directed edge in the updated residual operation graph and the topological structure of the updated residual operation graph; when the topological stability value is greater than or equal to the preset stability threshold, the residual flight path is determined to be feasible. This allows for real-time updating of the residual operation graph based on the dynamic risk weight of the directed edges, and determination of the feasibility of the residual flight path based on the topological stability value of the updated residual operation graph, thereby achieving real-time determination of the feasibility of low-altitude flight path operation.

[0075] In some embodiments, the actual flight position is projected onto a preset flight path of the aircraft to obtain a current reference point corresponding to the actual flight position, including:

[0076] Using the minimum distance projection algorithm, the projection parameters of the actual flight position onto the aircraft's preset flight path are obtained; based on the projection parameters, the current reference point corresponding to the actual flight position is obtained.

[0077] Specifically, let the set of discrete waypoints for the preset route be:

[0078] ;

[0079] in Indicates the starting point of the preset route. This represents the endpoint of the preset route, and the other points represent intermediate waypoints on the preset route. Let the Euclidean distance between adjacent waypoints be:

[0080] ;

[0081] Let the cumulative arc length be:

[0082] ;

[0083] The total length of the preset route is expressed as:

[0084] ;

[0085] During flight, the projection parameters of the actual flight position onto the aircraft's preset flight path. Represented as:

[0086] ;

[0087] in This indicates the actual flight position of the aircraft at time t. This indicates the position of the aircraft after traveling a distance *s* along the preset flight path from the starting point. The current reference point corresponding to the actual flight position is represented as:

[0088] ;

[0089] in This represents the cubic spline interpolation method. If the actual flight position falls between two discrete waypoints, a virtual reference point can be constructed using linear interpolation to ensure that the representation of the current point on a continuous flight path does not jump.

[0090] In some embodiments, redundant nodes are generated in the neighborhood space of the residual route, including:

[0091] Determine the neighborhood space of the remaining route; construct a cross-sectional plane within the neighborhood space; perform regular network sampling within the cross-sectional plane to generate regular redundant nodes; utilize a conditional generative adversarial network, taking the current environmental conditions as input, to generate enhanced redundant nodes.

[0092] Specifically, in the flight path curve Apply fixed arc length intervals Sampling is performed to obtain the set of master nodes, represented as:

[0093] ;

[0094] in Represented as:

[0095] ;

[0096] The remaining route is the unexecuted route from the current flight position to the destination. The neighborhood space of this remaining route is represented as follows:

[0097] ;

[0098] This neighborhood includes all routes whose distance does not exceed [the distance to the flight path]. The spatial points are candidate regions that allow the aircraft to perform lateral offsets and local maneuvers. A local 3D orthogonal basis is constructed at each master node. , Let the tangent vector be represented as:

[0099] ;

[0100] The unit tangent vector is represented as:

[0101] ;

[0102] Two tangent vectors are constructed using Gram-Schmidt orthogonalization. Orthogonal normal base , Thus in the master node Define a local cross-sectional plane perpendicular to the flight path direction. Within each cross-sectional plane, according to the sampling interval... Regular mesh sampling is performed to obtain regular redundant nodes. These regular redundant nodes are represented as follows:

[0103] ;

[0104] in and satisfy The set of redundant nodes in this rule is denoted as:

[0105] ;

[0106] Define the conditional probability distribution as:

[0107] ;

[0108] Where x represents the candidate node location, and c represents the local environmental conditions. These local environmental conditions can include wind field, rainfall, airspace density, no-fly zone boundaries, communication quality, etc. In conditional generative adversarial networks, the generator is represented as:

[0109] ;

[0110] The discriminator is represented as:

[0111] ;

[0112] The training data comes from historical feasible flight path points and their corresponding environmental conditions, and is represented as follows:

[0113] ;

[0114] The standard adversarial loss of conditional generative adversarial networks is defined as:

[0115] ;

[0116] To prevent the generated points from falling into high-risk areas, a safety constraint loss is added, which is expressed as:

[0117] ;

[0118] in For risk function, This represents the safety risk threshold. The total loss is expressed as:

[0119] ;

[0120] The training objective is represented as:

[0121] ;

[0122] After training is complete, during the inference phase, enhanced redundant nodes are generated by sampling and combining with conditions. These enhanced redundant nodes are represented as follows:

[0123] ;

[0124] Risk filtering and spatial deduplication are performed on the enhanced redundant node to obtain the final redundant enhancement set.

