FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003
COMPLETE SPECIFICATION (See Section 10 and Rule 13)
Title of invention:
SYSTEMS AND METHODS FOR UNSUPERVISED LEARNING BASED HOLDING LOOP COUNT ESTIMATION
IN AERIAL VEHICLES
Applicant
Tata Consultancy Services Limited A company Incorporated in India under the Companies Act, 1956
Having address:
Nirmal Building, 9th floor,
Nariman point, Mumbai 400021,
Maharashtra, India
Preamble to the description
The following specification particularly describes the invention and the manner in which it is to be performed.

TECHNICAL FIELD [001] The disclosure herein generally relates to estimating holding loop count in aerial vehicles, and, more particularly, to systems and methods for unsupervised learning based holding loop count estimation in aerial vehicles.
BACKGROUND
[002] Operations associated with aerial vehicles are affected due to one or more factors such as inclement weather, congestion, and the like. This may require the aerial vehicles to be diverted from a predetermined flight plan for the aerial vehicle from a source to destination and put in a holding loop. While roaming in the holding loop, the aerial vehicles consume significant amount of fuel attributing to high fuel cost. Further, the holding loops may cause time inefficiencies which may lead to delay, disruption in scheduling and maintenance, and increased operating cost. To be able to manage fuel consumption better, the aerial vehicles with holding loops are required to be differentiated from the aerial vehicles without holding loops. This helps in enabling proper scheduling of the aerial vehicle, optimized fuel consumption which can further bedrock for future prediction of holding loops of the aerial vehicles.
[003] Conventional methods for holding loop detection can identify the holding loops only when data of multiple aerial vehicles in same path is provided. Further, the conventional systems are dependent on flight plan of the aerial vehicles apriori which is further used to compute a deviation to detect the holding loop.
SUMMARY
[004] Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems.
[005] In an aspect, there is provided a processor implemented method, the method comprising: receiving, a plurality of data of a plurality of aerial vehicles, each of the plurality of aerial vehicles moving from one or more sources

to destinations, wherein the plurality of data comprises information pertaining to the one or more sources and destinations and a set of parameters obtained from one or more sensors associated with the plurality of aerial vehicles; parsing the plurality of received data to obtain a plurality of parsed files, wherein the plurality of parsed files are obtained by segregating the plurality of received data of the plurality of aerial vehicles such that each parsed file stores (i) meta information of the plurality of aerial vehicles and (ii) the set of parameters obtained from the one or more sensors associated with the plurality of aerial vehicles for a specific source-destination pair; deriving, using the set of parameters comprised in each parsed file, a set of features in accordance with one or more physics based principles.
[006] In accordance with an embodiment of the present disclosure, the set of parameters obtained from the one or more sensors comprises altitude(sy ), speed at ground(vx ), latitude, and longitude of the plurality of aerial vehicles. In accordance with an embodiment of the present disclosure, the set of derived features comprises displacement along x- axis ( sx ), velocity along y-axis (vy ), acceleration along x- axis (Ax ) and y-axis (Ay ), derivative of latitude, derivative of longitude, and angle of elevation or depression (θ) of the plurality of aerial vehicles. In accordance with an embodiment of the present disclosure, the step of deriving the set of features is preceded by sampling the plurality of parsed files using a mean median filtering technique to obtain uniform values of the set of parameters obtained from the one or more sensors.
[007] In accordance with an embodiment of the present disclosure, the method further comprising obtaining, by detecting zero crossings among a plurality of values of at least two features from the set of derived features, a reduced set of features; identifying, using a pattern recognition based unsupervised approach, a region of interest in the reduced set of features, wherein the region of interest is indicative of presence of a holding loop and identified when two transitions occur between three consecutive values of the obtained reduced set of features; and estimating, based on at least one of (i) the pattern

recognition based unsupervised approach and (ii) a graph based approach, holding loop counts of the plurality of aerial vehicles. In accordance with an embodiment of the present disclosure the graph based approach to estimate holding loop counts of the plurality of aerial vehicles comprises: assigning a node from a plurality of nodes to each value from a plurality of values of a specific derived feature selected from the set of derived features, wherein the node is assigned based on a comparison of each value of the specific derived feature with one or more thresholds, and wherein the specific derived feature is angle of elevation or depression; and determining one or more cyclic patterns of the plurality of nodes assigned to the plurality of values of the specific derived feature to be counted as the holding loop.
[008] In accordance with an embodiment of the present disclosure, the method further comprising: ensembling the holding loop counts obtained from both of the (i) the pattern recognition based unsupervised approach and (ii) the graph based approach to determine a final holding loop count of the plurality of aerial vehicles; and determining, duration of each holding loop of each of the plurality of aerial vehicles based on difference of start and end time of the holding loop.
[009] In another aspect, there is provided a system, the system comprising: one or more data storage devices operatively coupled to one or more hardware processors and configured to store instructions configured for execution via the one or more hardware processors to: receive, a plurality of data of a plurality of aerial vehicles, each of the plurality of aerial vehicles moving from one or more sources to destinations, wherein the plurality of data comprises information pertaining to the one or more sources and destinations and a set of parameters obtained from one or more sensors associated with the plurality of aerial vehicles; parse the plurality of received data to obtain a plurality of parsed files, wherein the plurality of parsed files are obtained by segregating the plurality of received data of the plurality of aerial vehicles such that each parsed file stores (i) meta information of the plurality of aerial vehicles and (ii) the set of parameters obtained from the one or more sensors associated with the plurality of

