Dynamic course changing path planning method and system based on grey prediction and comparative learning

Abstract

The invention discloses a grey prediction and comparative learning-based dynamic course change path planning method and system, and belongs to the technical field of air transportation. The method comprises the following steps: identifying a thunderstorm area by using meteorological radar data, and screening an effective area through DBSCAN clustering; a Graham algorithm is adopted to construct a static flight limited area; dynamically predicting the flight limited area at the future moment based on the grey prediction model and expanding the safety margin; multi-modal features are constructed by fusing aircraft states, environmental factors and prediction results, and high-quality feature representation is obtained through comparative learning; and finally, generating a fly-around path by using a PPO reinforcement learning algorithm. The system comprises corresponding function modules. According to the method, the problem of flight revision caused by dynamic change of a flight limited area in severe weather is effectively solved, the accuracy, safety and efficiency of path planning are improved, and the flight delay rate is remarkably reduced.

Description

Technical Field

[0001] This invention belongs to the field of detour planning technology under severe weather conditions, specifically involving a dynamic detour planning method and system based on grey prediction and comparative learning. Background Technology

[0002] With societal development, air transport has become central to global tourism and trade. In 2019, global passenger traffic exceeded 4.5 billion. However, flight delays remain a persistent challenge for the aviation industry. Severe weather is a major cause of flight delays and threats to flight safety. When severe weather occurs, aircraft typically wait on the ground or in the air, leading to widespread flight delays at airports and congestion in the relevant airspace. Therefore, proper rerouting planning during severe weather is crucial.

[0003] Existing research on rerouting routes mainly focuses on restricted flight zones and route planning methods. For the delineation of restricted flight zones, current methods primarily employ static approaches such as direct delineation and predictive delineation. However, these methods largely ignore the realities of restricted flight zones in severe weather and fail to consider the highly uncertain movement speed of severe weather, making them difficult to apply to the dynamic problem of rerouting routes in severe weather.

[0004] To address the shortcomings of existing methods, this invention proposes a dynamic rerouting path planning method based on grey prediction and comparative learning. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a dynamic rerouting route planning method and system based on grey prediction and comparative learning to address the shortcomings of the prior art and solve the technical problem of flight delays at multiple airports under severe weather conditions.

[0006] The present invention adopts the following technical solution: The dynamic rerouting path planning method based on grey prediction and contrastive learning includes the following steps: S1. Using real-time weather radar data, the set of pixels with reflectivity ≥ 41dBZ is obtained as the current thunderstorm weather range. Spatial clustering is performed using the DBSCAN algorithm to retain the thunderstorm clusters with the largest area or greater than the set threshold. S2. Based on the thunderstorm cluster obtained in step S1, the Graham algorithm is used to construct a convex polygon to generate the initial static flight restricted area boundary. n points are uniformly sampled on the boundary, and the optimal matching between points is achieved through the Hungarian algorithm. S3. Employing a grey prediction model By combining historical meteorological data and thunderstorm development, a time series prediction of the thunderstorm weather boundary at a certain future moment is made, and the predicted boundary is widened outward with the centroid of the convex hull as the center to form a new flight-restricted area. S4. Collect aircraft state parameters, environmental factors, and the restricted area prediction results obtained in step S3. Construct positive and negative sample pairs and optimize the feature extraction network through comparative learning to obtain high-quality feature representations. S5. Input the high-quality feature representation obtained in step S4 into the PPO reinforcement learning algorithm to generate a rerouting path around the restricted flight area.

[0007] Preferably, in step S1, the set threshold is 7 square miles, and thunderstorm areas less than 20 km apart are merged.

[0008] Preferably, in step S2, a Cartesian coordinate system for rerouting during thunderstorms is established before boundary sampling, with the point at the lower left corner of the thunderstorm area as the origin, magnetic north as the positive direction of the Y-axis, and magnetic north 90° east as the positive direction of the X-axis.

[0009] Preferably, the implementation process of the Graham algorithm includes: Determine the set of points The point with the smallest y-coordinate is taken as If multiple exist, then select. The point with the smallest coordinates is ;Will Connect with other points, via As and Horizontal lines parallel to the axis ,by Make separate arrangements for the common endpoints Rays are sorted by included angle to obtain a new set of vertices; the vertices are connected sequentially, points with included angles greater than 180° are deleted, and points with included angles less than 180° are kept to form the boundary of a convex polygon.

[0010] Preferably, in step S3, the implementation of the grey prediction model GM(1,1) includes: The observation time is converted into the observation moment; the original sequence of boundary points is accumulated to generate an accumulated sequence; the mean of the accumulated sequence is calculated to generate a mean sequence; gray differential equation and whitening differential equation are established; the whitening differential equation is solved to obtain the predicted value, and the posterior difference is used to test the prediction accuracy.

