Aircraft collision prediction method, device, equipment, medium and program product

Abstract

The present application provides a method, device, equipment, medium and program product for predicting collision of aircraft. The method includes: obtaining aircraft signals in the low-altitude area of ​​the target area collected by the collection equipment in the target area; identifying the data of aircraft currently flying in the low-altitude area based on the aircraft signals, the aircraft data including: the identity information, type and flight data of the aircraft; obtaining the predicted flight trajectory of the aircraft within a preset time period in the future based on the aircraft data and the meteorological data of the low-altitude area; determining whether there is a target aircraft with a collision risk based on the predicted flight trajectory of the aircraft within a preset time period in the future; if there is, outputting collision warning prompt information of the target aircraft. The method of the present application solves the technical problem that the existing collision prediction methods have low prediction result accuracy, resulting in poor flight safety of the aircraft.

Description

Technical Field

[0001] The present application relates to the field of aircraft safety and control technology, and in particular to an aircraft collision prediction method, device, equipment, medium and program product. Background Art

[0002] With the rapid development of the aviation industry, the number of aircraft is increasing, the density of air traffic is increasing, and the risk of collision between aircraft is also increasing. Therefore, the development of efficient and accurate aircraft collision prediction technology is of great significance to ensure aviation safety and reduce potential accidents.

[0003] Existing collision prediction technologies use different sensor data as the basis for collision prediction, each with its own advantages and limitations. However, it is difficult for aircraft to cope with complex and changing flight environments by relying solely on sensor data for collision prediction during flight.

[0004] In order to further improve the flight safety of aircraft, it is necessary to develop a collision prediction technology that can comprehensively consider multiple situations. Summary of the invention

[0005] The present application provides an aircraft collision prediction method, device, equipment, medium and program product to solve the technical problem that the existing collision prediction method has low prediction accuracy, resulting in poor flight safety of the aircraft.

[0006] In a first aspect, the present application provides a method for predicting a collision of an aircraft, comprising:

[0007] Acquire aircraft signals in the low-altitude area of ​​the target area collected by a collection device in the target area;

[0008] Identify the aircraft data currently flying in the low-altitude area according to the aircraft signal, the aircraft data including: aircraft identity information, type and flight data; the types include: unmanned aircraft, manned aircraft, and civil aircraft; the flight data includes: positioning data of the aircraft, speed data and angle data of the aircraft;

[0009] Obtaining a predicted flight trajectory of the aircraft within a preset time period in the future based on the aircraft data and the meteorological data of the low-altitude area;

[0010] Determining whether there is a target aircraft with a collision risk based on the predicted flight trajectory of the aircraft within a preset time period in the future;

[0011] If so, the collision warning prompt information of the target aircraft is output.

[0012] In an optional implementation, obtaining the predicted flight trajectory of the aircraft within a preset time period in the future according to the aircraft data and the meteorological data of the low-altitude area includes:

[0013] Generate a flight trajectory prediction task prompt according to the aircraft data, the low-altitude meteorological data, and the flight trajectory prediction task prompt template;

[0014] The flight trajectory prediction task prompt is input into the airspace management model to obtain the predicted flight trajectory of the aircraft within a preset time period in the future; wherein the airspace management model is obtained by fine-tuning a large-scale language model trained with a general corpus using a training data set, and the training data set includes a first training data subset; the first training data subset is constructed based on historical flight trajectory data of various types of aircraft, and meteorological data corresponding to the historical flight trajectory data.

[0015] In an optional implementation, the training data set further includes at least one of the following: a second training data subset and a third training data subset, wherein the second training data subset is constructed based on historical collision data of various types of aircraft, and the third training data subset is constructed based on simulated collision data of various types of aircraft;

[0016] The determining whether there is a target aircraft with a collision risk based on the predicted flight trajectory of the aircraft within a preset time period in the future includes:

[0017] Generate a collision prediction task prompt according to the predicted flight trajectory of the aircraft within a preset time period in the future and a collision prediction task prompt template;

[0018] The collision prediction task prompt is input into the airspace management model to obtain a prediction result of whether there is a target aircraft with a collision risk.

[0019] In an optional implementation, obtaining the predicted flight trajectory of the aircraft within a preset time period in the future according to the aircraft data and the meteorological data of the low-altitude area includes:

[0020] obtaining a flight plan for the aircraft;

[0021] Acquiring a planned flight trajectory of the aircraft according to a flight plan of the aircraft;

[0022] Acquire the meteorological change data of the aircraft during flight according to the meteorological data of the low-altitude area; the meteorological change data represent the change trend of each meteorological data when the aircraft flies to each flight position;

[0023] acquiring an actual flight trajectory of the aircraft according to the aircraft data;

[0024] The predicted flight trajectory of the aircraft within a preset time period in the future is obtained according to the actual flight trajectory, the planned flight trajectory, and the weather change data.

[0025] In an optional implementation, obtaining weather change data of the aircraft during flight includes:

[0026] The meteorological data of each predicted flight trajectory at the corresponding flight time is determined, and the meteorological change data of the aircraft is generated according to each meteorological data.

[0027] In an optional implementation manner, obtaining the predicted flight trajectory of the aircraft within a preset time period in the future according to the actual flight trajectory, the planned flight trajectory, and the weather change data includes:

[0028] Determining trajectory deviation data of a route segment flown by the aircraft according to the actual flight trajectory and the planned flight trajectory;

[0029] When the trajectory deviation data is less than a preset deviation threshold and the meteorological change data is less than a preset change threshold, determining the predicted flight trajectory of the aircraft at the future preset time period according to the planned flight trajectory; or,

[0030] When the trajectory deviation data is less than a preset deviation threshold and the meteorological change data is greater than or equal to a preset change threshold, determining a predicted flight trajectory of the aircraft at the future preset time period according to the planned flight trajectory, the meteorological data and a pre-trained trajectory adjustment model; or,

[0031] When the trajectory deviation data is greater than a preset deviation threshold, the predicted flight trajectory of the aircraft at the future preset time period is determined according to the aircraft data, the meteorological data, the actual flight trajectory, and the pre-trained trajectory prediction model.

[0032] In an optional implementation, determining the predicted flight trajectory of the aircraft at the future preset time period according to the planned flight trajectory, the meteorological data, and the pre-trained trajectory adjustment model includes:

[0033] Determine the initial flight trajectory of the aircraft at the preset future time period according to the planned flight trajectory, perform vector encoding on the initial flight trajectory, and obtain an initial trajectory vector corresponding to the aircraft;

[0034] Performing vector coding processing on each parameter included in the meteorological data to obtain a meteorological vector corresponding to the aircraft;

[0035] A pre-trained trajectory adjustment model is obtained, and the initial trajectory vector and the meteorological vector are input into the trajectory adjustment model to obtain a predicted flight trajectory output by the trajectory adjustment model.

[0036] In an optional implementation, determining a predicted flight trajectory of the aircraft for a preset time period in the future according to the aircraft data, the meteorological data, the actual flight trajectory and a pre-trained trajectory prediction model includes:

[0037] Determine the historical flight trajectory of the aircraft within a preset historical duration according to the actual flight trajectory, perform vector encoding on the historical flight trajectory, and obtain a historical trajectory vector corresponding to the aircraft;

[0038] Performing vector coding processing on each parameter included in the aircraft data and the meteorological data respectively to obtain an aircraft vector and a meteorological vector corresponding to the aircraft;

[0039] A pre-trained flight trajectory prediction model is obtained, and the historical trajectory vector, the aircraft vector, and the weather vector are input into the trajectory prediction model to obtain a predicted flight trajectory output by the model.

[0040] In an optional implementation, determining whether there is a target aircraft with a collision risk includes:

[0041] If the position difference between any of the predicted flight trajectories is zero, then determining that each of the aircraft corresponding to each of the predicted flight trajectories is a target aircraft with a collision risk;

[0042] If the position difference between any of the predicted flight trajectories is not zero but less than a preset position threshold, the volume data of each of the aircraft corresponding to each of the predicted flight trajectories is obtained, and it is determined whether there is a target aircraft with a collision risk based on each of the volume data and the position difference.

[0043] In an optional implementation, the volume data includes a bounding box size of the aircraft; the bounding box size is determined according to equipment data of the aircraft;

[0044] Determining whether there is a target aircraft with a collision risk according to each of the volume data and the position difference includes:

[0045] The sum of the sizes of the bounding boxes of the aircraft is calculated. If the sum of the sizes is smaller than the position difference, the aircraft corresponding to the predicted flight trajectories are determined to be target aircraft with a collision risk.

[0046] In an optional embodiment, the aircraft data further includes a flight mission;

[0047] After determining the target aircraft that presents a collision risk, the method further includes:

[0048] An avoidance aircraft among the target aircraft is determined according to the type and flight mission of each target aircraft, as well as an avoidance flight trajectory to be executed by the avoidance aircraft within the future preset time period.

[0049] In an optional implementation, determining the avoidance aircraft among the target aircraft includes:

[0050] If the types of the target aircraft are different, determining the avoidance aircraft according to the first priority corresponding to the type;

[0051] If the types of the target aircraft are the same, obtaining the flight missions of the target aircraft;

[0052] If the flight missions of the target aircraft are different, determining the avoidance aircraft according to the second priority corresponding to the flight missions;

[0053] If the flight missions of the target aircraft are the same, the avoidance aircraft is determined according to the actual flight trajectory of each target aircraft.