aerial vehicles for a specific source-destination pair; derive, using the set of parameters comprised in each parsed file, a set of features in accordance with one or more physics based principles.
[010] In accordance with an embodiment of the present disclosure, the set of parameters obtained from the one or more sensors comprises altitude(sy ), speed at ground(vx ), latitude, and longitude of the plurality of aerial vehicles. In accordance with an embodiment of the present disclosure, the set of derived features comprises displacement along x- axis (sx ), velocity along y-axis (vy ), acceleration along x- axis (Ax ) and y-axis (Ay ), derivative of latitude, derivative of longitude, and angle of elevation or depression (θ) of the plurality of aerial vehicles. In accordance with an embodiment of the present disclosure, the step of deriving the set of features is preceded by sampling the plurality of parsed files using a mean median filtering technique to obtain uniform values of the set of parameters obtained from the one or more sensors.
[011] In accordance with an embodiment of the present disclosure, the one or more processors 104 are further configured to obtain, by detecting zero crossings among a plurality of values of at least two features from the set of derived features, a reduced set of features; identify, using a pattern recognition based unsupervised approach, a region of interest in the reduced set of features, wherein the region of interest is indicative of presence of a holding loop and identified when two transitions occur between three consecutive values of the obtained reduced set of features; and estimate, based on at least one of (i) the pattern recognition based unsupervised approach and (ii) a graph based approach, holding loop counts of the plurality of aerial vehicles. In accordance with an embodiment of the present disclosure the graph based approach to estimate holding loop counts of the plurality of aerial vehicles comprises: assigning, a node from a plurality of nodes to each value from a plurality of values of a specific derived feature selected from the set of derived features, wherein the node is assigned based on a comparison of each value of the specific derived feature with one or more thresholds, and wherein the specific derived feature is angle of

elevation or depression; and determining one or more cyclic patterns of the plurality of nodes assigned to the plurality of values of the specific derived feature to be counted as the holding loop.
[012] In accordance with an embodiment of the present disclosure, the one or more processors 104 are further configured to ensemble the holding loop counts obtained from both of the (i) the pattern recognition based unsupervised approach and (ii) the graph based approach to determine a final holding loop count of the plurality of aerial vehicles; and determine, duration of each holding loop of each of the plurality of aerial vehicles based on difference of start and end time of the holding loop.
[013] In yet another aspect, there is provided a computer program product comprising a non-transitory computer readable medium having a computer readable program embodied therein, wherein the computer readable program, when executed on a computing device, causes the computing device to: receive, a plurality of data of a plurality of aerial vehicles, each of the plurality of aerial vehicles moving from one or more sources to destinations, wherein the plurality of data comprises information pertaining to the one or more sources and destinations and a set of parameters obtained from one or more sensors associated with the plurality of aerial vehicles; parse the plurality of received data to obtain a plurality of parsed files, wherein the plurality of parsed files are obtained by segregating the plurality of received data of the plurality of aerial vehicles such that each parsed file stores (i) meta information of the plurality of aerial vehicles and (ii) the set of parameters obtained from the one or more sensors associated with the plurality of aerial vehicles for a specific source-destination pair; derive, using the set of parameters comprised in each parsed file, a set of features in accordance with one or more physics based principles; obtain, by detecting zero crossings among a plurality of values of at least two features from the set of derived features, a reduced set of features; identify, using a pattern recognition based unsupervised approach, a region of interest in the reduced set of features, wherein the region of interest is indicative of presence of a holding loop and identified when two transitions occur between three consecutive values of the

obtained reduced set of features; and estimate, based on at least one of (i) the pattern recognition based unsupervised approach and (ii) a graph based approach, holding loop counts of the plurality of aerial vehicles. In accordance with an embodiment of the present disclosure the graph based approach to estimate holding loop counts of the plurality of aerial vehicles comprises: assigning a node from a plurality of nodes to each value from a plurality of values of a specific derived feature selected from the set of derived features, wherein the node is assigned based on a comparison of each value of the specific derived feature with one or more thresholds, and wherein the specific derived feature is angle of elevation or depression; and determining one or more cyclic patterns of the plurality of nodes assigned to the plurality of values of the specific derived feature to be counted as the holding loop.
[014] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
BRIEF DESCRIPTION OF THE DRAWINGS
[015] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles:
[016] FIG.1 illustrates an exemplary block diagram of a system for unsupervised learning based holding loop count estimation in aerial vehicles, in accordance with some embodiments of the present disclosure.
[017] FIG.2 illustrates an exemplary flow diagram of a processor implemented method for unsupervised learning based holding loop count estimation in aerial vehicles, in accordance with some embodiments of the present disclosure.
[018] FIG.3A through 3D illustrate different types of holding loops in path of aerial vehicles, in accordance with some embodiments of the present disclosure.