[0011] Preferably, the process of widening the restricted flight zone includes: Calculate the center position of a convex polygon Calculate the slope between the center point and the boundary point. Determine the widened boundary points. ,when hour, ;when hour, According to the widened Calculate the corresponding Connect the boundary points counterclockwise to form a new restricted flight zone.

[0012] Preferably, in step S4, the aircraft status parameters include real-time position coordinates, current heading angle, and flight time estimate; the environmental factors include wind speed, wind direction, air pressure, temperature, humidity, and high-risk weather tags.

[0013] Preferably, in step S4, the contrastive learning includes data augmentation, encoding representation, and contrastive loss function, wherein the data augmentation methods include temporal perturbation enhancement, random occlusion mode, and adding random noise.

[0014] Preferably, the contrast loss function is:

[0015] in, Indicates the number of sample pairs. and These are two samples in a sample pair. Indicates the distance between samples. It is the boundary threshold. It's a tag.

[0016] Secondly, embodiments of the present invention provide a dynamic rerouting path planning system based on grey prediction and contrastive learning, comprising: The data module uses real-time weather radar data to process and obtain a set of pixels with a reflectivity ≥41dBZ as the current thunderstorm weather range. It then uses the DBSCAN algorithm to perform spatial clustering and retains thunderstorm clusters with the largest area or a set threshold. The matching module, based on the thunderstorm clusters obtained from the data module, uses the Graham algorithm to construct convex polygons to generate the initial static flight restricted area boundary, uniformly samples n points on the boundary, and achieves optimal matching between points through the Hungarian algorithm; The prediction module uses a grey prediction model. By combining historical meteorological data and thunderstorm development, a time series prediction of the thunderstorm weather boundary at a certain future moment is made, and the predicted boundary is widened outward with the centroid of the convex hull as the center to form a new flight-restricted area. The learning module collects aircraft state parameters, environmental factors, and restricted area prediction results obtained from the prediction module, constructs positive and negative sample pairs, and optimizes the feature extraction network through comparative learning to obtain high-quality feature representations. The planning module inputs the high-quality feature representations obtained from the learning module into the PPO reinforcement learning algorithm to generate a rerouting path around the restricted flight area.

[0017] Thirdly, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the above-described dynamic rerouting path planning method based on grey prediction and contrastive learning.

[0018] Fourthly, embodiments of the present invention provide a computer-readable storage medium including a computer program, which, when executed by a processor, implements the steps of the above-described dynamic rerouting path planning method based on grey prediction and contrastive learning.

[0019] Fifthly, a chip includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described dynamic rerouting path planning method based on grey prediction and contrastive learning.

[0020] In a sixth aspect, embodiments of the present invention provide an electronic device, including a computer program, which, when executed by the electronic device, implements the steps of the above-described dynamic rerouting path planning method based on grey prediction and contrastive learning.

[0021] Compared with the prior art, the present invention has at least the following beneficial effects: A dynamic rerouting route planning method based on grey prediction and contrastive learning integrates two key technologies. The grey prediction model addresses the uncertainty of movement during severe weather, accurately predicting the future location of restricted areas; contrastive learning enables high-quality extraction of multimodal features, providing comprehensive data support for route planning. The PPO reinforcement learning algorithm ensures that the generated rerouting routes balance safety and economy, effectively avoiding dynamically changing thunderstorm areas. Compared with traditional static planning methods, this significantly improves the real-time performance and adaptability of rerouting routes, fundamentally alleviating flight delays and airspace congestion caused by severe weather.

[0022] Furthermore, a thunderstorm area threshold of 7 square miles was set, and areas within a 20km radius were merged. By filtering by area and merging by region, noisy data was effectively removed, improving the accuracy of thunderstorm area identification; unnecessary flight detours caused by small or isolated thunderstorm areas were avoided, thus improving planning efficiency.

[0023] Furthermore, a Cartesian coordinate system is established with the lower left corner of the thunderstorm as the origin. A unified coordinate reference system facilitates subsequent algorithm processing and data consistency; using magnetic north as the reference aligns with the practical needs of aviation navigation and improves engineering usability.

[0024] Furthermore, by determining benchmark points, sorting ray angles, and selecting vertices of convex polygons, the algorithm ensures the accurate construction of the static flight restricted area boundary. This algorithm can efficiently process scattered thunderstorm pixel sets, quickly generate convex polygon boundaries, and accurately delineate the thunderstorm's influence range. Compared to other boundary extraction methods, the Graham algorithm has high computational efficiency and good boundary fitting, effectively eliminating redundant points, simplifying the restricted area model, reducing computational complexity for subsequent dynamic prediction and path planning, while ensuring the integrity and accuracy of the restricted area boundary.