[0125] In some embodiments, a residual running graph is constructed based on redundant nodes and master nodes on the residual routes, including:

[0126] The redundant nodes and the main nodes on the residual routes are merged and deduplicated to obtain the node set of the residual routes; based on the node set of the residual routes, the residual candidate edge set is obtained; based on the residual candidate edge set, the residual running graph is constructed.

[0127] Specifically, the set of nodes of the remaining route is defined as follows:

[0128] ;

[0129] If any two nodes satisfy:

[0130] ;

[0131] Then, one of the nodes will be deleted or merged to avoid redundancy in the subsequent residual graph construction due to excessive density of local nodes. In addition, the path coordinates along the route are defined for each node, represented as:

[0132] ;

[0133] These path coordinates will be used to build the residual running graph later.

[0134] For any two nodes If the spatial distance between the two satisfies:

[0135] ;

[0136] And the path coordinates satisfy:

[0137] ;

[0138] Then it is believed arrive There is a potential forward connection relationship; construct candidate directed edges:

[0139] ;

[0140] Direction constraints here The purpose is to ensure that the direction of the edges in the graph is consistent with the direction of flight propulsion, avoiding meaningless backtracking edges. For local detours, a small number of lateral connecting edges with similar path coordinates need to be retained, but obvious reverse connections are not allowed in principle.

[0141] For each candidate edge Define a straight path as:

[0142] ;

[0143] This path indicates that the aircraft started from the node. Fly to the node The candidate geometric trajectory. Subsequent risk integrals will unfold along this curve. At time... The aircraft's current location is The corresponding path coordinates are Therefore, only those that satisfy the condition are retained. The nodes are added to the residual execution graph. The set of residual nodes is defined as:

[0144] ;

[0145] Furthermore, both ends of the node are kept in Candidate edges whose directions satisfy the forward constraints are used to obtain the set of residual candidate edges. After generating candidate edges, the residual running graph is constructed, represented as follows:

[0146] ;

[0147] in This residual operational graph represents the local directed graph structure from the current flight position to the target point that has not yet been executed and is available for online evaluation and repair. It is important to emphasize that the prefix portion of the graph that has already been executed is no longer included in the feasibility determination for the current cycle.

[0148] In some of these embodiments, the dynamic risk weight of each directed edge in the residual running graph is calculated, including:

[0149] The continuous risk field of the residual route is estimated using a trained deep neural network; based on the continuous risk field of the residual route, the dynamic risk weight of each directed edge in the residual flight graph is calculated.

[0150] Specifically, the continuous risk field of the residual route is defined as follows:

[0151] ;

[0152] The continuous risk field is decomposed into a weighted sum of several types of physical risks, expressed as:

[0153] ;

[0154] in It indicates meteorological risks, reflecting the impact of wind speed, wind direction, rainfall, visibility, turbulence, etc. on flight safety; It indicates the risk of static obstacles, reflecting the proximity of fixed obstacles such as buildings, towers, and mountains; It indicates the risk of airspace conflict, reflecting the probability of conflict caused by other aircraft, temporary activity airspace, etc.; It indicates regulatory risks and reflects constraints such as no-fly zones, height-restricted zones, and temporary control areas; This indicates communication risks, reflecting the control risks caused by degraded link quality or signal blind spots. For the weighting of various risks, the following conditions must be met:

[0155] ;

[0156] Considering that the risk distribution in real-world environments may be non-linear and the data sparse, a function approximator is used to learn the risk field. The input feature vector is defined as:

[0157] ;

[0158] By parameterizing functions:

[0159] ;

[0160] Predicted location At any moment The risk value. Among them... For network parameters, This can be achieved using multilayer perceptrons, spatiotemporal networks, or other regression models.

