Airport scene space similarity measurement method and system

Abstract

The invention discloses an airport scene space similarity measurement method and system. The method comprises the following steps: step 1, airport scene taxiway representation: generating vector representation of taxiway road sections and probability distribution between the road sections; 2, airport scene space similarity measurement: calculating airport scene space traffic similarity and space function similarity by utilizing probability distribution between road sections and cosine similarity expressed by taxiway road section vectors; and step 3, airport scene space similarity analysis: based on the calculation results of the space traffic similarity and the space function similarity, revealing the conventional running direction of the airport scene taxiway, and identifying road sections with similar functions. The method can reveal the spatial correlation characteristics of the road section, can be applied to complex airport scene operation mode mining and congestion prevention and control, provides a theoretical basis for airport operation management and optimization decision making, and fills the blank in the field of airport scene spatial similarity measurement.

Description

Technical Field

[0001] The present invention relates to the field of spatial similarity analysis, and in particular to a method and system for measuring the spatial similarity of airport scenes. Background Art

[0002] With the rapid growth in air transport demand, airport surface traffic systems are becoming increasingly complex. Simultaneously, they must handle multiple aircraft for arrival, departure, and taxi operations, resulting in dynamic, complex, and spatially heterogeneous surface traffic flows. Furthermore, airport surface traffic flows are not confined to a single taxiway but instead propagate along the entire system, influenced by various factors, including weather conditions and traffic control, presenting difficult-to-explore characteristics.

[0003] Existing airport surface traffic analysis methods often focus on analyzing the relationship between three traffic flow parameters using methods such as macroscopic fundamental diagrams (MFDs) and cellular transport models, thereby assessing airport surface operation patterns and congestion status. However, these analyses often overlook the spatial correlation and spatial coupling characteristics between airport taxiways, failing to fully analyze the transfer characteristics of traffic flows between taxiways. Therefore, this invention aims to reveal the spatial correlation characteristics of airport surface road segments based on a vector representation of taxiways, providing a theoretical basis for airport operations management and optimization decisions. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for measuring the spatial similarity of airport scenes, reveal the spatial traffic similarity and spatial function similarity of airport scenes, improve the analysis level of the current status of airport scene operation, and provide scientific guidance for airport operation and management.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for measuring spatial similarity of airport scenes comprises the following steps:

[0007] Step 1: Airport surface taxiway representation: Generate vector representation of taxiway segments and probability distribution between segments;

[0008] Step 2: Airport scene spatial similarity measurement: Using the probability distribution between road segments and the cosine similarity of taxiway segment vectors, the spatial traffic similarity and spatial function similarity of the airport scene are calculated respectively;

[0009] Step 3: Airport surface spatial similarity analysis: Based on the calculation results of spatial traffic similarity and spatial functional similarity, the commonly used operating directions of airport surface taxiways are revealed, and sections with similar functions are identified.

[0010] Furthermore, the step 1 specifically includes the following sub-steps:

[0011] Step 11: Use a natural language processing (NLP) algorithm to characterize the airport taxiways and generate a vector representation corresponding to each taxiway segment.

[0012] Step 12: Based on the vector representation corresponding to each coasting road segment generated in step 11, obtain the probability distribution results P(segment A, segment B), ..., P(segment X, segment Y) of each coasting road segment and other coasting road segments.

[0013] Furthermore, the step 2 specifically includes the following sub-steps:

[0014] Step 21: The probability distribution P(section A, section B) between sections is the probability of an aircraft transferring from section A to section B, which is the quantitative result of the spatial traffic similarity between sections A and B.

[0015] In step 22, based on the vector representation of the taxiway segment, the spatial functional similarity between the two taxiway segments is calculated using cosine similarity. The calculation formula is as follows:

[0016]

[0017] Among them, section A·section B is the inner product of the vector representations of the sliding road sections A and B; ||section A|| and ||section B|| are the modules of the vector representations of the sliding road sections A and B.

[0018] Furthermore, the step 3 specifically includes the following sub-steps:

[0019] Step 31, based on the spatial traffic similarity results of any two adjacent taxiway segments, mining the usual running directions of the airport taxiways;

[0020] In step 32, a certain taxiing road segment is taken as a research object, and the cosine similarity between the taxiing road segment and all other taxiing road segments is calculated. Then, by comparing the cosine similarity results, a road segment with a similar function to the research object is determined.

[0021] Furthermore, in step 31, the coasting road segments are divided into three categories: no traffic flow, one-way traffic flow, and two-way traffic flow.

[0022] Furthermore, the directions of the one-way traffic flow include from east to west, from west to east, from north to south and from south to north.

