Map-based location information display method and system

Abstract

The present application belongs to the field of flight navigation, and particularly relates to a map-based position information display method and system, aiming at solving the problem that the prior art uses static grid division and separated prediction model, which easily leads to system misalignment. The method of the present application comprises: in response to a position display request of an airborne device, determining a target region to be displayed and a scene semantic type; according to the scene semantic type, constructing an optimization function of multi-target cooperation, and obtaining optimal grid division parameters by solving the optimization function; according to the optimal grid division parameters, dividing the target region into multiple target grids, and constructing a space-time relationship graph; using a pre-trained space-time graph neural network model to predict access probability and access intensity; and according to the access probability and the access intensity, sequentially loading map information of each target grid in the airborne device. The method effectively improves the performance and efficiency of map information loading.

Description

Technical Field

[0001] This invention belongs to the technical field of flight navigation, and specifically relates to a map-based location information display method and system. Background Technology

[0002] Currently, airborne map information loading systems generally adopt a technical solution that combines static grid partitioning with a separate prediction model. In the grid partitioning stage, the target area is usually mechanically divided according to fixed size rules (such as equally spaced latitude and longitude grids or recursive quadtree partitioning), without considering the semantic characteristics differences of different scenarios; while the prediction model mostly uses independent spatial graph networks or temporal models.

[0003] The aforementioned technical solutions have limitations in practical applications. Static grid partitioning and separate prediction models can lead to the failure of spatiotemporal feature coordination, resulting in inaccurate system performance in dynamic environments. Specifically, the static rules in the grid partitioning process cannot adapt to changes in scene semantics, the separate spatial model ignores cross-regional migration patterns, and the temporal model cannot respond to the geographical diffusion effects of sudden disturbances. This ultimately leads to resource mismatch in loading decisions. For example, in critical navigation areas (such as airport approach airspace), there is a risk of information loss due to prediction errors, while low-value areas (such as uninhabited hills) continuously consume more than 60% of the GPU memory resources, severely restricting the aircraft's real-time obstacle avoidance capabilities and route planning efficiency. Summary of the Invention

[0004] To address the aforementioned problems in existing technologies, namely the failure of spatiotemporal feature coordination caused by static grid partitioning and separate prediction models during airborne map information loading, leading to inaccurate performance in dynamic environments, this invention proposes a map-based location information display method, comprising:

[0005] In response to the location display request from the airborne equipment, determine the target area to be displayed and the scene semantic type corresponding to the target area;

[0006] Based on the scene semantic type, a multi-objective collaborative optimization function is constructed, and the optimal grid partitioning parameters are obtained by solving the optimization function. The constraint objectives of the optimization function include a complexity minimization objective and a redundancy minimization objective. The complexity minimization objective is used to ensure that the computational load of the airborne equipment is lower than a preset threshold; the redundancy minimization objective is used to minimize the duplicate caching of map information between adjacent grids.

[0007] Based on the optimal grid division parameters, the target area is divided into multiple target grids, and a spatiotemporal relationship diagram of the multiple target grids is constructed by combining the map information of the target area.

[0008] Based on the spatiotemporal relationship graph, a pre-trained spatiotemporal graph neural network model is used to predict the access probability and access intensity of each target grid in a future preset time period. The spatiotemporal graph neural network model is a memory-enhanced spatiotemporal metagraph network. The memory-enhanced spatiotemporal metagraph network has a built-in inference engine and a multi-level memory pool. The inference engine is used to dynamically generate model parameters according to the scene semantic type of the target area. The multi-level memory pool is used to fuse real-time trajectory and historical event features, and to impose constraints based on preset aviation rules, and output prediction results that meet the constraints.

[0009] Based on the access probability and access intensity, map information for each target grid is sequentially loaded into the airborne device.

[0010] In some preferred embodiments, responding to the location display request from the airborne equipment and determining the target area to be displayed, and the scene semantic type corresponding to the target area, includes:

[0011] Receive display requests sent by airborne equipment, and parse the latitude and longitude range, altitude layer parameters, and display scale contained in the requests;

[0012] The three-dimensional spatial boundary is determined based on the latitude and longitude range and the altitude layer parameters. The three-dimensional spatial boundary includes a horizontal range boundary and a vertical altitude boundary.

[0013] The target area is determined based on the three-dimensional spatial boundary and the display scale.