[0161] The training samples are represented as follows:

[0162] ;

[0163] in This represents the true risk value obtained from historical samples or simulation annotations. The loss function is defined as:

[0164] ;

[0165] The training objective is represented as

[0166] ;

[0167] After training, use Online estimation of the continuous risk field of the remaining route.

[0168] For any edge in the residual running graph Its dynamic risk weight is defined as:

[0169] ;

[0170] in The parameterized straight path on this edge is represented as:

[0171] ;

[0172] If a discrete approximation is used, the dynamic risk weight can be written as:

[0173] ;

[0174] in This represents the number of discrete sampling points on the edge path.

[0175] In some embodiments, the edge set of the residual running graph is updated according to dynamic risk weights and a preset risk threshold to obtain an updated residual running graph, including:

[0176] When the dynamic risk weight is greater than the risk threshold, the directed edge corresponding to the dynamic risk weight is deleted from the edge set of the residual running graph to obtain the edge set of the updated residual running graph; based on the edge set of the updated residual running graph, the updated residual running graph is obtained.

[0177] Specifically, let the risk threshold be... When dynamic risk weights Greater than the risk threshold At that time, determine the directed edge corresponding to the dynamic risk weight. If the path is impassable at the current moment, remove the directed edge from the edge set of the residual graph. The resulting edge set of the updated residual graph is represented as follows:

[0178] ;

[0179] Based on the updated set of edges in the residual graph, the updated residual graph is obtained, represented as:

[0180] ;

[0181] in This represents the dynamic weighting function.

[0182] In some of these embodiments, Figure 3 This is a preferred flowchart of the low-altitude air route operation feasibility determination method in this embodiment, as follows: Figure 3 As shown, the method includes Figure 2 All the steps shown, in addition to:

[0183] Step S301: When the topological stability value is less than the stability threshold, locate the key failure region subgraph of the residual operation graph according to the dynamic risk weight.

[0184] Step S302: Using graph neural networks and reinforcement learning algorithms, local alternative paths are generated within the key failure region subgraph.

[0185] Step S303: The local alternative path is spliced ​​with the unfailed paths in the residual route to obtain the updated residual route.

[0186] Specifically, when the topologically stable value Less than the stability threshold At that time, the remaining route is determined to be infeasible. The standardized value of the edge risk is defined as:

[0187] ;

[0188] The standardized risk value of this edge is used to standardize risk comparisons between different edges.

[0189] The standardized risk gradient is defined as:

[0190] ;

[0191] This standardized risk gradient reflects the relative strength of edge risk growth.

[0192] The criticality score for each edge is defined as follows:

[0193] ;

[0194] in express Beside the bridge; express It belongs to the minimum cut set; express Located in the current reference path Up. The current reference path. It is the current reference optimal path from the current reference point to the destination on the current residual running graph. The purpose of adding scoring is to avoid over-processing edges that are irrelevant to the task.

[0195] Let the threshold be The set of critical failure edges is defined as follows:

[0196] ;

[0197] For the set of critical failure edges First, extract its endpoint set, which is represented as:

[0198] ;

[0199] Further expanding the neighborhood of this endpoint set yields the expanded set, denoted as:

[0200] ;

[0201] in express The order neighborhood expansion, the key failure region subgraph is defined as:

[0202] ;

[0203] in Represented as:

[0204] ;

[0205] The local incremental reconstruction process is modeled as a Markov decision process, represented as:

[0206] ;

[0207] in, For state space, For the action space, Let be the state transition probability. For the reward function, This is the discount factor. At time [time]... The local repair state is defined as:

[0208] ;

[0209] in This is a sub-map of the key region. For the local repair starting point boundary, To locally repair the endpoint boundary, This is the current node in the current repair process. This is the prefix for the currently generated local repair path. Nodes in the key region subgraph. The eigenvectors are defined as follows:

[0210] ;

[0211] in and Let represent the aggregate statistics of edge risk and edge risk gradient associated with this node, respectively. This indicates whether the node belongs to a critical area. Indicates whether it is a boundary node. The action space is a feasible edge of the current node, represented as:

[0212] ;

[0213] To avoid invalid actions, the action set must satisfy: Node Belongs to a local subgraph or its boundary connectivity set; edge The dynamic risk weight is less than or equal to the risk threshold; Not less than Or only a slight rollback is allowed; the path may reach the local repair endpoint boundary.