[0023] An airport scene spatial similarity measurement system, comprising:

[0024] Airport surface taxiway representation module, which generates vector representations of taxiway segments and probability distributions between segments;

[0025] Airport scene spatial similarity measurement module, used to measure the spatial traffic similarity and spatial function similarity of airport scenes;

[0026] The airport scene spatial similarity analysis module is used to explore the common operation directions of taxiways and identify sections with similar functions.

[0027] Beneficial effects: The present invention proposes an airport scene spatial similarity measurement method and system, which can realize the measurement and analysis of airport scene spatial traffic similarity and spatial function similarity, reveal the spatial correlation characteristics of traffic flow between road sections, and can be applied to complex airport scene operation mode mining and congestion prevention, providing a theoretical basis for airport operation management and optimization decision-making, filling the gap in the field of airport scene spatial similarity analysis technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flow chart of the airport scene spatial similarity measurement method;

[0029] Figure 2 is the spatial traffic similarity measurement result of airport surface taxiway segment 13;

[0030] Figure 3 The usual operating direction of the airport taxiway;

[0031] Figure 4 This section has a function similar to that of the airport taxiway section 13. DETAILED DESCRIPTION

[0032] The present invention will be further explained below with reference to the accompanying drawings.

[0033] In order to make the design objectives, technical routes and method advantages of the present invention easier to understand, the present invention is further explained in conjunction with the following drawings and embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0034] Example

[0035] This embodiment selects scene surveillance data of a typical busy airport, and the airport scene spatial similarity measurement method includes the following steps:

[0036] Step 1, airport taxiway representation, essentially generates vector representations of taxiway segments and the probability distribution between them. This includes the following sub-steps:

[0037] (1.1) Use natural language processing (NLP) algorithms to characterize the airport taxiways and generate vector representations corresponding to each taxiway segment.

[0038] (1.2) Based on the vector representation corresponding to each coasting road segment generated in step (1.1), obtain the probability distribution results P(section A, section B),…, P(section X, section Y) of each section and other sections.

[0039] In a specific application scenario, taking Shenzhen Bao'an International Airport as an example, a natural language processing (NLP) algorithm was used to generate vector representations of each taxiway segment and the probability distribution between segments, as shown in Tables 1 and 2.

[0040] Table 1. Vector representation of taxiway segments on airport surface in the embodiment

[0041]

[0042]

[0043] Table 2 Probability distribution results between taxiway segments at the airport in the embodiment

[0044]

[0045] Step 2, measuring the spatial similarity of the airport scene, essentially uses the probability distribution between road segments and the cosine similarity represented by the taxiway segment vector to calculate the spatial traffic similarity and spatial function similarity of the airport scene. It specifically includes the following sub-steps:

[0046] (2.1) The probability distribution P(section A, section B) between sections is the probability that an aircraft will transfer from section A to section B. For P(section A, section B) = 0.5, the probability that an aircraft will transfer from section A to section B is 0.5, which is the quantitative result of the spatial traffic similarity between sections A and B.

[0047] (2.2) Based on the vector representation of the taxiway segment, the cosine similarity is used to calculate the spatial functional similarity between two taxiway segments. The calculation formula is as follows:

[0048]

[0049] Wherein, Segment A·Segment B is the inner product of the vector representations of the coasting road segments A and B. |Segment A| and |Segment B| are the moduli of the vector representations of the coasting road segments A and B.

[0050] Specifically, the taxiway section 13 of Shenzhen Bao'an International Airport is taken as the research object, and the spatial traffic similarity measurement results between it and its adjacent taxiway sections are as follows: Figure 2 The spatial function similarity measurement results with other taxiway segments are shown in Table 3.

[0051] Table 3 Spatial function similarity measurement results of airport surface taxiway segment 13

[0052]

[0053]

[0054] Step 3, airport surface spatial similarity analysis, essentially reveals the common operating directions of airport taxiways based on the calculation results of spatial traffic similarity and spatial functional similarity, and identifies sections with similar functions. It specifically includes the following sub-steps:

[0055] (3.1) Based on the spatial traffic similarity between any two adjacent taxiway segments, the common operating directions of airport taxiways are discovered. Taxiway segments are then divided into three categories: no traffic flow, one-way traffic flow, and two-way traffic flow. One-way traffic flow directions include east to west, west to east, north to south, and south to north.

[0056] (3.2) Take a certain taxiing road segment as the research object and calculate its cosine similarity with all other taxiing road segments. Then, by comparing the cosine similarity results, identify the road segments with similar functions to the research object.

