WO2021122397A1 - Procédé et dispositif de supervision d'un système de pistage - Google Patents
Procédé et dispositif de supervision d'un système de pistage Download PDFInfo
- Publication number
- WO2021122397A1 WO2021122397A1 PCT/EP2020/085841 EP2020085841W WO2021122397A1 WO 2021122397 A1 WO2021122397 A1 WO 2021122397A1 EP 2020085841 W EP2020085841 W EP 2020085841W WO 2021122397 A1 WO2021122397 A1 WO 2021122397A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- quality
- service
- values
- tracking system
- determining
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G5/00—Traffic control systems for aircraft, e.g. air-traffic control [ATC]
- G08G5/0073—Surveillance aids
- G08G5/0078—Surveillance aids for monitoring traffic from the aircraft
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G5/00—Traffic control systems for aircraft, e.g. air-traffic control [ATC]
- G08G5/0043—Traffic management of multiple aircrafts from the ground
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/17—Function evaluation by approximation methods, e.g. inter- or extrapolation, smoothing, least mean square method
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G5/00—Traffic control systems for aircraft, e.g. air-traffic control [ATC]
- G08G5/0017—Arrangements for implementing traffic-related aircraft activities, e.g. arrangements for generating, displaying, acquiring or managing traffic information
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G5/00—Traffic control systems for aircraft, e.g. air-traffic control [ATC]
- G08G5/0017—Arrangements for implementing traffic-related aircraft activities, e.g. arrangements for generating, displaying, acquiring or managing traffic information
- G08G5/0021—Arrangements for implementing traffic-related aircraft activities, e.g. arrangements for generating, displaying, acquiring or managing traffic information located in the aircraft
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G5/00—Traffic control systems for aircraft, e.g. air-traffic control [ATC]
- G08G5/0017—Arrangements for implementing traffic-related aircraft activities, e.g. arrangements for generating, displaying, acquiring or managing traffic information
- G08G5/0026—Arrangements for implementing traffic-related aircraft activities, e.g. arrangements for generating, displaying, acquiring or managing traffic information located on the ground
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G5/00—Traffic control systems for aircraft, e.g. air-traffic control [ATC]
- G08G5/0047—Navigation or guidance aids for a single aircraft
- G08G5/0052—Navigation or guidance aids for a single aircraft for cruising
Definitions
- the invention relates generally to tracking systems, and in particular to a method and a device for monitoring a tracking system in air traffic.
- Air traffic tracking systems are conventionally used to manage air traffic and ensure its safety while ensuring minimum safety distances between aircraft to avoid collisions.
- a tracking system makes it possible to represent a situation in real time of the position of the planes and of their speed vectors.
- Air traffic control system is based on interactions between air traffic controllers, technical supervisors and the system. Air traffic controllers ensure safe distances between aircraft in the airspace of their sectors. Technical supervisors monitor the proper functioning of the air traffic control system. However, when the air traffic controller notices a degradation of the tracking system, the technical supervisor is often not able to identify the origin of this degradation or to remedy it. Conversely, when the technical supervisor observes technical facts, he is necessarily not in a position to determine the operational impact of these technical facts for the air traffic controller.
- the invention improves the situation.
- the invention proposes a method for supervising a tracking system capable of evaluating and supplying analysis information relating to the operation and the quality of service of an air traffic tracking system, in a current situation of the tracking system defined by the position and the speed vector of at least one aircraft moving in airspace, the position and the speed vector being determined from data from one or more sensors.
- the method comprises a step of determining performance indicator values representing the current situation, the method being characterized in that it comprises the steps of:
- the operating characteristic is representative of an operating abnormality of the tracking system, determining at least one performance indicator corresponding to the operating abnormality, and determining values of said at least one performance indicator corresponding to normal operation ;
- At least one measure of the quality of service is representative of a degradation of the quality of service, determine at least one indicator of performance corresponding to said degradation of the quality of service;
- the operating characteristic may be a state chosen from the group comprising a normal operating state, at least one known abnormal operating state, an unknown abnormal operating state, and an impossible operating state. , the operating characteristic being associated with an operating abnormality of the tracking system if said operating characteristic is not a normal operating state.