[0214] To ensure the strategy restores connectivity while maintaining low risk and low perturbation, the immediate reward for each step is defined as follows:

[0215] ;

[0216] in The connectivity reward is represented as:

[0217] ;

[0218] For stability rewards, it is represented as:

[0219] ;

[0220] The path length penalty is represented as:

[0221] ;

[0222] Risk penalties are represented as follows:

[0223] ;

[0224] The rollback penalty is represented as follows:

[0225] ;

[0226] When the local repair path is successfully connected and When the time is right, a termination reward is given to encourage the strategy to complete the bridging as quickly as possible.

[0227] To fully utilize the local graph structure, the local subgraph is input into the graph neural network encoder. Let the... Layer nodes are represented as Then message passing update is represented as:

[0228] ;

[0229] in Represents a node The neighborhood group, For learnable parameters, It is a non-linear activation function. After several layers of message passing, the local subgraph embedding is obtained. The policy network outputs the action probability, which is expressed as:

[0230] ;

[0231] in These are the policy network parameters.

[0232] The policy is trained using Proximal Policy Optimization (PPO) with the goal of maximizing long-term cumulative reward, expressed as:

[0233] ;

[0234] The objective function for PPO shearing is defined as:

[0235] ;

[0236] in The probability ratio is expressed as:

[0237] ;

[0238] The dominant function; The shearing threshold is used. Adding the value function loss and entropy regularization term, the total loss is obtained, expressed as:

[0239] ;

[0240] in The mean squared error loss of the value network; Let be the policy entropy.

[0241] When the strategy finds a path from the local subgraph arrive Feasible path When this feasible path is used, the failed segment in the original reference path is replaced to form a new residual repair path, which is represented as follows:

[0242] ;

[0243] in For currently executed or confirmed safe preliminary paths, From The original safe path to the destination.

[0244] For example, there is a critical edge in the current residual path. For high-risk failure edges, there are feasible redundant nodes in the local subgraph. The trained policy network selects action sequences: This means finding a bypass bridging path within a key local area.

[0245] In some embodiments, when the topological stability value is less than a stability threshold, the critical failure region subgraph of the residual operation graph is located based on dynamic risk weights, including:

[0246] When the topological stability value is less than the stability threshold, the set of critical failure edges in the residual running graph is located based on the dynamic risk weight; and the critical failure region subgraph of the residual running graph is located based on the set of critical failure edges in the residual running graph.

[0247] Specifically, when the topologically stable value Less than the stability threshold At that time, the remaining route is determined to be infeasible. The standardized value of the edge risk is defined as:

[0248] ;

[0249] The standardized risk value of this edge is used to standardize risk comparisons between different edges.

[0250] The standardized risk gradient is defined as:

[0251] ;

[0252] This standardized risk gradient reflects the relative strength of edge risk growth.

[0253] The criticality score for each edge is defined as follows:

[0254] ;

[0255] in express Beside the bridge; express It belongs to the minimum cut set; express Located in the current reference path Up. The current reference path. It is the current reference optimal path from the current reference point to the destination on the current residual running graph. The purpose of adding scoring is to avoid over-processing edges that are irrelevant to the task.

[0256] Let the threshold be The set of critical failure edges is defined as follows:

[0257] ;

[0258] For the set of critical failure edges First, extract its endpoint set, which is represented as:

[0259] ;

[0260] Further expanding the neighborhood of this endpoint set yields the expanded set, denoted as:

[0261] ;

[0262] in express The order neighborhood expansion, the key failure region subgraph is defined as:

[0263] ;

[0264] in Represented as:

[0265] ;

[0266] This embodiment also provides a low-altitude airway operation feasibility determination system, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. The terms "module," "unit," and "subunit," etc., used below refer to combinations of software and / or hardware that achieve a predetermined function. Although the system described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0267] Figure 4 This is a structural block diagram of the low-altitude airway operation feasibility assessment system in this embodiment, as shown below. Figure 4 As shown, the low-altitude route operation feasibility assessment system 40 includes: a residual route construction module 401, a residual flight map construction module 402, a residual flight map update module 403, a topology stability value calculation module 404, and a feasibility assessment module 405, wherein:

[0268] The residual route construction module 401 is used to obtain the actual flight position of the aircraft at the current moment, project the actual flight position onto the aircraft's preset route to obtain the current reference point corresponding to the actual flight position, and construct the residual route based on the current reference point.

