System, method, and computer program for flight vehicle path prediction

Abstract

The present invention relates to a technology for predicting, on the basis of an artificial intelligence model, a flight path along which a flight vehicle such as a UAM will fly. According to an aspect of the present invention, a method performed by a path prediction model training device comprises the steps of: obtaining multiple pieces of training data including performance information and fuel information of at least one flight vehicle, and weather information and route deviation information based on the location of the at least one flight vehicle; preprocessing the multiple pieces of training data; and training a path prediction model to predict the path of the flight vehicle, on the basis of the multiple processed pieces of training data.

Description

Aircraft path prediction system, method and computer program

[0001] The present invention relates to a technology for predicting the flight path of an aircraft such as a UAM based on an artificial intelligence model.

[0002] In air traffic control, predicting the path of aircraft such as urban air mobility (UAM) is crucial because it can prevent accidents by detecting risks of collisions and deviations in advance. While extensive research has recently focused on using deep learning to predict paths, existing methods rely solely on aircraft track information, such as position, direction, and speed, making it difficult to discern the pilot's flight intent. Furthermore, ground control struggles to obtain detailed aircraft flight information or pilot intent due to limitations in long-distance communication. Furthermore, relying on past flight trajectories to predict future paths complicates accurate path predictions.

[0003] The purpose of the present invention is to provide a technology for predicting a flight path based on various information such as performance, weather, and flight status of an aircraft.

[0004] The purpose of the present invention is to provide a technology for predicting a flight path based on an artificial intelligence model.

[0005] According to one aspect of the present invention, a method performed by a path prediction model learning device may include: a step of acquiring a plurality of learning data including performance information of at least one aircraft, fuel information, weather information according to the position of at least one aircraft, and route deviation information; a step of performing preprocessing on the plurality of learning data; and a step of learning a path prediction model to predict the path of the aircraft based on the plurality of preprocessed learning data.

[0006] In one embodiment, the step of performing preprocessing may include filtering at least one learning data set among a plurality of learning data sets that includes route deviation information exceeding a preset route deviation threshold value, and the step of training a route prediction model may include training a route prediction model based on the filtered plurality of learning data sets.

[0007] In one embodiment, the step of performing preprocessing may normalize performance information, fuel information, weather information according to the position of at least one aircraft, and route deviation information included in the plurality of learning data.

[0008] In one embodiment, the path prediction model may be a time series data prediction model.

[0009] In one embodiment, the time series data prediction model may be a Long Short Term Memory (LSTM) or a Recurrent Neural Network (RNN).

[0010] According to another aspect of the present invention, a method performed by an aircraft path prediction device may include: a step of acquiring performance information of a target aircraft, weather information at a first location of the target aircraft, and route deviation information; and a step of predicting a route of the target aircraft from preprocessed performance information, weather information, and route deviation information based on a pre-learned route prediction model.

[0011] In one embodiment, the method further includes a step of acquiring flight plan information of a target aircraft and flight information at a first location of the target aircraft, wherein the step of acquiring performance information of the target aircraft, weather information at the first location of the target aircraft, and route deviation information may calculate route deviation information based on the flight plan information and the flight information.

[0012] In one embodiment, the performance information includes at least one of maximum takeoff weight and cruise speed, the weather information includes at least one of wind direction and wind strength, the course deviation information is a course deviation degree evaluation index (Conformance metric) including at least one of Cross-Track Error (CTE), Along-Track Error (ATE), Horizontal Error (HE), Altitude Error (AE), Euclidean Error (EE), and Direction Angle Error (DAE), and the flight information may include at least one of latitude, longitude, bearing, speed, remaining fuel, and heading direction of the target aircraft.

[0013] According to another aspect of the present invention, a path prediction model learning device may include a memory including a command; and a processor that performs a method including a step of acquiring a plurality of learning data including performance information of at least one aircraft, fuel information, weather information according to a position of at least one aircraft, and route deviation information by executing the command; a step of performing preprocessing on the plurality of learning data; and a step of learning a path prediction model to predict a path of the aircraft based on the plurality of preprocessed learning data.

