System and method for determining runway configuration of an airport

Abstract

A system and method for determining a runway configuration for an airport. The method includes: retrieving (210) recorded surveillance data (214) including instances of aircraft positions (216) at an airport; determining (220) a plurality of three-dimensional surveillance cells (302) at each end (310a, 310b) of one or more runways (102) at an airport; calculating (230) a count of aircraft positions (216) within each surveillance cell (302); and determining (240) a current configuration of each runway (102) based on the counts calculated for the surveillance cells (302) for the runways (102). The predicted runway configuration for the airport can be used to update a flight plan for an aircraft to reduce overall flight duration and minimize fuel consumption.

Description

Technical Field

[0001] The present disclosure is included in the field of avionics and relates to a method for determining the configuration of an airport runway and a method for updating and improving an aircraft flight plan. Background Art

[0002] Knowing the runway configuration of an airport is considered critical information to provide accurate services and forecasts for efficient airline operations. Predicting the runway configuration of both the departure and destination airports has a direct impact on the flight performance of the aircraft, allowing for proper planning of the total flight duration and the fuel required to perform the flight.

[0003] While some airports update and publish runway configuration information to airlines, many others do not. Existing solutions for airports that do not report this information require geometric modeling of the trajectories of arriving and departing flights to find possible configurations for each runway. This approach has high computational cost and requires flights to land, making it inapplicable for real-time detection, such as during flight.

[0004] Therefore, there is a need to provide a computationally efficient method of determining the current runway configuration that overcomes existing shortcomings. Summary of the invention

[0005] The present disclosure relates to a system, a computer-implemented method, and a computer program product for determining a runway configuration of an airport and updating an aircraft flight plan based on the determined runway configuration.

[0006] The method used to determine the runway configuration uses a data-driven predictive algorithm that identifies the airport runway configuration based solely on automated aircraft surveillance reports and known airport layouts. The method does not rely on geometric calculations of aircraft trajectories to infer the airport's runway configuration; instead, the method uses only surveillance data from the terminal area.

[0007] The method includes a data collection step in which aircraft positions are obtained (e.g., as reported by ADS-B (Automatic Dependent Surveillance-Broadcast), secondary radar, or any other data source available at the airport). The method also includes defining a 3D grid centered on the airport and counting the number of position reports of aircraft in each surveillance cell of the grid over a period of time. The size and number of surveillance cells used for the data driven process and the selection of the 3D grid can be refined in an iterative process.

[0008] The method of the present disclosure allows flight planning to be improved by predicting airport runway configurations, thereby providing better estimates of actual flight time and required fuel. The present disclosure applies a machine learning approach to determine the runway configurations of both departure and destination airports. The goal is to minimize fuel consumption for approaching and taking off from the airport and taxiing to / from the gate once the airport configuration is known. By knowing the runway configuration, fuel use can be accurately estimated. The method uses a model in which the actual taxi distance can also be optimized in an iterative process using a controlled training data set.

[0009] A first aspect of the present disclosure relates to a method for determining a runway configuration of an airport. The method comprises: retrieving recorded surveillance data including instances of aircraft positions at the airport; determining a plurality of three-dimensional surveillance cells at each end of at least one runway of the airport; calculating a count of aircraft positions within each surveillance cell; and determining a current configuration of each runway based on the counts calculated for the surveillance cells of the runway.

[0010] The step of determining the surveillance unit may comprise the steps of: retrieving information about the spatial arrangement of runways at the airport; defining a three-dimensional grid on the airport, the grid being formed by a set of grid cells; and selecting the surveillance unit from among the grid cells.

[0011] The step of determining the current runway configuration for each runway may include pre-training based on a plurality of training monitoring data sets, the training monitoring data sets comprising a plurality of instances of all available configurations for the corresponding runway, wherein each monitoring data set may correspond to a different time period.

[0012] In an embodiment corresponding to a supervised learning method, the training is performed using a machine learning algorithm, wherein the machine learning algorithm is supplied with known runway configurations added to a training monitoring data set.