- the normal operating state may correspond to a situation known for a given configuration, the given configuration corresponding to a first configuration where said at least one aircraft is in cruising speed or to a second configuration where said at least one aircraft is approaching a given airport associated with a given positioning of one or more sensors and with a given type of weather for said given airport, said at least one known abnormal operating state being predefined, l 'impossible operating state corresponding to impossible combinations of performance indicator values.
- the step of determining performance indicator values representing the current situation may comprise the partitioning of the space of the performance indicator values into a plurality of areas comprising at least one area. of normality and at least one zone of known abnormality, each zone of the plurality of zones representing a class of normality and being associated with an operating state of the tracking system, said at least one zone of normality being associated with the state of normal operation, each of said at least one known abnormality zone being associated with a known abnormal operating state.
- the partitioning of the space of the values of performance indicators can comprise the determination of said at least one zone of normality, from standard data corresponding to performance indicator values obtained in normal operational situations, applying an unsupervised machine learning technique.
- the unsupervised machine learning technique can be chosen from a group comprising linear dimensionality reduction techniques, non-linear dimensionality reduction techniques, partitioning techniques and methods to core.
- the step of determining values of performance indicators representing the current situation may comprise determining a set of data from the standard data and a partitioning of the set of data into subsets by applying a supervised learning method, each subset corresponding to performance indicator values representing a known area of abnormality.
- the supervised learning method can be a data classification method.
- a measure of the quality of service can correspond to a multi-sensor tracking or to a single-sensor tracking, the measurement of the quality of service corresponding to the quality of the overall tracking for a plurality of sensors for multi-sensor tracking, and the tracking quality for a given sensor for single-sensor tracking.
- the values of performance indicators for the current situation can be included in a class of normality corresponding to the operating characteristic of the tracking system for the current situation, a measure of the quality of service of the tracking system being determined as a function of said normality class.
- the tracking system can be used for tracking said at least one aircraft in an airspace distributed over a plurality of spatial zones, a measure of the quality of service of the tracking system being determined by association with each of the plurality of spatial areas.
- the values of performance indicators may have at least one missing value, a measure of the quality of service. of the tracking system being determined in the presence of at least one missing value of performance indicators.
- the step of determining a performance indicator corresponding to the degradation of the quality of service can include determining an influence indicator associated with each performance indicator.
- each performance indicator can be associated with a set of executable evaluation processes to evaluate the state of a subsystem of the tracking system, the step of performing at at least one evaluation process comprising the execution of the evaluation processes associated with said at least one performance indicator corresponding to the abnormal operation of the tracking system and / or audited at least one performance indicator corresponding to the degradation of the quality of service.
- the invention further provides a device for supervising an air traffic tracking system, in a current situation of the tracking system defined by a position and a speed vector relating to at least one aircraft, the position and the vector speed being determined from data from one or more sensors, the device being configured to determine performance indicator values representing the current situation.
- the device is configured for:
- the operating characteristic is representative of an operating abnormality of the tracking system, determining at least one performance indicator corresponding to the operating abnormality, and determining values of said at least one performance indicator corresponding to normal operation ;
- At least one measure of the quality of service is representative of a degradation of the quality of service, determining at least one performance indicator corresponding to said degradation of the quality of service;
- the embodiments of the invention provide a double analysis of performance indicators calculated in quasi-real time making it possible to assess both the normal operation of a tracking system in a given situation and its compliance and quality of service.
- the embodiments of the invention make it possible to identify abnormal operation of the tracking system based on data flows generated in real time by means of unsupervised or semi-supervised machine learning techniques and generating an explanation for the abnormal operation, the explanation identifying characteristics and performance indicators explaining the detected abnormal operation.
- the embodiments of the invention make it possible to identify a degradation in the quality of service by means of supervised machine learning techniques and to generate an explanation in the event of a degradation in the quality of service of the tracking system. , the explanation of degradation of the quality of service identifying the indicators explaining the degradation of the quality of service.
- the embodiments of the invention make it possible to identify the root causes of operating abnormalities and degradation of the quality of service at the level of performance indicators and at the level of possible maintenance actions for an operator. of maintenance.