[0054] In an optional implementation, determining the avoidance flight trajectory to be executed by the avoidance aircraft within the future preset time period includes:

[0055] Determining a plurality of candidate avoidance flight trajectories for the avoidance aircraft within the future preset time period according to the predicted flight trajectories of each of the target aircraft;

[0056] The candidate avoidance flights are screened according to the actual flight trajectory of the avoidance aircraft, the aircraft data and the meteorological data to obtain the avoidance flight trajectory of the avoidance aircraft within a future preset time period.

[0057] In a second aspect, the present application provides a collision prediction device for an aircraft, comprising:

[0058] An aircraft signal acquisition module, used to acquire aircraft signals in the low-altitude area of ​​the target area acquired by acquisition equipment in the target area;

[0059] An aircraft data identification module, used to identify the aircraft data currently flying in the low-altitude area according to the aircraft signal, the aircraft data including: identity information, type and flight data of the aircraft; the types include: unmanned aircraft, manned aircraft, and civil aircraft; the flight data includes: positioning data of the aircraft, speed data and angle data of the aircraft;

[0060] A predicted flight trajectory determination module, used to obtain a predicted flight trajectory of the aircraft within a preset time period in the future based on the aircraft data and the meteorological data of the low-altitude area;

[0061] A target aircraft determination module, used to determine whether there is a target aircraft with a collision risk based on the predicted flight trajectory of the aircraft within a preset time period in the future;

[0062] The collision warning prompt information output module is used to output the collision warning prompt information of the target aircraft if it exists.

[0063] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0064] The memory stores computer instructions;

[0065] The processor executes the computer instructions stored in the memory to implement the method according to the first aspect.

[0066] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, they are used to implement the method described in the first aspect.

[0067] In a fifth aspect, the present application provides a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0068] The present application provides a method for predicting collision of aircraft, comprising: obtaining an aircraft signal in a low-altitude area of ​​a target area collected by a collection device in the target area; identifying the aircraft data currently flying in the low-altitude area according to the aircraft signal, wherein the aircraft data comprises: the identity information, type and flight data of the aircraft; the types comprise: unmanned aircraft, manned aircraft, and civil aircraft; the flight data comprises: the positioning data of the aircraft, the speed data and the angle data of the aircraft; obtaining the predicted flight trajectory of the aircraft within a preset time period in the future according to the aircraft data and the meteorological data of the low-altitude area; and based on the predicted flight trajectory of the aircraft within the preset time period in the future, Predict the flight trajectory and determine whether there is a target aircraft with a collision risk; if so, output collision warning prompt information of the target aircraft; in the above technical scheme, all aircraft signals in the low-altitude area of ​​the target area are collected by multiple acquisition devices to obtain aircraft data corresponding to each aircraft signal, so as to realize collision prediction based on all aircraft data in the low-altitude area and improve the reliability of the prediction result; further, the aircraft data and multiple source data of meteorological data in the low-altitude area are integrated to realize collision prediction of aircraft in the low-altitude area and improve the accuracy of the prediction result; and early warning prompts are given to target aircraft with a collision risk, and the occurrence of collision risk is avoided by avoidance, thereby improving the flight safety of aircraft. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0070] Figure 1 An application scenario diagram of the aircraft collision prediction method provided in this application;

[0071] Figure 2 A schematic diagram of a flow chart of a method for predicting a collision of an aircraft provided in an embodiment of the present application;

[0072] Figure 3 A schematic diagram of the structure of a collision prediction device for an aircraft provided in an embodiment of the present application;

[0073] Figure 4 It is a block diagram of an electronic device shown in an embodiment of the present application.

[0074] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0075] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0076] In the technical solution of this application, the collection, storage, use, processing, transmission, provision and disclosure of information such as aircraft data or user data involved shall comply with the provisions of relevant laws and regulations and shall not violate public order and good morals.

[0077] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0078] In some optional implementations, sensors such as radar, infrared sensors, and Automatic Dependent Surveillance-Broadcast (ADS-B) are widely used in aircraft collision warning and protection technologies. These sensors can obtain real-time data on the aircraft during flight and provide basic information for aircraft collision prediction during flight. For the flight safety of aircraft, collision prediction technology has high requirements for the accuracy and reliability of sensors. In addition, meteorological conditions in low altitudes change frequently, such as sudden strong winds and turbulence, which may also affect the flight path of the aircraft, and existing collision prediction technologies often lack consideration of these dynamic environmental factors, resulting in low prediction accuracy.

[0079] The aircraft collision prediction method provided in the present application is intended to solve the above technical problems of the prior art. Specifically, in the present application, by using a collection device including a plurality of sensors to collect aircraft signals corresponding to all aircraft flying in the low-altitude area in the target area, collision prediction of various types of aircraft flying in the low-altitude area is achieved; further, by combining the aircraft data corresponding to each aircraft signal and the multi-source data of meteorological data, collision prediction data is comprehensively realized to provide the accuracy and reliability of the prediction results. In this way, the technical problems in the prior art that collision prediction relying solely on sensor data is difficult to deal with complex types of aircraft and the accuracy of the obtained prediction results is low can be solved, and the technical effect of improving the accuracy of collision result prediction and flight safety can be achieved.

[0080] Figure 1An application scenario diagram of the aircraft collision prediction method provided in this application. The technical solution provided in this application can be applied to collision prediction of various types of aircraft flying in a preset airspace. In the development of the aviation transportation industry, the types and numbers of aircraft continue to increase, the density of air traffic is increasing, and the risk of collision between aircraft also increases. By collecting aircraft data of all aircraft in flight and meteorological data of the flight airspace, collision prediction of aircraft flying in the airspace can be achieved, which can reduce the probability of collision risk of aircraft during flight and improve flight safety.

[0081] See also Figure 1 The above application scenario involves multiple acquisition devices 11 and data processing devices 12; wherein the acquisition device 11 includes but is not limited to radar sensors, infrared sensors, ADS-B devices and remote identification (RID) devices. The data processing device can be an electronic device or a server with data receiving and data processing functions.

[0082] Specifically, the collection device 11 can be set in the target area where collision prediction is required, and through signal transmission with the onboard device of the aircraft, it can collect aircraft signals in the low altitude area of ​​the target area, and transmit the obtained aircraft signals to the data processing device 12.

[0083] The data processing device 12 processes the received aircraft signal to obtain processed aircraft data; wherein the aircraft data may include: aircraft identity information, type and flight data; types include: unmanned aircraft, manned aircraft, and civil aircraft; flight data includes: aircraft positioning data, aircraft speed data and angle data; at the same time, meteorological data of the low altitude area in the target area can also be collected based on the aircraft data; then, the predicted flight trajectory of the aircraft within a preset time period in the future is predicted based on the aircraft data and the meteorological data, and whether there is a target aircraft with a collision risk is determined based on the predicted flight trajectory; optionally, if so, a collision warning prompt information of the target aircraft is output.

[0084] In the above scheme, all aircraft signals in the low-altitude area of ​​the target area are collected through various collection devices to obtain aircraft data corresponding to each aircraft signal, so as to realize collision prediction based on all aircraft data in the low-altitude area and improve the reliability of the prediction results; further, the aircraft data and multiple source data of meteorological data in the low-altitude area are integrated to realize collision prediction of aircraft in the low-altitude area and improve the accuracy of the prediction results; and early warning prompts are given to target aircraft with collision risks, and the occurrence of collision risks is avoided through avoidance, thereby improving the flight safety of aircraft.

[0085] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0086] Figure 2 The flowchart of a method for predicting a collision of an aircraft provided in an embodiment of the present application is shown in FIG. The method can be executed by a collision prediction device of an aircraft, which can be a server or an electronic device. The following description will be made using an electronic device as an example. The method in this embodiment can be implemented by software, hardware, or a combination of software and hardware, such as Figure 2 As shown, the method comprises the following steps:

[0087] S210: Acquire aircraft signals in the low-altitude area of ​​the target area collected by a collection device in the target area.

[0088] In this embodiment, the target area can be understood as any geographical area where an aircraft can fly, including but not limited to urban airspace, mountain airspace, plain airspace, coastal airspace, and ocean airspace. Optionally, the target area can be highly divided according to the flight altitude of the aircraft to obtain different sub-airspaces. According to the flight parameters of different aircraft, various types of aircraft can be controlled to fly in different sub-airspaces to facilitate flight control.

[0089] In some exemplary embodiments, the target area may be divided into high-altitude airspace and low-altitude airspace. Low-altitude airspace generally refers to airspace with a flight altitude below FL245 (approximately 24,500 feet, 7,500 meters); the specific altitude limit may vary from region to region and is not limited thereto.

[0090] The acquisition device can be understood as a device used to collect aircraft signals generated by an aircraft during flight. For example, it can include but is not limited to radar sensors, infrared sensors, ADS-B devices, and RID devices. The above-mentioned acquisition device transmits signals with the onboard devices in the aircraft to collect the aircraft signals corresponding to the aircraft. For example, the aircraft signal can be a radar signal, an ADS-B broadcast signal, a transponder signal, etc.

[0091] Specifically, the RID acquisition device receives the remote identification (RID) signal of the drone, obtains the signal of the unmanned aircraft (such as a drone), and obtains the identity information, location, altitude, speed and other data of the unmanned aircraft.

[0092] ADS-B acquisition devices can collect and send signals from manned aircraft and civil aircraft. Specifically, they can obtain the identity information, location, altitude, speed and other data of manned aircraft and civil aircraft by receiving Automatic Dependent Surveillance-Broadcast (ADS-B) signals.

[0093] In this solution, the RID collection device can also be used to send unmanned aircraft signals; the ADS-B collection device can also be used to send manned aircraft signals, as well as civil aircraft signals; accordingly, the RID collection device in the collection equipment can also be used to receive manned aircraft signals, as well as civil aircraft signals, collected by the ADS-B collection device, and to send manned aircraft signals, as well as civil aircraft signals.