[019] FIGS.4A and 4B illustrate experimental results of a processor implemented method for unsupervised learning based holding loop count estimation in aerial vehicles, in accordance with some embodiments of the present disclosure.
DETAILED DESCRIPTION OF EMBODIMENTS [020] Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the scope of the disclosed embodiments. It is intended that the following detailed description be considered as exemplary only, with the true scope being indicated by the following claims.
[021] Considering the increased demand of easy and time efficient transportation, aerial vehicles such as commercial flights have been gaining traction as a faster mode of transportation. In multiple scenarios, the aerial vehicles are required to be diverted from their pre-determined path. For example, in case of aircrafts transporting passengers from a source to destination, a holding loop scenario may exist due to adverse weather, congestion at destination airports, and the like. In such cases, air traffic controller may give instructions to delay landing of the incoming aircrafts by traversing the incoming aircrafts into holding loops. In the context of the present disclosure, the expression ‘holding loop’ refers to a loop traversed by an aerial vehicle while it is in hold because it can’t be landed in destination at that time. However, existence of holding loops lead to delays, disruption in scheduling and maintenance, and increased operating cost of the aerial vehicles. To be able to manage fuel consumption better, the aerial vehicles with holding loops are required to be differentiated from the aerial vehicles without holding loops. Thus, it is necessary to detect the holding loops

and determine duration of the holding loops to ensure fuel optimization, proper scheduling and maintenance of the aerial vehicles at sources and destinations.
[022] State-of-the-art methods are not generic to any source-destination pairs and work only on a fixed source-destination pair and can identify the holding loop only if it exists in proximity to destination. Further, the state of the art methods are error prone, non-scalable, and do not accommodate anomalies in flight path of the aerial vehicle. Also, conventional methods provides only a probability of existence of the holding loop in the flight path of the aerial vehicles but cannot either confirm about it or count the number of the holding loops.
[023] The present disclosure has addressed these issues by providing unsupervised learning based approaches for estimating holding counts in the aerial vehicles. The method of present disclosure provides a scalable solution which is generic to any source-destination pairs and reduces manual intervention. Further, the method of present disclosure is not limited to identifying the holding loops with in a pre-defined or proximate distance from destination and can identify holding loop anywhere in the flight path of the aerial vehicles.
[024] Referring now to the drawings, and more particularly to FIGS. 1 through 4B, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments and these embodiments are described in the context of the following exemplary system and/or method.
[025] FIG.1 illustrates an exemplary block diagram of a system for unsupervised learning based holding loop count estimation in aerial vehicles, in accordance with some embodiments of the present disclosure. In an embodiment, the system 100 includes one or more processors 104, communication interface device(s) or input/output (I/O) interface(s) 106, and one or more data storage devices or memory 102 operatively coupled to the one or more processors 104. The one or more processors 104 that are hardware processors can be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, graphics controllers, logic circuitries, and/or any devices that manipulate signals based on operational

instructions. Among other capabilities, the processor(s) are configured to fetch and execute computer-readable instructions stored in the memory. In the context of the present disclosure, the expressions ‘processors’ and ‘hardware processors’ may be used interchangeably. In an embodiment, the system 100 can be implemented in a variety of computing systems, such as laptop computers, notebooks, hand-held devices, workstations, mainframe computers, servers, a network cloud and the like.
[026] I/O interface(s) 106 can include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface, and the like and can facilitate multiple communications within a wide variety of networks N/W and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular, or satellite. In an embodiment, the I/O interface(s) can include one or more ports for connecting a number of devices to one another or to another server.
[027] The memory 102 may include any computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes. In an embodiment, one or more modules (not shown) of the system 100 can be stored in the memory 102. The one or more modules (not shown) of the system 100 stored in the memory 102 may include routines, programs, objects, components, data structures, and so on, which perform particular tasks or implement particular (abstract) data types. In an embodiment, the memory 102 includes a data repository 108 for storing data processed, received, and generated as output(s) by the system 100.
[028] The data repository 108, amongst other things, includes a system database and other data. In an embodiment, the data repository 108 may be external (not shown) to the system 100 and accessed through the I/O interfaces 106. The memory 102 may further comprise information pertaining to input(s)/output(s) of each step performed by the processor 104 of the system 100

and methods of the present disclosure. In an embodiment, the system database stores information being processed at each step of the proposed methodology. The other data may include, data generated as a result of the execution of the one or more modules (not shown) of the system 100 stored in the memory 102. The generated data may be further learnt to provide improved learning in the next iterations to output desired results with improved accuracy.
[029] In an embodiment, an aerial vehicle 200 is configured to transmit and receive data from the system 100and sensors 202. In an embodiment, the system 100 may be separate and distinct from the aerial vehicle 200. For example, system 100 may be located at a land-based monitoring center. In at least one other embodiment, the system 100 may be an onboard component of the aerial vehicle 200, another aerial vehicle, watercraft, spacecraft (for example, a satellite), and/or the like. In an embodiment, the aerial vehicle 200 could be an aircraft, an unmanned aerial vehicle, drones, and/or the like. In an embodiment, sensors 200 may include but not limited to accelerometers, a global positioning system (GPS), radar, imaging system, an inertial measurement unit (IMU), LIDAR, and/or the like. In an embodiment, the sensors 202 may be separate and distinct from the aerial vehicle 200. In at least one other embodiment, the sensors may be onboard components of the aerial vehicle 200, another aerial vehicle, watercraft, spacecraft (for example, a satellite), and/or the like.
[030] FIG.2 illustrate an exemplary flow diagram of a computer implemented method 300 for unsupervised learning based holding loop count estimation in aerial vehicles in accordance with some embodiments of the present disclosure. In an embodiment, the system 100 includes one or more data storage devices or memory 102 operatively coupled to the one or more processors 104 and is configured to store instructions configured for execution of steps of the method 300 by the one or more processors 104. The steps of the method 300 will now be explained in detail with reference to the components of the system 100 of FIG.1. Although process steps, method steps, techniques or the like may be described in a sequential order, such processes, methods and techniques may be configured to work in alternate orders. In other words, any sequence or order of steps that may