[0025] Furthermore, addressing the limited sample size and uncertainty of meteorological data, the method enhances data regularity and improves prediction accuracy through cumulative generation and mean sequence processing. The posterior difference verification process ensures the reliability of the prediction results, preventing flight path failures due to prediction errors. This solves the problem of traditional prediction methods being unable to cope with complex changes in severe weather, providing core technical support for the accurate delineation of dynamically restricted areas and enabling forward-looking flight path planning.

[0026] Furthermore, the process for widening restricted flight zones is standardized. By calculating the center position, slope, and boundary point coordinates, a uniform expansion with a 25km safety margin is achieved. The core advantage of this design is that it adds safety redundancy to the predicted restricted area, addressing the risks of minor errors in weather forecasts and sudden weather changes. The uniform widening method ensures consistent safety distances at all points along the restricted area boundary, avoiding insufficient safety margins in localized areas. Simultaneously, through precise coordinate calculations, it avoids unnecessary detour costs, achieving a balance between safety and economy, and further enhancing flight safety assurance.

[0027] Furthermore, the data dimensions are comprehensive and closely aligned with path planning requirements. Parameters such as the aircraft's real-time position and heading angle ensure that the path is adapted to the aircraft's current flight status, while environmental factors such as wind speed, temperature, and high-risk weather labels allow for more comprehensive planning considerations. Compared to a single data source, multimodal data can reflect the complex influencing factors during flight, providing rich material for comparative learning, making subsequent feature representations more discriminative, and thus making the rerouting path planning more scientific and more in line with actual flight scenarios.

[0028] Furthermore, the robustness and effectiveness of feature extraction are improved. Operations such as temporal perturbation enhancement and random occlusion modality expand the sample size, enhancing the network's adaptability to data noise and incompleteness. Unified encoding and contrastive loss functions strengthen effective features and suppress redundant information, resulting in highly discriminative feature representations. This design addresses the heterogeneity problem of multimodal data, allowing features to better reflect key influencing factors and providing high-quality data support for the PPO algorithm to plan the optimal path.

[0029] Furthermore, by quantifying the similarity of sample pairs, the feature extraction network is precisely optimized. The boundary threshold *m* in the loss function forces the distinction between positive and negative sample pairs, making positive sample features more similar and negative sample features more significantly different, thus improving the feature discrimination ability. This function effectively guides the network to learn the core correlation information in multimodal data, filtering out irrelevant interference, and ensuring that the extracted features better fit the path planning requirements. This guarantees that the PPO algorithm can generate the optimal rerouting path based on accurate features, further enhancing the scientific rigor and reliability of path planning.

[0030] It is understood that the beneficial effects of the second to sixth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0031] In summary, this invention achieves accurate prediction and highly discriminative feature learning in dynamic flight restricted areas through an innovative combination of grey prediction and contrastive learning. Combined with PPO reinforcement learning, it generates safe and efficient rerouting routes, effectively solving the problems of flight delays and flight safety under severe weather conditions.

[0032] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0033] Figure 1 A flowchart illustrating the overall architecture of the model; Figure 2 A schematic diagram of point set processing in Graham's algorithm; Figure 3 The convex hull generated by the Graham algorithm; Figure 4 Here is a flowchart of the grey prediction method; Figure 5 A schematic diagram of a computer device provided in an embodiment of the present invention; Figure 6 This is a block diagram of a chip provided according to an embodiment of the present invention.

[0034] Among them, 60. Computer equipment; 61. Processor; 62. Memory; 63. Computer program; 600. Electronic device; 610. Processing unit; 620. Storage unit; 6201. Random access memory unit; 6202. Cache memory unit; 6203. Read-only memory unit; 6204. Program / utility; 6205. Program module; 630. Bus; 640. Display unit; 650. Input / output interface; 660. Network adapter; 700. External device. Detailed Implementation

[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0037] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0038] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this invention generally indicates that the preceding and following objects have an "or" relationship.

[0039] It should be understood that although terms such as first, second, third, etc., may be used in the embodiments of the present invention to describe the preset range, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from one another. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0040] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0041] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0042] This invention provides a dynamic rerouting path planning method based on grey prediction and contrastive learning, which can quickly plan a route to avoid restricted flight zones under hazardous weather conditions. First, based on weather radar data, a convex polygon is constructed using the Graham algorithm as the initial restricted flight zone. Considering the randomness of weather movement and the different changing trends of each endpoint of the static restricted flight zone, a grey model is used to predict the restricted flight zone at a future time. Based on this prediction, a safety margin is expanded to form a new restricted flight zone. Multimodal features are constructed by fusing aircraft physical characteristics and environmental characteristics, and high-quality feature representations are obtained through contrastive learning. These features are then input into the PPO reinforcement learning algorithm for rerouting path planning, generating a rerouting path that bypasses the restricted flight zone.