[0013] In another embodiment corresponding to the unsupervised learning method, training is performed as follows: a count of aircraft positions within each surveillance unit is calculated for each surveillance data set; the number of available configurations for each runway is indicated; and for each runway, the counts of the surveillance units associated with the runway are clustered to obtain a cluster corresponding to each available configuration.

[0014] According to an embodiment, the surveillance unit is a parallelepiped.The surveillance unit may have an axis of symmetry extending parallel to the associated runway.The size of the surveillance unit associated with the runway may be proportional to the length of the runway.

[0015] The present disclosure also relates to a method for updating an aircraft flight plan, including the previously discussed method of determining the current runway configuration of an airport and updating the flight plan of the aircraft based on the predicted airport runway configuration. The airport can be an arrival airport or a departure airport for the aircraft. In an embodiment, the update of the flight plan is performed before the aircraft departs, and the method includes updating the amount of fuel to be loaded on the aircraft based on the current runway configuration of the airport. In another embodiment, updating the flight plan includes updating the taxi plan of the aircraft at the airport. In another embodiment, the airport is an arrival airport for the aircraft, and the update of the flight plan is performed during the flight of the aircraft; and the update of the flight plan includes determining a flight path approaching the airport taking into account the current runway configuration of the airport.

[0016] Another aspect of the present disclosure relates to a system for determining a runway configuration of an airport. The system comprises a processing unit configured to perform the steps of the method as described above. The present disclosure also relates to a system for updating an aircraft flight plan, comprising a system for determining a runway configuration of an airport and a flight plan update module configured to receive a flight plan of the aircraft and calculate an updated flight plan for the aircraft based on the current runway configuration of the airport.

[0017] According to another aspect of the present disclosure, a computer program product for determining a runway configuration of an airport is provided. The computer program product includes at least one computer-readable storage medium storing a set of instructions, which, when executed by a processor, causes the processor to perform a method for determining an airport runway configuration. Another aspect of the present disclosure relates to a computer program product for updating an aircraft flight plan, including at least one computer-readable storage medium storing a set of instructions, which, when executed by a processor, causes the processor to perform a method for updating an aircraft flight plan as described above.

[0018] The features, functions, and advantages that have been discussed can be achieved independently in various embodiments or may be combined in yet other embodiments further details of which can be seen with reference to the following description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Below, there is described very briefly a series of drawings which contribute to a better understanding of the invention and which relate expressly to an embodiment of said invention presented as a non-limiting example thereof.

[0020] Figure 1 Indicates four different configurations of active runways.

[0021] Figure 2 A flow chart depicting a method for determining a runway configuration for an airport is depicted.

[0022] Figure 3 An exemplary embodiment of a surveillance unit defined at each end of a runway is shown.

[0023] FIG. 4A to FIG. 4D Several examples of different shapes, sizes and locations of surveillance units at each end of a runway are shown.

[0024] Figure 5A and Figure 5B The surveillance units selected for a three-runway airport are depicted in perspective and plan views.

[0025] Figure 6 An exemplary flow chart showing monitoring unit determination.

[0026] Fig. 7A and Figure 7B Aircraft positions for different runway configurations are depicted.

[0027] Figure 8 Flowchart showing the steps of counting the number of aircraft positions in each surveillance unit.

[0028] Fig.9A , Fig. 9B , Fig.10 and Fig.11 Represents different scenarios for updating the aircraft flight plan with the predicted runway configuration.

[0029] Fig.12 An exemplary implementation of a system for determining a runway configuration of an airport is depicted.

[0030] Fig.13 An exemplary implementation of a system for updating an aircraft flight plan is shown. DETAILED DESCRIPTION

[0031] The present disclosure relates to a computer-implemented method for predicting the runway configuration of a departure and / or destination airport and updating an aircraft flight plan based on the prediction.

[0032] The active runway 102 of an airport may be operated in several configurations. First, the runway 102 may be configured as an arrival (A) runway or a departure (D) runway. In addition, for each runway, the direction from which the aircraft 104 departs or arrives may also change over time, leaving four available configurations, such as Figure 1 The runway configuration is usually specified in the form {A1,A2,D1,D2}, where A1 and A2 are the arrival runways and D1 and D2 are the departure runways in the two available directions (s1,s2).