[0014] In some preferred embodiments, the method for obtaining the scene semantic type includes:

[0015] Call the preset geographic information database to extract the distribution characteristics of ground features and spatial attributes within the target area;

[0016] Based on the distribution characteristics of the ground features and the airspace attributes, the scene semantic type corresponding to the target area is determined. The scene semantic type includes urban airspace type, airway corridor type, and special control area type.

[0017] In some preferred embodiments, the step of constructing a multi-objective collaborative optimization function based on the scene semantic type, and obtaining the optimal mesh partitioning parameters by jointly solving the multi-objective collaborative optimization function through multiple constraints, includes:

[0018] Based on the semantic type of the scenario, select the associated optimization objective and constraint from the predefined set of optimization objectives and constraint sets;

[0019] Based on the selected optimization objectives and constraints, a multi-objective collaborative optimization function with grid partitioning parameters as decision variables is constructed.

[0020] The function is solved using a multi-constraint optimization algorithm, and the optimal meshing parameters that satisfy all constraints are output.

[0021] In some preferred embodiments, the step of dividing the target region into multiple target grids according to the optimal grid division parameters, and constructing a spatiotemporal relationship diagram of the multiple target grids in conjunction with the map information of the target region, includes:

[0022] According to the dynamic mesh partitioning strategy, the target region is decomposed into multiple non-overlapping target mesh sets;

[0023] Acquire historical trajectory data of airborne equipment and count the access frequency of each target grid at different time periods;

[0024] Establish a spatial adjacency matrix between grid cells and integrate periodic flow pattern data to generate a spatiotemporal relationship graph structure with spatiotemporal weights.

[0025] In some preferred embodiments, the step of using a pre-trained spatiotemporal graph neural network model to predict the access probability and access intensity of each target grid within a preset future time period includes:

[0026] Based on the constructed spatiotemporal relationship graph, the historical access sequence features of the target grid corresponding to each node in the graph and the spatial association features between nodes are extracted.

[0027] The historical access sequence features and spatial association features are input into the multi-level memory pool, which includes at least a short-term memory unit and a long-term memory unit, respectively used to cache real-time track data and fuse historical airspace events and typical traffic pattern features.

[0028] Based on the inference engine, combined with the scene semantic type of the current target region, model parameters suitable for the current scene are generated, and under the control of the inference engine, the output features of the multi-level memory pool are fused to perform forward inference calculation.

[0029] The preliminary results obtained from forward inference calculations are matched and verified with preset aviation rules, and outputs that do not meet the safety constraints are corrected.

[0030] The output is the access probability and access intensity of each target grid within a future preset time period, after correction to comply with the aforementioned aviation rule safety constraints.

[0031] In some preferred embodiments, the step of sequentially loading map information of each target grid into the airborne device according to the access probability and access intensity includes:

[0032] Based on a preset weighted fusion function, the access probability and access intensity of the same target grid are linearly weighted to determine the priority score corresponding to each target grid.

[0033] Determine the airspace control requirements for the target area and adjust the priority scores of the target grids in the airspace control areas within the target area;

[0034] The map loading sequence is obtained by arranging the target grids according to their priority scores.

[0035] According to the map loading sequence, the map information for each target grid is loaded.

[0036] In some preferred embodiments, loading map information for each target grid according to the map loading sequence includes:

[0037] The map loading sequence is analyzed, and the multiple target grids are divided into multiple loading batches, with each loading batch including at least one target grid.

[0038] Based on the hardware performance of the airborne equipment, a loading rate and rendering strategy are assigned to the target meshes of different loading batches one by one;

[0039] Based on the loading rate and rendering strategy, map information of each target grid is presented on the display interface of the airborne device.

[0040] In some preferred embodiments, the training process of the spatiotemporal graph neural network model includes:

[0041] Historical aeronautical data from multiple regions is collected, including grid access records of airspace at different scales;

[0042] The historical aeronautical data is standardized to eliminate differences in regional scale and grid density to obtain a training sample set. The standardization process is as follows: the physical dimensions of the grids in different airspaces are converted into dimensionless values ​​in a uniform scale coordinate system, and the grid access frequency is graded and normalized according to airspace type.

[0043] Based on the training sample set, the model is iteratively trained using the mini-batch gradient descent algorithm. In each round of training, the loss function value on the validation set is calculated, and the learning rate of the model is dynamically adjusted according to the loss function value until the preset training termination condition is met.

