Flight plan generation method and device for flight, computer device and storage medium

Abstract

This application relates to a method, apparatus, computer equipment, storage medium, and computer program product for generating flight plans. The method includes: acquiring the historical trajectory of a first flight and candidate runways for a second flight; selecting a takeoff runway for the second flight from the candidate runways based on the historical trajectory of the first flight; selecting a target departure procedure from the candidate departure procedures based on the matching degree between the planned trajectory of the candidate departure procedure and the historical trajectory of the first flight; and generating a flight plan for the second flight based on the takeoff runway of the second flight and the target departure procedure. This method can identify specific patterns in runway usage at various airports, select the flight plan for the second flight, and choose a target departure procedure from the candidate departure procedures, thus providing a more accurate flight plan. This approach mitigates the risk of overweight landings to some extent and avoids unnecessary fuel refueling waste.

Description

Technical Field

[0001] This application relates to the field of aviation technology, and in particular to a method, apparatus, computer equipment, storage medium, and computer program product for generating flight plans. Background Technology

[0002] In traditional technology, to ensure flight safety, except for some routes that have undergone risk assessment, airlines mainly use the arrival and departure procedures with the longest distance on the route for all flight releases, and release flights based on performance analysis of relatively conservative take-off and landing runways.

[0003] However, based on years of actual operation, due to the influence of operating modes, weather conditions and other factors, there are certain specific patterns in the use of runways by each airport. Using only fixed runways not only results in extra fuel waste, but also poses a risk of overweight landings in some cases. In other words, the current takeoff runway and departure procedure selection mode based on maximizing distance is not optimal for either operational safety or fuel efficiency.

[0004] To ensure flight safety, airlines typically use the maximum distance departure procedure for most flights on the departure direction of the route. However, based on years of operational experience, situations where the maximum distance departure procedure must be used are rare, and the crew's delivery of a departure procedure that is not accurate enough can, in some cases, lead to the risk of overweight landings. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for generating flight plans that can improve accuracy in addressing the aforementioned technical problems.

[0006] Firstly, this application provides a method for generating flight plans for airlines. The method includes:

[0007] Obtain the historical trajectory of the first flight and the candidate runways for the second flight;

[0008] Based on the historical trajectory of the first flight, the takeoff runway for the second flight is selected from the candidate runways;

[0009] Based on the matching degree between the planned trajectory of the candidate departure procedures and the historical trajectory of the first flight, the target departure procedure is selected from the candidate departure procedures;

[0010] Based on the takeoff runway of the second flight and the target departure procedure, a flight plan for the second flight is generated.

[0011] In one embodiment, the historical trajectory of the first flight includes first coordinate points and second coordinate points at different altitudes obtained in chronological order; the step of selecting the takeoff runway for the second flight from the candidate runways based on the historical trajectory of the first flight includes:

[0012] When the height of the first coordinate point is a preset height value, the difference in azimuth angle between the first coordinate point and the second coordinate point is used to query the difference in azimuth angle between the first coordinate point and the candidate runway entrance, and the runway azimuth angle query result is obtained.

[0013] Search for the candidate runway corresponding to the runway azimuth query result from the candidate runways;

[0014] Based on the candidate runways corresponding to the runway azimuth query results, the runway on the opposite side is selected as the takeoff runway for the second flight.

[0015] In one embodiment, selecting the takeoff runway for the second flight from the candidate runways based on the historical trajectory of the first flight further includes:

[0016] When the height of the first coordinate point and the height of the runway are greater than the preset height value, the distance between the first coordinate point and the runway entrance of each candidate runway is compared to obtain the distance comparison result.

[0017] Based on the distance comparison results, the candidate running tracks are screened.

[0018] The runway on the opposite side of the selected candidate runway will be chosen as the takeoff runway for the second flight.

[0019] In one embodiment, obtaining the historical trajectory of the first flight and the candidate runways of the second flight includes:

[0020] Obtain the historical trajectory of the first flight, and then obtain the second coordinate point from the historical trajectory;

[0021] Based on the distance between the second coordinate point and the runway entrance of each preset runway, preset runways on the same side as the trajectory of the first flight are selected.

[0022] The pre-defined runway on the same side as the historical trajectory of the first flight is used as the candidate runway for the second flight.

[0023] In one embodiment, obtaining the historical trajectory of the first flight and the candidate runways of the second flight includes:

[0024] Select waypoints sequentially from the original historical trajectory of the first flight;

[0025] The waypoints are verified based on their altitude data.

[0026] Generate a verified historical trajectory based on the verified waypoints;

[0027] The candidate runway for the second flight is obtained based on the verified historical trajectory.

[0028] In one embodiment, selecting the target departure procedure from the candidate departure procedures based on the matching degree between the planned trajectory of the candidate departure procedure and the historical trajectory of the first flight includes:

[0029] The system determines whether each point in the planned trajectory of the candidate departure procedure is a matching point by comparing it with the historical route points of the first flight.

[0030] The matching degree between the planned trajectory and the historical trajectory is calculated based on the number of matching points in the planned trajectory of the candidate departure procedure.

[0031] Based on the matching degree between the planned trajectory and the historical trajectory, a target exit procedure is selected from the candidate exit procedures.

[0032] In one embodiment, determining whether each point in the planned trajectory according to the candidate departure procedure matches the historical waypoints of the first flight, and whether each point in the planned trajectory is a matching point, includes:

[0033] If the number of historical waypoints of the first flight existing in the first matching range is greater than the number of the first preset waypoints, then the target point in the planned trajectory of the candidate departure procedure is the matching point;

[0034] If the number of historical waypoints of the first flight existing in the first matching range is less than or equal to the preset number of waypoints, then determine whether the target point is the matching point based on whether the historical waypoints of the first flight exist in the second matching range of each point in the planned trajectory of the candidate departure procedure; the second matching range is greater than the first matching range.

[0035] If the number of historical waypoints of the first flight existing in the first matching range is greater than the number of the first preset waypoints and less than or equal to the number of the second preset waypoints, and the departure point of the planned trajectory of the candidate departure procedure is different from the historical departure point of the first flight, then the target point is determined to be the matching point according to whether the historical waypoint of the first flight exists in the third matching range of each point in the planned trajectory of the candidate departure procedure; the third matching range is greater than the second matching range.

[0036] In one embodiment, the matching degree is determined based on the number of matching points between the planned trajectory and the historical trajectory of the first flight, and the departure point; selecting the target departure procedure from the candidate departure procedures according to the matching degree between the planned trajectory and the historical trajectory includes:

[0037] When there are multiple candidate departure procedures with the maximum number of matching points, obtain the historical departure point of the first flight and the departure point of the planned trajectory of the candidate departure procedure with the maximum number of matching points.

[0038] Determine whether the departure point of the planned trajectory is consistent with the historical departure point of the first flight;

[0039] If so, select the target departure procedure based on the candidate departure procedure that matches the historical departure point of the first flight;

[0040] If not, the target departure procedure is selected from the candidate departure procedures based on the flight departure distance between the historical departure point of the first flight and the departure airport, and the planned departure distance between the planned departure point of each candidate departure procedure and the departure airport.

[0041] In one embodiment, the method further includes:

[0042] When the departure airport of the second flight is the target airport, and the number of candidate departure procedures with the maximum number of matching points is less than or equal to the threshold number of candidate departure procedures, it is determined that the candidate departure procedure deviates from the target departure procedure of the second flight.

[0043] When the departure airport of the second flight is the target airport, and the number of matching points of the candidate departure procedure with the maximum number of matching points is less than or equal to the number of matching points threshold, and the departure point of the corresponding planned trajectory is inconsistent with the historical departure point of the first flight, it is determined that the candidate departure procedure deviates from the target departure procedure of the second flight.