[0014] According to another aspect of the present invention, there is provided a non-transitory computer-readable recording medium storing computer-executable instructions, which, when executed by a processor, cause the processor to perform a method comprising the steps of: acquiring a plurality of learning data including performance information of at least one aircraft, fuel information, weather information according to a position of the at least one aircraft, and route deviation information; performing preprocessing on the plurality of learning data; and training the path prediction model to predict the path of the aircraft based on the plurality of preprocessed learning data.

[0015] According to one aspect of the present invention, it becomes possible to predict a flight path based on various information such as performance, weather, and flight status of an aircraft.

[0016] In addition, according to another aspect of the present invention, a technology for predicting a flight path based on an artificial intelligence model is provided.

[0017] FIG. 1 is a drawing for explaining an aircraft path prediction system according to one embodiment of the present invention.

[0018] FIG. 2 is a block diagram of a path prediction model learning device according to one embodiment of the present invention.

[0019] Figure 3 is a flowchart of a path prediction model learning method according to one embodiment of the present invention.

[0020] Figure 4 is a block diagram of an aircraft path prediction device according to one embodiment of the present invention.

[0021] Figure 5 is a flowchart of a method for predicting an aircraft path according to one embodiment of the present invention.

[0022] FIG. 6 is a diagram for explaining an example of route deviation information according to one embodiment of the present invention.

[0023] FIG. 7 is a diagram for explaining path prediction by a prediction model according to one embodiment of the present invention.

[0024] FIG. 8 is a block diagram of an aircraft path prediction system according to another embodiment of the present invention.

[0025] The advantages and features of the present invention, and the methods for achieving them, will become clearer with reference to the embodiments described in detail below with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various forms. These embodiments are provided solely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the scope of the present invention is defined solely by the claims.

[0026] In describing embodiments of the present invention, specific descriptions of known functions or configurations will be omitted unless actually necessary. Furthermore, the terms described below are defined based on their functions in the embodiments of the present invention and may vary depending on the intent or custom of the user or operator. Therefore, their definitions should be based on the overall content of this specification.

[0027] The terms ‘…bu’, ‘…gi’, etc. used hereinafter mean a unit that processes at least one function or operation, which can be implemented by hardware, software, or a combination of hardware and software.

[0028] FIG. 1 is a drawing for explaining an aircraft path prediction system according to one embodiment of the present invention.

[0029] Referring to FIG. 1, an aircraft path prediction system (1000) according to another embodiment of the present invention may include a path prediction model learning device (1100) and an aircraft path prediction device (1200).

[0030] The path prediction model learning device (1100) can input learning data used to predict the path of an aircraft and train the path prediction model to predict the path of the aircraft based on the input learning data. The path prediction model learning device (1100) can transmit the learned path prediction model to the aircraft path prediction device (1200). The specific configuration and operation method of the path prediction model learning device (1100) will be described below with reference to FIGS. 2 and 3.

[0031] The flight path prediction device can acquire a learned path prediction model from a path prediction model learning device (1100). The flight path prediction device can input flight data of an aircraft whose path is to be predicted, and use the path prediction model to generate a predicted path, which is a predicted flight path of the aircraft, from the input flight data. The specific configuration and operation method of the flight path prediction device will be described below with reference to FIG. 4 and the drawings.

[0032] FIG. 2 is a block diagram of a path prediction model learning device according to one embodiment of the present invention.

[0033] Referring to FIG. 2, a path prediction model learning device (1100) according to one embodiment of the present invention may include an input unit (2100), a preprocessing unit (2200), a path prediction model learning unit (2300), and an output unit (2400).

[0034] The input unit (2100) can receive learning data used for learning a path prediction model.

[0035] In one embodiment, the training data may include performance information of at least one aircraft, position information of at least one aircraft over time, weather information, route deviation information, and the like.

[0036] In one embodiment, the performance information may refer to information about the specifications of the aircraft itself, such as the maximum take-off weight (MTOW), cruise speed, and travel distance of at least one aircraft.

[0037] In one embodiment, weather information may mean information about weather conditions such as wind direction, wind speed, and weather.

[0038] In one embodiment, the route deviation information may refer to a route deviation degree evaluation index (Conformance metric) including at least one of Cross-Track Error (CTE), Along-Track Error (ATE), Horizontal Error (HE), Altitude Error (AE), Euclidean Error (EE), and Direction Angle Error (DAE) for at least one aircraft, as illustrated in FIG. 6. In one embodiment, the learning data may further include flight information and flight plan information. Here, the flight information may refer to information regarding flight status such as latitude, longitude, direction, speed, heading direction, and remaining fuel amount of at least one aircraft. In addition, the flight plan information may refer to information such as a flight plan route, a departure point, and a destination point.

