Aircraft landing position prediction method and device, storage medium and program product

Abstract

The invention relates to the technical field of aviation, and particularly provides an aircraft landing position prediction method and device, a storage medium and a program product. The method comprises the following steps: acquiring flight state data of an aircraft in a preset time range before the current moment, wherein the flight state data is at least used for describing meteorological conditions of the aircraft; and predicting the flight state data by adopting a pre-trained neural network model to obtain a predicted landing position of the aircraft. The flight state data for predicting the landing position of the aircraft comprises the parameters for describing the meteorological conditions of the aircraft, so that the landing position of the aircraft is predicted by adopting the method provided by the invention; the method can adapt to the influence of changeable meteorological conditions on the aircraft landing distance in the aircraft landing process without inputting specific landing scene parameters.

Description

Technical Field

[0001] The present disclosure relates to the field of aviation technology, and in particular, to an aircraft landing position prediction method, device, storage medium, and program product. Background Art

[0002] Takeoff and landing are critical phases of flight. Due to the complex and uncertain factors of meteorological and signal environments, there are problems such as heavy workloads for pilots and high accident probabilities. Additionally, one of the challenges restricting the further development of the aviation industry is the increasing future aviation demand and the airport scheduling capabilities that are difficult to match. In response to the needs of reducing safety accidents during the aircraft landing phase and improving airspace operation efficiency, it is necessary to conduct research on landing distance prediction so that when a safety accident occurs to a pilot during the landing process, an early warning function and decision-making information support can be achieved, reducing the risk of the aircraft running off the runway.

[0003] Currently, the landing distance is derived by solving dynamic formulas, such as dimensionless equations for calculating the actual landing distance of an aircraft based on similarity theory and dimensional analysis methods. Such methods require input of specific landing scenario parameters and cannot adapt to the influence of changing meteorological conditions during the aircraft landing process on the landing distance. Summary of the Invention

[0004] In view of the above problems, the present disclosure is proposed. The present disclosure provides an aircraft landing position prediction method, device, storage medium, and program product.

[0005] According to a first aspect of the present disclosure, an aircraft landing position prediction method is provided, including:

[0006] Obtain flight state data of the aircraft within a preset time range before the current moment, where the flight state data is at least used to describe the meteorological conditions in which the aircraft is located;

[0007] Use a pre-trained neural network model to predict the flight state data to obtain the predicted landing position of the aircraft.

[0008] In addition, for the aircraft landing position prediction method according to the first aspect of the present disclosure, the flight state data includes:

[0009] Longitudinal landing distance influence parameters and lateral landing distance influence parameters;

[0010] The longitudinal landing distance influence parameters include the forward wind speed and lateral wind speed of the meteorological environment in which the aircraft is located, the current altitude of the aircraft, the altitude difference between the aircraft and the standard glide path, the current airspeed of the aircraft, the initial approach speed of the aircraft, and the difference in the forward direction between the aircraft and the standard glide path;

[0011] The lateral landing distance influencing parameters include the forward wind speed and lateral wind speed of the meteorological environment where the aircraft is located, the current altitude of the aircraft, the altitude difference between the aircraft and the standard glide path, the current airspeed of the aircraft, the starting approach speed of the aircraft, and the deviation distance between the aircraft and the runway center line.

[0012] In addition, according to the aircraft landing position prediction method of the first aspect of the present disclosure, the neural network model adopts a single-layer LSTM neural layer, and the number of LSTM neurons in the single-layer LSTM neural layer is 256.

[0013] In addition, according to the aircraft landing position prediction method of the first aspect of the present disclosure, the single-layer LSTM neural layer is connected with two fully connected layers;

[0014] Among them, the number of neurons in the first fully connected layer is 256, and the number of neurons in the second fully connected layer is 128.

[0015] In addition, according to the aircraft landing position prediction method of the first aspect of the present disclosure, the activation function of the neural network model is the tanh function.