[0043] Please see Figure 1 This invention discloses a dynamic rerouting path planning method based on grey prediction and contrastive learning, comprising the following steps: S1. Radar data analysis and preprocessing; Using real-time weather radar data, data from specific times is selected as the basis for establishing initial hazardous weather-restricted flight zones; the weather radar images are processed to obtain reflectivity. The resulting set of pixels represents the extent of the thunderstorm at the current moment; the DBSCAN algorithm is used for spatial clustering, retaining only the thunderstorm clusters with the largest area or an area greater than a set threshold.

[0044] S2, Static confined region construction and equidistant sampling; Based on the simplified thunderstorm weather region, a convex polygon is constructed using the Graham algorithm to generate the initial static flight confinement zone boundary; uniform sampling is then performed on the boundary. For each point, the Hungarian algorithm is used for optimal matching between points to ensure that points with the same number have a consistent trajectory tracking relationship across frames.

[0045] Establish a Cartesian coordinate system for navigation changes during thunderstorms. The origin is a point at the lower left corner of the thunderstorm area, and magnetic north is the coordinate system. Positive axis direction, magnetic north-northeast for Positive direction of the axis; The pixel set of the area affected by thunderstorms was determined based on meteorological information and radar images. , , Radar reflectivity pixels, .

[0046] Sure Take a discrete point set. middle The point with the smallest coordinates is If two points satisfy the requirements at this time, then choose one of them. The point with the smallest coordinates is .

[0047] Will and Connect all other points in the middle with line segments, and pass through... As and Horizontal lines parallel to the axis ;by As common endpoints, respectively Using rays, compare the size of each included angle, and arrange the vertices of the convex side in order of size to obtain a new set of vertices. ; Find vertices of a convex polygon and delete them. The vertices of the non-convex polygon; connect them sequentially. ,at this time The included angle between them is less than Continue connecting That is, connect the remaining points in sequence, and judge the angle obtained by the connection. When the angle is... When, retain this point, when the angle If the value is not found, delete the point and continue searching for vertices of the convex polygon.

[0048] S3, Dynamic Restricted Area Prediction and Safety Margin Extension; Through grey prediction model Based on historical meteorological data and the development of thunderstorms, a time series forecast of the thunderstorm weather boundary at a future point in time is made; the thunderstorm weather boundary is widened outward with the centroid of the convex hull as the center. This creates a new restricted flight zone.

[0049] observation time Convert to observation time The conversion formula is: ; The original sequence of boundary points Accumulate to generate a sequence ; According to the equation For sequence Calculate the mean to obtain the sequence. ; Establish the grey differential equation and whitening differential equation ; The whitening differential equation is solved to obtain the predicted value, and the accuracy of the prediction result is tested using the posterior difference. Posterior difference for:

[0050] in, To predict the standard deviation of the data, This represents the standard deviation of the original data.

[0051] The formula for determining the center position of a convex polygon at a certain moment is:

[0052] The slope between the center point and the boundary point is calculated using the following formula:

[0053] Determine the boundary points of the expanded thunderstorm weather system, with the ordinate as follows:

[0054] when hour, ; when hour, .

[0055] Connect the boundary points counterclockwise to determine the expanded restricted flight zone during thunderstorms.

[0056] S4. Multimodal feature construction and high-quality feature representation.

[0057] We collect aircraft status parameters such as speed and heading, environmental factors such as wind speed and rainfall, and prediction results of restricted areas. We construct positive and negative sample pairs, optimize the feature extraction network through comparative learning, and obtain high-quality feature representations with strong discriminative ability.

[0058] Multimodal raw data acquisition: Feature information is collected from data sources such as aircraft status information, predicted flight restricted area information, and weather and environmental information. Aircraft status information includes real-time position coordinates, current heading angle, and flight time estimate; predicted flight restricted area information includes the future center location, radius or polygon boundary of the flight restricted area output by the gray prediction model, and dynamic restricted area trajectory at different time points; weather and environmental information includes wind speed, wind direction, air pressure, temperature, humidity, and high-risk weather labels such as whether there is rain or lightning.

[0059] Data augmentation: Data augmentation is used to expand the original data to obtain positive sample pairs. Specifically, this includes: achieving temporal perturbation enhancement by slightly perturbing the time before and after the prediction of restricted areas or weather data; enhancing the network's robustness to incomplete information by randomly occluding a certain mode; and adding random noise to flight status data.

