Aircraft trajectory prediction method, device, terminal and storage medium

Abstract

The present invention discloses an aircraft trajectory prediction method, device, terminal and storage medium. The method acquires flight data of a target aircraft, determines the aircraft model of the target aircraft according to the flight data; determines a target trajectory prediction model corresponding to the target aircraft according to the aircraft model; and uses the target trajectory prediction model to perform trajectory prediction based on the flight data to determine the flight trajectory of the target aircraft. Since the present invention extracts the aircraft model corresponding to the target aircraft from the flight data of the target aircraft, and selects the corresponding trajectory prediction model as the target trajectory prediction model according to the aircraft model to predict the flight trajectory of the target aircraft, it can effectively solve the problem that the prior art does not perform classification training for specific aircraft models and flight dimensions, resulting in low trajectory prediction accuracy.

Description

Technical Field

[0001] The present invention relates to the field of flight safety technology, and in particular to an aircraft trajectory prediction method, device, terminal and storage medium. Background Art

[0002] In the low-altitude flight field, aircraft flying within 0-120m do not need to declare flight routes. In the absence of fixed routes, the uncertain routes may cause aircraft collisions, collisions with obstacles, or deviate from the routes and enter the no-fly zone due to various factors. In the controlled flight area of ​​120-300m, even if the aircraft has a fixed route, the trajectory will inevitably deviate, thus creating the possibility and risk of accidents.

[0003] With the development of deep learning and artificial intelligence, low-altitude aircraft trajectory prediction methods based on deep learning and artificial intelligence have made some progress. However, most of the existing deep learning methods train general models, and do not perform model classification training for specific aircraft models, flight dimensions, etc., resulting in low trajectory prediction accuracy.

[0004] Therefore, the existing technology still needs to be improved and developed. Summary of the invention

[0005] The technical problem to be solved by the present invention is that, in view of the above-mentioned defects of the prior art, a method, device, terminal and storage medium for predicting aircraft trajectories are provided, aiming to solve the problem that the prior art does not perform classification training for specific aircraft models and flight dimensions, resulting in low trajectory prediction accuracy.

[0006] The technical solution adopted by the present invention to solve the problem is as follows:

[0007] In a first aspect, an embodiment of the present invention provides a method for predicting an aircraft trajectory, wherein the method comprises:

[0008] Acquiring flight data of a target aircraft, and determining an aircraft model of the target aircraft according to the flight data;

[0009] Determine a target trajectory prediction model corresponding to the target aircraft according to the aircraft model;

[0010] The target trajectory prediction model is used to perform trajectory prediction based on the flight data to determine the flight trajectory of the target aircraft.

[0011] In one implementation method, determining a target trajectory prediction model corresponding to the target aircraft according to the aircraft model includes:

[0012] Determining whether there is a dedicated trajectory prediction model corresponding to the aircraft model;

[0013] If the dedicated trajectory prediction model corresponding to the aircraft model exists, selecting the dedicated trajectory prediction model corresponding to the aircraft model as the target trajectory prediction model;

[0014] If the dedicated trajectory prediction model corresponding to the aircraft model does not exist, a universal trajectory prediction model is selected as the target trajectory prediction model.

[0015] In one implementation method, the structure of the target trajectory prediction model includes: an input layer, a bidirectional gated recurrent unit layer, a long short-term memory network layer, a convolutional neural network layer, a fully connected layer, and an output layer.

[0016] In one implementation method, the method further includes:

[0017] Obtaining a validation set corresponding to the target trajectory prediction model, and determining a trajectory deviation threshold according to the validation set and the target trajectory prediction model;

[0018] Acquire an expected flight trajectory corresponding to the target aircraft, and determine a predicted trajectory deviation error according to the expected flight trajectory and the flight trajectory;

[0019] A flight trajectory detection result is determined according to the trajectory deviation threshold and the predicted trajectory deviation error.

[0020] In an implementation method, when the target trajectory prediction model is the dedicated trajectory prediction model, determining the trajectory deviation threshold according to the validation set and the target trajectory prediction model includes:

[0021] Using the dedicated trajectory prediction model to perform trajectory prediction based on the verification set to determine a number of first verification flight trajectories;

[0022] Determine a plurality of first trajectory deviation errors based on the verification set and the plurality of first verification flight trajectories;

[0023] A first safety factor and real-time environmental data are obtained, and the trajectory deviation threshold is determined according to each of the first trajectory deviation errors, the first safety factor, and the real-time environmental data.