[0044] If the departure airport of the second flight is not the target airport, the candidate departure procedure is determined to be deviated from the target departure procedure of the second flight based on whether the proportion of the number of matching points to the historical departure points of the first flight exceeds the corresponding threshold of the number of matching points.

[0045] In one embodiment, the method further includes a training step for the prediction model, the training step comprising:

[0046] The historical data of the first flight are resampled and randomly split according to the feature terms respectively;

[0047] Based on the data of resampled and randomly split feature terms, optimize at least one of the prediction models of the takeoff runway and the target departure procedure.

[0048] The optimized prediction model for the takeoff runway is used to predict the takeoff runway.

[0049] The optimized prediction model for the target exit procedure is used to predict the target exit procedure.

[0050] Secondly, this application also provides a flight plan generation apparatus for flights. The apparatus includes:

[0051] The data acquisition module is used to acquire the historical trajectory of the first flight and the candidate runways for the second flight;

[0052] The runway selection module is used to select the takeoff runway for the second flight from the candidate runways based on the historical trajectory of the first flight.

[0053] The departure procedure selection module is used to select a target departure procedure from the candidate departure procedures based on the matching degree between the planned trajectory of the candidate departure procedures and the historical trajectory of the first flight.

[0054] The plan generation module is used to generate a flight plan for the second flight based on the takeoff runway of the second flight and the target departure procedure.

[0055] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of generating flight plans for flights in any of the above embodiments.

[0056] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of generating flight plans for flights in any of the above embodiments.

[0057] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of generating flight plans for flights in any of the above embodiments.

[0058] The flight plan generation method, apparatus, computer equipment, storage medium, and computer program product described above select the takeoff runway for the second flight from the candidate runways based on the historical trajectory of the first flight, identifying a specific pattern in runway usage at each airport. Furthermore, based on the matching degree between the planned trajectory of the candidate departure procedures and the historical trajectory of the first flight, a target departure procedure is selected from the candidate departure procedures, ensuring that the departure of the second flight does not require using the departure procedure with the longest distance in the departure direction of that route. Therefore, based on the takeoff runway of the second flight and the target departure procedure, a more accurate flight plan can be generated, mitigating the risk of overweight landings to some extent and avoiding unnecessary fuel waste. Attached Figure Description

[0059] Figure 1 This is a diagram illustrating the application environment of a flight plan generation method in one embodiment.

[0060] Figure 2 This is a flowchart illustrating a method for generating flight plans for a flight in one embodiment;

[0061] Figure 3 This is a flowchart illustrating a flight plan generation method for a flight in another embodiment;

[0062] Figure 4 A schematic diagram of the runway selection process in one embodiment;

[0063] Figure 5 A flowchart illustrating the selection of a target departure procedure in one embodiment;

[0064] Figure 6 This is a schematic diagram of the prediction model training process in one embodiment;

[0065] Figure 7 This is a structural block diagram of a flight plan generation device for a flight in one embodiment;

[0066] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0068] The flight plan generation method for flights provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on another network server.

[0069] Terminal 102 can be one of various IoT devices, such as airborne equipment and airport equipment, or it can be a personal computer, laptop, smartphone, tablet, or portable wearable device. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers. This solution can be executed by at least one computer device, either terminal 102 or server 104.

[0070] In one embodiment, such as Figure 2 As shown, a method for generating flight plans is provided, which can be applied to... Figure 1 Taking terminal 102 as an example, let's explain from one perspective, including the following steps:

[0071] Step 202: Obtain the historical trajectory of the first flight and the candidate runways for the second flight.

[0072] The correspondence between Flight 1 and Flight 2 is determined by the construction and application of neural network models. This correspondence manifests as follows: the neural network model is built or invoked using data from Flight 1 to predict the data for Flight 2. Specifically, this correspondence can be illustrated in three aspects: Flight 1 can correspond to a flight number that has already executed a flight plan or is currently executing a flight plan within a certain time period; correspondingly, Flight 2 can correspond to a flight number whose flight plan is being generated. Flight 1 can be used to determine various feature terms of the neural network model; correspondingly, Flight 2 is used to obtain relevant data about the flight number for these feature terms. Flight 1 can also be used to obtain historical data to train, test, or validate a neural network model; correspondingly, Flight 2 is used to predict the flight path or departure procedures using the neural network model.

[0073] Flight 1 and Flight 2 are identified when they match any of the above correspondences. It can be understood that Flight 1 and Flight 2 can be the same flight number on different dates. For example, if the flight plan for Flight MF8555 on January 1st has already been executed, and the flight plan for Flight MF8555 on January 4th is being generated, then the data corresponding to Flight MF8555 on January 1st belongs to Flight 1. The neural network can be trained using the data corresponding to Flight MF8555 on January 1st to predict the data for Flight 2. The data for Flight 2 is Flight MF8555 on January 4th.

[0074] The historical trajectory of the first flight is obtained based on flight records from historical data. The data source for the historical trajectory can be any one of the following: the airline's own onboard QAR data, civil aviation information sharing data, or ADS-B data from FA purchased through GMP. Multiple data sources can be cross-verified to obtain more accurate data.

[0075] The flight track can be obtained by acquiring GNSS coordinate data from the start of the flight to the shutdown of the flight, based on the flight date and departure airport. GNSS coordinate data is coordinate data from the Global Navigation Satellite System. When there are inconsistencies in the GNSS coordinate data regarding flight data, corrections can be made using FA data (FlightAware). The historical track of the first flight must include at least two coordinate points at different altitudes, acquired in chronological order.

[0076] In one embodiment, the focus is on the discussion of candidate runways. Obtaining the historical trajectory of the first flight and the candidate runways of the second flight includes: obtaining the historical trajectory of the first flight and obtaining a second coordinate point from the historical trajectory; filtering the preset runways on the same side of the trajectory of the first flight based on the distance between the second coordinate point and the runway entrance of each preset runway; and using the preset runways on the same side of the historical trajectory of the first flight as candidate runways for the second flight.

[0077] The candidate runway for the second flight is the runway that needs to be considered when making the flight plan for the second flight. The candidate runway can be a preset runway in the departure airport where the second flight is located. The data of the preset runway is obtained from the interface by the airport four-letter code. The candidate runway can also be a runway on the same side selected from the preset runways based on the historical trajectory of the first flight.

[0078] Where the distance between the second coordinate point and the runway entrance of a certain preset runway is less than the distance between the second coordinate point and half of the runway entrances of each preset runway, then the preset runway is a same-side runway selected from the preset runways.

[0079] Step 204: Based on the historical trajectory of the first flight, select the takeoff runway for the second flight from the candidate runways.

[0080] The takeoff runway for the second flight can be predicted using a model constructed with a machine learning algorithm. This algorithm can be any regression model used for classification, including but not limited to decision tree models, random forest models, and extreme randomized trees (ET). Extreme randomized trees (ET) are particularly well-suited for handling discrete variable data. By incorporating resampling techniques and random splitting strategies, they effectively avoid overfitting and exhibit excellent performance when dealing with large datasets and high-dimensional features, demonstrating good generalization ability and high accuracy. The training steps for the ET prediction model include: resampling and randomly splitting the historical data of the first flight according to feature terms; optimizing the takeoff runway prediction model based on the resampled and randomly split feature terms; and using the optimized model to predict the takeoff runway.

