System for acquiring target sliding path

Abstract

The invention relates to a system for acquiring a target sliding path, and relates to the technical field of data processing, the system comprises a processor and a memory in which a computer program is stored, and when the computer program is executed by the processor, the following steps are realized: acquiring a target flight ID list corresponding to a target area and a specified feature information list corresponding to the target flight ID list, obtaining a target feature information list set corresponding to the target flight ID list, obtaining a preset specified time list corresponding to the target flight ID list according to the target feature information list set, and obtaining a target sliding path list corresponding to the target flight ID list through a multi-decoder attention model, according to the method, the problem of missing flight taxiing path nodes and time can be avoided during data preprocessing, the problem of data incompleteness is effectively solved, path adjustment is carried out by utilizing a related model based on the time of arriving at each taxiing node in the flight taxiing process, delay and resource waste are reduced, and taxiing efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a system for acquiring a target glide path. Background Art

[0002] At present, the global demand for air transport is recovering steadily, bringing more opportunities to the industry as well as new challenges. The expansion speed of airport infrastructure cannot keep up with the growth of air transport volume, and the contradiction between supply and demand of airport surface resources is becoming more and more serious. This not only reduces the airport taxiing speed and aviation economy, but also aggravates the congestion and queuing problems on the taxiway. At the same time, in actual operation, due to emergencies such as bad weather, mechanical failures and delays in boarding and disembarking of passengers, the actual arrival and departure times of flights often deviate from the planned time. Such delays will trigger a chain reaction and further affect the stability of airport ground operations. Therefore, predicting and optimizing the airport taxiing path can not only improve control efficiency, but also significantly reduce the delay rate. By accurately predicting the time when flights arrive at each key taxiing node, the taxiing trend of flights on the airport surface can be perceived in real time, potential delays and conflicts can be warned in advance, and measures to avoid conflicts and alleviate delays can be proposed, providing important decision-making basis for taxiway management. Summary of the invention

[0003] In view of the above technical problems, the technical solution adopted by the present invention is: a system for obtaining a target glide path, comprising: a processor and a memory storing a computer program, and when the computer program is executed by the processor, the following steps are implemented: S1, obtaining a target flight ID list corresponding to the target area and a specified feature information list corresponding to the target flight ID list, wherein the target flight ID list includes a plurality of target flight IDs, the specified feature information list includes a plurality of specified feature information, the specified feature information is flight taxi path information of the target flight corresponding to the target flight ID, and the flight taxi path information includes a critical taxi path of the flight in the target area and the time to reach each critical taxi node in the critical taxi path.

[0004] S2, obtaining a target feature information list set corresponding to the target flight ID list, wherein the target feature information list set includes several target feature information lists, each target flight ID corresponds to a target feature information list, and the target feature information list includes the target flight ID and the specified feature information corresponding to the target flight ID.

[0005] S3, according to the target feature information list set, obtain a preset designated time list corresponding to the target flight ID list, wherein the preset designated time list includes a plurality of preset designated times, and the preset designated time is the time when the target flight arrives at a preset key taxiing node.

[0006] S4. According to the target feature information list set and the preset designated time list, a target taxi path list corresponding to the target flight ID list is obtained through a multi-decoder attention model, wherein the target taxi path list includes a plurality of target taxi paths, each target flight ID corresponds to a target taxi path, and the target taxi path is a taxi path obtained by implementing a stop-and-wait strategy for the flight when there is a conflict between target flights.

[0007] Compared with the prior art, the present invention has obvious beneficial effects. By means of the above technical solution, the system for obtaining a target glide path provided by the present invention can achieve considerable technical advancement and practicality, and has wide industrial utilization value, and has at least the following beneficial effects: The present invention is a system for acquiring a target taxiing path, comprising: a processor and a memory storing a computer program. When the computer program is executed by the processor, the following steps are implemented: acquiring a target flight ID list corresponding to a target area and a designated feature information list corresponding to the target flight ID list, acquiring a target feature information list set corresponding to the target flight ID list, wherein the target feature information list includes a target flight ID and designated feature information corresponding to the target flight ID, acquiring a preset designated time list corresponding to the target flight ID list according to the target feature information list set, and acquiring a target taxiing path list corresponding to the target flight ID list through a multi-decoder attention model according to the target feature information list set and the preset designated time list. The present invention can avoid the problem of missing flight taxiing path nodes and time during data preprocessing, effectively solve the problem of incomplete data, ensure that each flight can match a complete taxiing path, so that the accuracy of subsequent taxiing situation prediction and taxiing path analysis is high, and use a relevant model to adjust the path based on the time when the flight arrives at each taxiing node during the taxiing process, reduce delays and resource waste, improve taxiing efficiency, shorten the waiting time of flights, and promote the optimization of the overall operation of the airport.