[0024] In an implementation method, when the target trajectory prediction model is the general trajectory prediction model, determining the trajectory deviation threshold according to the validation set and the target trajectory prediction model includes:

[0025] Using the general trajectory prediction model to perform trajectory prediction based on the verification set, and determining a plurality of second verification flight trajectories;

[0026] Determining a plurality of second trajectory deviation errors based on the verification set and the plurality of second verification flight trajectories;

[0027] A second safety factor is obtained, and the trajectory deviation threshold is determined according to each of the second trajectory deviation errors and the second safety factor.

[0028] In one implementation method, the method further includes:

[0029] When the expected flight trajectory of the target aircraft does not exist, determining an expected flight range according to the flight data and the trajectory deviation threshold;

[0030] The flight trajectory detection result is determined according to the flight trajectory and the expected flight range.

[0031] In a second aspect, an embodiment of the present invention further provides an aircraft trajectory prediction device, wherein the aircraft trajectory prediction device comprises:

[0032] A data acquisition module, used to acquire flight data of a target aircraft, and determine the aircraft model of the target aircraft according to the flight data;

[0033] A model determination module, used to determine a target trajectory prediction model corresponding to the target aircraft according to the aircraft model;

[0034] The trajectory prediction module is used to use the target trajectory prediction model to perform trajectory prediction based on the flight data to determine the flight trajectory of the target aircraft.

[0035] In a third aspect, an embodiment of the present invention further provides a terminal, comprising a memory and one or more processors; the memory stores one or more programs; the program comprises instructions for executing any of the above-described aircraft trajectory prediction methods; and the processor is used to execute the program.

[0036] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium on which a plurality of instructions are stored, wherein the instructions are suitable for being loaded and executed by a processor to implement any of the above-mentioned aircraft trajectory prediction methods.

[0037] Beneficial effects of the present invention: The embodiments of the present invention obtain flight data of a target aircraft, determine the aircraft model of the target aircraft according to the flight data; determine a target trajectory prediction model corresponding to the target aircraft according to the aircraft model; and use the target trajectory prediction model to perform trajectory prediction based on the flight data to determine the flight trajectory of the target aircraft. Since the present invention extracts the aircraft model corresponding to the target aircraft from the flight data of the target aircraft, and selects the corresponding trajectory prediction model as the target trajectory prediction model according to the aircraft model to predict the flight trajectory of the target aircraft, it can effectively solve the problem that the prior art does not perform classification training for specific aircraft models and flight dimensions, resulting in low trajectory prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 It is a flowchart of an aircraft trajectory prediction method provided by an embodiment of the present invention.

[0040] Figure 2 It is a schematic diagram of a specific implementation flow of the aircraft trajectory prediction method provided by an embodiment of the present invention.

[0041] Figure 3 It is a schematic diagram of the internal modules of the aircraft trajectory prediction device provided by an embodiment of the present invention.

[0042] Figure 4 It is a principle block diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The present invention discloses a method, device, terminal and storage medium for predicting aircraft trajectory. In order to make the purpose, technical solution and effect of the present invention clearer and more specific, the present invention is further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0044] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0045] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as herein.

[0046] In the low-altitude flight field, aircraft flying within 0-120m do not need to declare flight routes. In the absence of fixed routes, the uncertain routes may cause aircraft collisions, collisions with obstacles, or deviate from the routes and enter the no-fly zone due to various factors. In the controlled flight area of ​​120-300m, even if the aircraft has a fixed route, the trajectory will inevitably deviate, thus creating the possibility and risk of accidents.

[0047] With the development of deep learning and artificial intelligence, low-altitude aircraft trajectory prediction methods based on deep learning and artificial intelligence have made some progress. However, most of the existing deep learning methods train general models, and do not perform model classification training for specific aircraft models, flight dimensions, etc., resulting in low trajectory prediction accuracy.

[0048] In view of the above-mentioned defects of the prior art, the present invention provides an aircraft trajectory prediction method, which obtains flight data of a target aircraft, determines the aircraft model of the target aircraft according to the flight data; determines a target trajectory prediction model corresponding to the target aircraft according to the aircraft model; and uses the target trajectory prediction model to perform trajectory prediction based on the flight data to determine the flight trajectory of the target aircraft. Since the present invention extracts the aircraft model corresponding to the target aircraft from the flight data of the target aircraft, and selects the corresponding trajectory prediction model as the target trajectory prediction model according to the aircraft model to predict the flight trajectory of the target aircraft, it can effectively solve the problem that the prior art does not perform classification training for specific aircraft models and flight dimensions, resulting in low trajectory prediction accuracy.