[0039] In one embodiment, the input unit (2100) can receive data from an external device (not shown, e.g., a weather station, a control tower, an aircraft, etc.) using a wired or wireless communication method.

[0040] In one embodiment, the input unit (2100) can receive commands from a user required for learning a path prediction model using a user interface or the like (not shown).

[0041] The preprocessing unit (2200) can perform preprocessing on learning data used for learning a path prediction model.

[0042] In one embodiment, the preprocessing unit (2200) can preprocess the learning data by normalizing the learning data used for learning the prediction model.

[0043] In one embodiment, if the learning data does not include route deviation information, the preprocessing unit (2200) may generate route deviation information by calculating Cross-Track Error (CTE), Along-Track Error (ATE), Horizontal Error (HE), Altitude Error (AE), Euclidean Error (EE), and Direction Angle Error (DAE) based on flight information and flight plan information. At this time, the preprocessing unit (2200) may include the generated route deviation information in the learning data.

[0044] In one embodiment, the preprocessing unit (2200) may preprocess data of a target aircraft to be input into a learned prediction model by normalizing the data to predict the path of the target aircraft.

[0045] In one embodiment, the preprocessed learning data and the data of the target aircraft can be expressed as in the following mathematical expression 1.

[0046]

[0047] Here, the information included in the preprocessed learning data and the data of the target aircraft may include latitude, longitude, altitude, bearing, speed, remaining fuel, wind direction, wind strength, CTE (Cross Track Error), ATE (Along Track Error), AE (Altitude Error), MTOW (Maximum Take-off Weight), Cruise speed, and Range (movable distance), that is, 14 pieces of information. In addition, data up to the past time t-9 based on the reference time t are each can be expressed as

[0048] Additionally, the preprocessing unit (2200) can filter the training data by filtering out data that does not correspond to a normal flight state among the data included in the training data based on a preset threshold value. This is to exclude training data that does not include data regarding a normal flight state from the training of the path prediction model.

[0049] In one embodiment, the preprocessing unit (2200) may exclude learning data that includes course deviation information greater than a preset course deviation threshold value among course deviation information included in the learning data from the learning of the path prediction model. This is because course deviation information indicates course deviation compared to the flight plan, and therefore cannot be regarded as a normal flight state if it is greater than the course deviation threshold value. At this time, course deviation threshold values ​​may be set differently for each of the Cross-Track Error (CTE), Along-Track Error (ATE), Horizontal Error (HE), Altitude Error (AE), Euclidean Error (EE), and Direction Angle Error (DAE) included in the course deviation degree evaluation index (Conformance metric).

[0050] In one embodiment, the preprocessing unit (2200) may exclude learning data containing fuel information (e.g., remaining battery capacity) that is less than a preset fuel threshold from the learning of the path prediction model. This is because if the fuel level is not sufficient to reach the destination, an emergency landing must be made at a location other than the destination, and thus cannot be considered a normal flight state.

[0051] Additionally, the preprocessing unit (2200) can standardize the units of the learning data. Specifically, if the learning data includes information in the WGS84 coordinate system that uses units such as degrees, feet, and knots (kts), the preprocessing unit (2200) can convert the information into units such as meters (m), radian, and speed (m / s).

[0052] The path prediction model learning unit (2300) can learn a path prediction model used to predict the path of a target aircraft.

[0053] In one embodiment, the path prediction model may be a time series data prediction model that processes time series data, such as a Long Short Term Memory (LSTM) or a Recurrent Neural Network (RNN).

[0054] In one embodiment, the path prediction model may be a sequence-to-sequence model consisting of two modules, an encoder and a decoder. Each of the encoder and decoder may be composed of two layers of LSTM. The encoder receives time-series input data in chronological order and then passes it through two LSTM networks to output a hidden state in the form of a compressed form of all information. The decoder may sequentially predict the result by passing the last value of the input data and the output of the encoder through two LSTM networks.

[0055] In one embodiment, the path prediction model learning unit (2300) can train the path prediction model to predict the path along which the target aircraft flies based on learning data.