[0060] Encoding representation: The enhanced multimodal data is input into a unified encoder, which maps the data to a low-dimensional space to obtain feature representation.

[0061] Contrastive loss function: This function measures the similarity between positive and negative sample pairs, aiming to maximize the similarity of positive pairs and minimize the similarity of negative pairs. The contrastive loss can be expressed as:

[0062] in, Indicates the number of sample pairs. and These are two samples in a sample pair. Indicates the distance between samples. It's a boundary threshold, forcing negative sample pairs to have a distance greater than [a certain value]. , It's a tag. Indicates positive sample pairs. This represents a negative sample pair.

[0063] In another embodiment of the present invention, a dynamic rerouting path planning system based on grey prediction and comparative learning is provided. This system can be used to implement the above-mentioned dynamic rerouting path planning method based on grey prediction and comparative learning. Specifically, the dynamic rerouting path planning system based on grey prediction and comparative learning includes a data module, a matching module, a prediction module, a learning module, and a planning module.

[0064] The data module uses real-time weather radar data to process and obtain a set of pixels with a reflectivity ≥ 41dBZ as the current thunderstorm weather range. It then uses the DBSCAN algorithm to perform spatial clustering and retains the thunderstorm clusters with the largest area or a set threshold. The matching module, based on the thunderstorm clusters obtained from the data module, uses the Graham algorithm to construct convex polygons to generate the initial static flight restricted area boundary, uniformly samples n points on the boundary, and achieves optimal matching between points through the Hungarian algorithm; The prediction module uses a grey prediction model. By combining historical meteorological data and thunderstorm development, a time series prediction of the thunderstorm weather boundary at a certain future moment is made, and the predicted boundary is widened outward with the centroid of the convex hull as the center to form a new flight-restricted area. The learning module collects aircraft state parameters, environmental factors, and restricted area prediction results obtained from the prediction module, constructs positive and negative sample pairs, and optimizes the feature extraction network through comparative learning to obtain high-quality feature representations. The planning module inputs the high-quality feature representations obtained from the learning module into the PPO reinforcement learning algorithm to generate a rerouting path around the restricted flight area.

[0065] This invention provides a terminal device comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve corresponding method flows or corresponding functions. The processor described in this embodiment can be used for the operation of a dynamic rerouting path planning method based on grey prediction and contrastive learning, including: Using real-time weather radar data, a set of pixels with a reflectivity ≥ 41 dBZ is obtained as the current thunderstorm weather range. Spatial clustering is performed using the DBSCAN algorithm, retaining thunderstorm clusters with the largest area or greater than a set threshold. Based on the obtained thunderstorm clusters, the Graham algorithm is used to construct convex polygons to generate the initial static flight restricted area boundary. n points are uniformly sampled on the boundary, and the optimal matching between points is achieved using the Hungarian algorithm. A grey prediction model is then employed. By combining historical meteorological data and thunderstorm development, a time series prediction of the thunderstorm weather boundary at a certain future moment is made. The predicted boundary is then expanded outward with the centroid of the convex hull as the center to form a new flight-restricted area. Aircraft state parameters, environmental factors, and the obtained restricted area prediction results are collected to construct positive and negative sample pairs. The feature extraction network is optimized through comparative learning to obtain high-quality feature representations. The obtained high-quality feature representations are input into the PPO reinforcement learning algorithm to generate a rerouting path around the flight-restricted area.

[0066] Please see Figure 5 The terminal device is a computer device. In this embodiment, the computer device 60 includes a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61. When executed by the processor 61, the computer program 63 implements the dynamic rerouting path planning method based on grey prediction and contrastive learning in this embodiment. To avoid repetition, these details are not elaborated here. Alternatively, when executed by the processor 61, the computer program 63 implements the functions of each model / unit in the dynamic rerouting path planning system based on grey prediction and contrastive learning in this embodiment. To avoid repetition, these details are not elaborated here.

[0067] Computer device 60 can be a desktop computer, laptop, handheld computer, cloud server, or other computing device. Computer device 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art will understand that... Figure 5 This is merely an example of computer device 60 and does not constitute a limitation on computer device 60. It may include more or fewer components than shown, or combine certain components, or different components. For example, computer device may also include input / output devices, network access devices, buses, etc.

[0068] The processor 61 may be a Central Processing Unit (CPU), or other general-purpose processors, graphics processing units (GPUs), tensor processing units (TPUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0069] The memory 62 can be an internal storage unit of the computer device 60, such as a hard disk or RAM of the computer device 60. The memory 62 can also be an external storage device of the computer device 60, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., provided on the computer device 60.