[0094] Specifically, the RID collection device receives manned aircraft signals and civil aircraft signals collected by the ADS-B collection device through communication with the ADS-B collection device, and sends these signals to the data transmission layer.

[0095] On the basis of the above, the ADS-B collection device is also used to receive unmanned aircraft signals collected by the RID collection device and send unmanned aircraft signals.

[0096] Specifically, the ADS-B collection device receives the collected unmanned aerial vehicle signals through communication with the RID collection device, and sends these signals to the data transmission layer.

[0097] Among them, the intercommunication function between the RID collection device and the ADS-B collection device can enhance the flexibility and data redundancy of the system, and improve the reliability of the system and data accuracy.

[0098] In an optional implementation, the process of acquiring an aircraft signal in flight by means of a collection device may be as follows: an ADS-B OUT device (transmitter) on the aircraft acquires flight data such as the aircraft's position (latitude and longitude), speed, heading, altitude, call sign (such as ICAO code) through an onboard GPS system and other related sensors, and encodes the flight data into a standard ADS-B message format in the onboard ADS-B OUT device, and prepares to broadcast it via a radio frequency; an ADS-B IN system (receiver) on the ground or in the air receives the ADS-B signal broadcast from the aircraft through its equipped ADS-B device, thereby acquiring the aircraft signal of the aircraft.

[0099] In some other implementations, the aircraft signal can be obtained by receiving the electromagnetic wave signal transmitted to the aircraft and reflecting the received electromagnetic wave signal through a laser radar sensor, and the electromagnetic field signal reflected by the aircraft can be collected through high-precision sensors such as optical fiber electric field sensors and optical fiber magnetic field sensors to obtain the aircraft signal. It should be noted that the implementation method of obtaining the aircraft signal through the collection device is only an exemplary description of the technical solution of the present application, and is not intended to limit the technical solution of the present application.

[0100] S220. Identify data of aircraft currently flying in the low altitude area according to aircraft signals.

[0101] In this embodiment, the aircraft signals collected by the acquisition device can be subjected to signal analysis processing to obtain corresponding aircraft data. The aircraft data may include, but is not limited to: aircraft identity information, type and flight data; types include: unmanned aircraft, manned aircraft, and civil aircraft; flight data includes: aircraft positioning data, aircraft speed data and angle data.

[0102] The received radar signals can be analyzed to extract the distance, direction, altitude and identity information of the aircraft; the aircraft's location information (latitude and longitude, altitude), speed, heading, identification code (such as ICAO address), etc. can also be extracted by decoding the ADS-B broadcast signals.

[0103] It can be explained that since multiple sensors collect multiple signals at the same time, there may be redundant information after analyzing the aircraft signals collected by the collection equipment. The data from different collection devices can be fused and the redundant information after fusion can be eliminated to improve the accuracy and completeness of the data. In addition, the technical effect of improving data calculation efficiency can be achieved by reducing redundant data.

[0104] It should also be noted that in order to ensure the accuracy of the data processing results, the time of the acquisition equipment can be synchronized in advance to ensure that the data from different devices are processed under the same time reference to avoid time errors.

[0105] S230: Obtain the predicted flight trajectory of the aircraft within a preset time period in the future based on the aircraft data and the meteorological data of the low-altitude area.

[0106] In this embodiment, before performing collision prediction, flight trajectory prediction may be performed in advance for each aircraft flying in the low altitude area, and whether there is a collision risk may be determined based on the obtained predicted flight trajectory.

[0107] Optionally, most of the existing related technologies achieve aircraft flight trajectory prediction by integrating aircraft data collected by collection equipment. However, during the implementation of the above technology, the accuracy of collision prediction based on aircraft data collected by the equipment may be affected due to frequent changes in meteorological conditions in low-altitude areas, resulting in reduced flight safety.

[0108] In order to improve the accuracy of the prediction results, this embodiment may use meteorological data as one of the reference data for prediction when predicting the flight trajectory, thereby improving the accuracy of the predicted flight trajectory, thereby improving the accuracy of the collision prediction results.

[0109] Before making a flight trajectory prediction, you can obtain the weather data of the low-altitude area where the aircraft is flying. Optionally, you can access a website that provides weather data acquisition services and obtain the weather data of the low-altitude area where the aircraft is flying based on the weather data reading service provided by the website; of course, you can also collect weather data yourself through the weather data detection device pre-built in the target area and the aircraft to obtain weather data.

[0110] On this basis, any prediction method can be used to process aircraft data and meteorological data to achieve aircraft flight trajectory prediction.

[0111] It can be explained that meteorological data includes but is not limited to wind speed, wind direction data, temperature, air pressure data, as well as precipitation, visibility and other meteorological data.

[0112] In some optional implementations, the flight trajectory prediction method may include pre-established mathematical models, such as Kalman filtering, particle filtering, hidden Markov model, generalized pseudo-Bayes, interactive multi-model algorithm, etc. Optionally, a machine learning model may be used to implement flight trajectory prediction; for example, regression models such as local weighted linear regression and local weighted polynomial regression; another example is a neural network model pre-trained with historical flight trajectories, and a large-scale language model pre-trained with general corpus and fine-tuned with historical flight trajectories.

[0113] S240. Determine whether there is a target aircraft with a collision risk based on the predicted flight trajectory of the aircraft within a preset time period in the future.

[0114] In this embodiment, on the basis of obtaining the predicted flight trajectory of each aircraft, it is possible to determine whether each aircraft has overlapping flight trajectories within a preset time period in the future, or whether there are flight trajectories with deviations less than a preset threshold, and then obtain a prediction result of whether each aircraft will have a collision risk. Optionally, if any two or more aircraft have flight trajectories of the above situation within a preset time period in the future, it means that the two or more aircraft are target aircraft with a collision risk. Otherwise, it means that each aircraft in the low-altitude area will not be a target aircraft with a collision risk within a preset time period in the future.

[0115] S250: If so, output collision warning information of the target aircraft.

[0116] In this embodiment, when it is determined according to the above implementation that there is a target aircraft with a collision risk in the low airspace where the aircraft is flying within a preset time in the future, collision warning prompt information of the target aircraft is generated, and the above information is broadcast to the target aircraft, or all aircraft in the low airspace, so that each aircraft executes an avoidance strategy during the flight, reduces the risk of collision, and improves the flight safety of the aircraft. Optionally, the collision warning prompt information may include but is not limited to the target aircraft predicted to have a collision risk, the collision location, and the collision time.

[0117] The present application provides a method for predicting the collision of aircraft, including: obtaining aircraft signals in the low-altitude area of ​​the target area collected by a collection device in the target area; identifying aircraft data currently flying in the low-altitude area according to the aircraft signals, the aircraft data including: aircraft identity information, type and flight data; types including: unmanned aircraft, manned aircraft, and civil aircraft; flight data including: aircraft positioning data, aircraft speed data and angle data; obtaining a predicted flight trajectory of the aircraft within a preset time period in the future according to the aircraft data and the meteorological data of the low-altitude area; determining whether to Target aircraft with collision risk; if so, output collision warning prompt information of the target aircraft; in the above technical scheme, all aircraft signals in the low-altitude area of ​​the target area are collected by multiple acquisition devices, and the aircraft data corresponding to each aircraft signal is obtained, so that collision prediction can be made based on all aircraft data in the low-altitude area, thereby improving the reliability of the prediction results; further, the aircraft data and multiple source data of meteorological data in the low-altitude area are integrated to realize collision prediction of aircraft in the low-altitude area, thereby improving the accuracy of the prediction results; and warning prompts are given to target aircraft with collision risk, and the occurrence of collision risk is avoided by avoidance, thereby improving the flight safety of aircraft.

[0118] In air transportation, some types of aircraft, such as unmanned aircraft (consumer drones), can perform flight missions on their own without reporting flight plans to air traffic control departments before flying in pre-divided flight areas. Based on this, the flight trajectory of the aircraft within a preset time period in the future can be predicted through the aircraft data generated by the aircraft when flying on the flown route and the meteorological data corresponding to the flown route, and the predicted flight trajectory of the aircraft within a preset time period in the future can be obtained.

[0119] In some embodiments, the specific process of this embodiment for obtaining the predicted flight trajectory of the aircraft within a preset time period in the future based on the aircraft data and the meteorological data in the low-altitude area may include but is not limited to: generating a flight trajectory prediction task prompt based on the aircraft data, the meteorological data in the low-altitude area, and the flight trajectory prediction task prompt template; inputting the flight trajectory prediction task prompt into the airspace management model to obtain the predicted flight trajectory of the aircraft within a preset time period in the future.

[0120] The airspace management model is obtained by fine-tuning a large-scale language model trained using a general corpus using a training data set, and the training data set includes a first training data subset; the first training data subset is constructed based on historical flight trajectory data of various types of aircraft, and meteorological data corresponding to the historical flight trajectory data.

[0121] The general corpus mentioned above can be any general domain corpus that can be used for large-scale language model training. Large-scale language models trained with general corpus can handle a variety of natural language tasks, such as text summarization, keyword extraction, text prediction, data reasoning, etc., and can process input prediction tasks and output task prediction results.

[0122] By using the first training data subset to fine-tune the large-scale language model and deepening the large-scale language model's learning and cognition of knowledge related to flight trajectory prediction, the large-scale language model can accurately and quickly predict the flight trajectory when receiving a flight trajectory prediction task and output the prediction result.

[0123] It should be noted that this embodiment does not limit the specific architecture and number of parameters of the above-mentioned large-scale language model, nor the method of using the training dataset to fine-tune the large-scale language model trained with the general corpus.