be described does not necessarily indicate a requirement that the steps be performed in that order. The steps of processes described herein may be performed in any order practical. Further, some steps may be performed simultaneously.
[031] In accordance with an embodiment of the present disclosure, the one or more processors 104 are configured to receive, at step 302, a plurality of data of a plurality of aerial vehicles, each of the plurality of aerial vehicles moving from one or more sources to destinations, wherein the plurality of data comprises information pertaining to the one or more sources and destinations and a set of parameters obtained from one or more sensors 202 associated with the plurality of aerial vehicles. In an embodiment, the set of parameters obtained from one or more sensors 202 comprises altitude (sy ), speed at ground ( vx ), latitude value, and longitude value of the plurality of aerial vehicles. In an embodiment, the information pertaining to the one or more sources and one or more destinations may include but not limited to historical records of weather disturbances and real time weather prediction, historic records of occurrence of the holding loops in vicinity, time table information, and runway availability in real time, and number of aerial vehicles flying from and arriving at corresponding sources and destinations respectively.
[032] Further, at step 304, the one or more hardware processors 104 are configured to parse the plurality of received data to obtain a plurality of parsed files. In an embodiment, the plurality of parsed files are obtained by segregating the plurality of received data of the plurality of aerial vehicles such that each parsed file stores (i) meta information of the plurality of aerial vehicles and (ii) the set of parameters obtained from the one or more sensors associated with the plurality of aerial vehicles for a specific source-destination pair. In an embodiment, the plurality of received data is stored in a file which follows a naming convention and gives the meta information of the plurality of aerial vehicles such as source name, destination name, source code, start date, destination code, and/or the like for all sources and destinations. Thus, file names of the plurality of received data are parsed to obtain the plurality of parsed files

which store the meta information along with instantaneous sensor readings of all the respective aerial vehicles in a separate file in such a manner that all the information related to all the aerial vehicles from a particular source-destination pair can be found in a single file. Table 1 provides an example of the plurality of received data of the plurality of aerial vehicles.

Vehicle S. No File Name
371 UAE-
A388 A6EDA 20181124 165336 EG
LL_20181124 230800 OMDB 30 UA
E30-1542867931-airline-
0008 2914219 visium fuel timeseries
v8.2_FlightAware.csv
728 UAE-
A388_A6ED A_20181225_204318_EG
LL_20181226_025548_OMDB_4_UAE
4-1545546404-airline-
0033_3096345_visium_fuel_timeseries_
v8.2_FlightAware.csv
1086 UAE-
A388_A6EDD_20181122_163503_EG
LL_20181122_230013_OMDB_30_UA
E30-1542695132-airline-
0166_2891518_visium_fuel_timeseries_
v8.2_FlightAware.csv
1438 UAE-
A388_A6EDF_20181212_222034_EGL
L_20181213_042500_OMDB_6_UAE6
-1544423193-airline-
0006_3032440_visium_fuel_timeseries_
v8.2_FlightAware.csv

Sensor Parameters
( vx ) (sy ) Lat Long
3.1 2.7
541 175 93 2.43
67 0 75 3333
6.4 08
4.0 387 33 3.15
75 5 3 4167
10. 69
605 58 4.07
4.5 0 3 5
15.
5.3 64
916 678 16
67 7.5 7 4.5

1795 UAE-A388_A6EDF_20181214_165415_EGL L_20181214_230200_OMDB_30_UAE 30-1544595932-airline- 21.
0063_3042921_visium_fuel_timeseries_ 5.7 860 22 5.39
v8.2_FlightAware.csv 75 0 5 1667
2133 UAE-



A388_A6EDF_20181219_163937_EGL L_20181219_225500_OMDB_30_UAE 27.
30-1545027976-airline- 6.0 110 12
0222_3067216_visium_fuel_timeseries_ 333 37. 91 5.77
v8.2_FlightAware.csv 33 5 7 5
2473 UAE-