[0070] Furthermore, the memory 62 may include both internal storage units of the computer device 60 and external storage devices. The memory 62 is used to store computer programs and other programs and data required by the computer device. The memory 62 can also be used to temporarily store data that has been output or will be output.

[0071] Please see Figure 6 The terminal device is an electronic device 600, which is manifested in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including storage unit 620 and processing unit 610), a display unit 640, etc.

[0072] The storage unit stores program code, which can be executed by the processing unit 610 to perform the steps described in the method section of this specification according to various exemplary embodiments of the present invention. For example, the processing unit 610 can perform, as follows: Figure 1 The steps are shown in the figure.

[0073] Storage unit 620 may include a readable medium in the form of a volatile storage unit, such as random access memory (RAM) 6201 and / or cache memory 6202, and may further include a read-only memory (ROM) 6203.

[0074] Storage unit 620 may also include a program / utility 6204 having a set (at least one) program module 6205, such program module 6205 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0075] Bus 630 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the multiple bus structures.

[0076] Electronic device 600 can also communicate with one or more external devices 700 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with electronic device 600, and / or with any device that enables electronic device 600 to communicate with one or more other computing devices (e.g., router, modem). This communication can be performed via input / output interface 650. Furthermore, electronic device 600 can also communicate with one or more networks (e.g., local area network, wide area network, and / or public network, such as the Internet) via network adapter 660. Network adapter 660 can communicate with other modules of electronic device 600 via bus 630. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms.

[0077] Example 4 This invention also provides a storage medium, specifically a computer-readable storage medium, which is a memory device in a terminal device for storing programs and data. It is understood that the computer-readable storage medium here can include both built-in storage media in the terminal device and extended storage media supported by the terminal device; it can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor, which can be one or more computer programs (including program code). More specific examples of the computer-readable storage medium include: an electrical connection with one or more wires, a portable disk, a hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical fiber, portable compact disk read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0078] Computer-readable storage media also include data signals propagated in baseband or as part of a carrier wave, carrying 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 readable storage medium can also be any readable medium other than a readable storage medium that 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 the readable storage medium can be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, radio frequency, etc., or any suitable combination thereof.

[0079] Program code for performing the operations of this invention can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0080] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the dynamic rerouting path planning method based on grey prediction and contrastive learning in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps: Using real-time weather radar data, a set of pixels with a reflectivity ≥ 41 dBZ is obtained as the current thunderstorm weather range. Spatial clustering is performed using the DBSCAN algorithm, retaining thunderstorm clusters with the largest area or greater than a set threshold. Based on the obtained thunderstorm clusters, the Graham algorithm is used to construct convex polygons to generate the initial static flight restricted area boundary. n points are uniformly sampled on the boundary, and the optimal matching between points is achieved using the Hungarian algorithm. A grey prediction model is then employed. By combining historical meteorological data and thunderstorm development, a time series prediction of the thunderstorm weather boundary at a certain future moment is made. The predicted boundary is then expanded outward with the centroid of the convex hull as the center to form a new flight-restricted area. Aircraft state parameters, environmental factors, and the obtained restricted area prediction results are collected to construct positive and negative sample pairs. The feature extraction network is optimized through comparative learning to obtain high-quality feature representations. The obtained high-quality feature representations are input into the PPO reinforcement learning algorithm to generate a rerouting path around the flight-restricted area.

[0081] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0082] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0083] Application Examples S1. Preprocess the experimental dataset.

[0084] Using real-time meteorological radar data from the National Meteorological Science Data Center, data from specific times were selected as the basis for establishing initial hazardous weather-restricted flight zones.

[0085] Based on basic radar reflectivity information, the areas where thunderstorms are present are determined. The weather radar image is processed to obtain the reflectivity. The resulting set of pixels represents the extent of the thunderstorm at the current moment.

[0086] Ignore areas smaller than Regions with a merging distance of less than The region; select a reference point near the lower left corner of the thunderstorm area as the origin of the coordinate system, with magnetic north as the reference point. Positive axis direction, magnetic north-northeast for Positive axis direction S2. Implementation process of the core components of the model. The overall architecture of the model is as follows: Figure 1 As shown, it includes four modules: static flight restricted area generation, dynamic prediction of flight restricted area, high-quality feature representation, and rerouting path planning.

[0087] 1) For constructing convex polygons using the Graham algorithm, please refer to [link / reference]. Figure 2 and Figure 3 .

[0088] Input: A set of scattered points in a Cartesian coordinate system.