[0049] Exemplary methods:

[0050] like Figure 1 As shown, the method includes:

[0051] Step S100: Acquire flight data of a target aircraft, and determine the aircraft model of the target aircraft according to the flight data.

[0052] Specifically, the target aircraft includes civil aircraft, drones, helicopters, etc. The flight data of the target aircraft includes key information such as aircraft model, company, usage data, flight trajectory before the current moment, speed, altitude, attitude, etc. In this embodiment, Spark (an open source distributed computing system) is used to quickly collect, clean and format flight data to ensure data quality and consistency, and upload it to the aircraft source database. For trajectory prediction of low-altitude aircraft, Spark has the following advantages over other computing systems: it has good data processing capabilities and can process massive and complex trajectory data; through the built-in stream processing function, it can receive and process trajectory data in real time, thereby realizing real-time tracking and prediction of aircraft, and can provide early warning and decision support in time during the flight of the aircraft; it has good scalability and can dynamically increase computing resources according to demand to cope with the growth of data volume; it has fault tolerance and can automatically recover data and continue computing when a node fails, ensuring the stability of the system and the reliability of data; it has a large number of machine learning algorithms and data analysis tools, and has rich APIs (Application Programming Interfaces) and tools, which are conducive to the construction of trajectory prediction models.

[0053] like Figure 1 As shown, the method further comprises the following steps:

[0054] Step S200: determining a target trajectory prediction model corresponding to the target aircraft according to the aircraft model.

[0055] In simple terms, different aircraft models correspond to different trajectory prediction models. According to the target aircraft model, the trajectory prediction model corresponding to the target aircraft model is selected, that is, the target trajectory prediction model corresponding to the target aircraft, so as to perform model classification learning for a specific aircraft model and improve the accuracy and precision of the predicted trajectory.

[0056] In one implementation, determining a target trajectory prediction model corresponding to the target aircraft according to the aircraft model includes:

[0057] Determining whether there is a dedicated trajectory prediction model corresponding to the aircraft model;

[0058] If the dedicated trajectory prediction model corresponding to the aircraft model exists, selecting the dedicated trajectory prediction model corresponding to the aircraft model as the target trajectory prediction model;

[0059] If the dedicated trajectory prediction model corresponding to the aircraft model does not exist, a universal trajectory prediction model is selected as the target trajectory prediction model.

[0060] Specifically, Figure 2 As shown, the dedicated trajectory prediction model is a trajectory prediction model dedicated to a certain aircraft model, and different aircraft models correspond to different dedicated trajectory prediction models. The universal trajectory prediction model is a trajectory prediction model that can be used for different aircraft models or unrecognizable aircraft models. When there is a corresponding dedicated trajectory prediction model for the aircraft model, the accuracy of trajectory prediction of the dedicated trajectory prediction model is higher than that of the universal trajectory prediction model. Therefore, firstly, the corresponding dedicated trajectory prediction model is searched according to the aircraft model. If there is a dedicated trajectory prediction model corresponding to the aircraft model, the dedicated trajectory prediction model is used as the target trajectory prediction model; if there is no dedicated trajectory prediction model corresponding to the aircraft model, the universal trajectory prediction model is used as the target trajectory prediction model, thereby improving the accuracy of trajectory prediction and avoiding the problem of low prediction accuracy caused by only using a universal trajectory prediction model for trajectory prediction.

[0061] When there is no dedicated trajectory prediction model corresponding to an aircraft model, the flight data corresponding to the aircraft model is gradually collected to form a data set corresponding to the aircraft model. The dedicated trajectory prediction model corresponding to the aircraft model is obtained by training based on the data set, and the model library is expanded to enhance adaptability and specificity.

[0062] In one implementation, the structure of the target trajectory prediction model includes: an input layer, a bidirectional gated recurrent unit layer, a long short-term memory network layer, a convolutional neural network layer, a fully connected layer, and an output layer.