[0081] In one embodiment, the historical trajectory of the first flight includes first and second coordinate points at different altitudes, acquired sequentially over time. The altitude can be elevation, coordinate altitude, or the relative altitude of the coordinate point to the horizontal plane corresponding to the runway. Correspondingly, the preset altitude value is set according to elevation, coordinate altitude, or the altitude of the coordinate point to the horizontal plane corresponding to the runway. For example, when the altitude is set based on the relative altitude of the coordinate point to the horizontal plane corresponding to the runway, the preset altitude value is approximately 0.

[0082] Correspondingly, based on the historical trajectory of the first flight, the takeoff runway for the second flight is selected from the candidate runways. This includes: when the altitude of the first coordinate point is a preset altitude value, querying the difference in azimuth angle between the first coordinate point and the second coordinate point, and obtaining the runway azimuth angle query result; searching for the candidate runway corresponding to the runway azimuth angle query result from the candidate runways; and selecting the runway on the opposite side as the takeoff runway for the second flight based on the candidate runway corresponding to the runway azimuth angle query result.

[0083] The azimuth difference between the first and second coordinate points is obtained by performing a difference calculation on the two points. The difference calculation process can be as follows: calculate the azimuth of the first coordinate point relative to a preset direction, then calculate the azimuth of the second coordinate point relative to that preset direction; calculate the difference between the azimuth of the first and second coordinate points to obtain the azimuth difference value. It can be understood that the difference calculation can also be performed by calculating variance, standard deviation, data gradient, etc.

[0084] The runway azimuth query result refers to the runway threshold azimuth difference value that is closest to the azimuth difference value between the first coordinate point and the second coordinate point. The runway threshold azimuth difference value is calculated based on the difference between the azimuth of the first coordinate point and the coordinate azimuth of the runway threshold, and its calculation process is similar to that of the azimuth difference value between the first coordinate point and the second coordinate point.

[0085] The candidate runways corresponding to the runway azimuth query results are the runway gates closest to the aircraft representing the second flight when it leaves the runway. Correspondingly, based on the candidate runways corresponding to the runway azimuth query results, the runway on the opposite side is selected as the takeoff runway for the second flight. This includes: calculating the runway identification and transposition method on the opposite side based on the identification of the candidate runways corresponding to the runway azimuth query results, and selecting the takeoff runway for the second flight according to the runway identification and transposition method on the opposite side.

[0086] For example: In the historical trajectory of the first flight, select the first coordinate point with a non-zero altitude, and its adjacent coordinate points; label the adjacent coordinate points in chronological order as the first coordinate point P1 and the second coordinate point P2. If the altitude of the first coordinate point P1 is zero, meaning the first coordinate point P1 is a point on the runway, calculate the azimuth difference value Angle_0 between the first coordinate point P1 and the second coordinate point P2, and calculate the azimuth difference value Angle_X between the first coordinate point P1 and the runway thresholds of X candidate runways. The runway threshold N with the angle closest to Angle_0 is then determined. Runway threshold N is the closest runway threshold when the aircraft leaves the runway, but the actual flight enters the runway threshold N', which is obtained from runway threshold N by adding 18 to the runway markings and left-hand reversal.

[0087] Furthermore, when the height of the first coordinate point is greater than a preset height value, the distance between the first coordinate point and the runway openings of each candidate runway is compared to obtain a distance comparison result. Based on the distance comparison result, the candidate runways are filtered. The runway on the opposite side of the filtered candidate runway is selected as the takeoff runway for the second flight. The distance comparison result is used to determine the runway opening closest to the first coordinate point P1, and this runway opening is used to determine the filtered candidate runway. Then, the runway on the opposite side of the filtered candidate runway is selected as the takeoff runway for the second flight.

[0088] In one embodiment, the correction process for the second flight is described from the perspective of invoking a neural network model. When the terminal sends an instruction to invoke the neural network model, the consistency of the feature data of the first flight within the time interval between the departure time and the instruction sending time is verified; and / or, the consistency of the feature data of the second flight within the time interval is verified.

[0089] The feature data is set for modeling a specific airport and includes types such as historical flight data, flight number-related data, airline data, aircraft data, and meteorological data. The time period has multiple time lengths. First, the first flight and the second flight are determined according to the first time length. If the first time length does not determine the first flight and the second flight at the same time, the missing flight can be determined according to the second time length, which is longer than the first time length.

[0090] For example: Within 15 minutes before the instruction is sent, take the two flights with the most recent departure times within 15 minutes after the instruction is sent. If the characteristic data are inconsistent, the data is considered incorrect and corrected based on consistent characteristic data. If there are no enough departing flights within 15 minutes, take the two flights with the most recent departure / arrival times within 15 minutes before and after the instruction is sent. If they are inconsistent, the data is considered incorrect and corrected based on consistent characteristic data. If there are no enough flights within 15 minutes, take the two flights with the most recent departure / arrival times within 30 minutes before and after the instruction is sent. If they are inconsistent, the data is considered incorrect and corrected based on consistent characteristic data. If none of the above applies, FA data (FlightAware, flight tracking data) is used for correction.

[0091] Step 206: Select the target departure procedure from the candidate departure procedures based on the matching degree between the planned trajectory of the candidate departure procedure and the historical trajectory of the first flight.

[0092] Candidate departure procedures are optional departure procedures. Departure procedures, also known as Standard Instrument Departure (SID), are publicly released flight procedures used by aircraft conducting instrument flight after takeoff from an airport. There are two main types: pilot-guided SID and radar-guided SID. An airport may have many different SIDs, and even on the same runway, different departure procedures may guide aircraft to different departure points to exit the terminal control area.

[0093] The planned trajectory of the candidate departure procedure includes multiple waypoints and navigation points for each departure procedure. Each waypoint represents the coordinate point experienced by the corresponding flight. The matching degree between the planned trajectory and the historical trajectory of the first flight is calculated using waypoints, and the target departure procedure with a high matching degree is selected.

[0094] In one embodiment, the process of calculating the matching degree is described. Selecting a target departure procedure from the candidate departure procedures based on the matching degree between the planned trajectory of the candidate departure procedure and the historical trajectory of the first flight includes: determining whether each point in the planned trajectory of the candidate departure procedure is a matching point based on whether it matches the historical waypoints of the first flight; calculating the matching degree between the planned trajectory and the historical trajectory based on the number of matching points in the planned trajectory of the candidate departure procedure; and selecting the target departure procedure from the candidate departure procedures based on the matching degree between the planned trajectory and the historical trajectory.

[0095] Step 208: Generate the flight plan for the second flight based on the takeoff runway and target departure procedures of the second flight.

[0096] Since the takeoff runway and departure procedures for the second flight can be calculated based on neural network models, the calculation process is based on the flight's historical trajectory. Artificial intelligence algorithms are used to design and construct predictive models for the flight's takeoff runway and departure procedures, providing predictive references for takeoff runway and departure procedures to assist dispatchers in making flight plans. This breaks the current fixed selection of departure procedures in the terminal area and can effectively improve the accuracy and flexibility of flight plans.

[0097] A flight plan is a set of information provided by air traffic services to a specific flight operator to complete a single flight. Specifically, a flight plan can include parameters for any phase, such as taxiing, takeoff, climb, cruise, approach, descent, landing, and taxiing. These parameters can include speed, altitude, horizontal distance, etc. In this embodiment, the flight plan primarily covers the takeoff, climb, and part of the cruise phases.

[0098] In the flight plan generation method described above, the takeoff runway for the second flight is selected from candidate runways based on the historical trajectory of the first flight, revealing a specific pattern in runway usage at each airport. Furthermore, based on the matching degree between the planned trajectory of the candidate departure procedure and the historical trajectory of the first flight, a target departure procedure is selected from the candidate departure procedures, ensuring that the departure of the second flight does not require using the departure procedure with the longest distance in the departure direction of that route. Therefore, based on the takeoff runway and target departure procedure of the second flight, a more accurate flight plan can be generated, mitigating the risk of overweight landings to some extent and avoiding unnecessary fuel waste.