- the embodiments of the invention provide monitoring tools to the technical supervisor of a tracking system in air traffic allowing him to follow the evolution of the quality of service and to assess the impact. operational technical facts on the quality of service of the tracking system.
- the embodiments of the invention make it possible to link the measurement of the operational impact to the input data in order to allow the technical supervisor to identify the root causes of a possible degradation of the quality of service. and to have an explanation of the quality of service information provided.
- the joint use of supervised data (data tagged by an expert) and unsupervised (data flow generated by the sensors of a tracking system) make it possible to characterize the normal operating states and the operating states. operation corresponding to anomalies, detect abnormal situations and refine the model for evaluating the quality of service using unsupervised data.
- the explanation algorithms according to the embodiments of the invention make it possible to provide transparency to the user (controller or technical supervisor) and to identify the input data which explains an anomaly.
- Figure 1 is a flowchart representing a method of monitoring a tracking system, according to some embodiments of the invention.
- Figure 2 is a schematic view of an exemplary computerized system for implementing the method of monitoring a tracking system, according to certain embodiments of the invention.
- the embodiments of the invention provide a method of supervising a tracking system in air traffic at a current situation defined by the position and the speed vector of at least one aircraft from data generated in real time by one or more sensors.
- an aircraft can be any type of aircraft such as an airplane (airliner, military plane, private plane), a helicopter, a hot air balloon, or a drone.
- a sensor used in the tracking system can be a terrestrial, surface, or aerial sensor, such as:
- an air traffic control radar for example a primary radar or a secondary radar
- a multilateration system (using, for example, long-distance multilateration technology or even Wide Area Multilateration or WAM in English) made up of several beacons which receive the signals emitted by the transponder of an aircraft in order to locate it;
- ADS-C system (acronym for "Automatic Dependent Surveillance-Contract” in English) in which an aircraft uses its navigation systems to automatically determine and transmit its position to a processing center, or
- ADS-B system (acronym for 'Automatic Dependent Surveillance-Broadcast') in which an aircraft uses its navigation systems to automatically determine and broadcast its position as well as other information such as speed and flight sign.
- a given configuration corresponds to a first configuration where the at least one aircraft is cruising or to a second configuration where the at least one aircraft is approaching a given associated airport to a given positioning of one or more sensors and to a given type of weather encountered for the given airport.
- a given situation is defined by the position and the speed vector relating to at least one aircraft and is represented by values of the performance indicators evaluating the normal operation and the quality of service of the control system. tracking to the given situation.
- the embodiments of the invention provide a method of supervising a tracking system allowing the evaluation and the explanation of the normal operation and of the quality of service of a tracking system to a current situation, the current situation being defined by the position and speed vector relative to minus one aircraft operating in airspace.
- the position and the speed vector relating to at least one aircraft can be determined or estimated beforehand from data from one or more sensors implemented in the tracking system using an algorithm of tracking.
- the tracking algorithm can be a Kalman filter, according to different variations.
- the Kalman filter is used to determine the position, speed, and acceleration of the aircraft by iteratively estimating its position. At each iteration, the Kalman filter estimates a position of the aircraft at the current instant from a set of positions observed at previous instants corresponding to previous iterations. A correction step follows the estimation step to correct the predicted position using the current measurement.
- the tracking algorithm can be previously configured according to the given configuration, for example according to the characteristics of the sensors, or the type of weather.
- step 101 performance indicator values representing the current situation can be determined.
- the values of performance indicators can be determined over a sliding window of time.
- a set N ⁇ 1, ..., n ⁇ of performance indicators (also called 'performance metrics') for the measurement of the tracking quality of the tracking system.
- a performance indicator can be chosen depending on the application of the tracking system in the field of aviation.
- Examples of performance indicators include, without limitation:
- step 101 can include determining the reconstructed trajectory of at least one aircraft between the current instant and the current instant minus 1 hour 30 minutes using Q-Splines. Step 101 can further comprise determining the values of the metrics or performance indicators by integrating measurements between the current instant minus 15 minutes and the current instant minus 1 h 15 min.
- an operating characteristic also called "state of normality" of the tracking system for the current situation can be determined as a function of the performance indicator values determined in step 101.