[0016] In addition, according to the aircraft landing position prediction method of the first aspect of the present disclosure, before using the pre-trained neural network model to predict the flight state data and obtain the predicted landing position of the aircraft, it further includes:

[0017] Obtain model training data, where the model training data includes the historical flight state data of the aircraft within a set historical time range and the actual landing position corresponding to the historical flight state data, and the historical flight state data is at least used to describe the aircraft under different wind field conditions;

[0018] Use the neural network model to be trained to predict the model training data to obtain the predicted landing position;

[0019] Calculate the model loss based on the predicted landing position and the actual landing position, and use the model loss to optimize the parameters of the neural network model to be trained until the model converges.

[0020] According to the second aspect of the present disclosure, there is provided an aircraft landing position prediction device, including:

[0021] An acquisition module, configured to acquire the flight state data of the aircraft within a preset time range before the current moment, and the flight state data is at least used to describe the meteorological conditions where the aircraft is located;

[0022] A prediction module, configured to use a pre-trained neural network model to predict the flight state data to obtain the predicted landing position of the aircraft.

[0023] According to a third aspect of the present disclosure, there is provided an electronic device, including a memory, a processor, and a computer program stored on the memory, where the processor executes the computer program to implement the steps of the method described in the first aspect.

[0024] According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium, on which a computer program / instructions are stored, and when the computer program / instructions are executed by a processor, the steps of the method described in the first aspect are implemented.

[0025] According to a fifth aspect of the present disclosure, there is provided a computer program product, including a computer program / instructions, and when the computer program / instructions are executed by a processor, the steps of the method described in the first aspect are implemented.

[0026] As will be described in detail below, in the method for predicting an aircraft landing position according to an embodiment of the present disclosure, the flight state data for predicting the landing position of the aircraft includes parameters for describing the meteorological conditions in which the aircraft is located. Therefore, when using the method of the present application to predict the landing position of the aircraft, it is possible to adapt to the influence of the changing meteorological conditions during the aircraft landing process on the aircraft landing distance without inputting specific landing scenario parameters.

[0027] It should be understood that both the foregoing general description and the following detailed description are exemplary and are intended to provide further explanation of the claimed technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] By describing the embodiments of the present disclosure in more detail in conjunction with the drawings, the above and other objects, features, and advantages of the present disclosure will become more apparent. The drawings are used to provide a further understanding of the embodiments of the present disclosure, and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation to the present disclosure. In the drawings, the same reference numerals generally represent the same components or steps.

[0029] Figure 1 is a flowchart illustrating the application of the method for predicting an aircraft landing position according to an embodiment of the present disclosure.

[0030] Figure 2 is a schematic diagram illustrating the flight trajectory of an aircraft under windy conditions during the glide segment according to an embodiment of the present disclosure.

[0031] Figure 3 is a structural diagram illustrating the application of the device for predicting an aircraft landing position according to an embodiment of the present disclosure.

[0032] Figure 4 is a hardware block diagram illustrating an electronic device according to an embodiment of the present disclosure.

[0033] Figure 5It is a schematic diagram of a computer-readable storage medium according to an embodiment of the present disclosure. Detailed implementation manners

[0034] To make the objectives, technical solutions, and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure. Apparently, the described embodiments are only some of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure described and illustrated herein generally may be arranged and designed in a variety of different configurations. Therefore, the detailed description of the embodiments of the present disclosure provided herein is not intended to limit the scope of the claimed present disclosure, but merely represents selected embodiments of the present disclosure. All other embodiments obtained by those skilled in the art based on the embodiments of the present disclosure without creative efforts shall fall within the protection scope of the present disclosure.

[0035] It should be noted that like reference numerals and letters denote like items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0036] The term "and / or" in this document merely describes an association relationship and indicates that three relationships may exist. For example, A and / or B may represent: A exists alone, both A and B exist simultaneously, and B exists alone. In addition, the term "at least one" in this document means any one of multiple types or any combination of at least two of multiple types. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set composed of A, B, and C.