[0099] In one embodiment, such as Figure 3 As shown, a method for generating flight plans is provided, which can be applied to... Figure 1 Taking terminal 102 as an example, let's explain from another perspective, including the following steps:

[0100] Step 302: Obtain the historical trajectory of the first flight and the candidate runways for the second flight.

[0101] In one embodiment, obtaining the historical trajectory of a first flight and candidate runways for a second flight includes: sequentially selecting waypoints from the original historical trajectory of the first flight; verifying the waypoints based on their altitude data; generating a verified historical trajectory based on the verified waypoints; and obtaining candidate runways for the second flight based on the verified historical trajectory.

[0102] The original historical trajectory can be historical trajectory data verified from multiple data sources, or it can be historical trajectory data from a specific source. When there are multiple second flights, each flight is sorted according to its time series, and waypoints are selected sequentially. Supplementing this with protective verification of altitude data, obtaining candidate runways for the second flights based on the verified historical trajectories ensures the reliability of the second flight data. After obtaining the verified historical trajectories, candidate runways for the second flights can be obtained more accurately.

[0103] Step 304: Based on the historical trajectory of the first flight, select the takeoff runway for the second flight from the candidate runways.

[0104] Steps 302-304 can be referred to the embodiments of steps 202-204.

[0105] Step 306: Determine whether each point in the planned trajectory of the candidate departure procedure matches the historical route points of the first flight.

[0106] In one embodiment, determining whether each point in the planned trajectory is a matching point by checking whether it matches the historical waypoints of the first flight includes:

[0107] If the number of historical waypoints of the first flight in the first matching range is greater than the number of the first preset waypoints, then the target point in the planned trajectory of the candidate departure procedure is the matching point.

[0108] If the number of historical waypoints of the first flight in the first matching range is less than or equal to the number of preset waypoints, then determine whether the target point is a matching point based on whether there are historical waypoints of the first flight in the second matching range of each point in the planned trajectory of the candidate departure procedure; the second matching range is greater than the first matching range.

[0109] If the number of historical waypoints of the first flight in the first matching range is greater than the number of the first preset waypoints and less than or equal to the number of the second preset waypoints, and the departure point of the planned trajectory of the candidate departure procedure is different from the historical departure point of the first flight, then the target point is determined to be a matching point based on whether there are historical waypoints of the first flight in the third matching range of each point in the planned trajectory of the candidate departure procedure; the third matching range is greater than the second matching range.

[0110] The radii of the first, second, and third matching ranges increase sequentially, and the conditions for using each matching range differ, thus helping to more accurately determine whether each point matches and consequently calculate the matching degree more accurately. Specifically, the first matching range is 0.5 nautical miles, the second is 3 nautical miles, and the third is 5.5 nautical miles, with an error range of 0.5 nautical miles for the third matching range.

[0111] Step 308: Calculate the matching degree between the planned trajectory and the historical trajectory based on the number of matching points in the planned trajectory of the candidate departure procedure.

[0112] Step 310: Select the target exit procedure from the candidate exit procedures according to the matching degree between the planned trajectory and the historical trajectory.

[0113] In one embodiment, the matching degree is determined based on the number of matching points between the planned trajectory and the historical trajectory of the first flight, and the departure point; correspondingly, according to the matching degree between the planned trajectory and the historical trajectory, a target departure procedure is selected from the candidate departure procedures, including:

[0114] When there are multiple candidate departure procedures with the maximum number of matching points, obtain the historical departure point of the first flight and the departure point of the planned trajectory of the candidate departure procedure with the maximum number of matching points; determine whether the departure point of the planned trajectory is consistent with the historical departure point of the first flight.

[0115] If so, select the target departure procedure from the candidate departure procedures that match the historical departure point of the first flight; among them, when the candidate departure procedures that match the historical departure point of the first flight are the same, the target departure procedure shall be selected first; in addition, candidate departure procedures with PBN identifiers may also be selected first as the target departure procedure.

[0116] If not, select the target departure procedure from the candidate departure procedures based on the flight departure distance between the historical departure point and the departure airport of the first flight, and the planned departure distance between the planned departure point and the departure airport of each candidate departure procedure.

[0117] When the number of matching points between the planned trajectory and the historical trajectory of the first flight is at its maximum, and the departure point of the planned trajectory is consistent with the historical departure point, the target departure procedure with the highest matching degree with the second flight is obtained; when the number of matching points between the planned trajectory and the historical trajectory of the first flight is at its maximum, and the departure point of the planned trajectory is inconsistent with the historical departure point, the actual departure distance is calculated, and it is determined whether the actual departure point is consistent with the planned departure point.

[0118] Specifically, based on the historical departure point of the first flight and the departure airport, and the planned departure point of each candidate departure procedure and the planned departure airport, a target departure procedure is selected from the candidate departure procedures, including:

[0119] The first planned departure distance is calculated based on the first candidate departure procedure, and the second planned departure distance is calculated based on the second candidate departure procedure;

[0120] When the flight departure distance is less than the first planned departure distance and less than the second planned departure distance, the second candidate departure procedure is selected as the target departure procedure.

[0121] When the flight departure distance is greater than the first planned departure distance and the flight departure distance is greater than the second planned departure distance, the first candidate departure procedure is selected as the target departure procedure;

[0122] When the difference between the first planned departure distance and the second planned departure distance exceeds the planned departure distance range, and the flight departure distance is between the first planned departure distance and the second planned departure distance, the second planned departure distance is compared with the planned departure distance range based on the flight departure distance; if it is less than the range, the second candidate departure procedure is selected as the target departure procedure; if it is greater than the range, the first candidate departure procedure is selected as the target departure procedure.

[0123] When the difference between the first planned departure distance and the second planned departure distance is less than the planned departure distance range, and the flight departure distance is between the first planned departure distance and the second planned departure distance, compare the first difference between the flight departure distance and the first planned departure distance, and the second difference between the flight departure distance and the second planned departure distance; select the corresponding candidate departure procedure based on the smaller of the first difference and the second difference.

[0124] The selection of the corresponding candidate exit procedure based on the smaller of the first difference and the second difference means that: when the first difference is less than the second difference, the first candidate exit procedure is selected as the target exit procedure; when the first difference is greater than the second difference, the second candidate exit procedure is selected as the target exit procedure.

[0125] To more clearly illustrate the process of selecting the target departure procedure from the candidate departure procedures, the flight departure distance is taken as distance L, the first candidate departure procedure is determined as procedure A, the first planned departure distance is distance A, the second candidate departure procedure is determined as procedure B, and the first planned departure distance is distance B.

[0126] Correspondingly, when distance L is less than distance A and distance L is less than distance B, program B is selected as the target departure program;

[0127] When distance L is greater than distance A and distance L is greater than distance B, select program A as the target departure program;

[0128] When the difference between distance A and distance B is greater than the planned departure distance range (20 nautical miles), and distance L is between distance A and distance B, based on distance L, compare distance B with the sum of the planned departure distance range (20 nautical miles); if it is less than, procedure B is the target departure procedure; if it is greater than, procedure A is the target departure procedure.

[0129] When the difference between distance A and distance B is less than the planned departure distance range (20 nautical miles), and distance L is between distance A and distance B, the departure procedure corresponding to the closer distance (distance A or distance B) is taken as the target departure procedure code.