[0008] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the following specifically cites a preferred embodiment and describes it in detail with the accompanying drawings as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A flow chart of steps implemented when a processor of a system for acquiring a target glide path according to an embodiment of the present invention executes a computer program. DETAILED DESCRIPTION

[0010] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0011] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0012] Example This embodiment provides a system for acquiring a target glide path, the system comprising: a processor and a memory storing a computer program, and when the computer program is executed by the processor, the following steps are implemented: Figure 1 As shown: S1, obtaining a target flight ID list corresponding to the target area and a specified feature information list corresponding to the target flight ID list, wherein the target flight ID list includes a plurality of target flight IDs, the specified feature information list includes a plurality of specified feature information, the specified feature information is flight taxi path information of the target flight corresponding to the target flight ID, and the flight taxi path information includes a critical taxi path of the flight in the target area and the time to reach each critical taxi node in the critical taxi path.

[0013] Specifically, the target area is an airport where flights take off and land.

[0014] Specifically, each target flight ID corresponds to a specified feature information.

[0015] Furthermore, the target flight ID is a unique identifier representing the identity of the target flight generated based on the flight number, target date and target name corresponding to the target flight, wherein the target date is the date on which the target flight departs from the take-off airport, and the target name is the name of the take-off and landing airport of the target flight.

[0016] Specifically, S1 also includes the following steps: S11, obtain a list of candidate locations of the target area D = {D1, ..., D x , ..., D p} and the candidate region list H = {H1, ..., H y , ..., H q}, D x is the xth candidate location point, x=1…p, p is the number of candidate location points in the candidate location point list, H y is the y-th candidate region, y=1…q, q is the number of candidate regions in the candidate region list.

[0017] Specifically, the candidate location point is an end point of a runway in the target area.

[0018] Specifically, the candidate area is an area obtained by dividing the target area based on the distances between all the parking stands in the target area and the candidate location points. Those skilled in the art know that the target area can be divided according to the layout of the target area, which all falls within the protection scope of the present invention and will not be repeated here. For example, when the terminals in the target area are symmetrical, the target area is divided according to the candidate location points so that the parking stands are symmetrically distributed with respect to the flight buildings.

[0019] From the above, it can be seen that by dividing the area, the huge amount of data required to process all parking positions and runways can be effectively reduced, which not only reduces the computational complexity, but also makes the taxi path planning and clustering process more accurate and efficient.

[0020] S12, obtaining the target priorities of all target flight pairs, where the target priority F of the target flight pair consisting of the i-th target flight and the j-th target flight is ij Meet the following conditions: , β ij =min(leni,lenj),leni is the number of taxi points in the taxi path corresponding to the i-th target flight,lenj is the number of initial taxi points in the initial taxi path corresponding to the j-th target flight,α is 0 to β ij An integer of -1, X i，leni-α is the horizontal coordinate of the (leni-α)th initial taxiing point on the initial taxiing path corresponding to the i-th target flight, X j，lenj-α is the horizontal coordinate of the (lenj-α)th initial taxiing point on the initial taxiing path corresponding to the jth target flight, Y i，leni-α is the ordinate of the (leni-α)th initial taxiing point on the initial taxiing path corresponding to the i-th target flight, Y j，lenj-αis the ordinate of the (lenj-α)th initial taxiing point on the initial taxiing path corresponding to the jth target flight, min() is the minimum function, the value range of i is 1 to N, the value range of j is 1 to N, i≠j, and N is the number of target flight IDs in the target flight ID list.

[0021] Specifically, the initial glide path is the glide path of the target flight acquired through ADS-B technology.

[0022] Furthermore, the initial taxiing node is a taxiing node in an initial taxiing path.