[0063] Specifically, depending on the aircraft model of the target aircraft, both the dedicated trajectory prediction model and the general trajectory prediction model may be the target trajectory prediction model. Therefore, the structures of the dedicated trajectory prediction model and the general trajectory prediction model include: an input layer, a bidirectional gated recurrent unit layer, a long short-term memory network layer, a convolutional neural network layer, a fully connected layer, and an output layer. Among them, the input layer is used to preprocess the data, including normalization and time series window cutting, to adapt the neural network input. The bidirectional gated recurrent unit layer is used to capture the front-to-backward dependency relationship in the time series data, and by passing the data forward and backward, the past and future context information can be obtained at the same time. The long short-term memory network layer is used to process and predict the long time interval and delay in the time series data to enhance the model's memory and utilization ability of long-term dependency information. The convolutional neural network layer is used to process the output of the long short-term memory network layer through a one-dimensional convolution layer to extract high-level features from the sequence data to help the model capture local dependency characteristics. The fully connected layer and the output layer are used to convert the output of the convolutional neural network layer into the final prediction output.

[0064] In one implementation, the training method of the dedicated trajectory prediction model and the general trajectory prediction model includes: obtaining a number of historical flight data, determining a number of dedicated data sets corresponding to aircraft models based on the number of historical flight data and the aircraft models corresponding to each historical flight data, wherein the historical flight data includes key information such as aircraft model, company, usage data, flight trajectory, speed, altitude, and attitude; using each dedicated data set to train the initial trajectory prediction model respectively to determine the dedicated trajectory prediction model corresponding to each aircraft model, wherein the structure of the trajectory prediction model is the same as that of the target trajectory prediction model; determining a general data set based on the number of historical flight data, using the general data set to train the initial trajectory prediction model to determine the general trajectory prediction model.

[0065] like Figure 1 As shown, the method further comprises the following steps:

[0066] Step S300: Using the target trajectory prediction model to perform trajectory prediction based on the flight data to determine the flight trajectory of the target aircraft.

[0067] The trained target trajectory prediction model is used to learn the flight data of the target aircraft and predict the flight trajectory of the target aircraft at the next moment. The flight data is obtained in real time, and the flight trajectory of the target aircraft is continuously updated as the flight data is continuously updated.

[0068] In one implementation, the method further includes:

[0069] Obtaining a validation set corresponding to the target trajectory prediction model, and determining a trajectory deviation threshold according to the validation set and the target trajectory prediction model;

[0070] Acquire an expected flight trajectory corresponding to the target aircraft, and determine a predicted trajectory deviation error according to the expected flight trajectory and the flight trajectory;

[0071] A flight trajectory detection result is determined according to the trajectory deviation threshold and the predicted trajectory deviation error.

[0072] Specifically, the deviation between the predicted flight trajectory and the expected flight trajectory is judged by setting a trajectory deviation threshold. Since different aircraft models correspond to different target trajectory prediction models, the validation set data corresponding to different target trajectory prediction models are also different. For the target trajectory prediction model suitable for the target aircraft, the validation set corresponding to the target trajectory prediction model is obtained. The data in the validation set is historical flight data, including key information such as aircraft model, company, usage data, flight trajectory, speed, altitude, attitude, etc. The data in the validation set is different from the data in the data set for training the target trajectory prediction model. Through the target trajectory model learning of the validation set data, several flight trajectories are predicted, and the trajectory deviation threshold is calculated based on the predicted flight trajectory and the existing flight trajectory in the validation set.

[0073] If the target aircraft has submitted an expected flight trajectory before flight, the predicted flight trajectory is compared with the expected flight trajectory to obtain the predicted trajectory deviation error between the predicted flight trajectory and the expected flight trajectory. The predicted trajectory error is compared with the trajectory deviation threshold to determine the flight trajectory detection result.

[0074] Since the flight trajectory is generated in real time, if the predicted trajectory deviation error between the flight trajectory and the expected flight trajectory is greater than the trajectory deviation threshold, the flight trajectory detection result is determined to be abnormal, triggering a deviation alarm to remind relevant personnel to take corresponding measures; if the predicted trajectory deviation error between the flight trajectory and the expected flight trajectory is less than or equal to the trajectory deviation threshold, the predicted flight trajectory is corrected according to the expected flight trajectory, and the next trajectory prediction is performed based on the flight data acquired in real time.

[0075] In one implementation, when the target trajectory prediction model is the dedicated trajectory prediction model, determining a trajectory deviation threshold according to the validation set and the target trajectory prediction model includes:

[0076] Using the dedicated trajectory prediction model to perform trajectory prediction based on the verification set to determine a number of first verification flight trajectories;

[0077] Determine a plurality of first trajectory deviation errors based on the verification set and the plurality of first verification flight trajectories;

[0078] A first safety factor and real-time environmental data are obtained, and the trajectory deviation threshold is determined according to each of the first trajectory deviation errors, the first safety factor, and the real-time environmental data.