[0057] Based on the spatial traffic similarity of all adjacent taxiway segments, the conventional operating direction of the airport taxiway is obtained, such as Figure 3 As shown in Figure 1. Taking the taxiway segment 13 as the research object, we analyzed its spatial functional similarity and found that the segments with similar functions are as follows: Figure 4 shown.

[0058] Based on the above embodiment, the present invention further provides an airport scene spatial similarity measurement system, comprising an airport scene taxiway characterization module, an airport scene spatial similarity measurement module, and an airport scene spatial similarity analysis module, wherein:

[0059] The airport surface taxiway representation module is used to generate vector representations of taxiway segments and probability distributions between segments.

[0060] The airport scene spatial similarity measurement module is used to measure the airport scene spatial traffic similarity and spatial function similarity.

[0061] The airport scene spatial similarity analysis module is used to explore the common operation directions of taxiways and identify sections with similar functions.

[0062] The present invention uses the vector representation of taxiway segments obtained by the natural language processing (NLP) algorithm and the probability distribution between segments to carry out spatial similarity analysis of airport scenes. Based on the probability distribution between adjacent taxiway segments, the spatial traffic correlation of the airport scene is analyzed to reveal the usual running direction of the taxiways on the airport scene. By calculating the cosine similarity of the corresponding vectors of the taxiway segments, the spatial functional similarity between taxiway segments is quantified, thereby identifying segments with similar functions. The present invention can reveal the spatial correlation characteristics of segments and can be applied to the mining of operating modes and congestion prevention in complex airport scenes. It provides a theoretical basis for airport operation management and optimization decision-making, filling the gap in the field of spatial similarity measurement of airport scenes.

[0063] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for measuring spatial similarity of airport scenes, characterized by: The following steps are involved: Step 1: Airport surface taxiway representation: Generate vector representation of taxiway segments and probability distribution between segments; Step 2: Airport scene spatial similarity measurement: Using the probability distribution between road segments and the cosine similarity of taxiway segment vectors, the spatial traffic similarity and spatial function similarity of the airport scene are calculated respectively; Step 3: Airport surface spatial similarity analysis: Based on the calculation results of spatial traffic similarity and spatial functional similarity, the commonly used operating directions of airport surface taxiways are revealed, and sections with similar functions are identified.

2. The method for measuring spatial similarity of airport scenes according to claim 1, characterized in that: The step 1 specifically includes the following sub-steps: Step 11: Use a natural language processing algorithm to characterize the airport taxiways and generate a vector representation corresponding to each taxiway segment; Step 12: Based on the vector representation corresponding to each glide path segment generated in step 11, obtain the probability distribution results P(segment A, segment B), ..., P(segment X, segment Y) of each glide path segment and other glide path segments. 。 3. The method for measuring spatial similarity of airport scenes according to claim 1, characterized in that: The step 2 specifically includes the following sub-steps: Step 21 ， The probability distribution P(section A, section B) between sections is the probability of an aircraft transferring from section A to section B, which is the quantitative result of the spatial traffic similarity between sections A and B. In step 22, based on the vector representation of the taxiway segment, the spatial functional similarity between the two taxiway segments is calculated using cosine similarity. The calculation formula is as follows: Among them, section A·section B is the inner product of the vector representations of the sliding road sections A and B; ||section A|| and ||section B|| are the modules of the vector representations of the sliding road sections A and B.

4. The method for measuring spatial similarity of airport scenes according to claim 1, characterized in that: The step 3 specifically includes the following sub-steps: Step 31, based on the spatial traffic similarity results of any two adjacent taxiway segments, mining the usual running directions of the airport taxiways; In step 32, a certain taxiing road segment is taken as a research object, and the cosine similarity between the taxiing road segment and all other taxiing road segments is calculated. Then, by comparing the cosine similarity results, a road segment with a similar function to the research object is determined.

5. The method for measuring spatial similarity of airport scenes according to claim 4, characterized in that: In step 31, the coasting road segment is divided into three categories: no traffic flow, one-way traffic flow, and two-way traffic flow.

6. The method for measuring spatial similarity of airport scenes according to claim 5, characterized in that: The directions of the one-way traffic flow include from east to west, from west to east, from north to south and from south to north.

7. An airport scene spatial similarity measurement system, characterized by: include: Airport surface taxiway representation module, which generates vector representations of taxiway segments and probability distributions between segments; Airport scene spatial similarity measurement module, used to measure the spatial traffic similarity and spatial function similarity of airport scenes; The airport scene spatial similarity analysis module is used to explore the common operation directions of taxiways and identify sections with similar functions.

Images