[0044] This invention also proposes a map-based location information display system, comprising:

[0045] The data response module is configured to respond to the location display request of the airborne equipment, determine the target area to be displayed, and the scene semantic type corresponding to the target area;

[0046] The strategy calculation module is configured to construct a multi-objective collaborative optimization function based on the scene semantic type, and obtain the optimal grid partitioning parameters by solving the optimization function. The constraint objectives of the optimization function include a complexity minimization objective and a redundancy minimization objective. The complexity minimization objective is used to ensure that the computational load of the airborne equipment is lower than a preset threshold; the redundancy minimization objective is used to minimize the duplicate caching of map information between adjacent grids.

[0047] The strategy execution module is configured to divide the target area into multiple target grids according to the optimal grid division parameters, and to construct a spatiotemporal relationship diagram of the multiple target grids in combination with the map information of the target area;

[0048] The model prediction module is configured to predict the access probability and access intensity of each target grid within a preset time period based on the spatiotemporal relationship graph using a pre-trained spatiotemporal graph neural network model. The spatiotemporal graph neural network model is a memory-enhanced spatiotemporal metagraph network. The memory-enhanced spatiotemporal metagraph network has a built-in inference engine and a multi-level memory pool. The inference engine is used to dynamically generate model parameters according to the scene semantic type of the target area. The multi-level memory pool is used to fuse real-time flight track and historical event features, and to impose constraints based on preset aviation rules, and output prediction results that conform to the constraints.

[0049] The data loading module is configured to sequentially load map information of each target grid into the airborne device according to the access probability and access intensity.

[0050] The beneficial effects of this invention are:

[0051] Traditional methods employ fixed grid partitioning and static loading strategies, which are ill-suited to complex and ever-changing flight environments. This invention, by responding to location requests, first identifies the scene semantic type of the target area and dynamically generates optimal grid partitioning parameters accordingly. This allows the grid partitioning to adaptively adjust based on the complexity of the actual scene, achieving intelligent and adaptive airborne map information loading and significantly improving system performance and efficiency.

[0052] Based on this invention, by constructing a multi-objective optimization function with the core of minimizing complexity and redundancy, the optimal mesh parameters are solved. This ensures that the divided mesh can strictly control the computational load of the airborne equipment below a preset threshold, avoiding stuttering or delay due to excessive rendering pressure. It can also minimize data redundancy caching between adjacent meshes, saving valuable memory and storage space, thereby achieving the optimal balance of performance under resource constraints.

[0053] This invention constructs a spatiotemporal relationship graph of a grid and utilizes an innovative memory-enhanced spatiotemporal metagraph network model for prediction. This model not only captures the spatiotemporal dependencies between grids through a spatiotemporal graph neural network, but also enhances generalization ability by dynamically adjusting model parameters based on scene semantics through its built-in inference engine. Its multi-level memory pool, by fusing real-time flight track and historical event features and performing constraint corrections based on preset aviation rules, makes the prediction results more accurate. Based on this prediction result, map information is preloaded sequentially, ensuring that the data required by the user is ready before the view changes, completely eliminating the lag caused by real-time loading in traditional methods and providing a smoother visual experience. Attached Figure Description

[0054] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0055] Figure 1 This is a flowchart illustrating a map-based location information display method proposed in an embodiment of the present invention;

[0056] Figure 2 This is a schematic diagram of the structure of a computer system proposed in an embodiment of the present invention. Detailed Implementation

[0057] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0058] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0059] Reference Figure 1 ,like Figure 1 As shown, the present invention provides a map-based location information display method, including:

[0060] Step S10: Respond to the location display request from the airborne equipment, determine the target area to be displayed, and the scene semantic type corresponding to the target area;

[0061] Specifically, the location display request in this embodiment includes, but is not limited to, the following core input parameters:

[0062] Current status of airborne equipment: Real-time geolocation: longitude, latitude, altitude (provided by GPS / INS); Attitude and heading: pitch, roll, true heading / magnetic heading; Motion status: airspeed, ground speed, vertical rate; Display request parameters: Display mode: for example, map mode (MAP), planning mode (PLN), vertical situation display (VSD), airport map (APT), etc. The mode determines the processing logic; Range / zoom level: the requested display range, such as 5 nautical miles, 10 nautical miles, full range, etc.

[0063] In this embodiment, the target area refers to the geographical range ultimately displayed on the screen, and its determination is based on the following criteria:

[0064] Location-based centering region: Using the current latitude and longitude of the airborne equipment as the geometric center, calculate a rectangular latitude and longitude range (North-East-South-West) or a circular area based on the requested scaling level (e.g., 20 nautical miles radius).