[0079] For the dedicated trajectory prediction model, a number of first verification flight trajectories are obtained based on the prediction of the verification set using the dedicated trajectory prediction model, and each first verification flight trajectory is compared and calculated with the existing flight trajectory in the verification set to obtain a number of first trajectory deviation errors. For example, for flight data A in the verification set, trajectory prediction is performed based on flight data A using the dedicated trajectory prediction model to obtain a predicted flight trajectory B corresponding to A, and B is compared with the existing flight trajectory in A to obtain the first trajectory deviation error corresponding to B.

[0080] Considering that the general attributes of the dedicated trajectory prediction model, the corresponding aircraft model, flight mission type, flight route, performance, etc. are fixed, this embodiment focuses on the weights of the prohibited area, trajectory error tolerance, flight environment (wind speed, temperature, etc.), historical data and real-time data comparison, etc. to adjust the trajectory deviation threshold, so that the judgment interval of trajectory deviation is narrowed and the judgment is more accurate. This embodiment obtains the first safety factor and real-time environmental data, and calculates the trajectory deviation threshold according to the first safety factor, environmental data, and several first trajectory deviation errors, wherein the first safety factor is an adjustment factor, which is used to adjust the tolerance according to actual safety requirements, and the real-time environmental data is used to make dynamic adjustments according to changes in the flight environment.

[0081] Specifically, the calculation formula for determining the trajectory deviation threshold according to each first trajectory deviation error, the first safety factor and the real-time environmental data is: ;

[0082] in, is the trajectory deviation threshold of the dedicated trajectory prediction model, is the average trajectory deviation error of the dedicated trajectory prediction model on the validation set, is the standard deviation of the trajectory deviation error of the dedicated trajectory prediction model on the validation set, It is the adjustment coefficient for the dedicated trajectory model, that is, the first safety factor, which is used to adjust the tolerance according to the actual safety requirements to reflect higher prediction accuracy and safety requirements. It is a correction item or real-time environmental data, which is used to make dynamic adjustments based on the comparison results of real-time flight data and historical flight data and the changes in real-time environmental data.

[0083] In one implementation, when the target trajectory prediction model is the general trajectory prediction model, determining a trajectory deviation threshold according to the validation set and the target trajectory prediction model includes:

[0084] Using the general trajectory prediction model to perform trajectory prediction based on the verification set, and determining a plurality of second verification flight trajectories;

[0085] Determining a plurality of second trajectory deviation errors based on the verification set and the plurality of second verification flight trajectories;

[0086] A second safety factor is obtained, and the trajectory deviation threshold is determined according to each of the second trajectory deviation errors and the second safety factor.

[0087] For the general trajectory prediction model, the verification set corresponding to the general trajectory prediction model is used to predict the second verification flight trajectory, and the second trajectory deviation error is calculated based on the comparison between the second verification flight trajectory and the existing flight trajectory in the verification set, wherein the verification set of the general trajectory prediction model includes flight data of various types of aircraft models. For example, for flight data C in the verification set, the general trajectory prediction model is used to predict the flight trajectory D corresponding to C based on the flight data C, and the second trajectory deviation error corresponding to D is calculated by D and the existing flight trajectory in C.

[0088] Considering the general trajectory prediction model, the training set data comes from different aircraft, and the training data is numerous and complex, resulting in low prediction accuracy of the general trajectory prediction model. It is necessary to appropriately relax the tolerance and focus on the weight of general attributes such as general safety standards, flight mission types, aircraft drive types and performance to adjust the trajectory deviation threshold. In this embodiment, the trajectory deviation threshold is calculated according to the second safety factor and the second trajectory deviation error by obtaining a second safety factor and setting the second safety factor greater than the first safety factor.

[0089] Specifically, the calculation formula for determining the trajectory deviation threshold according to each second trajectory deviation error and the second safety factor is: ;

[0090] in, is the trajectory deviation threshold of the general trajectory prediction model, is the average trajectory deviation error of the general trajectory prediction model on the validation set, is the standard deviation of the trajectory deviation error of the general trajectory prediction model on the validation set, It is the adjustment coefficient for the general model, that is, the second safety factor, which is used to adjust the tolerance according to safety requirements.