[0124] In this embodiment, the prompt refers to a text prompt input into the large-scale language model when using the large-scale language model, which includes aircraft data corresponding to the aircraft whose flight trajectory is to be predicted, meteorological data, the range of prediction, and prompt information about how to predict and generate prediction results, so that the large-scale language model can better understand the prediction task it is processing and output the expected trajectory prediction results.

[0125] In this embodiment, the flight trajectory prediction task prompt template can be understood as a text prompt to assist in data input, and can also be interpreted as a prompt box for generating a flight trajectory prediction task prompt. The prompt box displays a pre-generated template text, as well as a prompt for the input position and content, so that after the user or data processing device that needs to perform trajectory prediction inputs data, a flight trajectory prediction task prompt corresponding to the prediction task can be generated.

[0126] Exemplarily, the flight trajectory prediction task prompt template may include the following content:

[0127] “The predicted flight trajectory of the aircraft when it flies in low altitude for a preset period of time in the future, predicted based on the input data;

[0128] Among them, aircraft data: ${inputA};

[0129] Weather data: ${inputB};

[0130] Prediction scope: ${inputC};

[0131] Output in the form of a trajectory graph. "

[0132] It can be explained that each ${xxxx} position in the flight trajectory prediction task prompt template is used to add the basic data required for flight trajectory prediction in this embodiment; in other words, by inputting each trajectory data required for prediction in the above positions, the flight trajectory prediction task prompt can be obtained. Specifically, the ${inputA} position is used to input aircraft data; the ${inputB} position is used to input meteorological data; and the ${inputC} position is used to input the duration of the prediction.

[0133] In some optional embodiments, the duration may be pre-set; for example, the template content may initially input "the predicted flight trajectory of the aircraft when flying in low altitude in the next 5 minutes based on the input data"; in this case, subsequent input content may not include the prediction range.

[0134] It can be understood that when generating corresponding prompts below, positions similar to ${xxxx} in the prompt template need to be replaced with corresponding parameter values ​​in the prediction task, and the specific replacement can be performed according to the actual situation.

[0135] Based on the generated flight trajectory prediction task prompts, they are input into the airspace management model generated after fine-tuning the large-scale language model, so as to realize flight trajectory prediction processing according to the input prompts, and output the predicted flight trajectory in the form of a template.

[0136] Based on the large-scale language model trained with general corpus obtained in the above implementation, the large-scale language model is fine-tuned using the second training data subset and the third training data subset, and the obtained airspace management model can also be used to perform collision prediction tasks. The second training data subset is constructed based on historical collision data of various types of aircraft, and the third training data subset is constructed based on simulated collision data of various types of aircraft.

[0137] It is understandable that since the amount of sample data generated by real aircraft collisions is relatively small, in order to enable the large-scale language model to deepen its learning and cognition of knowledge about collision prediction, simulated collision data can be constructed in advance, and the use of simulated data can make up for the small amount of sample data corresponding to real collisions. In this way, the large-scale language model trained with the second training data subset and the third training data subset can accurately and quickly process the collision prediction tasks input into the large model and output the prediction results.

[0138] In some embodiments, the specific process of this embodiment implementing the target aircraft at risk of collision based on the predicted flight trajectory of the aircraft within a preset time period in the future to determine whether there is a collision risk may include but is not limited to: generating a collision prediction task prompt based on the predicted flight trajectory of the aircraft within a preset time period in the future and a collision prediction task prompt template; inputting the collision prediction task prompt into the airspace management model to obtain a prediction result of whether there is a collision risk of the target aircraft.

[0139] In this implementation manner, the collision prediction task prompt template can be understood as a text prompt template that assists in generating a collision prediction task prompt.

[0140] For example, the collision prediction task prompt template may include the following content:

[0141] “Based on the input data, it is predicted whether there is a target aircraft with a collision risk when the aircraft flies in low altitude for a preset time in the future;

[0142] Among them, the predicted flight trajectory of the aircraft: ${inputA};

[0143] Output in the form of a trajectory graph. "

[0144] It should be understood that the aircraft data and predicted flight trajectory that can be input in the above-mentioned flight trajectory prediction task prompt template and collision prediction task prompt template can be data corresponding to multiple aircraft respectively, and there is no limit on the number of inputs.

[0145] Based on the generated collision prediction task prompts, they are input into the airspace management model generated after fine-tuning the large-scale language model to realize collision prediction processing according to the input prompts, and the prediction results are output in the form of the template after the prediction processing is completed.

[0146] Since large-scale language models have stronger learning and generalization capabilities, they can capture more subtle changes and trends in aircraft flight trajectory prediction and collision prediction, thereby improving the accuracy of predictions. Large-scale language models that have been fine-tuned using historical flight trajectory data can achieve more dynamic adjustments to the prediction process of flight trajectory prediction and collision prediction, making prediction results more accurate, greatly reducing the risk of aircraft collisions and ensuring flight safety.

[0147] In the aviation transportation industry, before the above-mentioned unmanned aircraft want to fly in the pre-divided control area, and before manned aircraft and medium and large aircraft such as civil aircraft fly in the suitable flight area or the control area, they need to report the flight plan to the air traffic management department in advance for flight approval, and perform the flight mission according to the planned flight trajectory approved during the flight. However, due to the high density of air traffic and the complex flight environment in low-altitude airspace, such as frequent weather changes, some aircraft may perform obstacle avoidance operations during flight, which may cause the trajectory of the aircraft to deviate during flight. In this case, in order to ensure the flight safety of the aircraft in the subsequent flight process, the flight trajectory of the aircraft can be predicted, and the collision prediction of the aircraft can be performed based on the predicted flight trajectory.

[0148] In this embodiment, in the process of predicting the flight trajectory of the arrayer, in addition to using aircraft data and meteorological data as basic information for prediction, the planned flight trajectory approved by the aircraft before the flight and the actual flight trajectory during the flight can also be combined to realize the flight trajectory prediction of the aircraft, so as to ensure the accuracy of the trajectory prediction results, and thus ensure the reliability of the collision prediction results.

[0149] Specifically, the specific process of obtaining the predicted flight trajectory of an aircraft within a preset time period in the future according to the aircraft data and the meteorological data in the low-altitude area implemented in this embodiment may include but is not limited to: obtaining the flight plan of the aircraft; obtaining the planned flight trajectory of the aircraft according to the flight plan of the aircraft; obtaining the meteorological change data of the aircraft during the flight according to the meteorological data in the low-altitude area; the meteorological change data characterizes the change trend corresponding to each meteorological data when the aircraft flies to each flight position; obtaining the actual flight trajectory of the aircraft according to the aircraft data; obtaining the predicted flight trajectory of the aircraft within a preset time period in the future according to the actual flight trajectory, the planned flight trajectory, and the meteorological change data.

[0150] It can be explained that the flight plan can be understood as the flight plan reported by the aircraft when the flight activity is approved. For example, the flight plan includes but is not limited to: flight time, location, altitude, speed, route and other information.

[0151] Specifically, the operators of unmanned aircraft, as well as the pilots of manned aircraft and civil aircraft, shall send the pre-designated flight plan to the air traffic management department for reporting and approval before the flight; the air traffic management department shall generate the planned flight trajectory of the aircraft based on the flight airspace in the received flight plan and the airspace traffic corresponding to the planned flight time, and shall provide feedback on the planned flight trajectory so that the operators of unmanned aircraft, as well as the pilots of manned aircraft and civil aircraft, can control the aircraft to perform flight missions based on the planned flight trajectory.

[0152] In order to further improve the accuracy of flight trajectory prediction, the meteorological data in the low-altitude area where the aircraft is flying can be analyzed for meteorological changes to obtain the meteorological change data of the aircraft during flight. It can be explained that the meteorological change data characterizes the change trend of each meteorological data when the aircraft flies to each flight position. At the same time, based on the aircraft data collected during the flight, the trajectory of the aircraft is analyzed to obtain the actual flight trajectory of the aircraft during the flight. On this basis, according to the actual flight trajectory, the planned flight trajectory, and the meteorological change data, the predicted flight trajectory of the aircraft within a preset time in the future is obtained; in this way, the trajectory of the aircraft is predicted by combining multiple data to further improve the accuracy of the flight trajectory prediction results.

[0153] Exemplarily, the process of obtaining the actual flight trajectory of an aircraft based on aircraft data may include, but is not limited to: for any moment, calling a pre-generated position calculation algorithm, substituting the positioning data, aircraft speed data and angle data at that moment as calculation parameters into the called position calculation algorithm for calculation, and obtaining the actual flight position at that moment. It is understandable that the position calculation is performed in the above manner at preset intervals to obtain the actual flight position corresponding to each flight moment, and a position set is generated for each flight moment and the actual flight position corresponding to each flight moment to obtain the actual flight trajectory of the aircraft. It can be explained that the above-mentioned method of obtaining the actual flight trajectory is only an exemplary implementation of the technical solution of the present application, and is not intended to limit the technical solution of the present application.

[0154] Specifically, in this embodiment, the specific process of obtaining the weather change data of the aircraft in real time during the flight may include: determining the weather data of each predicted flight trajectory at the corresponding flight time, and generating the weather change data of the aircraft based on the weather data.

[0155] During the flight of the aircraft, the meteorological data corresponding to the flight position of the aircraft can be read at preset intervals, and the obtained meteorological data can be analyzed to obtain the meteorological change data of the aircraft during the flight.

[0156] Exemplarily, when meteorological data corresponding to multiple flight times are obtained, the difference in meteorological data changes between any two adjacent flight times can be calculated, and each meteorological data change difference can be generated into a data set in time series. This set reflects the change trend of each meteorological data when the aircraft flies to each flight position.