[0065] Heading-based look-ahead region: Based on the current ground speed and heading, predict the position for a period of time in the future, so that the heading always points to the top of the screen (heading up mode).

[0066] Focus-based offset area: If the driver selects a specific target (such as ZOOM FPV or directly selects an airport), the coordinates of that target will be used as the display center.

[0067] In this embodiment, after determining the target geographical area, the scene semantics of the area are understood to determine how to optimally render and display information in order to highlight the key elements of the target area and downplay secondary information.

[0068] As a feasible implementation method, the elements contained in the target area can be determined by querying an airborne integrated database (such as navigation databases, terrain databases, and airport databases provided by Jeppesen, LIDO, etc.), thereby determining the scene semantic type.

[0069] Step S20: Based on the scene semantic type, construct a multi-objective collaborative optimization function, and obtain the optimal grid partitioning parameters by solving the optimization function. The constraint objectives of the optimization function include a complexity minimization objective and a redundancy minimization objective. The complexity minimization objective is used to ensure that the computational load of the airborne equipment is lower than a preset threshold; the redundancy minimization objective is used to minimize the duplicate caching of map information between adjacent grids.

[0070] In this embodiment, the mesh generation parameter refers to a multidimensional parameter used to define the mesh generation process, which includes at least:

[0071] G_size: Used to indicate the physical size of the underlying mesh;

[0072] LOD_levels: This indicates the number of levels of detail contained in each grid. Different LODs correspond to different map data resolutions.

[0073] In this embodiment, at least two constraint objectives are quantized and combined into a solvable optimization function.

[0074] It is easy to understand that the computational load of airborne equipment is strongly correlated with factors such as the number of mesh vertices to be processed and rendered, and the LOD complexity. Therefore, the number of vertices and LOD complexity of the mesh can be effectively determined based on its computational load.

[0075] It should be noted that the redundancy referred to in this embodiment mainly comes from:

[0076] Data overlap across grid boundaries: For seamless rendering, adjacent grids usually retain a certain overlap area at the boundary. The larger the overlap area, the higher the redundancy. Multiple LOD caches: Data at multiple different levels of detail (LOD) are usually cached in the same geographic area for quick switching. Obviously, the more LOD levels cached, the higher the redundancy.

[0077] Since two or more constraint objectives (reducing complexity, reducing redundancy) may conflict (for example, a finer mesh may reduce redundancy but increase complexity), this embodiment requires multi-objective collaborative optimization.

[0078] For example, two objective functions can be merged into a single overall objective function using a weighted summation method. The weight coefficients can be used to reflect the emphasis on the two objectives, and these weight coefficients can be dynamically set according to the semantic type of the scene, so that the mesh division can be fully adapted to the constraints of the scene semantics.

[0079] The optimization function in this embodiment is a typical constrained nonlinear optimization problem, which can be solved using a genetic algorithm (GA) or a particle swarm optimization (PSO) algorithm. This embodiment does not impose any restrictions on this.

[0080] Step S30: According to the optimal grid division parameters, the target area is divided into multiple target grids, and a spatiotemporal relationship diagram of the multiple target grids is constructed by combining the map information of the target area.

[0081] Specifically, the process of constructing the spatiotemporal relationship diagram in this step includes:

[0082] According to the dynamic grid partitioning strategy, the target area is decomposed into multiple non-overlapping target grid sets; historical trajectory data of airborne equipment is acquired, and the access frequency of each target grid in different time periods is statistically analyzed; a spatial adjacency matrix between grid units is established, and periodic traffic pattern data is integrated to generate a spatiotemporal relationship graph structure with spatiotemporal weights.

[0083] Step S40: Based on the spatiotemporal relationship graph, a pre-trained spatiotemporal graph neural network model is used to predict the access probability and access intensity of each target grid in a future preset time period. The spatiotemporal graph neural network model is a memory-enhanced spatiotemporal metagraph network. The memory-enhanced spatiotemporal metagraph network has a built-in inference engine and a multi-level memory pool. The inference engine is used to dynamically generate model parameters according to the scene semantic type of the target area. The multi-level memory pool is used to fuse real-time flight track and historical event features, and to constrain them based on preset aviation rules, and output prediction results that meet the constraints.