[0037] An aircraft running off the runway is the most common aircraft ground accident internationally. To improve the safety of civil aircraft during the landing phase and reduce the risk of the aircraft running off the runway, it is necessary to predict the landing distance required by the aircraft in the air, implement the early warning function for the aircraft running off the runway, and provide landing information support for the pilot. The prior art has problems of being unable to adapt to different landing conditions and unable to achieve dynamic prediction of the landing distance during the landing process.

[0038] In the related art, the landing distance is derived by solving the dynamic formula. For example, a dimensionless equation for calculating the actual landing distance of an aircraft is derived based on the similarity theory and dimensional analysis method. Such methods require input of specific landing scenario parameters and cannot adapt to the influence of variable meteorological conditions on the landing distance of the aircraft during the landing process.

[0039] To alleviate the technical problems existing in the related art, embodiments of the present disclosure provide an aircraft landing position prediction method, device, storage medium, and program product. In this method, the flight state data for predicting the landing position of the aircraft includes parameters for describing the meteorological conditions in which the aircraft is located. Therefore, by using the method of the present application to predict the landing position of the aircraft, it is possible to adapt to the influence of the changing meteorological conditions during the aircraft landing process on the landing distance of the aircraft without inputting specific landing scenario parameters.

[0040] To facilitate the understanding of this embodiment, first, a detailed introduction is given to an aircraft landing position prediction method disclosed in embodiments of the present disclosure. The execution subject of the aircraft landing position prediction method provided in embodiments of the present disclosure is generally an electronic device with certain computing capabilities, such as a terminal device, a server, or other processing devices. In some possible implementation manners, the aircraft landing position prediction method can be implemented by a processor invoking computer-readable instructions stored in a memory.

[0041] See Figure 1 As shown in the flowchart of the aircraft landing position prediction method provided in embodiments of the present disclosure, the method includes the following steps:

[0042] Step 101, obtain flight state data of the aircraft within a preset time range before the current moment, where the flight state data is at least used to describe the meteorological conditions in which the aircraft is located;

[0043] Step 102, use a pre-trained neural network model to predict the flight state data to obtain the predicted landing position of the aircraft.

[0044] In this embodiment, the preset time range is a time range defined or set in advance according to actual needs by a person. For example, the preset time range can be set to the first 10 s before the current moment.

[0045] According to the specifications of the International Civil Aviation Organization (ICAO), during a precision approach, an aircraft approaches and lands along the standard glide path according to the instrument landing system guidance signal. According to the definition in the report of the International Air Transport Association (IATA) in 2015, during takeoff and landing, an aircraft leaving one end or side of the runway surface on the runway surface is considered a runway deviation. It includes two types of events: (Veer off) turning and (overrun) overrun. Turning means the aircraft leaves one side of the runway. Overrun means the aircraft leaves the end of the runway. The aircraft landing process includes three stages: gliding, flaring, and rolling. In the present application, the landing position of the aircraft refers to the front-wheel landing position after the aircraft passes through the gliding and flaring stages.

[0046] In this embodiment, the meteorological conditions in which the aircraft is located include, but are not limited to, the wind field conditions in which the aircraft is located. The wind field conditions include turbulent wind and gust. Turbulent wind, also known as atmospheric turbulence, refers to the continuous random pulses superimposed on the constant wind (mean wind). It is generally considered that atmospheric turbulence is a stationary, homogeneous, ergodic and isotropic random process, and the statistical characteristics of this process do not change with time. Gust, also known as squall, is manifested as a deterministic change in wind speed. It mainly reflects the sharp change gradient of horizontal and vertical wind speeds. For example, sharp changes in wind speed will occur at the edges of rising warm air currents and descending cold air currents, mountains, cliffs, temperature change areas and the edges of storms.