A388_A6EDF_20181220_205630_EGL L_20181221_030500_OMDB_4_UAE4
-1545114399-airline- 3.1 2.7
0129_3072462_visium_fuel_timeseries_ 541 175 93 2.43
v8.2_FlightAware.csv 67 0 75 3333
Table 1
Here, the received data with file name “UAE-
A388_A6EDA_20181124_165336_EGLL_20181124_230800_OMDB_30_UAE 30-1542867931-airline-0008_2914219” is parsed to get source country UAE, start date 24-Nov-2018, source airport code EGLL, Destination airport OMDB, aerial vehicle number A388_A6EDA, departure time, and Airlines ID airline-0008.
[033] In accordance with an embodiment of the present disclosure, at step 306, the one or more processors 104 are configured to derive, using the set of parameters comprised in each parsed file, a set of features in accordance with one or more physics based principles. In an embodiment, the one or more physics based principles may include but not limited to differentiation and integration concepts, equation of motion for determining relationships between initial

velocity, final velocity, acceleration, displacement, and time. For example, if velocity is received as input, then acceleration can be determined by computing derivative of the velocity and likewise displacement can be determined by computing integration of the velocity. It must be appreciated that though these basic physics based principles are easy to be implemented, yet not explored for holding loop count estimation in the aerial vehicles. In an embodiment, the set of derived features are indicative of processed information captured in real time. In an embodiment, the set of derived features comprises displacement along x- axis (sx), velocity along y-axis (vy) acceleration along x- axis (Ax) and y-axis (Ay), derivative of latitude values, derivative of longitude values, and angle of elevation or depression (9) of the plurality of aerial vehicles. Here, (sx) is computed by integration of (vx), (vy) is computed by finding derivative of (sy), (Ax); and (Ay) are computed by integrating (vx) and (vy) respectively, and angle of elevation or depression (9) in the path of aerial vehicle is computed based on equation (1) provided below as:

[034] In an embodiment, the plurality of parsed files are ordered in a source-destination pairwise fashion but the set of parameters obtained from the one or more sensors 202 are not measured uniformly with respect to time due to several obstructions. These obstructions may include anomalies present in the flight path of the aerial vehicles, unusual patterns leading to false alarms, presence of noise in the plurality of received data, and or the like. Thus, the step of deriving the set of features is preceded by sampling the plurality of parsed files using a mean median filtering technique to obtain uniform values of the set of parameters obtained from one or more sensors. Since, the received and processed plurality of data is huge, a combined subset of values of the set of parameters obtained from the one or more sensors 202 and the set of derived features is provided in Table 2.

No.

▽Sx






Row vx Sy Sx Ux Ax
vy Ay Lat∂ ∂t ∂Long t∂ θ

727 726 725 724 723 402 401 400 399 398 397
556 778 111 3.15 778 8.508333 8.5 8.491667 8.508333 8.533333 8.483333
2033.333 2541.667 3066.667 3816.667 4675 37000 37000 37000 36912.5 36025 34912.5
2978.91 2975.918 2972.749 2969.593 2966.354 211.2833 202.7792 194.2833 185.7833 177.2625 168.7542
3.177778 3.161111 3.15 3.327778 3.55 8.5 8.491667 8.508333 8.533333 8.483333 8.625
-0.37222 0.016667 0.011111 -0.17778 -0.22
222 0.008333 0.008333 -0.01667 -0.025 0.05 -0.14167
2.991667 3.169444 3.155556 3.238889 3.438889 8.504167 8.495833 8.5 8.520833 8.508333 8.554167
-508.333 -525 -750 -858.333 -137.5 0 0 87.5 887.5 1112.5 1675
16.66667 225 108.3333 -720.833 558.3333 0 -87.5 -800 -225 -562.5 200
0.02583 0.040417 0.03735 -0.01813 -0.03109 -0.04125 0.03904 -0.0396 -0.03877 -0.02239 -0.01774
-0.0456 -0.03493 0.01801 0.056673 0.044227 -0.2206 0.2108 0.21792 0.219983 0.231467 0.22406
-89.6628 -89.6541 -89.7589 -89.7838 -88.7873 0 0 84.45154 89.44993 889.56182 89.7074

728 2.633333
333 1225
2981.629 2.805556 -0.17222 2.719444
-808.333 -300 0.023663
Table 2 [035] In accordance with an embodiment of the present disclosure, the one or more processors 104 are configured to obtain a reduced set of features, at step 308, by detecting zero crossings among a plurality of values of at least two features from the set of derived features. In an embodiment, the zero crossings are detected when two consecutive values of the at least two features from the set of derived features are of opposite sign (say + and -). From Table 2, zero crossings among a plurality of values of two features namely derivative of latitude values and derivative of longitude are detected. As can been seen in Table 2, in row numbers 400 and 401 and row numbers 724 and 725, change in sign of values of the derivative of latitude feature is detected. Thus, corresponding changes in the values of the derivative of latitude feature in row numbers 400 to 401 and 724 to 725 are considered as zero crossings and saved in a separate array. Similarly, in row numbers 401 and 402 and row numbers 725 and 726 of Table 2, change in sign of values of the derivative of longitude feature is detected and the corresponding changes in the values of the derivative of longitude feature in these rows are considered as zero crossings and stored in the separate array. In a similar way, zero crossings among a plurality of values are detected and stored in the separate array. The features and their corresponding values comprised in the separate array are referred as the reduced set of features. In an embodiment, the step of obtaining reduced features enables faster processing since holding loop counts are estimated using the reduced set of feature and whole data is not required to be processed for holding loop count estimation. Table 3 provide values of the set of reduced features obtained using the set of features comprised in Table 2.