[0084] Step S50: Based on the access probability and access intensity, sequentially load map information of each target grid into the airborne device.

[0085] Specifically, in step S40, a pre-trained spatiotemporal graph neural network model is used to predict the access probability and access intensity of each target grid within a preset future time period, including:

[0086] Based on the constructed spatiotemporal relationship graph, historical access sequence features and spatial association features between nodes corresponding to target grids are extracted. These features are then input into a multi-level memory pool, which includes at least short-term memory units and long-term memory units, used to cache real-time flight track data and fuse historical airspace events and typical traffic pattern features, respectively. Based on the inference engine and combined with the scene semantic type of the current target area, model parameters suitable for the current scene are generated. Under the control of the inference engine, the output features of the multi-level memory pool are fused for forward inference calculation. The preliminary results obtained from the forward inference calculation are matched and verified against preset aviation rules, and outputs that do not meet the safety constraints are corrected. Finally, the access probability and access intensity of each target grid within a future preset time period, after correction and compliance with the aviation rule safety constraints, are output.

[0087] More specifically, in the above embodiments, responding to the location display request of the airborne equipment and determining the target area to be displayed, and the scene semantic type corresponding to the target area, includes:

[0088] The system receives a display request from an airborne device, parses the latitude and longitude range, altitude layer parameters, and display scale contained in the request, determines the three-dimensional spatial boundary based on the latitude and longitude range and altitude layer parameters, the three-dimensional spatial boundary including the horizontal range boundary and the vertical altitude boundary, and determines the target area based on the three-dimensional spatial boundary and the display scale.

[0089] More specifically, in the above embodiments, the method for obtaining the scene semantic type includes:

[0090] A preset geographic information database is invoked to extract the distribution features of ground features and airspace attributes within the target area; based on the distribution features of ground features and airspace attributes, the scene semantic type corresponding to the target area is determined, including urban airspace type, airway corridor type, and special control area type.

[0091] More specifically, the step of constructing a multi-objective collaborative optimization function based on the scene semantic type, and obtaining the optimal mesh partitioning parameters by jointly solving the multi-objective collaborative optimization function through multiple constraints, includes:

[0092] Based on the semantic type of the scenario, optimization objectives and constraints associated with it are selected from the predefined set of optimization objectives and constraint sets; based on the selected optimization objectives and constraints, a multi-objective collaborative optimization function with grid partitioning parameters as decision variables is constructed; the function is solved using a multi-constraint optimization algorithm, and the optimal grid partitioning parameters that satisfy all constraints are output.

[0093] More specifically, in the above embodiments, the step of dividing the target region into multiple target grids according to the optimal grid division parameters, and constructing a spatiotemporal relationship diagram of the multiple target grids in conjunction with the map information of the target region, includes:

[0094] According to the dynamic grid partitioning strategy, the target area is decomposed into multiple non-overlapping target grid sets; historical trajectory data of airborne equipment is acquired, and the access frequency of each target grid in different time periods is statistically analyzed; a spatial adjacency matrix between grid units is established, and periodic traffic pattern data is integrated to generate a spatiotemporal relationship graph structure with spatiotemporal weights.

[0095] More specifically, in the above embodiments, the step of sequentially loading map information of each target grid into the airborne device according to the access probability and access intensity includes:

[0096] Based on a preset weighted fusion function, the access probability and access intensity of the same target grid are linearly weighted to determine the priority score corresponding to each target grid.

[0097] Determine the airspace control requirements for the target area and adjust the priority scores of the target grids in the airspace control areas within the target area;

[0098] The map loading sequence is obtained by arranging the target grids according to their priority scores.

[0099] According to the map loading sequence, the map information for each target grid is loaded.

[0100] More specifically, loading map information for each target grid according to the map loading sequence includes:

[0101] The map loading sequence is analyzed, and the multiple target grids are divided into multiple loading batches, with each loading batch including at least one target grid.

[0102] Based on the hardware performance of the airborne equipment, a loading rate and rendering strategy are assigned to the target meshes of different loading batches one by one;

[0103] Based on the loading rate and rendering strategy, map information of each target grid is presented on the display interface of the airborne device.

[0104] More specifically, in the above embodiments, the training process of the spatiotemporal graph neural network model includes:

[0105] Historical aeronautical data from multiple regions is collected, including grid access records of airspace at different scales;

[0106] The historical aeronautical data is standardized to eliminate differences in regional scale and grid density to obtain a training sample set. The standardization process is as follows: the physical dimensions of the grids in different airspaces are converted into dimensionless values ​​in a uniform scale coordinate system, and the grid access frequency is graded and normalized according to airspace type.

