AMDB-based taxiway-related navigation announcement intelligent classification and analysis method

Abstract

The invention discloses an AMDB (Airport Map Database)-based intelligent classification and analysis method for taxiway-related navigation announcements, and the method comprises the steps: carrying out the automatic classification of the navigation announcements, and carrying out the automatic classification of the navigation announcements; a navigation announcement key information extraction step used for extracting key element information from the navigation announcement data; a navigation announcement influence range analysis step used for calculating an influence area range of the navigation announcement; and a navigation announcement and AMDB data superposition step of superposing the influence range of the navigation announcement to an electronic map in combination with the AMDB data so as to realize the purpose of observing the influence condition of the scene in real time and the like. According to the method, automatic classification and key information extraction of the navigation announcements are realized by constructing the navigation announcement category library and the probability classification model, and the structured data is dynamically associated with the AMDB to generate a visual geographical influence range and time efficiency information.

Description

Technical Field

[0001] The present invention relates to the field of aviation information processing technology, and in particular to a method for intelligent classification, structured analysis and graphical display of NOTAMs based on an Airport Map Database (AMDB). Background Art

[0002] With the continuous development and expansion of aviation business and the increasing number of flights, the role of NOTAM data in production operations is becoming increasingly important. The requirements for the timeliness and accuracy of NOTAM in related fields are also increasing. This requires more efficient and accurate provision and processing of NOTAM data. However, the current NOTAM data still has some limitations in actual production, such as: Traditional rule-based NOTAM classification methods are inefficient and struggle to cope with the ever-increasing volume of data. They also suffer from low accuracy and cannot guarantee the correct handling of complex and ever-changing NOTAM content. Furthermore, the operational impact of traditional NOTAMs relies solely on manual processing and textual descriptions, without the ability to display them graphically or dynamically link them with AMDB data, limiting their application. Traditional NOTAMs exist in unstructured text format, and existing aviation information systems are unable to automatically generate structured data, limiting their application in scenarios such as flight planning and airport management. Summary of the Invention

[0003] In response to the aforementioned shortcomings of the prior art, the present invention provides an automatic classification and parsing method for taxiway-related NOTAMs based on AMDB, thereby improving the efficiency of classification, processing, and parsing of aviation intelligence. Furthermore, the method combines AMDB data to more intuitively and accurately display the impact of NOTAMs on aviation operations. Specifically, the method includes: An AMDB-based intelligent classification and parsing method for taxiway-related NOTAMs is characterized by comprising the following steps: Step 1: Classification model training: By testing and training known sample data, a NOTAM classification model is obtained; Step 2: NOTAM classification: Automatically classify NOTAMs using a classification model; Step 3: Extract key information from the NOTAM, using natural language analysis algorithms combined with keyword category settings to extract key element attributes and related data from the NOTAM; Step 4: Analyze and calculate the impact range of the NOTAM. By using PostGIS geospatial analysis algorithms, combined with AMDB data and key elements of the NOTAM, the impact range of the NOTAM is dynamically calculated. Step 5: Overlay the impact range data of the NOTAM with the AMDB data. Overlay the impact range data of the NOTAM onto the airport scene map produced using the AMDB data and mark it with a highlighted color to show the impact of the airport.

[0004] The specific calculation formula of the NOTAM classification model in step 1 is as follows: P(C|X)=P(X)P(X|C)P(C) Where P(C|X) is the probability that a NOTAM belongs to category C under the condition that certain features X of the NOTAM appear, i.e., the posterior probability. P(X|C) is the probability of the feature X corresponding to the NOTAM appearing when the NOTAM is classified as category C. P(C) is the prior probability that the NOTAM belongs to category C; P(X) is the marginal probability of the occurrence of feature X of the NOTAM; Since P(X) is the same for all categories, P(X) can be ignored when classifying, and only the size of P(X|C)P(C) needs to be compared; The specific classification rule formula is as follows: C^=argmaxCP(C)∏i=1nP(xi|C) Where C^ is the predicted NOTAM category and xi is the i-th component of the NOTAM feature X.

[0005] The key information extraction content of the NOTAM in step 3 specifically includes: NOTAM keyword preset: preset NOTAM keyword types according to NOTAM business rules and transmission specifications, providing a basis for extracting keywords from NOTAM; NOTAM keyword extraction: combining classification models with established NOTAM keyword rules to extract key information data from NOTAM information; Keyword marking and storage: marking and storing the keywords extracted from the navigation notice.