[0157] During the flight, if the meteorological data remains stable, that is, the change trend is not obvious, the aircraft can fly according to the planned flight trajectory. Therefore, when predicting the flight trajectory, only the planned flight trajectory can be used as a reference to improve the prediction efficiency. Optionally, if the meteorological data changes suddenly during the flight, it will affect the aircraft data of the aircraft during the flight, which will in turn affect its actual flight trajectory. For example, reduced visibility (due to weather conditions such as fog, smoke, rain or snow) may affect the flight trajectory, resulting in a trajectory deviation between the actual flight trajectory and the planned flight trajectory. Therefore, when predicting the flight trajectory, it is necessary to comprehensively consider the meteorological data, the actual flight trajectory and the planned flight trajectory to ensure the accuracy of the prediction results.

[0158] In summary, the specific process of acquiring the predicted flight trajectory of the aircraft within a preset time period in the future according to the actual flight trajectory, the planned flight trajectory, and the weather change data in this embodiment may include: determining the trajectory deviation data of the aircraft's flown section according to the actual flight trajectory and the planned flight trajectory; when the trajectory deviation data is less than the preset deviation threshold, and the weather change data is less than the preset change threshold, determining the predicted flight trajectory of the aircraft for a preset time period in the future according to the planned flight trajectory; or, when the trajectory deviation data is less than the preset deviation threshold, and the weather change data is greater than or equal to the preset change threshold, determining the predicted flight trajectory of the aircraft for a preset time period in the future according to the planned flight trajectory, the weather data, and the pre-trained trajectory adjustment model; or, when the trajectory deviation data is greater than the preset deviation threshold, determining the predicted flight trajectory of the aircraft for a preset time period in the future according to the aircraft data, the weather data, the actual flight trajectory, and the pre-trained trajectory prediction model.

[0159] In the process of predicting the flight trajectory of an aircraft, the trajectory deviation between the actual flight trajectory and the planned flight trajectory of the flown section can be calculated to determine the trajectory deviation data of the aircraft in the flown section; then, the corresponding prediction method for the aircraft flight trajectory prediction processing can be determined based on the deviation data of the flown section and the meteorological change data during the flight.

[0160] The trajectory deviation data of the aircraft is compared with a preset deviation threshold; optionally, if the trajectory deviation data is less than the preset deviation threshold, it means that the flight state of the aircraft is stable when flying in the flown section; further, the meteorological change data during the aircraft's flight is compared with the preset change threshold; optionally, if the trajectory deviation data is less than the preset change threshold, it means that the meteorological environment in the low altitude area is relatively stable during the aircraft's flight; in summary, it can be predicted that under the above circumstances, the aircraft can fly according to the planned flight trajectory within a preset time period in the future; therefore, the predicted flight trajectory of the aircraft within a preset time period in the future can be obtained by directly reading the data in the planned flight trajectory, so as to achieve fast and accurate prediction results.

[0161] Optionally, if the trajectory deviation data is greater than or equal to a preset change threshold, it means that the meteorological environment in the low-altitude area during the aircraft's flight is not very stable. In the subsequent flight, the actual flight trajectory of the aircraft may deviate from the planned flight trajectory. Therefore, when predicting the subsequent flight trajectory, on the basis of reading the planned flight trajectory, the changes in meteorological data can be considered, and the planned flight trajectory read can be adjusted through the preset trajectory adjustment model to obtain the predicted flight trajectory of the aircraft within a preset time period in the future.

[0162] It can be explained that the trajectory adjustment model can be a machine learning model (such as a regression model, a support vector machine, a neural network) or a trajectory prediction algorithm based on a physical model; optionally, the airspace management model obtained after fine-tuning the large-scale language model can also be used to adjust the trajectory model, which is not limited in this embodiment.

[0163] Specifically, the specific process of determining the predicted flight trajectory of an aircraft for a preset time in the future based on the planned flight trajectory, meteorological data and a pre-trained trajectory adjustment model in this embodiment may include: determining the initial flight trajectory of the aircraft for a preset time in the future based on the planned flight trajectory, vector encoding the initial flight trajectory to obtain an initial trajectory vector corresponding to the aircraft; vector encoding each parameter contained in the meteorological data to obtain a meteorological vector corresponding to the aircraft; obtaining a pre-trained trajectory adjustment model, inputting the initial trajectory vector and the meteorological vector into the trajectory adjustment model, and obtaining the predicted flight trajectory output by the trajectory adjustment model.

[0164] In order to facilitate the trajectory adjustment model to adjust the trajectory of the aircraft, before the planned flight trajectory and meteorological data for trajectory adjustment are input into the trajectory adjustment model, the above two data may be feature encoded in advance to obtain corresponding vectors, and the vectors may be input into the trajectory adjustment model.

[0165] Since the vector form after feature encoding is easier for the model to understand and process, the above scheme can better capture the patterns and regularities in the data by inputting the two types of data into the trajectory prediction model in the form of vectors for processing, thereby improving the accuracy and efficiency of the classification task; and the vector after feature encoding usually occupies less memory space, which can significantly reduce memory consumption, especially when processing large-scale data sets.

[0166] When the initial flight trajectory of the aircraft within a preset time period in the future is obtained according to the planned flight trajectory, the initial flight trajectory is encoded to obtain a high-dimensional vector.

[0167] Exemplarily, the initial flight trajectory before feature encoding includes: position (longitude, latitude, altitude), speed (x direction, y direction, z direction), acceleration (x direction, y direction, z direction), heading, and timestamp; the above initial flight trajectory is encoded using a preset encoding method, and the encoded initial trajectory vector may include: \[ \mathbf{v} =[\text{longitude}, \text{latitude}, \text{altitude}, \text{speed}_x, \text{speed}_y, \text{speed}_z, \text{acceleration}_x, \text{acceleration}_y, \text{acceleration}_z, \text{heading}, \text{timestamp}] \].

[0168] Optionally, while the initial flight trajectory is being vector-encoded, the meteorological data of the aircraft during the flight may also be encoded and processed to obtain the meteorological vector corresponding to the encoding.

[0169] It is understandable that the encoding methods used for encoding the initial flight trajectory and the meteorological data, and even other data that may appear below, may be the same or different, and there is no specific limitation on this.

[0170] After obtaining the initial trajectory vector and the meteorological vector, the two vectors are input into the trajectory adjustment model for trajectory adjustment processing to obtain the predicted flight trajectory after adjustment.

[0171] Optionally, if the trajectory deviation data is greater than or equal to a preset deviation threshold, it means that the aircraft has an unstable flight state when flying in the flown section, that is, the actual flight trajectory of the aircraft deviates from the planned flight trajectory due to obstacle avoidance or other factors such as weather during the flight; therefore, when performing subsequent flight trajectory prediction, directly reading the planned flight trajectory as the predicted flight trajectory within a preset time period in the future may have the problem of poor prediction accuracy; based on this, in order to obtain an accurate predicted flight trajectory, the aircraft can be directly processed according to the actual flight trajectory, the aircraft data in the current flight state, and the meteorological data through a preset trajectory prediction model to obtain the predicted flight trajectory within a preset time period in the future.

[0172] Specifically, in this embodiment, the specific process of determining the predicted flight trajectory of an aircraft for a preset time in the future based on aircraft data, meteorological data, actual flight trajectory and a pre-trained trajectory prediction model may include: determining the historical flight trajectory of the aircraft for a preset time in the past based on the actual flight trajectory, vector encoding the historical flight trajectory, and obtaining a historical trajectory vector corresponding to the aircraft; performing vector encoding processing on each parameter contained in the aircraft data and meteorological data, respectively, to obtain an aircraft vector and a meteorological vector corresponding to the aircraft; obtaining a pre-trained flight trajectory prediction model, inputting the historical trajectory vector, the aircraft vector and the meteorological vector into the trajectory prediction model, and obtaining a predicted flight trajectory output by the model.

[0173] In this embodiment, the above-mentioned implementation method can be referred to to perform feature coding processing on the acquired historical flight trajectory, meteorological data and aircraft data to obtain the historical trajectory vector, meteorological vector and aircraft vector corresponding to each data respectively; and the above-mentioned vectors are input into the trajectory prediction model for trajectory prediction processing to obtain the predicted flight trajectory output after prediction.

[0174] When the predicted flight trajectories of the aircraft within a preset time period in the future are obtained according to the above-mentioned embodiments, collision prediction is performed according to the predicted flight trajectories of the aircraft to obtain collision prediction results.

[0175] Specifically, the specific process of determining whether there is a target aircraft with a collision risk implemented in the embodiment of the present application may include: if the position difference between any predicted flight trajectories is zero, then determining that each aircraft corresponding to each predicted flight trajectory is a target aircraft with a collision risk; if the position difference between any predicted flight trajectories is not zero, but is less than a preset position threshold, then obtaining volume data of each aircraft corresponding to each predicted flight trajectory, and determining whether there is a target aircraft with a collision risk based on each volume data and the position difference.

[0176] In this embodiment, the predicted flight trajectory includes the predicted flight positions of the aircraft corresponding to multiple flight times within a preset time period in the future; if at any time within the preset time period in the future, if there are any two or more aircraft, and the corresponding predicted flight positions at this time are the same position, that is, the position difference between the two aircraft or more aircraft at this time is zero, then it means that at this time, the above two or more aircraft are target aircraft with a risk of collision.