[0033] Thus, the active runway 102 may be operated in four available configurations (A1, A2, D1, D2).Using automatic aircraft surveillance reports and known airport geometry, the method of the present disclosure predicts the current configuration of one or more runways of an airport.

[0034] The steps of this method are depicted in Figure 2 Specifically, the method includes a first step of retrieving (eg, obtaining) 210 recorded monitoring data 214 stored in a monitoring data warehouse 212 .

[0035] Surveillance data 214 includes multiple instances of aircraft positions 216 at an airport during takeoff or landing maneuvers on one or more runways at the airport. Aircraft positions 216 may be obtained via ADS-B, ASDI (industry aircraft situation display), secondary radar, ground radar, or any other data registry source available at the airport. Aircraft positions 216 include all necessary information to accurately locate the aircraft in space and time (e.g., altitude h, longitude λ, and latitude ) and time record t) as well as an aircraft identifier (ID) such as the tail number.

[0036] It can be in a given time period or several time periods (T1, T2, ..., T N ). Aircraft positions 216 may be registered for all airport runways or only for a particular runway or set of runways (e.g., after filtering is applied). Typically, instances of aircraft positions 216 correspond to different aircraft and include multiple positions registered over time for each aircraft involved. For example, aircraft positions may be collected every few seconds during landing or takeoff.

[0037] The method further includes the step of: performing a runway test at one or more runways 102 {R1, R2, ..., R m} The step of determining 220 a three-dimensional monitoring unit at each end of the runway 102. Figure 3 In the exemplary embodiment depicted, two surveillance cells 302 (e.g., 302a, 302b) are defined at each end of the runway 102. A simpler grid would employ only the four volumes depicted, two at each extreme of the runway, one near the ground and one above it. This basic grid captures the different slopes for landing and takeoff aircraft, making it possible to distinguish the direction of use of the runway. Figure 3 In an embodiment, the monitoring unit is a parallelepiped, and its size can be proportional to the runway length (LR). For example, the size of the parallelepiped (length L, width W and / or height H) can be such that L=a·LR, W=b·LR, and / or H=c·LR. In order to define the three-dimensional monitoring unit 302 around each runway 102, the spatial arrangement of the airport runways 102 must be known in advance.

[0038] like Figure 3As depicted in , the monitoring unit 302 can also be obtained by defining a main parallelepiped 304 centered above the runway 102 and discarding its central portion, so that two upper parts and two lower parts are defined at each end of the main parallelepiped 304, thereby obtaining four monitoring units 302 in the form of four smaller parallelepipeds. The dimensions (length TL, height TH and width W) of the main parallelepiped 304 can also be defined with respect to the runway length: TL=x·LR, W=b·LR, and / or TH=y·LR. The parameters a, b, c, x and y can be numbers or percentages related to the runway length LR and can be optimized in an iterative process using a controlled training data set.

[0039] exist Figure 3 In the embodiment of FIG. 3 , the monitoring unit 302 is represented as a rectangular parallelepiped. However, other shapes that define the volume of the three-dimensional monitoring unit 302 may be selected, such as a cylinder, a cube, or a hollow cylinder sector. In addition, although Figure 3 The embodiment of the present invention only shows two surveillance units 302 on each side of the runway 102, but more than two surveillance units 302 may be defined at each end of the runway 102. In an embodiment, at least two surveillance units 302 at different heights are defined at each end of the runway 102 so that an aircraft takeoff can be distinguished from an aircraft landing on the runway 102.

[0040] A monitoring unit 302 is defined at each end of the runway 102. The expression "at each end of the runway" may be understood as the monitoring unit 302 partially or completely covering the extreme area or end of the runway 102 as seen from a top view. Figure 3 Example of a top view Figure 4A In the embodiment, each monitoring unit 302 overlaps or includes one of the two ends (310a, 310b) of the runway 102 in plan view. Figure 4B Another embodiment using a cylinder instead of a parallelepiped as the monitoring unit 302 overlaps the end area of ​​the runway 102 is shown. Alternatively, "at each end of the runway" can also be understood as monitoring unit 302 such as Figure 4C adjoining the top as depicted in the example of Figure 4D As in the example of FIG. 5 , the outer region is located near the end (but not connected to or overlapping the end).