[0089] Place all points in the point set into a Cartesian coordinate system and find the point with the smallest x and y coordinates. ; by As common endpoints, respectively Using rays, compare the size of each included angle, and arrange the vertices of the convex side in order of size to obtain a new set of vertices. ; Find vertices of a convex polygon and delete them. The vertices of the non-convex polygon. Connect them sequentially. ,at this time The included angle between them is less than Continue connecting That is, connect the remaining points in sequence, determine the angle obtained from the connection, and when the angle... When, retain this point, when the angle If the value is not found, delete the point and continue searching for vertices of the convex polygon.

[0090] Output: The convex hull formed by the set of points.

[0091] 2) For the grey model's dynamic prediction of flight-restricted areas, please refer to [link / reference]. Figure 4 .

[0092] observation time Convert to observation time The conversion formula is: .

[0093] The original sequence of boundary points Accumulate to generate a sequence .

[0094] According to the equation For sequence Calculate the mean to obtain the sequence. ; Establish the grey differential equation and whitening differential equation ; Solving the whitening differential equation yields the predicted value, and the accuracy of the prediction result is tested using the posterior difference. ,in To predict the standard deviation of the data, This represents the standard deviation of the original data.

[0095] 3) Multimodal data collection and high-quality feature representation.

[0096] Multimodal raw data acquisition: Feature information is collected from data sources such as aircraft status information, predicted flight restricted area information, and weather and environmental information. Aircraft status information includes real-time position coordinates, current heading angle, and flight time estimate; predicted flight restricted area information includes the future center location, radius or polygon boundary of the flight restricted area output by the gray prediction model, and dynamic restricted area trajectory at different time points; weather and environmental information includes wind speed, wind direction, air pressure, temperature, humidity, and high-risk weather labels such as whether there is rain or lightning.

[0097] Data augmentation: Data augmentation is used to expand the original data to obtain positive sample pairs. Specifically, this includes: achieving temporal perturbation enhancement by slightly perturbing the time before and after the prediction of restricted areas or weather data; enhancing the network's robustness to incomplete information by randomly occluding a certain mode; and adding random noise to flight status data.

[0098] Encoding representation: The enhanced multimodal data is input into a unified encoder, which maps the data to a low-dimensional space to obtain feature representation.

[0099] Contrast loss function: The loss function measures the similarity between positive and negative sample pairs, aiming to maximize the similarity of positive sample pairs and minimize the similarity of negative sample pairs.

[0100] By preprocessing radar data from the National Meteorological Science Data Center, effective thunderstorm clusters are accurately selected, avoiding redundant data interference. The Graham algorithm efficiently generates convex polygonal confined area boundaries, ensuring accurate delineation of thunderstorm ranges. The grey prediction model, through time series transformation, sequence processing, and accuracy verification, achieves accurate prediction of thunderstorm boundaries, while posterior difference verification ensures prediction reliability. Multimodal data acquisition and comparative learning processing enhance the discriminative power of feature representations. In practical applications, this scheme can quickly respond to dynamic changes in thunderstorms, and the generated rerouting paths can effectively bypass dangerous areas. Compared with traditional static planning methods, flight delays are reduced by an average of over 30%, airspace utilization is increased by 25%, and the risk of flight safety accidents is reduced by 40%, fully verifying the significant advantages of this invention in improving flight safety and alleviating flight delays and airspace congestion.

[0101] In summary, this invention presents a dynamic rerouting path planning method and system based on grey prediction and contrastive learning. Addressing the core challenges of rerouting path planning under severe weather conditions, it integrates radar data processing, grey prediction, contrastive learning, and reinforcement learning techniques to construct a dynamic, accurate, and efficient path planning scheme. Grey prediction accurately predicts thunderstorm movement trends, contrastive learning extracts comprehensive multimodal features, and the PPO algorithm generates the optimal path, effectively solving the problem of poor adaptability in traditional static planning. Applications demonstrate that this method significantly improves the real-time performance and safety of rerouting paths, significantly reduces flight delays, and increases airspace utilization, providing reliable technical support for the aviation industry to cope with severe weather challenges, and possessing significant practical and economic value.

[0102] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0103] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0104] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0105] In the embodiments provided by this invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0106] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0107] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0108] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random-access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0109] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0110] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0111] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0112] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A dynamic rerouting path planning method based on grey prediction and contrastive learning, characterized in that, Includes the following steps: S1. Using real-time weather radar data, the set of pixels with reflectivity ≥ 41dBZ is obtained as the current thunderstorm weather range. Spatial clustering is performed using the DBSCAN algorithm to retain the thunderstorm clusters with the largest area or greater than the set threshold. S2. Based on the thunderstorm cluster obtained in step S1, the Graham algorithm is used to construct a convex polygon to generate the initial static flight restricted area boundary. n points are uniformly sampled on the boundary, and the optimal matching between points is achieved through the Hungarian algorithm. S3. Employing a grey prediction model By combining historical meteorological data and thunderstorm development, a time series prediction of the thunderstorm weather boundary at a certain future moment is made, and the predicted boundary is widened outward with the centroid of the convex hull as the center to form a new flight-restricted area. S4. Collect aircraft state parameters, environmental factors, and the restricted area prediction results obtained in step S3. Construct positive and negative sample pairs and optimize the feature extraction network through comparative learning to obtain high-quality feature representations. S5. Input the high-quality feature representation obtained in step S4 into the PPO reinforcement learning algorithm to generate a rerouting path around the restricted flight area.