[0177] Since the position difference is calculated based on the center point of the aircraft and the collision prediction is performed when the position prediction is performed, different volume data corresponds to different aircraft in actual flight. Based on this, it can be understood that although the position difference between aircraft is not zero at a certain moment, there may still be a risk of collision between aircraft during actual flight. Therefore, in the collision prediction process, a pre-set position threshold can be obtained. When it is determined that the position difference between any number of aircraft at a certain moment is less than the position threshold, it means that there is a high probability of a collision risk between the above-mentioned aircraft. In order to obtain accurate prediction results, the volume data of each aircraft corresponding to each predicted flight trajectory can be obtained, and the target aircraft with a collision risk can be determined based on each volume data and the position difference.

[0178] It may be explained that the volume data includes the size of the bounding box of the aircraft; the size of the bounding box is determined according to the equipment data of the aircraft.

[0179] Correspondingly, the specific process of implementing the present embodiment to determine whether there is a target aircraft with a collision risk based on each volume data and position difference may include but is not limited to: determining whether there is a target aircraft with a collision risk based on each volume data and position difference, including: calculating the sum of the sizes of the bounding boxes of each aircraft, if the sum of the sizes is less than the position difference, determining that each aircraft corresponding to each predicted flight trajectory is a target aircraft with a collision risk.

[0180] In this embodiment, for any aircraft whose position difference is less than the position threshold, the bounding box distance of the aircraft from the center of the aircraft to the edge of the bounding box is determined, the sum of the bounding box distances between any two of the above aircraft is calculated, and the sum of the bounding box distances is compared with the preset position difference; optionally, if it is less than, it means that the two aircraft are bounding boxes with a risk of collision, otherwise, it means that there is no risk of collision between the two aircraft.

[0181] On the basis of determining the target aircraft with collision risk according to the above implementation method, collision warning prompt information is transmitted to each target aircraft so that each target aircraft can avoid it and ensure flight safety. In some applications, when there are a large number of target aircraft with collision risk, if each target aircraft executes an avoidance strategy, it is easy to have a new collision risk with other aircraft; therefore, in this embodiment, it is proposed that when transmitting collision warning prompt information to the target aircraft, the avoidance strategy generated according to the collision wind direction is also transmitted to the target aircraft at the same time, so that each target aircraft can execute the avoidance strategy in an orderly manner and ensure the flight safety of each aircraft in the airspace.

[0182] Optionally, after determining the target aircraft with a collision risk, the technical solution of this embodiment also includes: determining the avoidance aircraft among the target aircraft, and the avoidance flight trajectory to be performed by the avoidance aircraft within a preset time period in the future, based on the type and flight mission of each target aircraft.

[0183] Among them, flight missions can be interpreted as the tasks performed by aircraft during flight, including but not limited to commercial aviation, scientific research, rescue operations, and drone missions. Avoidance aircraft can be interpreted as aircraft that changes the planned flight trajectory during flight to achieve self-help obstacle avoidance and avoid the risk of collision.

[0184] In this embodiment, based on the type and flight mission corresponding to the target aircraft, an avoidance strategy corresponding to the target aircraft when avoiding collision risk is generated, that is, the avoidance aircraft that needs to be avoided among the target aircraft is determined, as well as the avoidance flight trajectory to be executed by the avoidance aircraft within a preset time period in the future.

[0185] Specifically, the specific implementation method of determining the avoidance aircraft among the target aircraft in this embodiment may include but is not limited to: if the types of the target aircraft are different, determining the avoidance aircraft according to the first priority corresponding to the type; if the types of the target aircraft are the same, obtaining the flight mission of each target aircraft.

[0186] If the flight missions of the target aircraft are different, the avoidance aircraft is determined based on the second priority corresponding to the flight missions; if the flight missions of the target aircraft are the same, the avoidance aircraft is determined based on the actual flight trajectories of the target aircraft.

[0187] Exemplarily, when determining the avoidance aircraft among the target aircraft, the aircraft type of each target aircraft is compared, and the first priority corresponding to each aircraft type is obtained, and the target aircraft are prioritized; for example, the target aircraft ranked from low to high priority are: unmanned aircraft, manned aircraft, and civil aircraft.

[0188] Optionally, if there are multiple target aircraft of the same type during the sorting process, for example, there are three unmanned aircraft; obtain the flight missions of the three aircraft respectively, which are logistics express transportation and medical emergency transportation; in this case, the priorities of the unmanned aircraft are sorted from low to high as logistics express transportation and medical emergency transportation; among them, if there are two unmanned aircraft that are logistics express transportation aircraft, the actual flight trajectories of the two aircraft can be obtained, the flight trajectories can be analyzed, and the aircraft with the lower priority among the two can be determined, and the aircraft can be determined as the avoidance aircraft.

[0189] It should be understood that the above-mentioned process of determining the avoiding aircraft is an exemplary explanation of the present embodiment and is not intended to be a limitation of the present embodiment; in actual flight, other methods may also be used for determination, and of course the number of determined avoiding aircraft is not limited to one.

[0190] In the above implementation mode, priority rules are used to ensure that aircraft with emergency and high-priority missions are given priority in avoidance, reduce collision risks, and consider multiple factors (type, mission, trajectory), so as to flexibly respond to complex air traffic situations and improve the efficiency and coordination of air traffic management.

[0191] Specifically, the specific implementation process of determining the avoidance flight trajectory of the avoiding aircraft within a preset time period in the future in this embodiment may include but is not limited to: determining multiple candidate avoidance flight trajectories of the avoiding aircraft within a preset time period in the future based on the predicted flight trajectory of each target aircraft; screening among the candidate avoidance flights based on the actual flight trajectory, aircraft data and meteorological data of the avoiding aircraft to obtain the avoidance flight trajectory of the avoiding aircraft within the preset time period in the future.

[0192] Specifically, the relative speed and flight direction of the target aircraft are determined according to the predicted trajectory, and then the avoidance direction of the aircraft is determined; for example, when two target aircraft approach each other on the same horizontal plane, if the avoidance aircraft needs to perform vertical avoidance, the flight altitude of the avoidance aircraft can be changed to avoid collision. Since there may be other aircraft at other altitudes, the avoidable flight altitude of the avoidance aircraft can be determined in combination with the predicted flight trajectories of other aircraft, and multiple candidate avoidance flight trajectories can be determined according to the avoidable flight altitudes.

[0193] Furthermore, by comprehensively considering the aircraft data of the avoiding aircraft, the actual flight trajectory, and the current weather conditions of the flight, trajectory screening is performed among multiple candidate avoidance flight trajectories to determine a more appropriate avoidance flight trajectory.

[0194] To further improve the flight safety system of aircraft, when formulating the avoidance strategy, it can be synchronized with the aircraft for rapid and accurate execution. Specifically, the pilot or the autopilot system can be notified through the onboard system or the ground control center to execute the avoidance strategy, or the pilot manually or the autopilot system adjusts the flight parameters (altitude, direction, speed) according to the avoidance strategy. When performing an avoidance flight, the avoidance effect can be monitored in real time to ensure that the avoidance action is effective and avoid new risks. And, after confirming that the avoidance is successful, gradually restore to the planned flight trajectory, that is, fly at the original flight altitude, angle and speed. Optionally, after a successful avoidance, the avoidance can also be evaluated and recorded to optimize future avoidance strategies.

[0195] Figure 3A schematic diagram of the structure of a collision prediction device for an aircraft provided in an embodiment of the present application. Figure 3 The device includes: an aircraft signal acquisition module 310, an aircraft data identification module 320, a predicted flight trajectory determination module 330, a target aircraft determination module 340 and a collision warning prompt information output module 350; wherein,

[0196] The aircraft signal acquisition module 310 is used to acquire aircraft signals in the low-altitude area of ​​the target area acquired by acquisition equipment in the target area;

[0197] The aircraft data identification module 320 is used to identify the aircraft data currently flying in the low-altitude area according to the aircraft signal, the aircraft data includes: the identity information, type and flight data of the aircraft; the type includes: unmanned aircraft, manned aircraft, and civil aircraft; the flight data includes: the positioning data of the aircraft, the speed data and the angle data of the aircraft;

[0198] The predicted flight trajectory determination module 330 is used to obtain the predicted flight trajectory of the aircraft within a preset time period in the future based on the aircraft data and the meteorological data of the low-altitude area;

[0199] A target aircraft determination module 340 is used to determine whether there is a target aircraft with a collision risk based on the predicted flight trajectory of the aircraft within a preset time period in the future;

[0200] The collision warning prompt information output module 350 is used to output the collision warning prompt information of the target aircraft if it exists.

[0201] In some optional implementations, the predicted flight trajectory determination module includes:

[0202] The task prompt generation submodule is used to generate a flight trajectory prediction task prompt based on aircraft data, low-altitude meteorological data, and a flight trajectory prediction task prompt template;

[0203] The predicted flight trajectory generation submodule is used to input the flight trajectory prediction task prompt into the airspace management model to obtain the predicted flight trajectory of the aircraft within a preset time in the future; wherein the airspace management model is obtained by fine-tuning a large-scale language model trained with a general corpus using a training data set, and the training data set includes a first training data subset; the first training data subset is constructed based on historical flight trajectory data of various types of aircraft, and meteorological data corresponding to the historical flight trajectory data.

[0204] In some optional embodiments, the training data set further includes at least one of the following: a second training data subset and a third training data subset, the second training data subset being constructed based on historical collision data of various types of aircraft, and the third training data subset being constructed based on simulated collision data of various types of aircraft;

[0205] Target aircraft identification module, including:

[0206] A task prompt generation submodule is used to generate a collision prediction task prompt based on the predicted flight trajectory of the aircraft within a preset time in the future and the collision prediction task prompt template;

[0207] The collision prediction generation submodule is used to input the collision prediction task prompt into the airspace management model to obtain the prediction result of whether there is a target aircraft with a collision risk.