[0041] exist Figure 5A and Figure 5B , respectively, show a three-dimensional view and a top view of an airport with three runways (102a, 102b, 102c) and its spatial arrangement. The airport has two parallel long runways (102a, 102b) and one intersecting short runway (102c). In this embodiment, Figure 6As reflected in the flowchart of FIG. 6 , the step of determining the monitoring unit 302 includes retrieving 602 information about the spatial arrangement of the airport runways (e.g., the length, orientation and starting point of each runway obtained from the airport layout 600 or any other equivalent information required to spatially locate the runways), and defining 604 a three-dimensional grid 500 ( Figure 5B Although using a rectangular parallelepiped or a cube as the basic grid unit 502 may be advantageous for computational efficiency, the shape of the grid unit 502 may also be different.

[0042] A grid 500 is defined over the area covering the airport so that it contains at least (in a top view) all relevant airport runways. In this exemplary embodiment, the grid 500 is formed by a matrix of 9×9×3 grid cells 502 (9 rows, 9 columns and a height of 3 cells), but in a real scenario it may include many more grid cells 502 (tens, hundreds or even thousands of cells). A monitoring cell 302 is then selected 606 from among the cells 502 of the grid 500. In Figure 5A In the example of , sixteen grid cells 502 have been identified as being associated with traffic counting (i.e., identified as surveillance cells 302) in order to detect the configuration of the airport runways: two surveillance cells 302 at each end of the parallel long runways (102a, 102b) and four surveillance cells at each end of the intersecting short runway (102c).

[0043] Once the monitoring units are selected 606, the spatial arrangement 222 of the monitoring units may be determined 608; specifically, if i monitoring units are calculated, the monitoring units 302 are spatially defined, so that the volumes {V1, V2, ..., V i}.

[0044] In such Figure 5A In the example depicted in , only the first two heights of the matrix 500 are considered relevant to the selection of the monitoring cell 302, but different selections of cells may also be considered (e.g., considering cells at the third height of the matrix). As previously explained, the monitoring cell 302 may:

[0045] - Partially or completely cover (in plan view) the end of the runway.

[0046] -Abutting against the end of the runway.

[0047] - Close to the end of the runway (and just outside the end).

[0048] return Figure 2 The method of determining the runway configuration further comprises calculating 230 using the retrieved monitoring data 214 for a given time period (Tx ) or a number of time periods (eg, T1 and T2) in which the number of aircraft positions 216 contained in each monitoring unit 302 is determined.

[0049] Fig. 7A Aircraft positions 216 registered during different time periods when the runway 102 is operated in two different configurations (A1, D1) in the first direction s1 are shown. The instances of aircraft positions 216 generally correspond to samples collected from multiple different aircraft 104, and some of the samples may relate to the same aircraft, reflecting departure or arrival processes at different times. Fig. 7A The samples shown in the lower left edge of correspond to aircraft positions 216 recorded when runway 102 operates in configuration A1 (i.e., arrival, direction s1), with most samples being within corresponding lower surveillance unit 302a. The samples in the upper right region relate to aircraft positions 216 registered when runway 102 operates in configuration D1 (i.e., departure, direction s1), with most samples being contained within corresponding upper surveillance unit 302b.

[0050] exist Figure 7B In the embodiment, instances of the aircraft position 216 are registered when the runway 102 is operated in two different configurations (A2, D2) in a second direction s2 opposite to the first direction s1. Figure 7B The aircraft position 216 on the lower left side of the runway 102 was recorded when the runway 102 was running in configuration D2 (departure, direction s2), and most of the samples were contained in the corresponding upper monitoring unit 302b. The aircraft position 216 on the upper right side was registered when the runway 102 was running in configuration A2 (arrival, direction s2), and most of the aircraft positions 216 were contained in the corresponding lower monitoring unit 302a.