[0047] In this embodiment, the landing position of the aircraft includes the longitudinal landing position and the lateral landing position of the aircraft. Among them, the longitudinal landing position of the aircraft is affected by the parameters affecting the longitudinal landing distance of the aircraft, and the lateral landing position of the aircraft is affected by the parameters affecting the lateral landing distance of the aircraft.

[0048] Among them, the parameters affecting the longitudinal landing distance include the forward wind speed and lateral wind speed of the meteorological environment in which the aircraft is located, the current altitude of the aircraft, the height difference between the aircraft and the standard glide path, the current airspeed of the aircraft, the starting approach speed of the aircraft, and the difference in the forward direction between the aircraft and the standard glide path.

[0049] The parameters affecting the lateral landing distance include the forward wind speed and lateral wind speed of the meteorological environment in which the aircraft is located, the current altitude of the aircraft, the height difference between the aircraft and the standard glide path, the current airspeed of the aircraft, the starting approach speed of the aircraft, and the deviation distance from the runway center line.

[0050] Among them, the glide path refers to the flight path of the aircraft relative to the ground when the aircraft is approaching and landing. The quality of the landing system channel environment directly determines the guiding ability of the system. The instrument landing system glide slope beacon generates a glide path by radiating horizontally polarized radio waves through an antenna based on the characteristics of a good conductor on the horizontal ground and using the mirror principle. The glide path structure of the instrument landing system is a very important technical parameter of the glide system.

[0051] The final landing point of the aircraft is affected by various factors such as the weight of the aircraft, the attitude of the aircraft, and meteorological conditions. Therefore, parameters such as the current altitude of the flight used to characterize the weight of the aircraft and the height difference between the aircraft and the standard glide path are introduced into the flight state data of the aircraft. At the same time, parameters such as the current airspeed of the aircraft and the starting approach speed of the aircraft used to characterize the attitude of the aircraft are introduced, as well as the forward wind speed and lateral wind speed characterizing meteorological conditions.

[0052] The meteorological conditions that affect the final landing point of an aircraft are mainly wind. Therefore, during the process of introducing the above-mentioned flight state data, the current wind speed and lateral wind speed in the longitudinal landing distance influence parameter and the lateral landing distance influence parameter in the flight state data are also classified by type. In this embodiment, the types of wind that affect the aircraft mainly include gusts and turbulent winds. The introduction of gusts and turbulent winds will not be elaborated here, and reference can be made to the foregoing content.

[0053] It should be understood that the current wind speed and lateral wind speed in the aircraft state data collected within a certain time range can reflect the type of wind. For example, if the current wind speed and lateral wind speed in the aircraft state data collected within a certain period of time basically do not change with time, then it can be determined that the type of wind in the aircraft state data collected during this period is turbulent wind. If the current wind speed and lateral wind speed in the aircraft state data collected within a certain period of time change rapidly with time, then it can be determined that the type of wind in the aircraft state data collected during this period is gust.

[0054] In this embodiment, when the longitudinal landing distance influence parameter and the lateral landing distance influence parameter are included in the aircraft state data, a neural network model is set for the longitudinal landing distance influence parameter and the lateral landing distance influence parameter respectively, that is, a longitudinal prediction model is set for the longitudinal landing distance influence parameter, and a lateral prediction model is set for the lateral landing distance influence parameter. The longitudinal prediction model predicts the longitudinal landing distance influence parameter to predict the longitudinal landing position of the aircraft, and the lateral prediction model predicts the lateral landing distance influence parameter to predict the lateral landing position of the aircraft. Then, by combining the longitudinal landing position and the lateral landing position, the predicted landing position of the aircraft can be obtained.

[0055] In this embodiment, the glide path refers to the flight path of the aircraft relative to the ground when the aircraft is approaching for landing. The quality of the landing system channel environment directly determines the guiding ability of the system. The instrument landing system glide slope beacon generates a glide path by radiating horizontally polarized radio waves through an antenna based on the characteristics of a good conductor on a horizontal ground and using the mirror image principle. The glide path structure of the instrument landing system is a very important technical parameter of the glide system. Please refer to Figure 2 , Figure 2 This is the flight trajectory of the aircraft in the glide segment with wind conditions shown in this application. A standard glide path is shown in this aircraft trajectory diagram.