[0269] The residual operation graph construction module 402 is used to generate redundant nodes in the neighborhood space of the residual route; and to construct the residual operation graph based on the redundant nodes and the main nodes on the residual route; the residual operation graph includes several directed edges.

[0270] The residual running graph update module 403 is used to calculate the dynamic risk weight of each directed edge in the residual running graph; based on the dynamic risk weight and the preset risk threshold, it updates the edge set of the residual running graph to obtain the updated residual running graph.

[0271] The topology stability value calculation module 404 is used to calculate the topology stability value of the updated residual running graph based on the dynamic risk weights and the topology of the updated residual running graph.

[0272] The feasibility determination module 405 is used to determine the feasibility of the residual route when the topology stability value is greater than or equal to the preset stability threshold.

[0273] Figure 5 This is a preferred structural block diagram of the low-altitude airway operation feasibility determination system in this embodiment, as shown below. Figure 5 As shown, the low-altitude air route operation feasibility assessment system 40 includes... Figure 4 All the modules shown, in addition to the critical failure submap location module 501 and the residual route update module 502, include:

[0274] The critical failure subgraph localization module 501 is used to locate the critical failure region subgraph of the residual operation graph according to the dynamic risk weight when the topological stability value is less than the stability threshold.

[0275] The residual route update module 502 is used to generate local alternative paths in the critical failure area subgraph using graph neural networks and reinforcement learning algorithms; and to concatenate the local alternative paths with the unfailed paths in the residual routes to obtain the updated residual routes.

[0276] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0277] It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit it. All other embodiments derived by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0278] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0279] Obviously, the accompanying drawings are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar situations based on these drawings without any creative effort. Furthermore, it is understood that although the work done in this development process may be complex and lengthy, for those skilled in the art, certain design, manufacturing, or production modifications made based on the technical content disclosed in this application are merely conventional technical means and should not be considered as insufficient disclosure of this application.

[0280] The term "embodiment" in this application refers to a specific feature, structure, or characteristic described in connection with an embodiment that may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily imply the same embodiment, nor does it imply that it is mutually exclusive with or independent of other embodiments. It will be clearly or implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0281] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of patent protection. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the appended claims.

Claims

1. A method for determining the feasibility of low-altitude air route operation, characterized in that, include: Obtain the actual flight position of the aircraft at the current moment, project the actual flight position onto the preset flight path of the aircraft, and obtain the current reference point corresponding to the actual flight position; Construct the residual route based on the current reference point; Redundant nodes are generated in the neighborhood space of the residual route; a residual operation graph is constructed based on the redundant nodes and the main nodes on the residual route; the residual operation graph includes several directed edges; Calculate the dynamic risk weight of each directed edge in the residual running graph; Based on the dynamic risk weights and preset risk thresholds, the edge set of the residual running graph is updated to obtain the updated residual running graph; Based on the dynamic risk weight of each directed edge in the updated residual running graph and the topology of the updated residual running graph, calculate the topological stability value of the updated residual running graph. When the topology stability value is greater than or equal to a preset stability threshold, the residual route is determined to be feasible.

2. The method for determining the feasibility of low-altitude air route operation according to claim 1, characterized in that, The step of projecting the actual flight position onto the aircraft's preset flight path to obtain the current reference point corresponding to the actual flight position includes: The projection parameters of the actual flight position projected onto the preset flight path of the aircraft are obtained by using the minimum distance projection algorithm; Based on the projection parameters, the current reference point corresponding to the actual flight position is obtained.