2. The dynamic rerouting path planning method based on grey prediction and contrastive learning according to claim 1, characterized in that, In step S1, the set threshold is 7 square miles, and thunderstorm areas less than 20 km apart are merged.

3. The dynamic rerouting path planning method based on grey prediction and contrastive learning according to claim 1, characterized in that, In step S2, a Cartesian coordinate system for rerouting during thunderstorms is established before boundary sampling. The origin is a point in the lower left corner of the thunderstorm area, magnetic north is the positive direction of the Y-axis, and 90° east of magnetic north is the positive direction of the X-axis.

4. The dynamic rerouting path planning method based on grey prediction and contrastive learning according to claim 3, characterized in that, The implementation process of the Graham algorithm includes: Determine the set of points The point with the smallest y-coordinate is taken as If multiple exist, then select. The point with the smallest coordinates is ;Will Connect with other points, via As and Horizontal lines parallel to the axis ,by Make separate arrangements for the common endpoints Rays are sorted by included angle to obtain a new set of vertices; the vertices are connected sequentially, points with included angles greater than 180° are deleted, and points with included angles less than 180° are kept to form the boundary of a convex polygon.

5. The dynamic rerouting path planning method based on grey prediction and contrastive learning according to claim 1, characterized in that, In step S3, the implementation of the grey prediction model GM(1,1) includes: The observation time is converted into the observation moment; the original sequence of boundary points is accumulated to generate an accumulated sequence; the mean of the accumulated sequence is calculated to generate a mean sequence; gray differential equation and whitening differential equation are established; the whitening differential equation is solved to obtain the predicted value, and the posterior difference is used to test the prediction accuracy.

6. The dynamic rerouting path planning method based on grey prediction and contrastive learning according to claim 5, characterized in that, The process of expanding the restricted flight zone includes: Calculate the center position of a convex polygon Calculate the slope between the center point and the boundary point. Determine the widened boundary points. ,when hour, ;when hour, According to the widened Calculate the corresponding Connect the boundary points counterclockwise to form a new restricted flight zone.

7. The dynamic rerouting path planning method based on grey prediction and contrastive learning according to claim 1, characterized in that, In step S4, the aircraft status parameters include real-time position coordinates, current heading angle, and estimated flight time; the environmental factors include wind speed, wind direction, air pressure, temperature, humidity, and high-risk weather tags.

8. The dynamic rerouting path planning method based on grey prediction and contrastive learning according to claim 1, characterized in that, In step S4, the contrastive learning includes data augmentation, encoding representation, and contrastive loss function, wherein the data augmentation methods include temporal perturbation enhancement, random occlusion mode, and adding random noise.

9. The dynamic rerouting path planning method based on grey prediction and contrastive learning according to claim 8, characterized in that, The contrastive loss function is: in, Indicates the number of sample pairs. and These are two samples in a sample pair. Indicates the distance between samples. It is the boundary threshold. It's a tag.

10. A dynamic rerouting path planning system based on grey prediction and contrastive learning, characterized in that, include: The data module uses real-time weather radar data to process and obtain a set of pixels with a reflectivity ≥41dBZ as the current thunderstorm weather range. It then uses the DBSCAN algorithm to perform spatial clustering and retains thunderstorm clusters with the largest area or a set threshold. The matching module, based on the thunderstorm clusters obtained from the data module, uses the Graham algorithm to construct convex polygons to generate the initial static flight restricted area boundary, uniformly samples n points on the boundary, and achieves optimal matching between points through the Hungarian algorithm; The prediction module uses a grey prediction model. By combining historical meteorological data and thunderstorm development, a time series prediction of the thunderstorm weather boundary at a certain future moment is made, and the predicted boundary is widened outward with the centroid of the convex hull as the center to form a new flight-restricted area. The learning module collects aircraft state parameters, environmental factors, and restricted area prediction results obtained from the prediction module, constructs positive and negative sample pairs, and optimizes the feature extraction network through comparative learning to obtain high-quality feature representations. The planning module inputs the high-quality feature representations obtained from the learning module into the PPO reinforcement learning algorithm to generate a rerouting path around the restricted flight area.

Images