[0056] In this embodiment, the neural network model uses a single-layer LSTM neural layer, and the number of LSTM neurons in the single-layer LSTM neural layer is 256. The single-layer LSTM neural layer is connected to two fully connected layers; among them, the number of neurons in the first fully connected layer is 256, and the number of neurons in the second fully connected layer is 128. The activation function of the neural network model is the tanh function.

[0057] In an alternative embodiment, before predicting the flight state data using a pre-trained neural network model to obtain the predicted landing position of the aircraft, the neural network model can also be trained. Therefore, before predicting the flight state data using a pre-trained neural network model to obtain the predicted landing position of the aircraft, the method may further include the following steps:

[0058] Obtain model training data, where the model training data includes the historical flight state data of the aircraft within a set historical time range and the actual landing position corresponding to the historical flight state data. The historical flight state data is at least used to describe the aircraft under different wind field conditions;

[0059] Use the neural network model to be trained to predict the model training data to obtain the predicted landing position;

[0060] Calculate the model loss based on the predicted landing position and the actual landing position, and use the model loss to optimize the parameters of the neural network model to be trained until the model converges.

[0061] In the technical solution provided in this embodiment, the flight state data for predicting the landing position of the aircraft includes parameters for describing the meteorological conditions of the aircraft. Therefore, when using the method of the present application to predict the landing position of the aircraft, it is possible to adapt to the influence of the changing meteorological conditions during the aircraft landing process on the landing distance of the aircraft without inputting specific landing scenario parameters.

[0062] For ease of understanding, the following describes the solution of the present application with a specific embodiment.

[0063] According to the specifications of the International Civil Aviation Organization (ICAO), during a precision approach, an aircraft approaches and lands along the standard glide path according to the instrument landing system guidance signal. According to the definition in the 2015 report of the International Air Transport Association (IATA), during takeoff and landing, an aircraft leaving one end or side of the runway surface is considered a runway deviation. It includes two types of events: veer off and overrun.

[0064] Veer off means that the aircraft leaves one side of the runway. Overrun means that the aircraft leaves the end of the runway. The aircraft landing process includes three stages: glide, flare, and rollout. In the present application, the landing position of the aircraft refers to the position where the front wheel touches the ground after the aircraft passes through the glide and flare stages.

[0065] To obtain the longitudinal and lateral touchdown positions of the aircraft, two databases need to be established in the present application, which are respectively used for the distance of the aircraft's lateral deviation from the runway center line and the distance of the aircraft's longitudinal deviation from the standard landing point.

[0066] The variables shown in Table 1 below will affect the longitudinal landing distance of the aircraft, and the variables shown in Table 2 below will affect the lateral landing distance of the aircraft.

[0067]

[0068] The final landing point of the aircraft is affected by various factors such as the aircraft weight, aircraft attitude, and meteorological conditions. However, the most influential factor is the wind. To improve the adaptability of this algorithm during the actual landing process of the aircraft, flight data with two common wind fields, namely gusts and turbulent winds, appearing at various stages of the aircraft landing are simulated in the aircraft landing data, as shown in Table 3.

[0069]

[0070] The database used in this algorithm is collected based on the B737-800 simulation model platform. A standard glide angle of 2.5 degrees is given during each landing approach. Starting from a height of 500 meters, an attempt is made to capture the instrument landing signal, and the approach is made along the standard glide path until the aircraft touches down and lands, collecting one set of flight simulation data.

[0071] Considering that the final landing position of the aircraft is strongly correlated with the state of the aircraft at past moments, therefore, a neural network that can extract the characteristics of time series information is considered for use in predicting the landing position of the aircraft. LSTM, short for Long Short Term Memory, is a special recurrent neural network that can analyze the input at each moment using time series.