[0006] It also includes the automatic retrieval and processing steps of navigation notices. By retrieving the latest navigation notice data in real time, it uses an automatic classification model to classify and process it, and classify the data.

[0007] It also includes a real-time monitoring step for the effectiveness of the navigation notice, and controls the display and removal of graphic data related to the affected area on the map component according to the current real-time effectiveness of the navigation notice.

[0008] It also includes data format conversion and storage steps, which convert and store relevant geographic spatial information data into a unified data format to generate a universal data format.

[0009] The present invention combines AMDB data and intelligently classifies and analyzes traditional NOTAM data on the basis of AM data, solving the problem that traditional taxiway-related NOTAMs cannot be automatically classified and analyzed, and improving the classification, processing and analysis efficiency of NOTAMs. By dynamically calculating the impact area of ​​the NOTAM and superimposing it with AMDB data, the NOTAM data is intuitively presented to related applications, realizing a graphical presentation of the NOTAM impact area. The generated general structured data can provide data support for external systems and offer more accurate and diversified aviation intelligence data services for related fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 The main functions and application diagram of the automatic classification and parsing method of NOTAM Figure 2 Sample data for training the automatic classification model of NOTAM Figure 3 Extraction of flow charts for key elements of NOTAM Figure 4 Labeling diagrams for NOTAM categories Figure 5 Sample diagram of the extraction result data for key elements of NOTAM Figure 6 Schematic diagram showing the overlay of AMDB and NOTAM data DETAILED DESCRIPTION The technical solution of the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0011] Figure 1 This is a diagram of the main functions and applications of the automatic classification and parsing method of navigation notices, including the main data processing steps and application scenarios of the data included in this invention.

[0012] First, classification model training is carried out, mainly through a large amount of testing and training on sample data to form a classification model; secondly, NOTAM classification is carried out, and the classification model is used to automatically classify NOTAMs; then NOTAM keywords are extracted, and the natural language analysis algorithm is combined with the setting of keyword categories to extract key element attributes and related data from the NOTAM; then the NOTAM impact range is calculated, and the impact area of ​​the NOTAM is dynamically calculated using the PostGIS geospatial analysis algorithm combined with AMDB data and NOTAM key element information; then the NOTAM impact range data is superimposed with the AMDB data, and the NOTAM impact range data is superimposed on the airport scene map created using AMDB data to show the impact of the airport; finally, through unified data format conversion and storage, universal spatial data that can be used by various platforms is generated. This data can provide data support for external systems and applications, and the NOTAM data can be applied more widely, accurately and intuitively in various business systems.

[0013] 1. The specific steps of NOTAM classification model training are as follows: See Figure 2 , for the training sample data diagram, firstly, a batch of historical NOTAM data are sorted out as sample data. These sample data are selected from the historical real NOTAM data as the basis for model training. If necessary, these sample data are cleaned and converted; then, according to the characteristics and sending specifications of the NOTAM, the feature data are extracted and the corresponding NOTAM category is marked.

[0014] The method of training a classification algorithm model uses a probability distribution algorithm to train a model on sample data, continuously calculates the probability of the corresponding navigation notice category appearing when certain known features appear, and infers the category to which the sample data belongs through probability, and summarizes a series of category prediction and classification rules, that is, when certain features exist, what category the corresponding navigation notice is most likely to be, and obtains a category prediction model. Then, based on this model, new sample data is predicted, and the probability of the navigation notice category appearing under the condition that certain features appear is calculated. Finally, a navigation notice classification model is formed, and based on this, subsequent intelligent automatic classification of navigation notice data is carried out, thereby achieving the purpose of automatic classification of navigation notice data.

[0015] The specific calculation formula is as follows: P(C|X)=P(X)P(X|C)P(C) P(C|X) is the probability that a NOTAM belongs to category C under the condition that certain features X of the NOTAM appear, that is, the posterior probability.

[0016] P(X|C) is the probability of the feature X corresponding to the NOTAM appearing under the condition that the NOTAM belongs to category C.

[0017] P(C) is the prior probability that the NOTAM belongs to category C.

[0018] P(X) is the marginal probability that feature X of the NOTAM will occur.

[0019] Since P(X) is the same for all categories, we can ignore P(X) when performing classification and only need to compare the sizes of P(X|C)P(C).