Row No.
∂Lat
∂t ∂Long
∂t

400 -0.0396 0.21792
401 0.03904 0.2108
402 0.04125 -0.2206
724 -0.01813 0.056673
725 0.03735 0.01801
726 0.040417 -0.03493
Table 3 [036] In accordance with an embodiment of the present disclosure, at step 310, the one or more processors 104 are configured to identify, using a pattern recognition based unsupervised approach, a region of interest in the reduced set of features. In an embodiment, the region of interest is indicative of presence of a holding loop and identified when two transitions occur between three consecutive values of the obtained reduced set of features. In other words, if elements of the separate array or the reduced set of features are of pattern +-+ or -+-, then the corresponding values are identified as the region of interest. As can be seen in Table 3, values of derivative of latitude feature makes a pattern +-+ in row numbers 402, 724 and 725 and thus identified as a region of interest. Similarly, values of derivative of longitude feature makes a pattern -+- in row numbers 401, 402 and 724 and thus identified as the region of interest. Here, Table 3 provides only a subset of reduced features identified based on the set of features of Table 2. However, the received and processed data is too large, so for the large data, a window of size 3 is defined for identifying the region of interest from the set of reduced feature based on the patterns +-+ and -+-. Initially, first three elements of the separate array are analyzed to identify the region of interest. If the region of interest is not identified in the first three elements, then, pointer of the separate array is increased by 1 and the next three elements of the separate array are analyzed for identifying the region of interest. This process continues until all of the region of interests are identified. In an embodiment, a holding loop is generally a pattern/path over which an aerial vehicle is flown. In an embodiment, the holding loop may include a route having multiple legs and turns, each of

which provides a preset distance and time of flight. FIG.3A through 3D illustrate different types of holding loops in path of the aerial vehicles, in accordance with some embodiments of the present disclosure. For example, as can be seen in FIG.3A through 3D, the holding loop may resemble but not limited to a racetrack, a spiral shape, a circle, a zig-zag, and a combination thereof. In an embodiment, the pattern based unsupervised approach does not require any labelled data and training. It may be understood to a person skilled in the art that unsupervised approach has not been implemented for holding loop count estimation in any of the state of the art methods.
[037] In accordance with an embodiment of the present disclosure, at step 312, the one or more processors 104 are configured to estimate, based on at least one of (i) the pattern recognition based unsupervised approach and (ii) a graph based approach, holding loop counts of the plurality of aerial vehicles. In an embodiment, the holding loop counts in aerial vehicles can be estimated using two approaches, wherein first approach is the pattern based unsupervised approach in which the region of interests are identified and marked/counted as the holding loop. Further, second approach for estimating count of the holding loops is the graph based approach. In accordance with an embodiment of the present disclosure, the graph based approach to estimate holding loop counts of the plurality of aerial vehicles utilizes the one or more hardware processers 104 configured to assign, a node from a plurality of nodes to each value from a plurality of values of a specific derived feature selected from the set of derived features, wherein the node is assigned based on a comparison of each value of the specific derived feature with one or more thresholds, and wherein the specific derived feature is angle of elevation or depression. Further, the one or more processors 104 are configured to determine one or more cyclic patterns of the plurality of nodes assigned to the plurality of values of the specific derived feature to be counted as the holding loop. In an embodiment, the one or more thresholds refer to one or more predefined ranges of degree of the angle of elevation or depression (θ ). For example, if the value of angle of elevation or depression ( θ ) is less than 45 degree, then a node ‘a’ is assigned to it. Similarly, if the value of

angle of elevation or depression ( θ ) is equal to 45 degree, then a node ‘b’ is assigned. In a similar way, node ‘c’ is assigned for 45° > θ > 90°, node ‘d’ is
assigned for θ = 90° , node ‘e’ is assigned for 90° > θ > 135° , node ‘f’ is
assigned for 135° > θ > 270° , node ‘g’ is assigned for θ = 270° , node ‘h’ is
assigned for 270° > θ > 315° , and node ‘i’ is assigned for 315°> θ > 360° .
[038] Since, the received and processed plurality of data is huge, a subset of values of the angle of elevation or depression ( θ ) feature is provided in Table 4.

Row No. Angle of elevation/depression( θ ) in degrees Assigned node
24 89.59836 c
25 89.69559 c
26 76.68498841 c
27 23.2 a
28 0 a
29 0 a
30 45 b
31 45 b
32 47.2 c
33 54.75243 c
Table 4
Referring to Table 4, pattern of the nodes assigned to all rows is identified as ‘cccaaabbcc’. In an embodiment, repeated nodes from the pattern are considered

as a single node to give a final pattern. For example, the final pattern for ‘cccaaabbcc’ is c-a-b-c. Further, it is determined if the final patterns are cyclic patterns or not, wherein the cyclic patterns refer to the patterns in which starting node and ending node remains same. As can be seen, the final pattern c-a-b-c starts and end with node ‘c’, thus, it is identified as the cyclic pattern. In an embodiment, each cyclic pattern is counted as the holding loop. In a similar way, other cyclic patterns are identified to estimate holding loop counts of the plurality of aerial vehicle.
[039] In an embodiment, the holding loops can be counted using any one of the two approaches or using both of them. In accordance with an embodiment of the present disclosure, in case of estimating the holding loop counts using both the approaches, the one or more processors 104 are configured to ensemble the holding loop counts obtained from both (i) the pattern recognition based unsupervised approach and (ii) the graph based approach to determine a final holding loop count of the plurality of aerial vehicles. In another embodiment, an average of the loop counts obtained from both the approaches is determined and reported as the final loop count of the plurality of aerial vehicles. In accordance with an embodiment of the present disclosure, the one or more processors 104 are further configured to determine duration of each holding loop of each of the plurality of aerial vehicles based on difference of start and end time of the holding loop.
[040] In an embodiment, the method/system of present disclosure determines holding loop patterns from the graph based approach, holding loop trends, sudden changes in altitude (Alternatively referred as deviation from normal altitude) which is obtained from the set of derived features, and anomalies in the path of the plurality of aerial vehicles. Further, the method of present disclosure may help in providing advisory for passengers, connection planning of the aerial vehicles, schedule re-planning of the aerial vehicles, fuel recommendation, high fuel consumption causality analysis, and flying policies and enforcement.