[0051] exist Figure 8 , a flowchart depicts an embodiment of the steps involved in calculating 230 a count of aircraft positions 216 within each surveillance unit 302. Aircraft positions 216 stored in surveillance data repository 212 may be grouped in different time periods (T1, T2, ..., T N ). The first step includes selecting 802 one or more of these time periods and retrieving 804 the aircraft position corresponding to the given time period (using, for example, the time record t of the registered aircraft position 216), obtaining the selected aircraft position 216. For each aircraft position 216, the registered aircraft position (h, λ, ) and each monitoring unit 302 {SC1, SC2, ..., SC i} spatial arrangement 222{V1,V2,…,V i} to compare 808. If the aircraft position (h, λ, ) in the volume {V1, V2, ..., V i} (e.g., within a surveillance unit), the count associated with the surveillance unit 302 is updated 810 (increased by 1). Finally, once all selected aircraft positions 216 have been checked 812, the counts 232 {N1, N2, ..., N} of aircraft positions calculated for each surveillance unit are obtained. i}.

[0052] exist Figure 2 In the last step of the flowchart of FIG. 1 , counts 232 of aircraft positions reported in the various surveillance units are supplied to an algorithm for determining 240 the current configuration 250 {C1, C2, …, C} of the various runways {1, 2, …, m} involved based on the counts calculated by the surveillance units 302 for the corresponding runways 102. m}. Select the configuration among all available configurations (A1, A2, D1, D2 or Not used if the runway is not open).

[0053] In order to determine the current runway configuration, pre-training is performed based on a plurality of training monitoring data sets comprising instances of aircraft positions registered for all configurations available for each runway. Each monitoring data set typically corresponds to a different time period during which the runway configuration may have changed. The training may correspond to a supervised learning method (for each instance of the aircraft position, the runway configuration is known) or to an unsupervised learning method (the runway configuration at the recording time t of the instance is unknown).

[0054] In a supervised learning approach, training is performed using a machine learning algorithm that is fed with known runway configurations that are added to a training surveillance data set. The model feature is the number of registered aircraft positions within surveillance cells 302. For each time period and count of each surveillance cell, the known configuration (added to the data set) is the dependent variable and the count of the surveillance cell is the independent variable. After many known configurations and time periods, the machine learning algorithm is used to learn the pattern so that with only the counts 232 of the surveillance cells, the algorithm can output the runway configuration 250 {C1, C2, ..., C m} for all runways {R1, R2, …, R m}.

[0055] The size and shape of the monitoring cells 302 may also be optimized. To find the geometric parameters defining the grid and the monitoring cells (in Figure 3 In the example of {a,b,c,x,y}), the model is trained using a special controlled dataset consisting of data from known time periods of operation at each of the available runway configurations (A1,A2,D1,D2). i ,b i,c i ,x i ,y i} iterations. The geometric parameters that more clearly assign the correct configuration to each data set in each time period are selected. Once the geometric parameters are optimized, deterministic training can be performed with the selected geometric parameters for a real training data set (including, for example, several different dates with varying configurations).

[0056] In an unsupervised learning approach, multiple time periods are clustered together to eventually find multiple clusters that are equivalent to the different airport configurations used. For each runway, the model feature is the number of registered aircraft positions 216 within the monitoring unit 302. Figure 3 In the embodiment of the invention, for one runway, there are four features (because there are four monitoring units). Since the training is unsupervised, the trainer indicates the number of possible configurations to expect for each runway (e.g., A1, A2, D1, D2, or not used). The clustering algorithm is fed many instances of the four features at different intervals and tries to find clusters. Thus, unlike supervised training, the configurations are unknown, but the algorithm finds them by clustering. The clustering algorithm tries to find explicit clusters for five different scenarios: A1, D1 ( Fig. 7A )、A2、D2( Figure 7B ) or not operating (feature is zero). Then, for each cluster, the centroid is found and the reference values ​​of these four features are used to classify future situations. The centroid represents a typical configuration. Once the algorithm is trained, with only the count of monitored cells, the algorithm can output which cluster the configuration belongs to and know the cluster of the configuration that can be used.