[0020] The specific classification rule formula is as follows: C^=argmaxCP(C)∏i=1nP(xi|C) where C^ is the predicted NOTAM category and xi is the i-th component of the NOTAM feature X.

[0021] According to this formula, after multiple calculations, the category with the highest comparison probability is considered to be the category to which the navigation should belong, thereby achieving the classification of navigation notices.

[0022] II. Steps for classifying NOTAMs: Used to automatically identify and classify NOTAM data to improve the efficiency and accuracy of NOTAM data classification.

[0023] It also includes automatic retrieval and processing of navigation notices, which retrieves the latest navigation notice data in real time, classifies it using an automatic classification model, and labels the data with categories.

[0024] 3. Steps for extracting key information from NOTAMs: Used to extract key element information from NOTAM data for subsequent spatial analysis and structured data generation.

[0025] Figure 3 A flowchart for extracting key elements from NOTAMs is created. First, some keyword categories are preset based on the characteristics of the NOTAMs. Then, based on the classification results of the NOTAMs, keyword information and related attribute values ​​are extracted from the NOTAMs to prepare for subsequent spatial analysis and calculations. Specifically, the following steps are involved: NOTAM keyword preset: According to the business rules and sending specifications of NOTAM, NOTAM keyword types are preset to provide a basis for extracting keywords from NOTAM.

[0026] NOTAM keyword extraction: Combining the classification model with established NOTAM keyword rules and leveraging natural language processing technology, we extract key element data from NOTAM information, such as taxiway numbers, directional words, quantifiers, unit words, parking stand numbers, and reason explanations.

[0027] Keyword tagging and storage: Tags and stores keywords extracted from the NOTAM to provide data support for further spatial analysis and calculation of the impact range of the NOTAM.

[0028] Figure 4 A NOTAM category labeling chart is created. During the NOTAM classification process, NOTAM data is tagged with categories to facilitate more convenient use of the data in subsequent steps.

[0029] Figure 5 This is a sample diagram of the key element extraction result data of the NOTAM. Combining the classification results of the NOTAM and the preset keyword element information, natural language analysis technology is used to extract the key element information and related data values ​​from the NOTAM content, preparing for the next step of spatial analysis.

[0030] IV. Calculation steps for analyzing the impact range of NOTAM: Used to dynamically calculate the impact area of ​​a NOTAM by combining AMDB data and key elements of the NOTAM through PostGIS spatial analysis algorithms, and generate a common geographic information data format, including: Based on key elements of a NOTAM, combined with AMDB data, spatial analysis algorithms are used to analyze and calculate the precise impact area of ​​the NOTAM, such as a construction area, a taxiway section, or a controlled area. This module applies optimized spatial analysis algorithms and designs corresponding functions. These functions, combined with practical application scenarios, calculate the actual impact area of ​​the NOTAM and generate geographic information data in a custom format. The analysis results are converted into and stored in the universal GeoJSON format for easy use in data presentation or to provide data support for external systems.

[0031] The following examples illustrate the technical implementation process of the present invention: A NOTAM states: "Closure of Taxiway Y3 between Taxiways T4 and T6." Based on the categorized NOTAMs, a series of keyword categories are set, such as taxiway number, stand number, and location word. In this example, the key element is the taxiway number.

[0032] 1. Generate different regular expressions for different NOTAM types based on the preset keyword categories. Taking the above telegram content as an example, the generated regular expression is: ([AZ]\\d+). The corresponding taxiway number can be obtained through the following code: public class Main { public static void main(string[] args) { string text="Close taxiway Y3 between taxiways T4 and T6"; Pattern pattern=Pattern.compile(“([AZ]\\d+)”); Matcher matcher=pattern.matcher(text); while(matcher.find()) { System.out.println(matcher.group(1)); } / / Output: / / T4 / / T6 / / Y3 } } 2. Combined with the AMDB database, the spatial data related to the three taxiways T4, T6, and Y3 are queried and counted as Line1, Line2, and Line3 respectively.

[0033] 3. Use the PostGIS spatial calculation functions ST_ClosestPoint(Line1, Line3) and ST_ClosestPoint(Line2, Line3) to calculate the intersection point Point1 of taxiway T4 and taxiway Y3, and the intersection point Point2 of taxiway T6 and taxiway Y3, respectively. Then use the PostGIS spatial calculation function ST_LineSubstring(Line3, Point1, Point2) to calculate the partial data of taxiway Y3 located between Point1 and Point2, recorded as geomResult, which is the affected area described in this telegram.