[0130] Furthermore, to ensure the safety of the second flight, the method determines whether a candidate departure procedure is suitable as the target departure procedure for the second flight by judging whether it deviates from the target departure procedure of the second flight; based on this, the method also includes:

[0131] When the departure airport of the second flight is the target airport, and the number of candidate departure procedures with the maximum number of matching points is less than or equal to the threshold number of candidate departure procedures, it is determined that the candidate departure procedure deviates from the target departure procedure of the second flight.

[0132] When the departure airport of the second flight is the target airport, and the number of matching points of the candidate departure procedure with the maximum number of matching points is less than or equal to the threshold of the number of matching points, and the departure point of the corresponding planned trajectory is inconsistent with the historical departure point of the first flight, it is determined that the candidate departure procedure deviates from the target departure procedure of the second flight.

[0133] If the departure airport of the second flight is not the target airport, the candidate departure procedure is determined to be deviated from the target departure procedure of the second flight by whether the proportion of the number of matching points to the historical departure points of the first flight exceeds the corresponding threshold of the number of matching points.

[0134] Specifically, the departure airport is the airport from which the second flight departs. Once the departure airport has been modeled using a neural network, the target airport is determined. The target airport can then collect feature data according to the neural network model and use this data to make predictions, accurately determining whether each candidate departure procedure can be used as the target departure procedure for the second flight. When the departure airport is not the target airport, the degree of matching based on the number of matching points is used to determine whether the candidate departure procedure deviates from the target departure procedure for the second flight.

[0135] In one embodiment, the target airport is Xiamen Airport, the number of matching points for each airport is i, where i is an integer; the maximum number of matching points is imax, the threshold for the number of candidate departure procedures is 1, and the threshold for the number of matching points is 2; correspondingly, it includes:

[0136] If the departure airport of the second flight is Xiamen Airport, and the number of candidate departure procedures in IMAX is less than or equal to 1, it is determined that the candidate departure procedure deviates from the target departure procedure of the second flight, and the candidate departure procedure is denoted as "ATC DIRECT".

[0137] If the departure airport of the second flight is Xiamen Airport, and the number of candidate departure procedures in IMAX is less than or equal to 2, and the departure points are inconsistent, it is identified as the target departure procedure for the second flight, and each candidate departure procedure is recorded as "ATC DIRECT".

[0138] When the departure airport of the second flight is not Xiamen's airport, the departure procedure is judged based on the proportion of the historical departure points of the first flight. If the ratio of the matching point to the planned point is less than 0.7, it is considered a deviation from the target departure procedure of the second flight, and each candidate departure procedure is recorded as "ATC DIRECT". If the ratio of the matching point to the planned point is greater than 0.7, it is determined that the candidate departure procedure has not deviated from the target departure procedure of the second flight, and the target departure procedure of the second flight can be selected from the candidate departure procedures according to the matching degree of the number of matching points. Here, deviating from the target departure procedure of the second flight means that there is no target departure procedure for the second flight among the candidate departure procedures.

[0139] Step 312: Generate the flight plan for the second flight based on the takeoff runway and target departure procedures of the second flight.

[0140] Step 312 can be referred to the embodiment of step 208.

[0141] The method further includes: resampling and randomly splitting the historical data of the first flight according to the feature terms respectively; optimizing the prediction model of the target departure procedure according to the data of the resampled and randomly split feature terms; wherein the optimized prediction model of the target departure procedure is used to predict the target departure procedure.

[0142] In one embodiment, the application scenario will be used to illustrate the significance of the above-mentioned prediction model.

[0143] Balancing operational safety and fuel efficiency has always been a challenge for civil aviation development. Statistics show that in 2020, the entire civil aviation industry consumed approximately 25 million tons of fuel, resulting in carbon emissions of about 80 million tons. Achieving coordinated development of transportation safety and green, low-carbon practices in the aviation industry is a major challenge for airlines. However, some documents point out the need to "promote the deep integration of civil aviation with new technologies such as artificial intelligence, big data, and the Internet of Things, and leverage digital technology to empower technological innovation in the civil aviation industry." After years of information-based operations, airlines have accumulated massive amounts of operational data. Our direction of action is to enable data-driven intelligent decision-making, explore potential fuel-saving opportunities while ensuring flight safety, and obtain the optimal solution that balances safety margins and fuel efficiency.

[0144] Current analysis of flight takeoff and departure procedures suffers primarily from low accuracy. This is manifested in inefficient data integration, unreliable analytical methods, and the fact that the release system can only utilize a single, fixed route. To ensure flight safety, except for certain routes that have undergone risk assessment, airlines use the departure procedure with the longest possible distance along that route for all flight departures. However, based on years of operational experience, situations where the maximum distance departure procedure has to be used are rare. This fails to provide the crew with a more effective and accurate departure procedure, resulting in wasted fuel and, in some cases, the risk of overweight landings. Therefore, the current departure selection model is not optimal for either operational safety or fuel efficiency.

[0145] The survey found that the selection of departure procedures is highly correlated with factors such as weather conditions at the takeoff / landing airport and along the route, the first departure point, the current runway usage, air traffic control, third-party activities, and route congestion. It is a complex combination decision problem, and traditional analysis methods are not effective in quantifying the impact of these factors on departure procedures.

[0146] Therefore, this solution requires separate modeling for takeoff runway prediction and departure procedure prediction. First, the takeoff runway and departure procedure for the first flight must be determined separately. Then, these are combined with flight plans, aeronautical information, weather data, and other information for the second flight to form two datasets. These datasets are then fed into a machine learning algorithm to train a prediction model, which is finally published to the service. Thus, through the computational process related to this artificial intelligence algorithm and the historical operational big data of the first flight, a prediction model for flight takeoff runways and departure procedures is designed and constructed. This provides dispatchers with predictive references for takeoff runways and departure procedures when creating flight plans, breaking the current fixed selection of departure procedures in the terminal area. This effectively improves the accuracy and flexibility of flight plans and overcomes various problems in the aforementioned application scenarios. The research results can also be extended to civil aviation control units in due course, providing auxiliary references for control scheme formulation and decision-making, effectively improving the rational utilization rate of terminal area airspace resources and enhancing the level of refined operation.

[0147] Furthermore, such as Figure 4 The diagram illustrates the steps involved in the predictive model for takeoff runways, including the following:

[0148] Step 401: Obtain historical trajectory GNSS coordinate data for the period from start to finish of each flight, based on the flight date and departure airport.

[0149] Step 402: Filter the historical trajectories by departure location, sort the filtered historical trajectories by time, and process the historical trajectories sorted by time according to steps 3-13 below. The historical trajectories being processed are the historical trajectories of the first flight.

[0150] Step 403: From the historical trajectory of the first flight, find a coordinate point with a non-zero altitude, and mark the adjacent coordinate points of the first coordinate point P1 and the second coordinate point P2 in chronological order; wherein, the adjacent coordinate points of the first coordinate point are preferably the coordinate points of the previous time point of the first coordinate point.

[0151] Step 404: Obtain runway data from the interface using the airport's four-letter code (assuming 2X runway gates are obtained);

[0152] Step 405: Calculate the distance between the second coordinate point P2 and the latitude and longitude coordinates of the runway entrance, and keep the X runway entrances that are closest to each other (select runways on the same side);

[0153] Step 406: If the height of the first coordinate point P1 is zero, then the first coordinate point P1 is a point on the runway. Calculate the azimuth difference value Angle_0 between P1 and P2, calculate the azimuth difference value Angle_X between P1 and each runway opening, and find the runway opening N whose angle is closest to Angle_0.

[0154] Step 407: If the altitude of P1 is greater than zero, calculate the nearest runway gate N to P1. N is the nearest runway gate when the aircraft leaves the runway.

[0155] Step 408: The second flight actually enters the runway N' opposite to runway N. N' is obtained through +18 and L / R transpose.