[0072] In this application, what needs to be achieved is to predict the position of the aircraft at future moments through the state of the aircraft at past moments. Therefore, through the LSTM neural network, the input data is the time series data from t - N to t, with a length of N; the moment S is the moment when the aircraft is expected to touch down, which is defined in practice as the moment when the aircraft height is 0.5 meters or the moment when the front wheel of the aircraft touches the ground.

[0073] The neural network model used in the present invention has an input variable of 7 * N, where N takes 10. The input variables of the longitudinal prediction model and the lateral prediction model are shown in Table 1 and Table 2 respectively. A single-layer LSTM neural layer is used, and the number of LSTM neurons is 256; two fully connected layers are used after the LSTM layer. The number of neurons in the first layer is 256, and the number of neurons in the second layer is 128. The tanh function is used as the activation function, and the output variable is the touchdown position of the aircraft.

[0074] To improve the training efficiency of the neural network and eliminate the influence of the dimensions of different input variables, all flight data is normalized. The normalization method used in the present invention is Min-Max Normalization. The conversion function is as follows:

[0075]

[0076] where max is the maximum value of the sample data and min is the minimum value of the sample data. The data values are mapped to the range [0 - 1] through the conversion function.

[0077] Tables 4 and 5 are the longitudinal prediction data packet and the lateral prediction data packet respectively, where V wind_x is the forward wind speed, V wind_y is the lateral wind speed, H is the current altitude of the aircraft, DH is the altitude difference between the aircraft and the standard glide path, V air is the current airspeed of the aircraft, V app is the approach speed of the aircraft at the start. In Table 4, DX is the difference in the forward direction between the aircraft and the standard glide path. In Table 5, DX is the deviation distance between the aircraft and the runway centerline, and Y is the deviation value between the actual landing point of the aircraft and the standard landing point position.

[0078]

[0079]

[0080] This application uses four performance indicators, namely Mean Squared Error (MSE), Root-Mean-Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE), to evaluate the accuracy of the neural network model.

[0081] The specific calculation formulas are as follows:

[0082]

[0083] In the formulas: is the original value of the landing distance; y i is the predicted value of the landing distance.

[0084] In the longitudinal channel, the model has a total of 89,543 training data. Among them, 70% is used as the training set, 15% is used as the test set, and 15% is used as the validation set. Another 5,000 pieces of data are used to test the model accuracy. In the lateral channel, the model has a total of 132,066 training data. Among them, 70% is used as the training set, 15% is used as the test set, and 15% is used as the validation set. Another 5,000 pieces of data are used to test the model accuracy. The longitudinal landing distance model has MSE = 48.3950, RMSE = 13.7257, MAE = 4.7345, and MAPE = 0.2939. The MSE of the lateral deviation distance model is 13.5336, RMSE = 6.5980, MAE = 2.0196, and MAPE = 4.8759. Considering that the actual runway length is generally not less than 2,000 meters and the width is generally not less than 40 meters, it is considered that the prediction deviation of this prediction algorithm is relatively accurate.

[0085] The embodiment of the present disclosure also provides an aircraft landing position prediction device, which is used to execute the aircraft landing position prediction method provided in any one of the above embodiments. As Figure 3 shown, the device includes:

[0086] An acquisition module 31, configured to acquire flight state data of the aircraft within a preset time range before the current moment, where the flight state data is at least used to describe the meteorological conditions where the aircraft is located;

[0087] A prediction module 32, configured to use a pre-trained neural network model to predict the flight state data and obtain the predicted landing position of the aircraft.

[0088] In some embodiments, the flight state data includes:

[0089] Longitudinal landing distance influence parameters and lateral landing distance influence parameters;

[0090] The longitudinal landing distance influence parameters include the forward wind speed and lateral wind speed of the meteorological environment where the aircraft is located, the current altitude of the aircraft, the height difference between the aircraft and the standard glide path, the current airspeed of the aircraft, the starting approach speed of the aircraft, and the forward direction difference between the aircraft and the standard glide path;

[0091] The lateral landing distance influence parameters include the forward wind speed and lateral wind speed of the meteorological environment where the aircraft is located, the current altitude of the aircraft, the height difference between the aircraft and the standard glide path, the current airspeed of the aircraft, the starting approach speed of the aircraft, and the deviation distance between the aircraft and the runway center line.