[0056] In one embodiment, the path prediction model learning unit (2300) can learn the path prediction model based on preprocessed learning data.

[0057] In one embodiment, the path prediction model learning unit (2300) uses MSE loss in the learning process of the prediction model, and can perform learning with 3000 epochs, and the learning rate can be adjusted between 0.001 and 0.01.

[0058] The output unit (2400) can transmit the path prediction model to an external device (not shown) using a wired or wireless communication method.

[0059] In one embodiment, the output unit (2400) can transmit the learned path prediction model to the aircraft path prediction device (1200).

[0060] Figure 3 is a flowchart of a path prediction model learning method according to one embodiment of the present invention.

[0061] Hereinafter, the above method is described as an example performed by the path prediction model learning device (1100) illustrated in FIG. 2.

[0062] Referring to FIG. 3, in step S3100, the path prediction model learning device (1100) can obtain learning data used for learning the path prediction model.

[0063] In one embodiment, the training data may include performance information of at least one aircraft, position information of at least one aircraft over time, weather information, route deviation information, and the like.

[0064] In one embodiment, the performance information may refer to information about the specifications of the aircraft itself, such as the maximum take-off weight (MTOW), cruise speed, and travel distance of at least one aircraft.

[0065] In one embodiment, weather information may mean information about weather conditions such as wind direction, wind speed, and weather.

[0066] In one embodiment, the course deviation information may mean a course deviation degree evaluation index (Conformance metric) including at least one of Cross-Track Error (CTE), Along-Track Error (ATE), Horizontal Error (HE), Altitude Error (AE), Euclidean Error (EE), and Direction Angle Error (DAE) for at least one aircraft, as illustrated in FIG. 6.

[0067] In one embodiment, the training data may further include flight information and flight plan information. Here, the flight information may refer to information regarding the flight status of at least one aircraft, such as latitude, longitude, bearing, speed, heading, and remaining fuel. Furthermore, the flight plan information may refer to information such as the flight plan route, departure point, and destination.

[0068] In one embodiment, the path prediction model learning device (1100) can receive data from an external device (not shown, for example, a weather station, a control tower, an aircraft, etc.) using a wired or wireless communication method.

[0069] In step S3200, the path prediction model learning device (1100) can perform preprocessing on learning data used for learning the path prediction model.

[0070] In one embodiment, the path prediction model learning device (1100) can preprocess learning data by normalizing the learning data used for learning the prediction model.

[0071] In one embodiment, the path prediction model learning device (1100) can generate path deviation information by calculating Cross-Track Error (CTE), Along-Track Error (ATE), Horizontal Error (HE), Altitude Error (AE), Euclidean Error (EE), and Direction Angle Error (DAE) based on flight information and flight plan information when the learning data does not include path deviation information.

[0072] In one embodiment, the path prediction model learning device (1100) can preprocess data of a target aircraft to be input into a learned prediction model by normalizing the data to predict the path of the target aircraft.

[0073] Additionally, the path prediction model learning device (1100) can filter the learning data by filtering out data that does not correspond to a normal flight state among the data included in the learning data based on a preset threshold value. This is to exclude learning data that does not include data regarding a normal flight state from learning the path prediction model.

[0074] In one embodiment, the path prediction model learning device (1100) may exclude learning data that includes path deviation information greater than a preset path deviation threshold from among the path deviation information included in the learning data from the path prediction model learning. This is because path deviation information indicates a deviation from the flight plan, and therefore, if it is greater than the path deviation threshold, it cannot be considered a normal flight state.

[0075] In one embodiment, the path prediction model learning device (1100) may exclude learning data containing fuel information (e.g., remaining battery capacity) that is less than a preset fuel threshold from learning the path prediction model. This is because, if the fuel level is not sufficient to fly to the destination, an emergency landing must be made at a location other than the destination, and thus, it cannot be considered a normal flight state.

[0076] Additionally, the path prediction model learning device (1100) can standardize the units of learning data. Specifically, if the learning data includes information of the WGS84 coordinate system that uses units such as degrees, feet, and knots (kts), the path prediction model learning device (1100) can convert it into units such as meters (m), radian, and speed (m / s).

[0077] In step S3300, the path prediction model learning device (1100) can learn a path prediction model used to predict the path of the target aircraft.