[0107] Based on the training sample set, the model is iteratively trained using the mini-batch gradient descent algorithm. In each round of training, the loss function value on the validation set is calculated, and the learning rate of the model is dynamically adjusted according to the loss function value until the preset training termination condition is met.

[0108] The following is for reference. Figure 2 It shows a schematic diagram of the structure of a computer system suitable for implementing the methods and system embodiments of this application. Figure 2 The server shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0109] like Figure 2As shown, the computer system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes based on programs stored in Read Only Memory (ROM) 302 or programs loaded from storage section 308 into Random Access Memory (RAM) 303. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0110] The following components are connected to the input / output interface 305: an input section 306 including a keyboard, mouse, etc.; an output section 307 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 303 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the input / output interface 305 as needed. A removable medium 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 310 as needed so that computer programs read from it can be installed into the storage section 308 as needed.

[0111] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by central processing unit 301, it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof.

[0112] More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0113] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0114] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0115] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.

[0116] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.

[0117] The technical solution of the present invention has now been described in conjunction with the preferred embodiments shown in the accompanying drawings.

[0118] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for displaying location information based on a map, characterized in that, The method includes: In response to the location display request from the airborne equipment, the target area to be displayed and the scene semantic type corresponding to the target area are determined. The scene semantic type includes urban airspace type, airway corridor type and special control area type. Based on the scene semantic type, a multi-objective collaborative optimization function is constructed, and the optimal grid partitioning parameters are obtained by solving the optimization function. The constraint objectives of the optimization function include a complexity minimization objective and a redundancy minimization objective. The complexity minimization objective is used to ensure that the computational load of the airborne equipment is below a preset threshold; the redundancy minimization objective is used to minimize the duplicate caching of map information between adjacent grids. Obtaining the optimal grid partitioning parameters includes: selecting an associated optimization objective and constraint condition from a predefined set of optimization objectives and constraint conditions based on the scene semantic type; constructing a multi-objective collaborative optimization function with grid partitioning parameters as decision variables based on the selected optimization objectives and constraints; and solving the function using a multi-constraint optimization algorithm to output the optimal grid partitioning parameters that satisfy all constraints. Based on the optimal grid division parameters, the target area is divided into multiple target grids, and a spatiotemporal relationship diagram of the multiple target grids is constructed by combining the map information of the target area. Based on the spatiotemporal relationship graph, a pre-trained spatiotemporal graph neural network model is used to predict the access probability and access intensity of each target grid in a future preset time period. The spatiotemporal graph neural network model is a memory-enhanced spatiotemporal metagraph network. The memory-enhanced spatiotemporal metagraph network has a built-in inference engine and a multi-level memory pool. The inference engine is used to dynamically generate model parameters according to the scene semantic type of the target area. The multi-level memory pool is used to fuse real-time trajectory and historical event features, and to impose constraints based on preset aviation rules, and output prediction results that meet the constraints. Based on the access probability and access intensity, map information for each target grid is sequentially loaded into the airborne device.

2. The map-based location information display method according to claim 1, characterized in that, The response to the location display request from the airborne equipment determines the target area to be displayed and the scene semantic type corresponding to the target area, including: Receive display requests sent by airborne equipment, and parse the latitude and longitude range, altitude layer parameters, and display scale contained in the requests; The three-dimensional spatial boundary is determined based on the latitude and longitude range and the altitude layer parameters. The three-dimensional spatial boundary includes a horizontal range boundary and a vertical altitude boundary. The target area is determined based on the three-dimensional spatial boundary and the display scale.

3. The map-based location information display method according to claim 2, characterized in that, The method for obtaining the scene semantic type includes: Call the preset geographic information database to extract the distribution characteristics of ground features and spatial attributes within the target area; Based on the distribution characteristics of the ground features and the spatial attributes, the scene semantic type corresponding to the target area is determined.

4. The map-based location information display method according to claim 1, characterized in that, The step of dividing the target region into multiple target grids according to the optimal grid division parameters, and constructing a spatiotemporal relationship diagram of the multiple target grids in conjunction with the map information of the target region, includes: According to the optimal grid partitioning parameters, the target region is decomposed into multiple non-overlapping target grid sets; Acquire historical trajectory data of airborne equipment and count the access frequency of each target grid at different time periods; Establish a spatial adjacency matrix between grid cells and integrate periodic flow pattern data to generate a spatiotemporal relationship graph structure with spatiotemporal weights.