[0057] The method for determining the runway configuration of an airport may be used to update a flight plan of an aircraft based on the predicted runway configuration of the airport. Fig.9A , Fig. 9B , Fig.10 and Fig.11 Different scenarios for updating an aircraft flight plan are shown.

[0058] exist Fig.9A and Fig. 9B In the example of , the runway configuration is applied during the pre-flight phase to select an updated flight path and / or an updated amount of fuel to be loaded. The airport for which the runway configuration is predicted may be the aircraft's arrival airport 902 ( Fig.9A ) or departure airport 904( Fig. 9B ). Knowing the runway configuration of arrival airport 902 or departure airport 904 may mean that updated flight path 920 is shorter than original flight path 910, thereby requiring less time and fuel to reach the destination.

[0059] like Fig.10As depicted in the example of , the determination of the current runway configuration of the airport may also be applied during pre-flight to estimate the correct taxi time. Knowing the runway configuration of the airport (departure or arrival airport) may mean a shorter / longer taxi-out / in time than the reference time, thereby adjusting the target off-block time and the arrival schedule to account for the correct taxi time adjustment, departure delays may be avoided. The taxi plan (1002a or 1002b) of the aircraft is updated based on the current runway configuration and the selected runway 102.

[0060] Fig.11 1100 to update the flight path in advance. In many cases, the aircraft follows the original flight path 1110 until the flight plan is updated (last minute flight path adjustment 1120) close to arrival at airport 902. However, knowing in advance the change in the runway configuration at arrival airport 902 can be used to change the original flight path 1110 to an updated flight path 1130 (advance flight path adjustment) in advance to avoid the less efficient last minute flight path adjustment 1120. This update will mean a shorter flight path than waiting for a last minute adjustment.

[0061] Fig.12 A system for determining the runway configuration of an airport is represented. The system substantially performs the steps of the method described previously. In an embodiment, the system is installed on an aircraft. The system comprises a processing unit 1200, such as a processor, a microcontroller or a computer or any other similar electronic device equipped with data processing means. The processing unit 1200 comprises:

[0062] An aircraft position retrieval module 1210 configured to retrieve the recorded surveillance data 214 including instances of aircraft positions 216 for airports.

[0063] A surveillance unit determination module 1220 configured to determine a plurality of three-dimensional surveillance units 302 {SC1, SC2, ..., SC i Module 1220 determines the volumes {V1, V2, ..., V i} spatial arrangement 222.

[0064] - A count calculation module 1230 configured to calculate a count 232 {N1, N2, ..., N i}.

[0065] A runway configuration prediction module 1240 configured to predict the current configuration 250 of each runway 102 based on the counts 232 calculated by the monitoring unit 302 for the runways 102. In an embodiment, the prediction is performed based on a machine learning algorithm such as an artificial neural network, clustering or a support vector machine. To this end, the processing unit 1200 may include a machine learning module 1242 that applies machine learning in a pre-training step using a plurality of training monitoring data sets 1244, the training monitoring data sets 1244 comprising examples of aircraft positions for different configurations of each runway 102.

[0066] Fig.13 A system 1300 for updating an aircraft flight plan is depicted. Fig.9A , Fig. 9B , Fig.10 and Fig.11 The system can be used when the flight path or taxi plan needs to be adjusted (e.g., enhanced or updated to adapt to changing conditions) in the different scenarios described in . The system is preferably installed on the aircraft for autonomous real-time calculation and execution of updated and optimized flight plans.

[0067] The system used to update the aircraft's flight plan uses Fig.12 The system 1300 for flight plan updating includes a flight plan updating module 1310 (which may be a part of the processing unit 1200 or a separate component, such as Fig.13 ), which is configured to receive the flight plan 1320 of the aircraft and calculate an updated flight plan 1330 based on the runway configuration 250 of the predicted airport (which can be the arrival airport 902 or the departure airport 904 of the aircraft). The updated flight plan 1330 can be sent to the flight management system (FMS) 1340 of the aircraft for execution (e.g., execution of an updated flight path or an updated taxi plan). In an embodiment, the processing unit 1200 and the flight plan update module 1310 can be part of the FMS 1340 so that both the calculation and execution of the updated flight plan are performed by the FMS 1340.