[0078] In one embodiment, the path prediction model may be a time series data prediction model that processes time series data, such as a Long Short Term Memory (LSTM) or a Recurrent Neural Network (RNN).

[0079] In one embodiment, the path prediction model may be a sequence-to-sequence model consisting of two modules, an encoder and a decoder. Each of the encoder and decoder may be composed of two layers of LSTM. The encoder receives time-series input data in chronological order and then passes it through two LSTM networks to output a hidden state in the form of a compressed form of all information. The decoder may sequentially predict the result by passing the last value of the input data and the output of the encoder through two LSTM networks.

[0080] In one embodiment, the path prediction model learning device (1100) can train the path prediction model to predict the path along which the target aircraft flies based on learning data.

[0081] In one embodiment, the path prediction model learning device (1100) can learn the path prediction model based on preprocessed learning data.

[0082] In one embodiment, the path prediction model learning device (1100) uses MSE loss in the learning process of the prediction model, can learn with 3000 epochs, and the learning rate can be adjusted between 0.001 and 0.01.

[0083] Figure 4 is a block diagram of an aircraft path prediction device according to one embodiment of the present invention.

[0084] Referring to FIG. 4, an aircraft path prediction device (1200) according to one embodiment of the present invention may include an input unit (4100), a preprocessing unit (4200), a path prediction unit (4300), and an output unit (4400).

[0085] The input unit (4100) can receive data of a target aircraft that is the subject of flight path prediction.

[0086] In one embodiment, the data of the target aircraft may include performance information of the target aircraft, weather information based on the location of at least one aircraft, route deviation information, and the like.

[0087] In one embodiment, the data of the target aircraft may include performance information of the target aircraft, position information of at least one aircraft over time, weather information, route deviation information, and the like.

[0088] In one embodiment, the performance information of the target aircraft may mean information about the specifications of the aircraft itself, such as the maximum take-off weight (MTOW), cruise speed, and travel distance of the target aircraft.

[0089] In one embodiment, the meteorological information of the target aircraft may refer to information about meteorological conditions such as wind direction, wind strength, and weather.

[0090] In one embodiment, the course deviation information may mean a course deviation degree evaluation index (Conformance metric) including at least one of Cross-Track Error (CTE), Along-Track Error (ATE), Horizontal Error (HE), Altitude Error (AE), Euclidean Error (EE), and Direction Angle Error (DAE) for the target aircraft.

[0091] In one embodiment, the target aircraft's data may further include flight information and flight plan information. Here, the flight information may refer to information regarding the target aircraft's flight status, such as its latitude, longitude, bearing, speed, heading, and remaining fuel level. Furthermore, the flight plan information may refer to information such as the flight plan route, departure point, and destination.

[0092] In one embodiment, the input unit (4100) can receive data from an external device (not shown, e.g., a weather station, a control tower, an aircraft, etc.) using a wired or wireless communication method.

[0093] In one embodiment, the input unit (4100) may receive commands from a user necessary to predict the path of an aircraft using a user interface or the like (not shown).

[0094] Additionally, the input unit (4100) can obtain a pre-learned path prediction model from a prediction model learning device.

[0095] The preprocessing unit (4200) can preprocess the data of the target aircraft to be input into the learned prediction model by normalizing the data to predict the path of the target aircraft.

[0096] In one embodiment, the data of the target aircraft can be expressed as in the above mathematical expression 1.

[0097] Additionally, the preprocessing unit (4200) can standardize the units of data of the target aircraft. Specifically, if the training data includes information of the WGS84 coordinate system that uses units such as degrees, feet, and knots (kts), the preprocessing unit (4200) can convert it into units such as meters (m), radian, and speed (m / s).

[0098] The path prediction unit (4300) can generate an expected flight path of the target aircraft from data of the target aircraft based on a pre-learned path prediction model.

[0099] In one embodiment, the path prediction unit (4300) can predict the path of a target aircraft from preprocessed aircraft data using a learned path prediction model.

[0100] In one embodiment, the path of the target aircraft predicted by the path prediction model can be expressed as in mathematical equation 2 below.

[0101]

[0102] Prediction results by the prediction model It can contain three pieces of information, namely latitude, longitude, and altitude. As shown in Fig. 7, the data (input data) of the target aircraft up to the past time t-9 based on the reference time t When input, the prediction model outputs the latitude, longitude, and altitude of the target aircraft from time t+1 to t+5. By predicting this, it becomes possible to predict the flight path as a result.