[0023] S13, based on the target priorities between all target flight pairs, clustering is performed using the K-Means algorithm to obtain a target clustering result list set DH = {DH1, ..., DH y , ..., DH q}, DH y ={DH y1 , ..., DH yx , ..., DH yp}, where DH yx The clustering result is that the parking position of the flights taking off and landing in the y-th candidate area is in the y-th candidate area, and the flights taking off and landing in the y-th candidate area fly in and out of the runway where the x-th candidate position point is located.

[0024] S14, based on the DH, a number of key glide paths are obtained, wherein the key glide paths are glide paths within the target area extracted from the DH.

[0025] Based on the above content, it can be seen that by realizing clustering within and between regions, the designated flight information corresponding to the target flight is obtained, and the missing data is filled in. During data preprocessing, the problem of missing flight taxi path nodes and time is avoided, and the continuity of the data is guaranteed, so that the subsequent taxiing situation prediction and taxiing path analysis are more accurate.

[0026] Specifically, S1 also includes the following steps: S01, obtain the critical taxi path L corresponding to the i-th target flight ID i =(L i1 , ..., L ie , ..., L if（i） ), L ie is the e-th critical taxiing node in the critical taxiing path corresponding to the i-th target flight ID, e=1…f(i), f(i) is the number of critical taxiing nodes in the critical taxiing path corresponding to the i-th target flight ID.

[0027] S02, when L ieWhen the corresponding key time is missing, obtain L ie The corresponding critical time T ie , where T ie Meet the following conditions: T ie =T i（e-1） + (T i / T 0 i )×(T i（e+1） -T i（e-1） ), T i（e-1） is the time for the i-th target flight to fly to the (e-1)th critical taxiing node, T i is the total taxiing time of the i-th target flight in its corresponding critical taxiing path, T 0 i is the average taxiing time of all target flights that are consistent with the critical taxiing path corresponding to the i-th target flight, T i（e+1） is the time it takes for the i-th target flight to fly to the (e+1)th critical taxiing node.

[0028] Specifically, the critical time is the time it takes for the target flight to fly to a critical taxiing node in the critical taxiing path.

[0029] S03, when L i1 To L ie When the corresponding key time is missing, obtain L ie The corresponding critical time T ie , where T ie Meet the following conditions: T ie =T i（e+1） -T i e（e+1） , T i e（e+1） is the average taxiing time of the i-th target flight from the e-th key taxiing node to the e+1-th key taxiing node.

[0030] S2, obtaining a target feature information list set corresponding to the target flight ID list, wherein the target feature information list set includes several target feature information lists, each target flight ID corresponds to a target feature information list, and the target feature information list includes the target flight ID and the specified feature information corresponding to the target flight ID.

[0031] Specifically, the target feature information list includes several target feature information, and the target feature information is the data under the target feature, wherein the target feature includes: flight number, target date of the flight, location of the flight, parking position of the flight, runway port, data sending time, flight ID, and initial taxi path of the flight.

[0032] Furthermore, the flight location information includes the longitude and latitude of the flight during taxiing.

[0033] Furthermore, when the flight is an inbound flight, the parking position of the flight is the starting point of the flight, and when the flight is a outbound flight, the parking position of the flight is the end point of the flight.

[0034] Furthermore, when the flight is an inbound flight, the runway port is the end point of the flight, and when the flight is a outbound flight, the runway port is the starting point of the flight.

[0035] Furthermore, the data sending time is the time of each data transmission.

[0036] S3, according to the target feature information list set, obtain a preset designated time list corresponding to the target flight ID list, wherein the preset designated time list includes a plurality of preset designated times, and the preset designated time is the time when the target flight arrives at a preset key taxiing node.

[0037] Specifically, the preset specified time is predicted by the trained PatchTST model.

[0038] Furthermore, S3 also includes the following steps: S31, based on the target feature information list set, obtain several single-variable flight time series corresponding to each flight.

[0039] S32, grouping the single variable flight time series by flight and dividing it into multiple patches, wherein the single variable flight time series is the flight time series corresponding to each target feature.

[0040] Specifically, those skilled in the art know that the length of the flight time series can be set according to actual needs.

[0041] S33, the data generated after each patch is converted by the linear mapping matrix and the position encoding matrix is ​​input into the Transformer model, wherein the linear mapping matrix is ​​used to convert the patch data into feature identifiers suitable for model processing, and the position editing matrix enhances the modeling ability of the sequence time dimension by introducing time order information.