[0092] In some embodiments, the neural network model uses a single-layer LSTM neural layer, and the number of LSTM neurons in the single-layer LSTM neural layer is 256.

[0093] In some embodiments, the single-layer LSTM neural layer is connected to two fully connected layers;

[0094] Among them, the number of neurons in the first fully connected layer is 256, and the number of neurons in the second fully connected layer is 128.

[0095] In some embodiments, the activation function of the neural network model is the tanh function.

[0096] In some embodiments, the device is further configured to:

[0097] Before using the pre-trained neural network model to predict the flight state data to obtain the predicted landing position of the aircraft, obtain model training data, where the model training data includes historical flight state data of the aircraft within a set historical time range and the actual landing position corresponding to the historical flight state data, and the historical flight state data is at least used to describe the aircraft under different wind field conditions;

[0098] Use the neural network model to be trained to predict the model training data to obtain a predicted landing position;

[0099] Calculate a model loss based on the predicted landing position and the actual landing position, and use the model loss to optimize the parameters of the neural network model to be trained until the model converges.

[0100] The aircraft landing position prediction device provided by the embodiments of the present disclosure and the aircraft landing position prediction method provided by the embodiments of the present disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, run, or implemented by them.

[0101] The embodiments of the present disclosure also provide an electronic device to execute the above-mentioned aircraft landing position prediction method. Please refer to Figure 4 It shows a schematic diagram of an electronic device provided by some embodiments of the present disclosure. As Figure 4 shown, the electronic device 4 includes: a processor 400, a memory 401, a bus 402, and a communication interface 403. The processor 400, the communication interface 403, and the memory 401 are connected through the bus 402; a computer program that can run on the processor 400 is stored in the memory 401, and when the processor 400 runs the computer program, it executes the aircraft landing position prediction method provided by any of the foregoing embodiments of the present disclosure.

[0102] Among them, the memory 401 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the device network element and at least one other network element is realized through at least one communication interface 403 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used.

[0103] The bus 402 can be an ISA bus, a PCI bus, an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. Among them, the memory 401 is used to store a program. After receiving an execution instruction, the processor 400 executes the program. The aircraft landing position prediction method disclosed in any implementation manner of the foregoing embodiments of the present disclosure can be applied to the processor 400 or implemented by the processor 400.

[0104] The processor 400 may be an integrated circuit chip with signal processing capabilities. In the implementation process, the steps of the above method can be completed by the integrated logic circuit in the hardware of the processor 400 or the instructions in software form. The above-mentioned processor 400 can be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present disclosure. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present disclosure can be directly embodied as being executed and completed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory 401, and the processor 400 reads the information in the memory 401 and combines its hardware to complete the steps of the above method.

[0105] The electronic device provided by the embodiments of the present disclosure and the aircraft landing position prediction method provided by the embodiments of the present disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, run or implemented by them.

[0106] The embodiments of the present disclosure also provide a computer-readable storage medium corresponding to the aircraft landing position prediction method provided in the foregoing embodiments. Please refer to Figure 5 , which shows that the computer-readable storage medium is an optical disc 30, on which a computer program (i.e., a program product) is stored. When the computer program is run by a processor, it will execute the aircraft landing position prediction method provided in any of the foregoing embodiments.

[0107] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory, or other optical and magnetic storage media, which will not be elaborated here one by one.

[0108] The computer-readable storage medium provided in the above embodiments of the present disclosure and the aircraft landing position prediction method provided in the embodiments of the present disclosure are based on the same inventive concept and have the same beneficial effects as the methods adopted, run, or implemented by the application programs stored therein.