[0208] In some optional implementations, the predicted flight trajectory determination module includes:

[0209] The flight trajectory acquisition submodule is used to obtain the flight plan of the aircraft;

[0210] A planned flight trajectory acquisition submodule is used to acquire the planned flight trajectory of the aircraft according to the flight plan of the aircraft;

[0211] The meteorological change data acquisition submodule is used to obtain the meteorological change data of the aircraft during the flight according to the meteorological data in the low-altitude area; the meteorological change data represents the change trend of each meteorological data when the aircraft flies to each flight position;

[0212] The actual flight trajectory acquisition submodule is used to obtain the actual flight trajectory of the aircraft based on the aircraft data;

[0213] The predicted flight trajectory acquisition submodule is used to obtain the predicted flight trajectory of the aircraft within a preset time period in the future based on the actual flight trajectory, the planned flight trajectory, and the weather change data.

[0214] In some optional implementations, the meteorological change data acquisition module includes:

[0215] The weather change data acquisition unit is used to determine the weather data of each predicted flight trajectory at the corresponding flight time, and generate the weather change data of the aircraft according to each weather data.

[0216] In some optional implementations, the predicted flight trajectory acquisition submodule includes:

[0217] A trajectory deviation data acquisition unit, used to determine the trajectory deviation data of the flight section of the aircraft according to the actual flight trajectory and the planned flight trajectory;

[0218] A first predicted flight trajectory determination unit is configured to determine a predicted flight trajectory of the aircraft for a preset time period in the future according to the planned flight trajectory when the trajectory deviation data is less than a preset deviation threshold and the meteorological change data is less than a preset change threshold; or

[0219] a second predicted flight trajectory determination unit, configured to determine a predicted flight trajectory of the aircraft for a preset time period in the future according to the planned flight trajectory, the meteorological data and the pre-trained trajectory adjustment model when the trajectory deviation data is less than a preset deviation threshold and the meteorological change data is greater than or equal to a preset change threshold; or

[0220] The third predicted flight trajectory determination unit is used to determine the predicted flight trajectory of the aircraft for a preset time period in the future according to the aircraft data, meteorological data, the actual flight trajectory, and the pre-trained trajectory prediction model when the trajectory deviation data is greater than a preset deviation threshold.

[0221] In some optional implementations, the second predicted flight trajectory determination unit includes:

[0222] An initial trajectory vector acquisition subunit is used to determine the initial flight trajectory of the aircraft for a preset time in the future according to the planned flight trajectory, perform vector encoding on the initial flight trajectory, and obtain an initial trajectory vector corresponding to the aircraft;

[0223] A meteorological vector acquisition subunit is used to perform vector encoding processing on various parameters contained in the meteorological data to obtain a meteorological vector corresponding to the aircraft;

[0224] The predicted flight trajectory determination subunit is used to obtain a pre-trained trajectory adjustment model, input the initial trajectory vector and the meteorological vector into the trajectory adjustment model, and obtain the predicted flight trajectory output by the trajectory adjustment model.

[0225] In some optional implementations, the third predicted flight trajectory determination unit includes:

[0226] A historical trajectory vector determination subunit is used to determine the historical flight trajectory of the aircraft within a preset historical duration according to the actual flight trajectory, and to perform vector encoding on the historical flight trajectory to obtain a historical trajectory vector corresponding to the aircraft;

[0227] A meteorological vector and aircraft vector determination subunit, used to perform vector encoding processing on each parameter contained in the aircraft data and the meteorological data, respectively, to obtain an aircraft vector and a meteorological vector corresponding to the aircraft;

[0228] The predicted flight trajectory determination subunit is used to obtain a pre-trained flight trajectory prediction model, input the historical trajectory vector, aircraft vector and weather vector into the trajectory prediction model, and obtain the predicted flight trajectory output by the model.

[0229] In some optional implementations, the target aircraft determination module includes:

[0230] A first target aircraft determination submodule, configured to determine that each aircraft corresponding to each predicted flight trajectory is a target aircraft with a collision risk if the position difference between any predicted flight trajectories is zero;

[0231] The second target aircraft determination submodule is used to obtain the volume data of each aircraft corresponding to each predicted flight trajectory if the position difference between any predicted flight trajectories is not zero but less than a preset position threshold, and determine whether there is a target aircraft with a collision risk based on each volume data and the position difference.

[0232] In some optional embodiments, the volume data includes a bounding box size of the aircraft; the bounding box size is determined according to equipment data of the aircraft;

[0233] The second target aircraft determination submodule includes:

[0234] The target aircraft determination unit is used to calculate the sum of the sizes of the bounding boxes of each aircraft. If the sum of the sizes is smaller than the position difference, each aircraft corresponding to each predicted flight trajectory is determined to be a target aircraft with a collision risk.

[0235] In some optional embodiments, the aircraft data also includes flight missions;

[0236] The device also includes:

[0237] The avoidance aircraft and avoidance flight trajectory determination module is used to determine the avoidance aircraft among the target aircraft after determining the target aircraft with a collision risk, and the avoidance flight trajectory to be executed by the avoidance aircraft within a preset time period in the future according to the type and flight mission of each target aircraft.

[0238] In some optional implementations, the avoidance aircraft and avoidance flight trajectory determination module includes:

[0239] a first avoidance aircraft determination submodule, configured to determine the avoidance aircraft according to a first priority corresponding to the type if the target aircraft are of different types;

[0240] A second avoidance aircraft determination submodule, configured to obtain a flight mission of each target aircraft if the type of each target aircraft is the same;

[0241] a third avoiding aircraft determining submodule, configured to determine the avoiding aircraft according to the second priority corresponding to the flight mission if the flight missions of the target aircraft are different;

[0242] The fourth avoiding aircraft determining submodule is used to determine the avoiding aircraft according to the actual flight trajectory of each target aircraft if the flight missions of the target aircraft are the same.

[0243] In some optional implementations, the avoidance aircraft and avoidance flight trajectory determination module includes:

[0244] A candidate avoidance flight trajectory determination submodule is used to determine a plurality of candidate avoidance flight trajectories of the avoidance aircraft within a preset time period in the future according to the predicted flight trajectory of each target aircraft;

[0245] The avoidance flight trajectory determination submodule is used to screen among the candidate avoidance flights according to the actual flight trajectory of the avoidance aircraft, aircraft data and meteorological data, and obtain the avoidance flight trajectory of the avoidance aircraft within a preset time period in the future.

[0246] Figure 4 It is a block diagram of an electronic device shown in an embodiment of the present application, and the device may be a computer, a digital broadcast terminal, etc.

[0247] See also Figure 4 , the device 800 may include one or more of the following components: a processing component 802 , a memory 804 , a power component 806 , a multimedia component 808 , an audio component 810 , an input / output interface 812 , a sensor component 814 , and a communication component 816 .

[0248] The processing component 802 generally controls the overall operation of the device 800, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 802 may include one or more modules to facilitate the interaction between the processing component 802 and other components. For example, the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0249] The memory 804 is configured to store various types of data to support operations on the device 800. Examples of such data include instructions for any application or method operating on the device 800, contact data, phone book data, messages, pictures, videos, etc. The memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0250] The power supply component 806 provides power to the various components of the device 800. The power supply component 806 can include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 800.

[0251] The multimedia component 808 includes a screen that provides an output interface between the device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0252] The audio component 810 is configured to output and / or input audio signals. For example, the audio component 810 includes a microphone (MIC), and when the device 800 is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 804 or sent via the communication component 816. In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

[0253] The input / output interface 812 provides an interface between the processing component 802 and the peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0254] The sensor assembly 814 includes one or more sensors for providing various aspects of status assessment for the device 800. For example, the sensor assembly 814 can detect the open / closed state of the device 800, the relative positioning of components, such as the display and keypad of the device 800, and the sensor assembly 814 can also detect the position change of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, the orientation or acceleration / deceleration of the device 800, and the temperature change of the device 800. The sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 814 may also include an optical sensor, such as a solid image (Complementary Metal Oxide Semiconductor, CMOS) sensor or a semiconductor image (Charge-coupled Device, CCD) sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0255] The communication component 816 is configured to facilitate wired or wireless communication between the device 800 and other devices. The device 800 can access a wireless network based on a communication standard, such as WiFi, 4G or 5G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0256] In an exemplary embodiment, the device 800 may be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0257] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, and the instructions can be executed by a processor 820 of the device 800 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0258] A non-transitory computer-readable storage medium, when instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to execute the above-mentioned aircraft collision prediction method.

[0259] An embodiment of the present application also provides a chip for running instructions, which is used to execute the technical solution of the aircraft collision prediction method in the above embodiment.

[0260] An embodiment of the present application further provides a computer-readable storage medium, in which computer instructions are stored. When the computer instructions are executed on a computer, the computer executes the technical solution of the aircraft collision prediction method of the above embodiment.

[0261] An embodiment of the present application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When at least one processor executes the computer program, it can implement the technical solution of the collision prediction method of the aircraft in the above embodiment.