[0042] S34, calculates multi-head attention to integrate the correlation between different patches, wherein the relationship between each patch is modeled through the multi-head attention mechanism.

[0043] S35, obtaining a preset designated time list through a Flatten operation.

[0044] S4. According to the target feature information list set and the preset designated time list, a target taxi path list corresponding to the target flight ID list is obtained through a multi-decoder attention model, wherein the target taxi path list includes a plurality of target taxi paths, each target flight ID corresponds to a target taxi path, and the target taxi path is a taxi path obtained by implementing a stop-and-wait strategy for the flight when there is a conflict between target flights.

[0045] Specifically, the input of the multi-decoder attention model is a list set of target feature information of all target flights and a list of preset designated times of all target flights. In the encoder part, the designated feature information corresponding to all target flight IDs is embedded as a feature vector and used as input for the multi-decoder attention model to make decisions. In the decoder part, the target feature information of the current target flight is used as input, and the target taxi path of each target flight is finally output.

[0046] Furthermore, when , determine the conflict between the i-th target flight and the j-th target flight at time t, where Q t ij is the distance between the i-th target flight and the j-th target flight at time t, Q t uv is the distance between the u-th target flight and the v-th target flight at time t, the value range of u is 1 to N-1, the value range of v is u+1 to N, and t is a certain moment when the target flight is taxiing in the target area.

[0047] From the above content, it can be seen that by dynamically adjusting the waiting time of flights during taxiing, ensuring that sufficient safety distance is always maintained between flights, the impact on taxiing efficiency is reduced. This can not only effectively reduce the probability of taxiing conflicts, but also significantly improve the safety and overall efficiency of airport operations.

[0048] The present embodiment provides a system for acquiring a target taxiing path, comprising: a processor and a memory storing a computer program. When the computer program is executed by the processor, the following steps are implemented: acquiring a target flight ID list corresponding to a target area and a designated feature information list corresponding to the target flight ID list, acquiring a target feature information list set corresponding to the target flight ID list, wherein the target feature information list includes the target flight ID and the designated feature information corresponding to the target flight ID, acquiring a preset designated time list corresponding to the target flight ID list according to the target feature information list set, and acquiring a target taxiing path list corresponding to the target flight ID list through a multi-decoder attention model according to the target feature information list set and the preset designated time list. The present invention can avoid the problem of missing flight taxiing path nodes and time during data preprocessing, effectively solve the problem of incomplete data, ensure that each flight can match a complete taxiing path, so that the subsequent taxiing situation prediction and taxiing path analysis are more accurate, and use relevant models to adjust the path based on the time when the flight arrives at each taxiing node during the taxiing process, reduce delays and resource waste, improve taxiing efficiency, shorten the waiting time of flights, and promote the optimization of the overall operation of the airport.

[0049] Although some specific embodiments of the present invention have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It should also be understood by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. A system for obtaining a target glide path, characterized in that: The system comprises: a processor and a memory storing a computer program, and when the computer program is executed by the processor, the following steps are implemented: S1, obtaining a target flight ID list corresponding to the target area and a specified feature information list corresponding to the target flight ID list, wherein the target flight ID list includes a plurality of target flight IDs, the specified feature information list includes a plurality of specified feature information, the specified feature information is flight taxi path information of the target flight corresponding to the target flight ID, and the flight taxi path information includes a critical taxi path of the flight in the target area and a time to reach each critical taxi node in the critical taxi path; S2, obtaining a target feature information list set corresponding to the target flight ID list, wherein the target feature information list set includes a plurality of target feature information lists, each target flight ID corresponds to a target feature information list, and the target feature information list includes the target flight ID and the specified feature information corresponding to the target flight ID; S3, according to the target feature information list set, obtaining a preset designated time list corresponding to the target flight ID list, wherein the preset designated time list includes a plurality of preset designated times, and the preset designated time is the time when the target flight arrives at a preset key taxiing node; S4. According to the target feature information list set and the preset designated time list, a target taxi path list corresponding to the target flight ID list is obtained through a multi-decoder attention model, wherein the target taxi path list includes a plurality of target taxi paths, each target flight ID corresponds to a target taxi path, and the target taxi path is a taxi path obtained by implementing a stop-and-wait strategy for the flight when there is a conflict between target flights.