[0156] Furthermore, the method also includes a scheme for correcting the feature data of the first flight and the feature data of the second flight, including:

[0157] Step 409: Within 15 minutes before the instruction is sent, take the two flights with the most recent departure times within 15 minutes after the instruction. If the feature data are inconsistent, the data is judged as erroneous and corrected according to the consistent feature data.

[0158] Step 410: If there are not enough departing flights within 15 minutes, take the two flights with the most recent departure / arrival times within the previous 15 minutes and the next 15 minutes respectively. If they are inconsistent, they are judged as erroneous data and corrected according to the consistent feature data.

[0159] Step 411: If there are not enough flights within 15 minutes, take the two flights with the most recent departure / arrival times within the previous 30 minutes and the next 30 minutes respectively. If they are inconsistent, the data is judged as erroneous and corrected according to the consistency feature data. If the above is not the case, take FA data (FlightAware, flight tracking data) for correction.

[0160] Step 412: If the above conditions are not met, then retrieve the FA data;

[0161] Step 413: Output the takeoff runway for the second flight.

[0162] Furthermore, such as Figure 5 The diagram illustrates the steps of the prediction model for the exit procedure, including:

[0163] Step 501: Collect data from various points along the flight paths of each flight. The data sources for each point include one or more of the following: airline-owned onboard QAR data, civil aviation information sharing data, and ADS-B data from aircraft manufacturers (FAs) purchased through GMP.

[0164] Step 502: Transfer the historical trajectories from the airborne QAR data, civil aviation information sharing data, and FA data. The historical trajectories include the latitude and longitude data of each coordinate point sorted by time data, as well as the altitude data of the coordinate points used to verify the latitude and longitude data.

[0165] Step 503: Sort the original historical trajectory of each flight according to time, select waypoints in the time-sorted historical trajectory, and supplement with protective verification of altitude data to ensure the reliability of time data and historical trajectory, so as to obtain the verified historical trajectory.

[0166] Step 504: Verify the historical trajectories from multiple data sources, cross-verify them, remove unreliable data sources, and calculate the usable data using the exit procedure matching algorithm.

[0167] 4.1 Following all candidate departure procedures, take the corresponding waypoints of the planned trajectory one by one. If there is a historical waypoint of the first flight within the first matching range (e.g., 0.5 nautical miles) of the target point among the waypoints, then the target point is the matching point.

[0168] The detailed process for confirming matching points, as described in sections 4.2 and 4.1, includes:

[0169] 4.2.1. If the number of historical waypoints of the first flight that the target point exists in the first matching range is less than or equal to the number of preset waypoints (e.g., 1), then the coverage area changes from the first matching range to the second matching range (3NM).

[0170] 4.2.2. If the number of historical waypoints of the first flight within the first matching range is greater than the number of the first preset waypoints (e.g., 1), and less than or equal to the number of the second preset waypoints (e.g., 3), and the departure point of the planned trajectory of the candidate departure procedure is different from the historical departure point of the first flight, then the coverage area changes from the first matching range to the third matching range (5.5NM); where 0.5NM is the error range; after consulting with the air traffic control unit, 5NM is the more commonly used offset data.

[0171] 4.3 All candidate departure procedures are analyzed, and the one with the most matching points is selected as the target departure procedure. The departure point corresponding to this candidate departure procedure is the actual departure point of the second flight. The departure distance of this candidate departure procedure is denoted as the planned departure distance corresponding to the target departure procedure. The distance between the actual departure point of the second flight and the historical departure point of the first flight (often the last QAR point) is calculated and denoted as the departure point difference.

[0172] 4.4. The detailed judgment process of the program with the most matching points in 4.3 includes:

[0173] 4.4.1. If the number of candidate departure procedures with the maximum number of matching points is less than or equal to the threshold number of candidate departure procedures (1), no further matching will be performed. The candidate departure procedure is deemed to have deviated from the target departure procedure of the second flight, and the candidate departure procedure is recorded as "ATC DIRECT".

[0174] 4.4.2. If the number of candidate departure procedures with the maximum number of matching points is greater than the threshold (1) for the number of candidate departure procedures, calculate the actual departure distance and determine whether the departure point of the planned trajectory is consistent with the historical departure point of the first flight; if consistent, it is the target departure procedure; if inconsistent, it is determined that the candidate departure procedure deviates from the target departure procedure of the second flight.

[0175] 4.4.3. If the number of matching points of the candidate departure procedure with the maximum number of matching points is less than or equal to the threshold (2), and the departure point of the planned trajectory is inconsistent with the historical departure point of the first flight, the candidate departure procedure is deemed to have deviated from the target departure procedure of the second flight, and the candidate departure procedure is recorded as "ATC DIRECT".

[0176] 4.4.4. When the departure airport is not the target airport (such as the airport in Xiamen), the judgment is made according to whether the proportion of the number of matching points to the historical departure points of the first flight exceeds the corresponding threshold for the number of matching points. If the ratio of matching points to planned points is less than 0.7, the candidate departure procedure is considered to have deviated from the target departure procedure of the second flight.

[0177] In sections 4.5 and 4.3, when there are multiple candidate exit procedures with the maximum number of matching points, the procedure with the same exit point as the planned procedure is selected first; the procedure with the PBN identifier is also selected first. If none of the conditions are met, a judgment is made based on distance calculation.

[0178] In sections 4.6 and 4.5, the specific steps for determining the departure distance are as follows: the departure distance is taken as distance L; the first candidate departure procedure is determined as procedure A, and the first planned departure distance is distance A; the second candidate departure procedure is determined as procedure B, and the first planned departure distance is distance B.

[0179] 4.6.1. When distance L is less than distance A and distance L is less than distance B, select the code of program B as the code of the target departure program;

[0180] 4.6.2. When distance L is greater than distance A and distance L is greater than distance B, the code of program A is selected as the code of the target departure program;

[0181] 4.6.3. When the difference between distance A and distance B is greater than the planned departure distance range (20 nautical miles), and distance L is between distance A and distance B, based on distance L, compare distance B with the sum of the planned departure distance range (20 nautical miles); if it is less than, the code of procedure B is the code of the target departure procedure; if it is greater than, the code of procedure A is the code of the target departure procedure.

[0182] 4.6.4. When the difference between distance A and distance B is less than the planned departure distance range (20 nautical miles), and distance L is between distance A and distance B, the departure procedure corresponding to the closer distance is taken as the target departure procedure code.

[0183] Step 505: Predict the target departure procedure as the actual departure procedure for the second flight.

[0184] In one embodiment, the prediction model is described using multiple feature terms, and data is collected according to these feature terms to achieve modeling. The target term predicted by the prediction model is at least one of the takeoff runway and the target departure procedure for the second flight.

[0185] The data for the feature items includes flight-related planning information, such as flight date, airline, departure / arrival airport, estimated departure time, and aircraft type; aeronautical information, including total airport traffic and arrival / departure traffic; and weather information, including wind speed, wind direction, visibility, cloud base height, cloud cover, and special weather phenomena from METAR and TAF reports. The model feature items and their preset formats are shown in Table 1.

[0186] Table 1

[0187]

[0188]

[0189] Furthermore, such as Figure 6 As shown, the training method of the prediction model is explained.