3. The method for determining the feasibility of low-altitude air route operation according to claim 1, characterized in that, The process of generating redundant nodes in the neighborhood of the remaining route includes: Determine the neighborhood space of the remaining route; within the neighborhood space, construct a cross-sectional plane; Regular network sampling is performed within the cross-sectional plane to generate regular redundant nodes; an enhanced redundant node is generated using a conditional generative adversarial network with the current environmental conditions as input.

4. The method for determining the feasibility of low-altitude air route operation according to claim 1, characterized in that, The step of constructing a residual flight path based on the redundant nodes and the master nodes on the residual routes includes: The redundant nodes and the main nodes on the residual route are merged and deduplicated to obtain the node set of the residual route. Based on the set of nodes of the remaining routes, a set of candidate edges for the remaining routes is obtained; based on the set of candidate edges for the remaining routes, the remaining route graph is constructed.

5. The method for determining the feasibility of low-altitude air route operation according to claim 1, characterized in that, The calculation of the dynamic risk weight of each directed edge in the residual running graph includes: The continuous risk field of the residual route is estimated using a trained deep neural network; Based on the continuous risk field of the residual route, the dynamic risk weight of each directed edge in the residual flight graph is calculated.

6. The method for determining the feasibility of low-altitude air route operation according to claim 1, characterized in that, The step of updating the edge set of the residual running graph according to the dynamic risk weight and the preset risk threshold to obtain the updated residual running graph includes: When the dynamic risk weight is greater than the risk threshold, the directed edge corresponding to the dynamic risk weight is deleted from the edge set of the residual running graph to obtain the edge set of the updated residual running graph. The updated residual running graph is obtained based on the edge set of the updated residual running graph.

7. The method for determining the feasibility of low-altitude air route operation according to claim 1, characterized in that, Also includes: When the topological stability value is less than the stability threshold, the key failure region subgraph of the residual operation graph is located according to the dynamic risk weight. Using graph neural networks and reinforcement learning algorithms, local alternative paths are generated within the critical failure region subgraph; these local alternative paths are then concatenated with the unfailed paths in the residual route to obtain the updated residual route.

8. The method for determining the feasibility of low-altitude air route operation according to claim 7, characterized in that, When the topological stability value is less than the stability threshold, the critical failure region subgraph of the residual operation graph is located according to the dynamic risk weight, including: When the topological stability value is less than the stability threshold, the set of critical failure edges of the residual running graph is located according to the dynamic risk weight. Based on the set of critical failure edges in the residual running graph, locate the critical failure region subgraph of the residual running graph.

9. A low-altitude air route operation feasibility assessment system, characterized in that, It includes a residual route construction module, a residual flight path construction module, a residual flight path update module, a topology stability value calculation module, and a feasibility determination module, among which: The residual route construction module is used to obtain the actual flight position of the aircraft at the current moment, project the actual flight position onto the preset route of the aircraft to obtain the current reference point corresponding to the actual flight position, and construct the residual route based on the current reference point; The residual operation graph construction module is used to generate redundant nodes in the neighborhood space of the residual route; and to construct a residual operation graph based on the redundant nodes and the main nodes on the residual route; the residual operation graph includes several directed edges. The residual running graph update module is used to calculate the dynamic risk weight of each directed edge in the residual running graph; and update the edge set of the residual running graph according to the dynamic risk weight and the preset risk threshold to obtain the updated residual running graph. The topology stability value calculation module is used to calculate the topology stability value of the updated residual running graph based on the dynamic risk weight of each directed edge in the updated residual running graph and the topology of the updated residual running graph. The feasibility determination module is used to determine the feasibility of the residual route when the topology stability value is greater than or equal to a preset stability threshold.

10. The low-altitude air route operation feasibility assessment system according to claim 9, characterized in that, It also includes a critical failure submap location module and a residual route update module, wherein: The critical failure subgraph localization module is used to locate the critical failure region subgraph of the residual operation graph according to the dynamic risk weight when the topological stability value is less than the stability threshold. The residual route update module is used to generate local alternative paths within the critical failure region subgraph using graph neural networks and reinforcement learning algorithms; and to concatenate the local alternative paths with the unfailed paths in the residual routes to obtain the updated residual routes.

Images