2. The system for acquiring a target glide path according to claim 1, characterized in that: The target flight ID is a unique identifier representing the identity of the target flight generated based on the flight number, target date and target name corresponding to the target flight, wherein the target date is the date on which the target flight departs from the take-off airport, and the target name is the name of the take-off and landing airport of the target flight.

3. The system for acquiring a target glide path according to claim 1, characterized in that: S1 also includes the following steps: S11, obtain a list of candidate locations of the target area D = {D1, ..., D x , ..., D p } and the candidate region list H = {H1, ..., H y , ..., H q }, D x is the xth candidate location point, x=1…p, p is the number of candidate location points in the candidate location point list, H y is the y-th candidate region, y=1…q, q is the number of candidate regions in the candidate region list; S12, obtaining the target priorities of all target flight pairs, where the target priority F of the target flight pair consisting of the i-th target flight and the j-th target flight is ij Meet the following conditions: , β ij =min(leni,lenj),leni is the number of initial taxi points in the initial taxi path corresponding to the i-th target flight,lenj is the number of initial taxi points in the initial taxi path corresponding to the j-th target flight,α is 0 to β ij An integer of -1, X i，leni-α is the horizontal coordinate of the (leni-α)th initial taxiing point on the initial taxiing path corresponding to the i-th target flight, X j，lenj-α is the horizontal coordinate of the (lenj-α)th initial taxiing point on the initial taxiing path corresponding to the jth target flight, Y i，leni-α is the ordinate of the (leni-α)th initial taxiing point on the initial taxiing path corresponding to the i-th target flight, Y j，lenj-α is the ordinate of the (lenj-α)th initial taxiing point on the initial taxiing path corresponding to the jth target flight, min() is the minimum value function, the value range of i is 1 to N, the value range of j is 1 to N, i≠j, and N is the number of target flight IDs in the target flight ID list; S13, based on the target priorities between all target flight pairs, clustering is performed using the K-Means algorithm to obtain a target clustering result list set DH = {DH1, ..., DH y , ..., DH q }, DH y ={DH y1 , ..., DH yx , ..., DH yp }, where DH yx The clustering result is that the parking position of the flights taking off and landing in the yth candidate area is in the yth candidate area and the flights taking off and landing in the yth candidate area fly in and out of the runway where the xth candidate position point is located; S14, based on the DH, a number of key glide paths are obtained, wherein the key glide paths are glide paths within the target area extracted from the DH.

4. The system for acquiring a target glide path according to claim 3, characterized in that: The candidate area is an area obtained by dividing the target area based on the distances between all parking spaces in the target area and the candidate location points.

5. The system for acquiring a target glide path according to claim 1, characterized in that: S1 also includes the following steps: S01, obtain the critical taxi path L corresponding to the i-th target flight ID i =(L i1 , ..., L ie , ..., L if（i） ), L ie is the e-th critical taxiing node in the critical taxiing path corresponding to the i-th target flight ID, e=1…f(i), f(i) is the number of critical taxiing nodes in the critical taxiing path corresponding to the i-th target flight ID; S02, when L ie When the corresponding key time is missing, obtain L ie The corresponding critical time T ie , where T ie Meet the following conditions: T ie =T i（e-1） + (T i / T 0 i )×(T i（e+1） -T i（e-1） ), T i（e-1） is the time for the i-th target flight to fly to the (e-1)th critical taxiing node, T i is the total taxiing time of the i-th target flight in its corresponding critical taxiing path, T 0 i is the average taxiing time of all target flights that are consistent with the critical taxiing path corresponding to the i-th target flight, T i（e+1） The time for the i-th target flight to fly to the (e+1)th critical taxiing node; S03, when L i1 To L ie When the corresponding key time is missing, obtain L ie The corresponding critical time T ie , where T ie Meet the following conditions: T ie =T i（e+1） -T i e（e+1） , T i e（e+1） is the average taxiing time of the i-th target flight from the e-th key taxiing node to the e+1-th key taxiing node.

6. The system for acquiring a target glide path according to claim 1, characterized in that: The target feature information list includes several target feature information, and the target feature information is data under the target feature, wherein the target feature includes: flight number, target date of the flight, location of the flight, parking position of the flight, runway port, data sending time, flight ID, and initial taxi path of the flight.

7. The system for acquiring a target glide path according to claim 1, characterized in that: The preset specified time is predicted by the trained PatchTST model.

Images