[0034] 4. Use the postgis data conversion function ST_AsGeoJSON(geomResult) to convert the calculation results into geojson data. The data example is as follows: { { "type". "Feature”, "geometry":{ “type”:"Linestring", “coordinates”:[ [ 116.40630148691915, 39.50001127825618 ], [ 116,40626918281346, 39.500215778361834 ], [ 116.40615209852452, 39.50095697165819 ], [ 116.4061303386332, 39.5010947201453 ] ] } } "id":0 } 5. Then overlay the geojson data onto the map platform, and the rendering effect is as shown in the attached Figure 6 shown.

[0035] The foregoing is merely an embodiment of the present invention, and the commonly known specific technical solutions and / or features of the solutions are not described in detail herein. It should be noted that those skilled in the art may make various modifications and improvements without departing from the technical solution of the present invention, and such modifications and improvements should also be considered within the scope of protection of the present invention and will not affect the effectiveness of the implementation of the present invention or the practical application of the patent.

Claims

1. An AMDB-based intelligent classification and parsing method for taxiway-related NOTAMs, characterized by: The steps include: Step 1: Classification model training: By testing and training known sample data, a NOTAM classification model is obtained; Step 2: NOTAM classification: Automatically classify NOTAMs using a classification model; Step 3: Extract key information from the NOTAM, using natural language analysis algorithms combined with keyword category settings to extract key element attributes and related data from the NOTAM; Step 4: Analyze and calculate the impact range of the NOTAM. By using PostGIS geospatial analysis algorithms, combined with AMDB data and key elements of the NOTAM, the impact range of the NOTAM is dynamically calculated. Step 5: Overlay the impact range data of the NOTAM with the AMDB data. Overlay the impact range data of the NOTAM onto the airport scene map produced using the AMDB data and mark it with a highlighted color to show the impact of the airport.

2. The method for intelligent classification and analysis of taxiway-related NOTAMs based on AMDB according to claim 1, characterized in that: The specific calculation formula of the NOTAM classification model obtained in step 1 is as follows: P(C|X)=P(X)P(X|C)P(C) Where P(C|X) is the probability that a NOTAM belongs to category C under the condition that certain features X of the NOTAM appear, i.e., the posterior probability. P(X|C) is the probability of the feature X corresponding to the NOTAM appearing when the NOTAM is classified as category C. P(C) is the prior probability that the NOTAM belongs to category C; P(X) is the marginal probability of the occurrence of feature X of the NOTAM; Since P(X) is the same for all categories, P(X) can be ignored when classifying, and only the size of P(X|C)P(C) needs to be compared; The specific classification rule formula is as follows: C^=argmaxCP(C)∏i=1nP(xi|C) Where C^ is the predicted NOTAM category and xi is the i-th component of the NOTAM feature X.

3. The method for intelligent classification and analysis of taxiway-related NOTAMs based on AMDB according to claim 1 or 2, characterized in that: The key information extraction content of the NOTAM in step 3 specifically includes: NOTAM keyword preset: preset NOTAM keyword types according to NOTAM business rules and transmission specifications, providing a basis for extracting keywords from NOTAM; NOTAM keyword extraction: combining classification models with established NOTAM keyword rules to extract key information data from NOTAM information; Keyword marking and storage: marking and storing the keywords extracted from the navigation notice.

4. The method for intelligent classification and analysis of taxiway-related NOTAMs based on AMDB according to claim 3, characterized in that: It also includes automatic retrieval and processing steps for NOTAM data, which retrieves the latest NOTAM data in real time, classifies it using an automatic classification model, and labels the data with categories.

5. The method for intelligent classification and analysis of taxiway-related NOTAMs based on AMDB according to claim 4, characterized in that: The method also includes a real-time monitoring step for the effectiveness of the navigation notice, and controls the display and removal of graphic data related to the affected area on the map component according to the current real-time effectiveness of the navigation notice.

6. The method for intelligent classification and analysis of taxiway-related NOTAMs based on AMDB according to claim 5, characterized in that: It also includes data format conversion and storage steps, which convert and store relevant geographic spatial information data into a unified data format to generate a universal data format.

Images