[0091] In one implementation, the method further includes:

[0092] When the expected flight trajectory of the target aircraft does not exist, determining an expected flight range according to the flight data and the trajectory deviation threshold;

[0093] The flight trajectory detection result is determined according to the flight trajectory and the expected flight range.

[0094] Specifically, if the target aircraft is a private drone or an emergency aircraft, there may be no expected flight trajectory. When the expected flight trajectory of the target aircraft does not exist, the aircraft model corresponding to the target aircraft and the position, speed, attitude, etc. of the target aircraft at the current moment are determined based on the flight data to estimate the position range corresponding to the target aircraft at the next moment. At the same time, the expected flight range of the target aircraft is determined based on the trajectory deviation threshold and the position range. When the predicted flight trajectory is within the expected flight range, the flight trajectory detection result is determined to be normal. When the predicted flight trajectory is not within the expected flight range, the flight trajectory detection result is determined to be abnormal.

[0095] Based on the above embodiments, the present invention also provides an aircraft trajectory prediction device, such as Figure 3 As shown, the device comprises:

[0096] The data acquisition module 01 is used to acquire the flight data of the target aircraft and determine the aircraft model of the target aircraft according to the flight data;

[0097] A model determination module 02 is used to determine a target trajectory prediction model corresponding to the target aircraft according to the aircraft model;

[0098] The trajectory prediction module 03 is used to use the target trajectory prediction model to perform trajectory prediction based on the flight data to determine the flight trajectory of the target aircraft.

[0099] Based on the above embodiment, the present invention further provides a terminal, whose principle block diagram can be shown as follows: Figure 4 As shown. The terminal includes a processor, a memory, a network interface, and a display screen connected through a system bus. Among them, the processor of the terminal is used to provide computing and control capabilities. The memory of the terminal includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the terminal is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the aircraft trajectory prediction method is implemented. The display screen of the terminal can be a liquid crystal display screen or an electronic ink display screen.

[0100] Those skilled in the art will understand that Figure 4The principle block diagram shown in the figure is only a block diagram of a partial structure related to the scheme of the present invention, and does not constitute a limitation on the terminal to which the scheme of the present invention is applied. The specific terminal may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0101] In one implementation, one or more programs are stored in the memory of the terminal, and the one or more programs are configured to be executed by one or more processors, and the one or more programs include instructions for performing an aircraft trajectory prediction method.

[0102] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0103] In summary, the present invention discloses an aircraft trajectory prediction method, device, terminal and storage medium, wherein the method obtains flight data of a target aircraft, determines the aircraft model of the target aircraft according to the flight data; determines a target trajectory prediction model corresponding to the target aircraft according to the aircraft model; and uses the target trajectory prediction model to perform trajectory prediction based on the flight data to determine the flight trajectory of the target aircraft. Since the present invention extracts the aircraft model corresponding to the target aircraft from the flight data of the target aircraft, and selects the corresponding trajectory prediction model as the target trajectory prediction model according to the aircraft model to predict the flight trajectory of the target aircraft, it can effectively solve the problem that the prior art does not perform classification training for specific aircraft models and flight dimensions, resulting in low trajectory prediction accuracy.

[0104] It should be understood that the application of the present invention is not limited to the above examples. For ordinary technicians in this field, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A method for predicting an aircraft trajectory, characterized in that: The method comprises: Acquiring flight data of a target aircraft, and determining an aircraft model of the target aircraft according to the flight data; Determine a target trajectory prediction model corresponding to the target aircraft according to the aircraft model; Using the target trajectory prediction model to perform trajectory prediction based on the flight data to determine the flight trajectory of the target aircraft; The method further comprises: Obtaining a validation set corresponding to the target trajectory prediction model, and determining a trajectory deviation threshold according to the validation set and the target trajectory prediction model; Acquire an expected flight trajectory corresponding to the target aircraft, and determine a predicted trajectory deviation error according to the expected flight trajectory and the flight trajectory; Determining a flight trajectory detection result according to the trajectory deviation threshold and the predicted trajectory deviation error; When the target trajectory prediction model is a dedicated trajectory prediction model, determining the trajectory deviation threshold according to the verification set and the target trajectory prediction model includes: Using the dedicated trajectory prediction model to perform trajectory prediction based on the verification set to determine a number of first verification flight trajectories; Determine a plurality of first trajectory deviation errors based on the verification set and the plurality of first verification flight trajectories; Acquire a first safety factor and real-time environmental data, and determine the trajectory deviation threshold according to each of the first trajectory deviation errors, the first safety factor, and the real-time environmental data; The calculation formula of the trajectory deviation threshold is: ； in, is the trajectory deviation threshold, is the average value of the trajectory deviation error of the dedicated trajectory prediction model on the validation set, is the trajectory deviation error standard deviation of the dedicated trajectory prediction model on the validation set, is the first safety factor, Used to dynamically adjust according to the real-time environmental data.