EXPERIMENTAL OBSERVATIONS
[041] FIGS.4A and 4B illustrate experimental results of a processor implemented method for unsupervised learning based holding loop count estimation in aerial vehicles, in accordance with some embodiments of the present disclosure. As depicted in FIG. 4A, experiments were performed on 17518 flights travelling in 438 different sectors. It was observed from the experiments that almost 19% of the aerial vehicles, which means 3053 flights out of 17518, were having holding loops in their path. On the other hand, 58% of the sectors had at least one aerial vehicle affected by the holding loops. Further, from the experiments, it was observed that average holding loop count was 1.65 for all the aerial vehicles that were put in the holding loop.
[042] FIG. 4B shows confusion matrices to determine accuracy of holding loop predictor for holding loop detection and holding loop count estimation. It is observed that the holding loop predictor works with 98.5% accuracy in worst case when tested over 193 flights in the hithrow to Dubai sector. Further, it can be seen from the confusing matrix for the holding loop detection that out of 60 flights having a holding loop, one could not be detected correctly. But the holding loop count estimation for aerial vehicles with more than one holding loop works with 100% accuracy as shown in confusion matrix for count of holding loops.
[043] The method of the present disclosure provides a scalable solution for holding loop count estimation in the aerial vehicles which is generic to any source-destination pairs and reduces manual intervention. Further, the method of present disclosure is not limited to identifying the holding loops with in a pre-defined or proximate distance from destination and can identify holding loop anywhere in the flight path of the aerial vehicle. Furthermore, the method of present disclosure obtains a reduced set of features which helps in faster processing of data since not all data is required to be processed for holding loop count estimation.
[044] The written description describes the subject matter herein to enable any person skilled in the art to make and use the embodiments. The scope

of the subject matter embodiments is defined by the claims and may include
other modifications that occur to those skilled in the art. Such other
modifications are intended to be within the scope of the claims if they have similar elements that do not differ from the literal language of the claims or if they include equivalent elements with insubstantial differences from the literal language of the claims.
[045] It is to be understood that the scope of the protection is extended to such a program and in addition to a computer-readable means having a message therein; such computer-readable storage means contain program-code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The hardware device can be any kind of device which can be programmed including e.g. any kind of computer like a server or a personal computer, or the like, or any combination thereof. The device may also include means which could be e.g. hardware means like e.g. an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of hardware and software means, e.g. an ASIC and an FPGA, or at least one microprocessor and at least one memory with software processing components located therein. Thus, the means can include both hardware means and software means. The method embodiments described herein could be implemented in hardware and software. The device may also include software means. Alternatively, the embodiments may be implemented on different hardware devices, e.g. using a plurality of CPUs.
[046] The embodiments herein can comprise hardware and software elements. The embodiments that are implemented in software include but are not limited to, firmware, resident software, microcode, etc. The functions performed by various components described herein may be implemented in other components or combinations of other components. For the purposes of this description, a computer-usable or computer readable medium can be any apparatus that can comprise, store, communicate, propagate, or transport the

program for use by or in connection with the instruction execution system, apparatus, or device.
[047] The illustrated steps are set out to explain the exemplary
embodiments shown, and it should be anticipated that ongoing technological
development will change the manner in which particular functions are performed.
These examples are presented herein for purposes of illustration, and not
limitation. Further, the boundaries of the functional building blocks have been
arbitrarily defined herein for the convenience of the description. Alternative
boundaries can be defined so long as the specified functions and relationships
thereof are appropriately performed. Alternatives (including equivalents,
extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope of the disclosed embodiments. Also, the words “comprising,” “having,” “containing,” and “including,” and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.
[048] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile

memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
[049] It is intended that the disclosure and examples be considered as exemplary only, with a true scope of disclosed embodiments being indicated by the following claims.