[0190] Considering that the majority of influencing variables in the prediction of takeoff runway and departure procedures are discrete numerical or categorical variables, and that the Extremely Randomized Trees (ET) algorithm in machine learning is well-suited for processing discrete variable data, effectively avoiding overfitting by introducing resampling techniques and a random splitting strategy, it exhibits excellent performance when handling large-scale data volumes and high-dimensional features, demonstrating good generalization performance and high accuracy. Therefore, the ET algorithm is adopted as the base model. The data calculated by the base model involves vectors containing numerical feature representations of flight plans, aeronautical information, and weather information, as well as the data and corresponding processes of the second flight's takeoff runway and target departure procedures combined with these vectors, resulting in prediction models for the takeoff runway and departure procedures. The specific steps are as follows:

[0191] Step 601: According to the steps corresponding to the prediction models of takeoff runway and departure procedure, obtain the first flight takeoff runway related data (ACTRWY) and departure procedure related data (ACTPGM) of the designated airport, and generate datasets A1 and A2 as the target columns of the prediction model;

[0192] Step 602: Collect flight plans, aeronautical information, and weather information for the first flight, integrate them according to the preset format of the feature items, and create datasets D1 and D2 as feature items of the model;

[0193] Step 603: Merge A1 and A2 with D1 and D2 respectively according to flight ID, update datasets D1 and D2, perform data preprocessing, including data integration, data cleaning and feature construction, one-hot encoding, remove problematic data, and obtain the updated dataset;

[0194] Step 604: Based on the updated datasets D1 and D2, respectively, use the ET algorithm to train the takeoff runway prediction model and the departure procedure prediction model, and perform five-fold cross-validation and hyperparameter optimization.

[0195] Step 605: Output the prediction model with the best comprehensive score as the takeoff runway prediction model and the departure procedure prediction model, respectively.

[0196] Step 606: Publish the takeoff runway prediction model and departure procedure prediction model output in step 605 as services, and provide APIs for application systems to call. When calling, input the data corresponding to the second flight into the called model according to the preset format of the feature data.

[0197] The data for the second flight can be updated feature data obtained from the first flight, or it can be updated feature data obtained from the second flight.

[0198] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0199] Based on the same inventive concept, this application also provides a flight plan generation apparatus for implementing the flight plan generation method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more flight plan generation apparatus embodiments provided below can be found in the limitations of the flight plan generation method described above, and will not be repeated here.

[0200] In one embodiment, such as Figure 8 As shown, a flight plan generation device is provided, comprising: a data acquisition module 702, a runway selection module 704, a departure procedure selection module 706, and a plan generation module 708, wherein:

[0201] The data acquisition module 702 is used to acquire the historical trajectory of the first flight and the candidate runways for the second flight;

[0202] Runway selection module 704 is used to select the takeoff runway for the second flight from the candidate runways based on the historical trajectory of the first flight;

[0203] The departure procedure selection module 706 is used to select a target departure procedure from the candidate departure procedures based on the matching degree between the planned trajectory of the candidate departure procedures and the historical trajectory of the first flight.

[0204] The plan generation module 708 is used to generate a flight plan for the second flight based on the takeoff runway of the second flight and the target departure procedure.

[0205] In one embodiment, the historical trajectory of the first flight includes first coordinate points and second coordinate points at different altitudes obtained in chronological order; the runway selection module 704 includes:

[0206] Angle query unit is used to query the difference in azimuth angle between the first coordinate point and the candidate runway entrance when the height of the first coordinate point is a preset height value, based on the difference in azimuth angle between the first coordinate point and the second coordinate point, to obtain the runway azimuth angle query result.

[0207] The runway calculation unit is used to search for the candidate runway corresponding to the runway azimuth query result from the candidate runways;

[0208] The runway selection unit is used to select the runway on the opposite side as the takeoff runway for the second flight based on the candidate runways corresponding to the runway azimuth query results.

[0209] In one embodiment, the runway selection module 704 further includes:

[0210] The spacing comparison unit is used to compare the spacing between the first coordinate point and the runway entrance of each candidate runway when the height of the first coordinate point and the height of the runway are greater than the preset height value, and to obtain the spacing comparison result.

[0211] A runway screening unit is used to screen the candidate runways based on the spacing comparison results;

[0212] The takeoff runway selection unit is used to select the runway on the opposite side of the selected candidate runways as the takeoff runway for the second flight.

[0213] In one embodiment, the data acquisition module 702 is used for:

[0214] Obtain the historical trajectory of the first flight, and then obtain the second coordinate point from the historical trajectory;

[0215] Based on the distance between the second coordinate point and the runway entrance of each preset runway, preset runways on the same side as the trajectory of the first flight are selected.

[0216] The pre-defined runway on the same side as the historical trajectory of the first flight is used as the candidate runway for the second flight.

[0217] In one embodiment, the data acquisition module 702 is configured to:

[0218] Select waypoints sequentially from the original historical trajectory of the first flight;

[0219] The waypoints are verified based on their altitude data.

[0220] Generate a verified historical trajectory based on the verified waypoints;

[0221] The candidate runway for the second flight is obtained based on the verified historical trajectory.

[0222] In one embodiment, the departure procedure selection module 706 includes:

[0223] The matching point determination unit is used to determine whether each point in the planned trajectory of the candidate departure procedure is a matching point by comparing it with the historical route points of the first flight.

[0224] The matching degree calculation unit is used to calculate the matching degree between the planned trajectory and the historical trajectory based on the number of matching points in the planned trajectory of the candidate departure procedure;

[0225] The target departure procedure selection unit is used to select a target departure procedure from the candidate departure procedures according to the matching degree between the planned trajectory and the historical trajectory.

[0226] In one embodiment, the matching point determination unit is specifically used for:

[0227] If the number of historical waypoints of the first flight existing in the first matching range is greater than the number of the first preset waypoints, then the target point in the planned trajectory of the candidate departure procedure is the matching point;

[0228] If the number of historical waypoints of the first flight existing in the first matching range is less than or equal to the preset number of waypoints, then determine whether the target point is the matching point based on whether the historical waypoints of the first flight exist in the second matching range of each point in the planned trajectory of the candidate departure procedure; the second matching range is greater than the first matching range.

[0229] If the number of historical waypoints of the first flight existing in the first matching range is greater than the number of the first preset waypoints and less than or equal to the number of the second preset waypoints, and the departure point of the planned trajectory of the candidate departure procedure is different from the historical departure point of the first flight, then the target point is determined to be the matching point according to whether the historical waypoint of the first flight exists in the third matching range of each point in the planned trajectory of the candidate departure procedure; the third matching range is greater than the second matching range.

[0230] In one embodiment, the matching degree is determined based on the number of matching points between the planned trajectory and the historical trajectory of the first flight, and the departure point; the target departure procedure selection unit is specifically used for:

[0231] When there are multiple candidate departure procedures with the maximum number of matching points, obtain the historical departure point of the first flight and the departure point of the planned trajectory of the candidate departure procedure with the maximum number of matching points.

[0232] Determine whether the departure point of the planned trajectory is consistent with the historical departure point of the first flight;

[0233] If so, select the target departure procedure based on the candidate departure procedure that matches the historical departure point of the first flight;

[0234] If not, the target departure procedure is selected from the candidate departure procedures based on the flight departure distance between the historical departure point of the first flight and the departure airport, and the planned departure distance between the planned departure point of each candidate departure procedure and the departure airport.

[0235] In one embodiment, the target departure procedure selection unit is further configured to:

[0236] When the departure airport of the second flight is the target airport, and the number of candidate departure procedures with the maximum number of matching points is less than or equal to the threshold number of candidate departure procedures, it is determined that the candidate departure procedure deviates from the target departure procedure of the second flight.

[0237] When the departure airport of the second flight is the target airport, and the number of matching points of the candidate departure procedure with the maximum number of matching points is less than or equal to the number of matching points threshold, and the departure point of the corresponding planned trajectory is inconsistent with the historical departure point of the first flight, it is determined that the candidate departure procedure deviates from the target departure procedure of the second flight.