2. The method for predicting the trajectory of an aircraft according to claim 1, characterized in that: Determining a target trajectory prediction model corresponding to the target aircraft according to the aircraft model includes: Determining whether there is a dedicated trajectory prediction model corresponding to the aircraft model; If the dedicated trajectory prediction model corresponding to the aircraft model exists, selecting the dedicated trajectory prediction model corresponding to the aircraft model as the target trajectory prediction model; If the dedicated trajectory prediction model corresponding to the aircraft model does not exist, a universal trajectory prediction model is selected as the target trajectory prediction model.

3. The method for predicting the trajectory of an aircraft according to claim 1, characterized in that: The structure of the target trajectory prediction model includes: an input layer, a bidirectional gated recurrent unit layer, a long short-term memory network layer, a convolutional neural network layer, a fully connected layer and an output layer.

4. The method for predicting the trajectory of an aircraft according to claim 2, characterized in that: When the target trajectory prediction model is the general trajectory prediction model, determining the trajectory deviation threshold according to the verification set and the target trajectory prediction model includes: Using the general trajectory prediction model to perform trajectory prediction based on the verification set, and determining a plurality of second verification flight trajectories; Determining a plurality of second trajectory deviation errors based on the verification set and the plurality of second verification flight trajectories; A second safety factor is obtained, and the trajectory deviation threshold is determined according to each of the second trajectory deviation errors and the second safety factor.

5. The method for predicting the trajectory of an aircraft according to claim 1, characterized in that: The method further comprises: When the expected flight trajectory of the target aircraft does not exist, determining an expected flight range according to the flight data and the trajectory deviation threshold; The flight trajectory detection result is determined according to the flight trajectory and the expected flight range.

6. An aircraft trajectory prediction device, characterized in that: The device comprises: A data acquisition module, used to acquire flight data of a target aircraft, and determine the aircraft model of the target aircraft according to the flight data; A model determination module, used to determine a target trajectory prediction model corresponding to the target aircraft according to the aircraft model; A trajectory prediction module, used to use the target trajectory prediction model to perform trajectory prediction based on the flight data to determine the flight trajectory of the target aircraft; The apparatus also includes a method for performing the following: Obtaining a validation set corresponding to the target trajectory prediction model, and determining a trajectory deviation threshold according to the validation set and the target trajectory prediction model; Acquire an expected flight trajectory corresponding to the target aircraft, and determine a predicted trajectory deviation error according to the expected flight trajectory and the flight trajectory; Determining a flight trajectory detection result according to the trajectory deviation threshold and the predicted trajectory deviation error; When the target trajectory prediction model is a dedicated trajectory prediction model, determining the trajectory deviation threshold according to the verification set and the target trajectory prediction model includes: Using the dedicated trajectory prediction model to perform trajectory prediction based on the verification set to determine a number of first verification flight trajectories; Determine a plurality of first trajectory deviation errors based on the verification set and the plurality of first verification flight trajectories; Acquire a first safety factor and real-time environmental data, and determine the trajectory deviation threshold according to each of the first trajectory deviation errors, the first safety factor, and the real-time environmental data; The calculation formula of the trajectory deviation threshold is: ； in, is the trajectory deviation threshold, is the average value of the trajectory deviation error of the dedicated trajectory prediction model on the validation set, is the trajectory deviation error standard deviation of the dedicated trajectory prediction model on the validation set, is the first safety factor, Used to dynamically adjust according to the real-time environmental data.

7. A terminal, characterized in that: The terminal includes a memory and one or more processors; the memory stores one or more programs; the program contains instructions for executing the aircraft trajectory prediction method as described in any one of claims 1-5; and the processor is used to execute the program.

8. A computer-readable storage medium having a plurality of instructions stored thereon, characterized in that: The instructions are suitable for being loaded and executed by a processor to implement the steps of the aircraft trajectory prediction method described in any one of claims 1-5 above.

Images