[0262] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0263] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for predicting a collision of an aircraft, characterized in that: The method comprises: Acquire aircraft signals in the low-altitude area of ​​the target area collected by a collection device in the target area; the collection device integrates a RID collection device and an ADS-B collection device; the RID collection device is used to collect unmanned aircraft signals; the ADS-B collection device is used to collect manned aircraft signals and civil aviation aircraft signals; Identify the aircraft data currently flying in the low-altitude area according to the aircraft signal, the aircraft data including: aircraft identity information, type and flight data; the types include: unmanned aircraft, manned aircraft, and civil aircraft; the flight data includes: positioning data of the aircraft, speed data and angle data of the aircraft; Obtaining a predicted flight trajectory of the aircraft within a preset time period in the future based on the aircraft data and the meteorological data of the low-altitude area; Determining whether there is a target aircraft with a collision risk based on the predicted flight trajectory of the aircraft within a preset time period in the future; If so, output collision warning prompt information of the target aircraft; Determining, according to the type and flight mission of each of the target aircraft, an avoidance aircraft among the target aircraft, and an avoidance flight trajectory to be executed by the avoidance aircraft within the future preset time period; Wherein, determining the avoidance aircraft among the target aircraft comprises: If the types of the target aircraft are different, determining the avoidance aircraft according to the first priority corresponding to the type; If the types of the target aircraft are the same, obtaining the flight missions of the target aircraft; If the flight missions of the target aircraft are different, determining the avoidance aircraft according to the second priority corresponding to the flight missions; If the flight missions of the target aircraft are the same, the avoidance aircraft is determined according to the actual flight trajectory of each target aircraft.

2. The method according to claim 1, characterized in that The step of obtaining a predicted flight trajectory of the aircraft within a preset time period in the future according to the aircraft data and the meteorological data of the low-altitude area includes: Generate a flight trajectory prediction task prompt according to the aircraft data, the low-altitude meteorological data, and the flight trajectory prediction task prompt template; The flight trajectory prediction task prompt is input into the airspace management model to obtain the predicted flight trajectory of the aircraft within a preset time period in the future; wherein the airspace management model is obtained by fine-tuning a large-scale language model trained with a general corpus using a training data set, and the training data set includes a first training data subset; the first training data subset is constructed based on historical flight trajectory data of various types of aircraft, and meteorological data corresponding to the historical flight trajectory data.

3. The method according to claim 2, characterized in that The training data set further includes at least one of the following: a second training data subset and a third training data subset, wherein the second training data subset is constructed based on historical collision data of various types of aircraft, and the third training data subset is constructed based on simulated collision data of various types of aircraft; The determining whether there is a target aircraft with a collision risk based on the predicted flight trajectory of the aircraft within a preset time period in the future includes: Generate a collision prediction task prompt according to the predicted flight trajectory of the aircraft within a preset time period in the future and a collision prediction task prompt template; The collision prediction task prompt is input into the airspace management model to obtain a prediction result of whether there is a target aircraft with a collision risk.

4. The method according to claim 1, characterized in that The step of obtaining a predicted flight trajectory of the aircraft within a preset time period in the future according to the aircraft data and the meteorological data of the low-altitude area includes: obtaining a flight plan for the aircraft; Acquiring a planned flight trajectory of the aircraft according to a flight plan of the aircraft; Acquire the meteorological change data of the aircraft during flight according to the meteorological data of the low-altitude area; the meteorological change data represent the change trend of each meteorological data when the aircraft flies to each flight position; acquiring an actual flight trajectory of the aircraft according to the aircraft data; The predicted flight trajectory of the aircraft within a preset time period in the future is obtained according to the actual flight trajectory, the planned flight trajectory, and the weather change data.

5. The method according to claim 4, characterized in that Obtaining weather change data of the aircraft during flight, including: The meteorological data of each predicted flight trajectory at the corresponding flight time is determined, and the meteorological change data of the aircraft is generated according to each meteorological data.

6. The method according to claim 4, characterized in that The step of obtaining a predicted flight trajectory of the aircraft within a preset time period in the future according to the actual flight trajectory, the planned flight trajectory, and the weather change data includes: Determining trajectory deviation data of a route segment flown by the aircraft according to the actual flight trajectory and the planned flight trajectory; When the trajectory deviation data is less than a preset deviation threshold and the meteorological change data is less than a preset change threshold, determining the predicted flight trajectory of the aircraft at the future preset time period according to the planned flight trajectory; or, When the trajectory deviation data is less than a preset deviation threshold and the meteorological change data is greater than or equal to a preset change threshold, determining a predicted flight trajectory of the aircraft at the future preset time period according to the planned flight trajectory, the meteorological data and a pre-trained trajectory adjustment model; or, When the trajectory deviation data is greater than a preset deviation threshold, the predicted flight trajectory of the aircraft at the future preset time period is determined according to the aircraft data, the meteorological data, the actual flight trajectory, and the pre-trained trajectory prediction model.

7. The method according to claim 6, characterized in that Determining a predicted flight trajectory of the aircraft at the future preset time period according to the planned flight trajectory, the meteorological data and the pre-trained trajectory adjustment model includes: Determine the initial flight trajectory of the aircraft at the preset future time period according to the planned flight trajectory, perform vector encoding on the initial flight trajectory, and obtain an initial trajectory vector corresponding to the aircraft; Performing vector coding processing on each parameter included in the meteorological data to obtain a meteorological vector corresponding to the aircraft; A pre-trained trajectory adjustment model is obtained, and the initial trajectory vector and the meteorological vector are input into the trajectory adjustment model to obtain a predicted flight trajectory output by the trajectory adjustment model.

8. The method according to claim 6, characterized in that Determining a predicted flight trajectory of the aircraft for a preset time in the future according to the aircraft data, the meteorological data, the actual flight trajectory and the pre-trained trajectory prediction model includes: Determine the historical flight trajectory of the aircraft within a preset historical duration according to the actual flight trajectory, perform vector encoding on the historical flight trajectory, and obtain a historical trajectory vector corresponding to the aircraft; Performing vector coding processing on each parameter included in the aircraft data and the meteorological data respectively to obtain an aircraft vector and a meteorological vector corresponding to the aircraft; A pre-trained flight trajectory prediction model is obtained, and the historical trajectory vector, the aircraft vector, and the weather vector are input into the trajectory prediction model to obtain a predicted flight trajectory output by the model.

9. The method according to any one of claims 1 to 8, characterized in that: Determine if there is a target aircraft that presents a risk of collision, including: If the position difference between any of the predicted flight trajectories is zero, then determining that each of the aircraft corresponding to each of the predicted flight trajectories is a target aircraft with a collision risk; If the position difference between any of the predicted flight trajectories is not zero but less than a preset position threshold, the volume data of each of the aircraft corresponding to each of the predicted flight trajectories is obtained, and it is determined whether there is a target aircraft with a collision risk based on each of the volume data and the position difference.

10. The method according to claim 9, characterized in that The volume data includes the size of the bounding box of the aircraft; the size of the bounding box is determined according to the equipment data of the aircraft; Determining whether there is a target aircraft with a collision risk according to each of the volume data and the position difference includes: The sum of the sizes of the bounding boxes of the aircraft is calculated. If the sum of the sizes is smaller than the position difference, the aircraft corresponding to the predicted flight trajectories are determined to be target aircraft with a collision risk.

11. The method according to claim 1, characterized in that Determining the avoidance flight trajectory to be executed by the avoidance aircraft within the future preset time period includes: Determining a plurality of candidate avoidance flight trajectories for the avoidance aircraft within the future preset time period according to the predicted flight trajectories of each of the target aircraft; The candidate avoidance flights are screened according to the actual flight trajectory of the avoidance aircraft, the aircraft data and the meteorological data to obtain the avoidance flight trajectory of the avoidance aircraft within a future preset time period.

12. An aircraft collision prediction device, characterized in that: The device comprises: An aircraft signal acquisition module is used to acquire aircraft signals in the low-altitude area of ​​the target area acquired by an acquisition device in the target area; the acquisition device is integrated with a RID acquisition device and an ADS-B acquisition device; the RID acquisition device is used to acquire unmanned aircraft signals; the ADS-B acquisition device is used to acquire manned aircraft signals and civil aviation aircraft signals; An aircraft data identification module, used to identify the aircraft data currently flying in the low-altitude area according to the aircraft signal, the aircraft data including: identity information, type and flight data of the aircraft; the types include: unmanned aircraft, manned aircraft, and civil aircraft; the flight data includes: positioning data of the aircraft, speed data and angle data of the aircraft; A predicted flight trajectory determination module, used to obtain a predicted flight trajectory of the aircraft within a preset time period in the future based on the aircraft data and the meteorological data of the low-altitude area; A target aircraft determination module, used to determine whether there is a target aircraft with a collision risk based on the predicted flight trajectory of the aircraft within a preset time period in the future; A collision warning prompt information output module is used to output the collision warning prompt information of the target aircraft if it exists; An avoidance aircraft and avoidance flight trajectory determination module, for determining, after determining target aircraft with a collision risk, an avoidance aircraft among the target aircraft and an avoidance flight trajectory to be executed by the avoidance aircraft within a preset time period in the future according to the type and flight mission of each target aircraft; Among them, the avoidance aircraft and avoidance flight trajectory determination module includes: a first avoidance aircraft determination submodule, configured to determine the avoidance aircraft according to a first priority corresponding to the type if the target aircraft are of different types; A second avoidance aircraft determination submodule, configured to obtain a flight mission of each target aircraft if the type of each target aircraft is the same; a third avoiding aircraft determining submodule, configured to determine the avoiding aircraft according to the second priority corresponding to the flight mission if the flight missions of the target aircraft are different; The fourth avoiding aircraft determining submodule is used to determine the avoiding aircraft according to the actual flight trajectory of each target aircraft if the flight missions of the target aircraft are the same.

13. An electronic device, characterized in that: include: A processor and a memory communicatively connected to the processor; The memory stores computer instructions; When executing the computer instructions, the processor is used to implement the collision prediction method for an aircraft according to any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, which are used to implement the aircraft collision prediction method according to any one of claims 1 to 11 when executed by a processor.

15. A computer program product, characterized in that The invention comprises a computer program, which, when executed by a processor, implements the collision prediction method for an aircraft according to any one of claims 1 to 11.

Images