WE CLAIM:
1. A processor implemented method, comprising:
receiving (302), a plurality of data of a plurality of aerial vehicles, each of the plurality of aerial vehicles moving from one or more sources to destinations, wherein the plurality of data comprises information pertaining to the one or more sources and destinations and a set of parameters obtained from one or more sensors associated with the plurality of aerial vehicles;
parsing (304) the plurality of received data to obtain a plurality of parsed files, wherein the plurality of parsed files are obtained by segregating the plurality of received data of the plurality of aerial vehicles such that each parsed file stores (i) meta information of the plurality of aerial vehicles and (ii) the set of parameters obtained from the one or more sensors associated with the plurality of aerial vehicles for a specific source-destination pair;
deriving (306), using the set of parameters comprised in each parsed file, a set of features in accordance with one or more physics based principles;
obtaining (308), by detecting zero crossings among a plurality of values of at least two features from the set of derived features, a reduced set of features;
identifying (310), using a pattern recognition based unsupervised approach, a region of interest in the reduced set of features, wherein the region of interest is indicative of presence of a holding loop and identified when two transitions occur between three consecutive values of the obtained reduced set of features; and
estimating (312), based on at least one of (i) the pattern recognition based unsupervised approach and (ii) a graph based approach, holding loop counts of the plurality of aerial vehicles;

2. The processor implemented method as claimed in claim 1, wherein the step of deriving the set of features is preceded by sampling the plurality of parsed files using a mean median filtering technique to obtain uniform values of the set of parameters obtained from the one or more sensors.
3. The processor implemented method as claimed in claim 1, wherein the set of parameters obtained from the one or more sensors comprises altitude(sy ), speed at ground(vx ), latitude, and longitude of the plurality of aerial vehicles.
4. The processor implemented method as claimed in claim 1, wherein the set of derived features comprises displacement along x- axis (sx ), velocity along y-axis (vy ), acceleration along x- axis ( Ax) and y-axis (Ay ), derivative of latitude, derivative of longitude, and angle of elevation or depression (θ) of the plurality of aerial vehicles.
5. The processor implemented method as claimed in claim 1, wherein the graph based approach to estimate holding loop counts of the plurality of aerial vehicles comprises:
assigning a node from a plurality of nodes to each value from a plurality of values of a specific derived feature selected from the set of derived features, wherein the node is assigned based on a comparison of each value of the specific derived feature with one or more thresholds, and wherein the specific derived feature is angle of elevation or depression; and
determining one or more cyclic patterns of the plurality of nodes assigned to the plurality of values of the specific derived feature to be counted as the holding loop.
6. The processor implemented method as claimed in claim 1, further
comprising:

ensembling the holding loop counts obtained from both (i) the pattern recognition based unsupervised approach and (ii) the graph based approach to determine a final holding loop count of the plurality of aerial vehicles; and
determining, duration of each holding loop of each of the plurality of aerial vehicles based on difference of start and end time of the holding loop.
7. A system (100), comprising: one or more data storage devices (102)
operatively coupled to one or more hardware processors (104) and configured to store instructions configured for execution via the one or more hardware processors to:
receive, a plurality of data of a plurality of aerial vehicles, each of the plurality of aerial vehicles moving from one or more sources to destinations, wherein the plurality of data comprises information pertaining to the one or more sources and the one or more destinations and a set of parameters obtained from one or more sensors associated with the plurality of aerial vehicles;
parse the plurality of received data to obtain a plurality of parsed files, wherein the plurality of parsed files are obtained by segregating the plurality of received data of the plurality of aerial vehicles such that each parsed file stores (i) meta information of the plurality of aerial vehicles and (ii) the set of parameters obtained from the one or more sensors associated with the plurality of aerial vehicles for a specific source-destination pair;
derive, using the set of parameters comprised in each parsed file, a set of features in accordance with one or more physics based principles;
obtain, by detecting zero crossings among a plurality of values of at least two features from the set of derived features, a reduced set of features;

identify, using a pattern recognition based unsupervised approach, a region of interest in the reduced set of features, wherein the region of interest is indicative of presence of a holding loop and identified when two transitions occur between three consecutive values of the obtained reduced set of features; and
estimating, based on at least one of (i) the pattern recognition based unsupervised approach and (ii) a graph based approach, holding loop counts of the plurality of aerial vehicles;
8. The system as claimed in claim 7, wherein the step of deriving the set of features is preceded by sampling the plurality of parsed files using a mean median filtering technique to obtain uniform values of the set of parameters obtained from the one or more sensors.
9. The system as claimed in claim 7, wherein the set of parameters obtained from the one or more sensors comprises altitude( ), speed at ground( ), latitude, and longitude of the plurality of aerial vehicles.
10. The system as claimed in claim 7, wherein the set of derived features comprises displacement along x- axis (sx ), velocity along y-axis (vy ), acceleration along x- axis (Ax ) and y-axis (Ay ), derivative of latitude, derivative of longitude, and angle of elevation or depression (θ) of the plurality of aerial vehicles.
11. The system as claimed in claim 7, wherein the graph based approach to estimate holding loop counts of the plurality of aerial vehicles comprises:
assigning a node from a plurality of nodes to each value from a plurality of values of a specific derived feature selected from the set of derived features, wherein the node is assigned based on a comparison of each value of the specific derived feature with one or more thresholds, and wherein the specific derived feature is angle of elevation or depression; and

determining one or more cyclic patterns of the plurality of nodes assigned to the plurality of values of the specific derived feature to be counted as the holding loop.
12. The system as claimed in claim 1, further comprising:
ensembling the holding loop counts obtained from both (i) the pattern recognition based unsupervised approach and (ii) the graph based approach to determine a final holding loop count of the plurality of aerial vehicles; and
determining, duration of each holding loop of each of the plurality of aerial vehicles based on difference of start and end time of the holding loop.