[0109] It should be noted that:

[0110] In the specification provided here, a large number of specific details are described. However, it can be understood that the embodiments of the present disclosure can be practiced without these specific details. In some instances, well-known structures and technologies are not shown in detail so as not to obscure the understanding of this specification.

[0111] Similarly, it should be understood that, in order to streamline the present disclosure and help understand one or more of the various inventive aspects, in the foregoing description of the exemplary embodiments of the present disclosure, the various features of the present disclosure are sometimes grouped together into a single embodiment, figure, or description thereof. However, the disclosed method should not be construed as reflecting the following schematic: that the claimed present disclosure requires more features than those expressly recited in each claim. Rather, as reflected in the following claims, the inventive aspects lie in less than all the features of the single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into the detailed description, where each claim itself serves as a separate embodiment of the present disclosure.

[0112] In addition, those skilled in the art will understand that although some of the embodiments described herein include certain features included in other embodiments rather than other features, the combination of features of different embodiments is meant to be within the scope of the present disclosure and forms different embodiments. For example, in the following claims, any one of the claimed embodiments can be used in any combination.

[0113] As described above, the above are only the preferred specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present disclosure should be covered by the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. An aircraft landing position prediction method, characterized in that, Including: Obtain flight state data of the aircraft within a preset time range before the current moment, where the flight state data is at least used to describe the meteorological conditions in which the aircraft is located; Use a pre-trained neural network model to predict the flight state data to obtain the predicted landing position of the aircraft.

2. The method according to claim 1, characterized in that, The flight state data includes: Longitudinal landing distance influence parameters and lateral landing distance influence parameters; The longitudinal landing distance influence parameters include the forward wind speed and lateral wind speed of the meteorological environment in which the aircraft is located, the current altitude of the aircraft, the height difference between the aircraft and the standard glide path, the current airspeed of the aircraft, the starting approach speed of the aircraft, and the difference in the forward direction between the aircraft and the standard glide path; The lateral landing distance influence parameters include the forward wind speed and lateral wind speed of the meteorological environment in which the aircraft is located, the current altitude of the aircraft, the height difference between the aircraft and the standard glide path, the current airspeed of the aircraft, the starting approach speed of the aircraft, and the deviation distance from the runway center line.

3. The method according to claim 1, wherein The neural network model uses a single-layer LSTM neural layer, and the number of LSTM neurons in the single-layer LSTM neural layer is 256.

4. The method according to claim 3, characterized in that The single-layer LSTM neural layer is connected to two fully connected layers; Among them, the number of neurons in the first fully connected layer is 256, and the number of neurons in the second fully connected layer is 128.

5. The method according to claim 3 or 4, characterized in that, The activation function of the neural network model is the tanh function.

6. The method according to claim 1, wherein Before using the pre-trained neural network model to predict the flight state data to obtain the predicted landing position of the aircraft, it further includes: Obtain model training data, where the model training data includes the historical flight state data of the aircraft within a set historical time range and the actual landing position corresponding to the historical flight state data, and the historical flight state data is at least used to describe different wind field conditions in which the aircraft is located; Use the neural network model to be trained to predict the model training data to obtain the predicted landing position; Calculate the model loss based on the predicted landing position and the actual landing position, and use the model loss to optimize the parameters of the neural network model to be trained until the model converges.

7. An aircraft landing position prediction device, characterized in that, Including: An acquisition module for obtaining flight state data of the aircraft within a preset time range before the current moment, where the flight state data is at least used to describe the meteorological conditions in which the aircraft is located; A prediction module for using a pre-trained neural network model to predict the flight state data to obtain the predicted landing position of the aircraft.

8. An electronic device, comprising a memory, a processor, and a computer program stored on the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1-6.

9. A computer-readable storage medium having computer programs / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, the steps of the method according to any one of claims 1-6 are implemented.

10. A computer program product, comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, the steps of the method according to any one of claims 1-6 are implemented.

Images