Claims

1. A method for updating an aircraft flight plan, the method comprising the following steps: retrieving (210) recorded surveillance data (214) including an instance of an aircraft position (216) at an airport; Determining (220) a plurality of three-dimensional surveillance units (302) at each end (310a, 310b) of at least one runway (102) of the airport; calculating (230) a count of aircraft positions (216) within each surveillance unit (302); determining (240) a current configuration of each runway (102) based on the counts calculated by the monitoring unit (302) for the runway (102); updating a flight plan for the aircraft based on a current runway configuration for the airport; as well as The updated flight plan is autonomously executed by a system installed on the aircraft.

2. The method according to claim 1, wherein: The step of determining (220) a monitoring unit comprises: Retrieving (602) information about the spatial arrangement of airport runways; defining (604) a three-dimensional grid (500) over the airport, the grid being formed from a set of grid cells (502); and The monitoring unit (302) is selected (606) from among the grid units (502).

3. The method according to claim 1 or 2, wherein: The step of determining the current runway configuration of each runway (102) includes pre-training based on a plurality of training monitoring data sets (1244), the training monitoring data sets (1244) including instances of all of the plurality of available configurations for the corresponding runway.

4. The method according to claim 3, wherein: The training is performed using a machine learning algorithm, wherein the machine learning algorithm is supplied with known runway configurations that are added to the training monitoring data set (1244).

5. The method according to claim 3, wherein: The training is performed by the following steps: For each surveillance data set, calculating a count of aircraft positions within each surveillance unit; Indicate the number of available configurations for each runway; as well as For each runway, the counts of the surveillance units associated with the runway are clustered to obtain clusters corresponding to each available configuration.

6. The method according to claim 1 or 2, wherein: The monitoring unit has an axis of symmetry which extends parallel to the associated runway.

7. The method according to claim 1 or 2, wherein: The monitoring unit (302) associated with the runway (102) is a parallelepiped whose dimensions are proportional to the length (LR) of the runway (102).

8. The method according to claim 1, wherein: The airport is an arrival airport (902) or a departure airport (904) of the aircraft, and wherein the step of updating the flight plan includes updating a taxi plan (1002a, 1002b) of the aircraft at the airport.

9. The method according to claim 8, wherein: The step of executing the updated flight plan includes executing an updated taxi plan.

10. The method according to claim 1, wherein: The airport is an arrival airport for the aircraft, and the step of updating the flight plan is performed during the flight of the aircraft; and wherein the step of updating the flight plan includes determining an updated flight path approaching the airport taking into account the current runway configuration of the airport (1130).

11. The method according to claim 10, wherein: The step of executing the updated flight plan includes executing the updated flight path.

12. A system for updating an aircraft flight plan, the system comprising: A processing unit (1200), the processing unit (1200) is configured to: retrieving recorded surveillance data (214) including an instance of an aircraft position (216) at an airport; Determining a plurality of three-dimensional surveillance units (302) at each end (310a, 310b) of at least one runway (102) of the airport; calculating a count (232) of aircraft positions (216) within each surveillance unit (302); as well as determining a current configuration (250) of each runway (102) based on the count (232) calculated by the monitoring unit (302) for the runway (102); and a flight plan update module (1310) configured to receive a flight plan (1320) for an aircraft and calculate an updated flight plan (1330) for the aircraft based on a current runway configuration of the airport, Therein, the system is installed on an aircraft and is configured to autonomously execute an updated flight plan (1330).

13. The system of claim 12, further comprising a flight management system (1340) of the aircraft, the flight management system (1340) configured to execute the updated flight plan (1330).

14. The system according to claim 12 or 13, wherein: The system is configured to execute an updated flight path.

15. The system according to claim 12 or 13, wherein: The system is configured to execute the updated taxi plan.

16. A computer-readable storage medium storing a set of instructions, wherein: The instructions, when executed by a processor, cause the processor to perform the method according to any one of claims 1 to 11.

Images