5. The map-based location information display method according to claim 1, characterized in that, The method of using a pre-trained spatiotemporal graph neural network model to predict the access probability and access intensity of each target grid within a preset future time period includes: Based on the constructed spatiotemporal relationship graph, the historical access sequence features of the target grid corresponding to each node in the graph and the spatial association features between nodes are extracted. The historical access sequence features and spatial association features are input into the multi-level memory pool, which includes at least a short-term memory unit and a long-term memory unit, respectively used to cache real-time track data and fuse historical airspace events and typical traffic pattern features. Based on the inference engine, combined with the scene semantic type of the current target region, model parameters suitable for the current scene are generated, and under the control of the inference engine, the output features of the multi-level memory pool are fused to perform forward inference calculation. The preliminary results obtained from forward inference calculation are matched and verified with preset aviation rules, and outputs that do not conform to the preset aviation rules are corrected. The output is the access probability and access intensity of each target grid within a future preset time period, after correction to conform to the preset aviation rules.

6. The map-based location information display method according to claim 1, characterized in that, The step of sequentially loading map information of each target grid into the airborne device based on the access probability and access intensity includes: Based on a preset weighted fusion function, the access probability and access intensity of the same target grid are linearly weighted to determine the priority score corresponding to each target grid. Determine the airspace control requirements for the target area and adjust the priority scores of the target grids in the airspace control areas within the target area; The map loading sequence is obtained by arranging the target grids according to their priority scores. According to the map loading sequence, the map information for each target grid is loaded.

7. The map-based location information display method according to claim 6, characterized in that, The step of loading map information for each target grid according to the map loading sequence includes: The map loading sequence is analyzed, and the multiple target grids are divided into multiple loading batches, with each loading batch including at least one target grid. Based on the hardware performance of the airborne equipment, a loading rate and rendering strategy are assigned to the target meshes of different loading batches one by one; Based on the loading rate and rendering strategy, map information of each target grid is presented on the display interface of the airborne device.

8. The map-based location information display method according to claim 1, characterized in that, The training process of the spatiotemporal graph neural network model includes: Historical aeronautical data from multiple regions is collected, including grid access records of airspace at different scales; The historical aeronautical data is standardized to eliminate differences in regional scale and grid density to obtain a training sample set. The standardization process is as follows: the physical dimensions of the grids in different airspaces are converted into dimensionless values ​​in a uniform scale coordinate system, and the grid access frequency is graded and normalized according to airspace type. Based on the training sample set, the model is iteratively trained using the mini-batch gradient descent algorithm. In each round of training, the loss function value on the validation set is calculated, and the learning rate of the model is dynamically adjusted according to the loss function value until the preset training termination condition is met.

9. A map-based location information display system, applied to the method described in any one of claims 1-8, characterized in that, The system includes: The data response module is configured to respond to the location display request of the airborne equipment, determine the target area to be displayed, and the scene semantic type corresponding to the target area; The strategy calculation module is configured to construct a multi-objective collaborative optimization function based on the scene semantic type, and obtain the optimal grid partitioning parameters by solving the optimization function. The constraint objectives of the optimization function include a complexity minimization objective and a redundancy minimization objective. The complexity minimization objective is used to ensure that the computational load of the airborne equipment is lower than a preset threshold; the redundancy minimization objective is used to minimize the duplicate caching of map information between adjacent grids. The strategy execution module is configured to divide the target area into multiple target grids according to the optimal grid division parameters, and to construct a spatiotemporal relationship diagram of the multiple target grids in combination with the map information of the target area; The model prediction module is configured to predict the access probability and access intensity of each target grid within a preset time period based on the spatiotemporal relationship graph using a pre-trained spatiotemporal graph neural network model. The spatiotemporal graph neural network model is a memory-enhanced spatiotemporal metagraph network. The memory-enhanced spatiotemporal metagraph network has a built-in inference engine and a multi-level memory pool. The inference engine is used to dynamically generate model parameters according to the scene semantic type of the target area. The multi-level memory pool is used to fuse real-time flight track and historical event features, and to impose constraints based on preset aviation rules, and output prediction results that conform to the constraints. The data loading module is configured to sequentially load map information of each target grid into the airborne device according to the access probability and access intensity.

Images