[0103] The output unit (4400) can output the results of predicting the flight path for the target aircraft in a form that can be visually and audibly recognized by the user.

[0104] In one embodiment, the output unit (4400) can transmit the result of predicting the flight path to an external device (not shown) using a wired or wireless communication method.

[0105] Figure 5 is a flowchart of a method for predicting an aircraft path according to one embodiment of the present invention.

[0106] Hereinafter, the above method is described as an example performed by the aircraft path prediction device (1200) illustrated in FIG. 4.

[0107] Referring to FIG. 5, in step S5100, the aircraft path prediction device (1200) can obtain the path prediction model learned in the path prediction model learning device (1100) from the path prediction model learning device (1100).

[0108] In step S5200, the aircraft path prediction device (1200) can obtain data of the target aircraft.

[0109] In one embodiment, the data of the target aircraft may include performance information of the target aircraft, weather information based on the location of at least one aircraft, route deviation information, and the like.

[0110] In one embodiment, the data of the target aircraft may include performance information of the target aircraft, position information of at least one aircraft over time, weather information, route deviation information, and the like.

[0111] In one embodiment, the performance information of the target aircraft may mean information about the specifications of the aircraft itself, such as the maximum take-off weight (MTOW), cruise speed, and travel distance of the target aircraft.

[0112] In one embodiment, the meteorological information of the target aircraft may refer to information about meteorological conditions such as wind direction, wind strength, and weather.

[0113] In one embodiment, the course deviation information may mean a course deviation degree evaluation index (Conformance metric) including at least one of Cross-Track Error (CTE), Along-Track Error (ATE), Horizontal Error (HE), Altitude Error (AE), Euclidean Error (EE), and Direction Angle Error (DAE) for the target aircraft.

[0114] In one embodiment, the target aircraft's data may further include flight information and flight plan information. Here, the flight information may refer to information regarding the target aircraft's flight status, such as its latitude, longitude, bearing, speed, heading, and remaining fuel level. Furthermore, the flight plan information may refer to information such as the flight plan route, departure point, and destination.

[0115] In step S5300, the aircraft path prediction device (1200) can preprocess the data of the target aircraft to be input into the learned prediction model by normalizing the data to predict the path of the target aircraft.

[0116] In one embodiment, the data of the target aircraft can be expressed as in the above mathematical expression 1.

[0117] Additionally, the aircraft path prediction device (1200) can standardize the units of data of the target aircraft. Specifically, the preprocessing unit (4200) can convert, when the training data includes information of the WGS84 coordinate system that uses units such as degrees, feet, and knots, into units such as meters (m), radian, and speed (m / s).

[0118] In step S5400, the aircraft path prediction device (1200) can generate an expected flight path of the target aircraft from data of the target aircraft based on a pre-learned path prediction model.

[0119] In one embodiment, the aircraft path prediction device (1200) can predict the path of a target aircraft from preprocessed aircraft data using a learned path prediction model.

[0120] In one embodiment, the path of the target aircraft predicted by the path prediction model can be expressed as in the above mathematical expression 2.

[0121] FIG. 8 is a block diagram of an aircraft path prediction system according to another embodiment of the present invention.

[0122] As illustrated in FIG. 8, each of the path prediction model learning device (1100) and the aircraft path prediction device (1200) included in the aircraft path prediction system (1000) may include at least one element of a processor (8100), a memory (8200), a storage unit (8300), a user interface input unit (8400), and a user interface output unit (8500), which may communicate with each other via a bus (8600). In addition, each of the path prediction model learning device (1100) and the aircraft path prediction device (1200) included in the aircraft path prediction system (1000) may also include a network interface (8700) for connecting to a network. The processor (8100) may be a CPU or a semiconductor device that executes processing instructions stored in the memory (8200) and / or the storage (8300). The memory (8200) and storage (8300) may include various types of volatile / non-volatile memory media. For example, the memory may include ROM (8240) and RAM (8250).

[0123] The devices described above may be implemented as hardware components, software components, and / or a combination of hardware components and software components. For example, the devices and components described in the embodiments may be implemented using one or more general-purpose computers or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing instructions and responding. The processing device may execute an operating system (OS) and one or more software applications running on the operating system.