[0238] If the departure airport of the second flight is not the target airport, the candidate departure procedure is determined to be deviated from the target departure procedure of the second flight based on whether the proportion of the number of matching points to the historical departure points of the first flight exceeds the corresponding threshold of the number of matching points.

[0239] In one embodiment, the apparatus further includes a training module for the prediction model, the training module being used to:

[0240] The historical data of the first flight are resampled and randomly split according to the feature terms respectively;

[0241] Based on the data of resampled and randomly split feature terms, optimize at least one of the prediction models of the takeoff runway and the target departure procedure.

[0242] The optimized prediction model for the takeoff runway is used to predict the takeoff runway.

[0243] The optimized prediction model for the target exit procedure is used to predict the target exit procedure.

[0244] The various modules in the flight plan generation device for the aforementioned flights can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0245] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for generating flight plans. The display unit of the computer device is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0246] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0247] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0248] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0249] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0250] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0251] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0252] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0253] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for generating flight plans, characterized in that, The method includes: Obtain the historical trajectory of the first flight and the candidate runways for the second flight; the historical trajectory of the first flight includes first and second coordinate points at different altitudes obtained in chronological order. When the altitude of the first coordinate point is a preset altitude value, the azimuth difference between the first coordinate point and the second coordinate point is used to query the azimuth difference between the first coordinate point and the candidate runway to obtain the runway azimuth query result; the candidate runway corresponding to the runway azimuth query result is searched from the candidate runways; and the runway on the opposite side is selected as the takeoff runway for the second flight based on the candidate runway corresponding to the runway azimuth query result. Determine whether each point in the planned trajectory of the candidate departure procedure is a matching point by comparing it with the historical waypoints of the first flight; calculate the matching degree between the planned trajectory and the historical trajectory based on the number of matching points in the planned trajectory of the candidate departure procedure; and select the target departure procedure from the candidate departure procedures according to the matching degree between the planned trajectory and the historical trajectory. Based on the takeoff runway of the second flight and the target departure procedure, a flight plan for the second flight is generated.

2. The method according to claim 1, characterized in that, The method further includes: When the height of the first coordinate point and the height of the runway are greater than the preset height value, the distance between the first coordinate point and the runway entrance of each candidate runway is compared to obtain the distance comparison result. Based on the distance comparison results, the candidate running tracks are screened. The runway on the opposite side of the selected candidate runway will be chosen as the takeoff runway for the second flight.

3. The method according to claim 1, characterized in that, The acquisition of the historical trajectory of the first flight and the candidate runways of the second flight includes: Obtain the historical trajectory of the first flight, and then obtain the second coordinate point from the historical trajectory; Based on the distance between the second coordinate point and the runway entrance of each preset runway, preset runways on the same side as the trajectory of the first flight are selected. The pre-defined runway on the same side as the historical trajectory of the first flight is used as the candidate runway for the second flight.

4. The method according to claim 1, characterized in that, The acquisition of the historical trajectory of the first flight and the candidate runways of the second flight includes: Select waypoints sequentially from the original historical trajectory of the first flight; The waypoints are verified based on their altitude data. Generate a verified historical trajectory based on the verified waypoints; The candidate runway for the second flight is obtained based on the verified historical trajectory.

5. The method according to claim 1, characterized in that, The determination of whether each point in the planned trajectory according to the candidate departure procedure matches the historical route points of the first flight, includes: If the number of historical waypoints of the first flight existing in the first matching range is greater than the number of the first preset waypoints, then the target point in the planned trajectory of the candidate departure procedure is the matching point; If the number of historical waypoints of the first flight existing in the first matching range is less than or equal to the preset number of waypoints, then determine whether the target point is the matching point based on whether the historical waypoints of the first flight exist in the second matching range of each point in the planned trajectory of the candidate departure procedure; the second matching range is greater than the first matching range. If the number of historical waypoints of the first flight existing in the first matching range is greater than the number of the first preset waypoints and less than or equal to the number of the second preset waypoints, and the departure point of the planned trajectory of the candidate departure procedure is different from the historical departure point of the first flight, then the target point is determined to be the matching point according to whether the historical waypoint of the first flight exists in the third matching range of each point in the planned trajectory of the candidate departure procedure; the third matching range is greater than the second matching range.

6. The method according to claim 1, characterized in that, The matching degree is determined based on the number of matching points and departure points between the planned trajectory and the historical trajectory of the first flight; the step of selecting a target departure procedure from the candidate departure procedures according to the matching degree between the planned trajectory and the historical trajectory includes: When there are multiple candidate departure procedures with the maximum number of matching points, obtain the historical departure point of the first flight and the departure point of the planned trajectory of the candidate departure procedure with the maximum number of matching points. Determine whether the departure point of the planned trajectory is consistent with the historical departure point of the first flight; If so, select the target departure procedure based on the candidate departure procedure that matches the historical departure point of the first flight; If not, the target departure procedure is selected from the candidate departure procedures based on the flight departure distance between the historical departure point of the first flight and the departure airport, and the planned departure distance between the planned departure point of each candidate departure procedure and the departure airport.

7. The method according to claim 1, characterized in that, The method further includes: When the departure airport of the second flight is the target airport, and the number of candidate departure procedures with the maximum number of matching points is less than or equal to the threshold number of candidate departure procedures, it is determined that the candidate departure procedure deviates from the target departure procedure of the second flight. When the departure airport of the second flight is the target airport, and the number of matching points of the candidate departure procedure with the maximum number of matching points is less than or equal to the number of matching points threshold, and the departure point of the corresponding planned trajectory is inconsistent with the historical departure point of the first flight, it is determined that the candidate departure procedure deviates from the target departure procedure of the second flight. If the departure airport of the second flight is not the target airport, the candidate departure procedure is determined to be deviated from the target departure procedure of the second flight based on whether the proportion of the number of matching points to the historical departure points of the first flight exceeds the corresponding threshold of the number of matching points.

8. The method according to claim 1, characterized in that, The method further includes a training step for the prediction model, the training step comprising: The historical data of the first flight are resampled and randomly split according to the feature terms respectively; Based on the data of resampled and randomly split feature terms, optimize at least one of the prediction models of the takeoff runway and the target departure procedure. The optimized prediction model for the takeoff runway is used to predict the takeoff runway. The optimized prediction model for the target exit procedure is used to predict the target exit procedure.

9. A flight plan generation device for flights, characterized in that, The device includes: The data acquisition module is used to acquire the historical trajectory of the first flight and the candidate runways of the second flight; the historical trajectory of the first flight includes first coordinate points and second coordinate points with different altitudes acquired in chronological order. The runway selection module is used to, when the altitude of the first coordinate point is a preset altitude value, query the difference in azimuth angle between the first coordinate point and the second coordinate point and the runway opening of the candidate runway based on the difference in azimuth angle between the first coordinate point and the second coordinate point, and obtain the runway azimuth angle query result; search for the candidate runway corresponding to the runway azimuth angle query result from the candidate runways; and select the runway on the opposite side as the take-off runway for the second flight based on the candidate runway corresponding to the runway azimuth angle query result. The departure procedure selection module is used to determine whether each point in the planned trajectory of the candidate departure procedure is a matching point according to whether each point in the planned trajectory matches the historical waypoints of the first flight; calculate the matching degree between the planned trajectory and the historical trajectory based on the number of matching points in the planned trajectory of the candidate departure procedure; and select the target departure procedure from the candidate departure procedures according to the matching degree between the planned trajectory and the historical trajectory. The plan generation module is used to generate a flight plan for the second flight based on the takeoff runway of the second flight and the target departure procedure.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images