[0124] Additionally, the processing device may access, store, manipulate, process, and generate data in response to the execution of software. For ease of understanding, the processing device is sometimes described as being used alone; however, those skilled in the art will appreciate that the processing device may include multiple processing elements and / or multiple types of processing elements. For example, the processing device may include multiple processors, or a processor and a controller. Other processing configurations, such as parallel processors, are also possible.

[0125] Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing device to perform a desired operation or may, independently or collectively, command the processing device. The software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave, for interpretation by the processing device or for providing instructions or data to the processing device. The software may also be distributed over networked computer systems and stored or executed in a distributed manner. The software and data may be stored on one or more computer-readable recording media.

[0126] The above description is merely an illustrative example of the technical idea of ​​the present invention, and those skilled in the art will appreciate that various modifications and variations can be made without departing from the essential quality of the present invention. Therefore, the embodiments disclosed in this specification are intended to illustrate rather than limit the technical idea of ​​the present invention, and the scope of the technical idea of ​​the present invention is not limited by these embodiments. The scope of protection of the present invention should be interpreted by the following claims, and all technical ideas within a scope equivalent thereto should be interpreted as being included in the scope of the rights of the present invention.

Claims

1. A method performed by a path prediction model learning device, A step of acquiring a plurality of learning data including performance information of at least one aircraft, fuel information, weather information according to the location of the at least one aircraft, and route deviation information; A step of performing preprocessing on the plurality of learning data; and A step of training the path prediction model to predict the path of an aircraft based on a plurality of preprocessed learning data; A method comprising:

2. In paragraph 1, The step of performing the above preprocessing is: Filtering at least one learning data among the plurality of learning data that includes route deviation information exceeding a preset route deviation threshold, The step of training the above path prediction model is: A method for training the path prediction model based on the filtered plurality of learning data.

3. In paragraph 1, The step of performing the above preprocessing is: A method for normalizing performance information, fuel information, weather information according to the location of at least one aircraft, and route deviation information included in the plurality of learning data.

4. In paragraph 1, The above path prediction model is a time series data prediction model.

5. In paragraph 4, The above time series data prediction model is a method of LSTM (Long Short Term Memory) or RNN (Recurrent Neural Network).

6. A method performed by an aircraft path prediction device, A step of acquiring performance information of a target aircraft, weather information at a first location of the target aircraft, and route deviation information; and A step of predicting the path of the target aircraft from preprocessed performance information, weather information, and route deviation information based on the path prediction model according to Article 1; A method for predicting an aircraft path, comprising:

7. In paragraph 6, Further comprising a step of obtaining flight plan information of the target aircraft and flight information at the first location of the target aircraft, The step of obtaining performance information of the target aircraft, weather information at the first location of the target aircraft, and route deviation information is as follows: A method for calculating the route deviation information based on the above flight plan information and the above flight information.

8. In paragraph 7, The above performance information includes at least one of maximum takeoff weight and cruising speed, The above weather information includes at least one of wind direction and wind strength, The above route deviation information is a route deviation degree evaluation index (conformance metric) including at least one of CTE (Cross-Track Error), ATE (Along-Track Error), HE (Horizontal Error), AE (Altitude Error), EE (Euclidean Error), and DAE (Direction Angle Error). A method wherein the flight information includes at least one of latitude, longitude, bearing, speed, fuel level, and heading direction of the target aircraft.

9. A non-transitory computer-readable recording medium storing computer-executable instructions, The above computer-executable instructions, when executed by a processor, A step of acquiring a plurality of learning data including performance information of at least one aircraft, fuel information, weather information according to the location of the at least one aircraft, and route deviation information; A step of performing preprocessing on the plurality of learning data; and A method comprising a step of training the path prediction model to predict the path of an aircraft based on a plurality of preprocessed learning data, wherein the processor performs the method. Non-transitory computer-readable recording medium.

10. Memory containing instructions; and A processor that performs a method including: acquiring a plurality of learning data including performance information of at least one aircraft, fuel information, weather information according to the location of the at least one aircraft, and route deviation information by executing the above command; performing preprocessing on the plurality of learning data; and training a path prediction model to predict the path of the aircraft based on the plurality of preprocessed learning data; A path prediction model learning device including:

Images