- step 103 The objective of step 103 is to identify the state of normality of the current situation represented by the values of the performance indicators determined in step 101.
- the operating characteristic may be a state chosen from a group comprising a normal operating state (or normal state), at least one known abnormal operating state (or even q 3 1 operating states known abnormal), an unknown abnormal operating condition, and an impossible operating condition, the operating characteristic being associated with an operating abnormality of the tracking system if the operating characteristic is not a normal operating condition.
- the normal operating state corresponds to normal values of performance indicators and to standard situations recorded in a given configuration for which the tracking system is functioning normally with a well-configured tracking and no problems identified on the sensors of the tracking system.
- Each known abnormal operating state is predefined and corresponds to an abnormal situation identified and defined by a business expert.
- the abnormal impossible operating state corresponds to impossible combinations of the values of the performance indicators, i.e. incompatible values on the the indicators.
- the unknown abnormal operating state corresponds to other situations which are a priori possible and abnormal, although it is impossible to identify which situations they correspond to.
- Performance indicator values can be represented by dots in a multidimensional space of performance indicator values and can be separated or grouped into different partitions.
- step 101 can comprise the partitioning of the space of the performance indicator values into a plurality of zones comprising at least one zone of normality and at least one zone of known abnormality, each zone of the plurality of areas representing a class of normality and being associated with an operating state of the tracking system such that the at least one normality area is associated with the normal operating state and each of the abnormality areas known abnormal operating state is associated with a known abnormal operating state among the previously identified known abnormal operating states.
- An operating state can be associated with several zones of normality.
- Known abnormality areas can be non-disjoint and the normality area must not intersect an abnormality area.
- Partitioned areas may further include an area associated with the impossible operating state and an area associated with the unknown abnormal operating state.
- the partitioning of the space of performance indicator values into zones can be based on both unsupervised and supervised machine learning techniques.
- the partitioning of the space of the values of performance indicators into zones can comprise the determination of the at least one zone of normality, from standard unsupervised data corresponding to values. performance indicators obtained in normal operational situations, by applying an unsupervised machine learning technique.
- the identification of the underlying structure of the clouds of points representing the values of the performance indicators associated with these normal operational situations makes it possible to determine, within the space of the performance indicator values, at least one zone of normality representative only normal situations, to identify areas of high density of normal situations which bring a strong certainty of normality in their neighborhood, and to identify anomalies, or rare points, which bring little certainty as to normality of their neighborhood.
- the unsupervised machine learning technique can be chosen from a group comprising linear dimensionality reduction techniques (eg principal component analysis), non-dimensionality reduction techniques.
- linear e.g. autoencoders and variety learning
- data partitioning techniques or 'clustering' in English e.g. hierarchical clustering algorithms and the 'Density-based Spatial Clustering of Applications with Noise 'or DBSCAN
- kernel methods eg kernel clustering and kernel PCR method.
- the generated standard data can be used to determine at least one known abnormality area in the space of performance indicator values.
- step 101 may comprise determining a set of data from the standard data generated and the partitioning of the set of data into subsets by applying a method of supervised learning, each subset corresponding to performance indicator values representing a known abnormality area.
- determining the dataset may include modifying the standard data through transformations so that the dataset matches values of known anomalous performance indicators.
- the transformations of the standard data may consist in applying similar disturbers to disturbance audits.
- the supervised learning method can be a method of classifying data making it possible to identify the underlying structures of each of the sets of situations corresponding to the same state of known abnormality.
- an uncertainty indicator can be associated with the supervised learning method to quantify the uncertainty of the outputs generated by the learning method.
- the determination of at least one zone of normality can be based on a set of vectors of performance indicators recovered from data flows generated by the system in nominal situation corresponding to a well-adjusted tracking and to no incident.
- the set of performance indicator vectors can be processed first by applying dimensionality reduction using an autoencoder, and then classified using a uni-class classification algorithm.
- a plurality of known abnormal operating states can be identified over the life of the tracking system, including an abnormal operating condition, poor calibration and an abnormal operating condition. solar irruption.
- the determination of the known abnormal areas corresponding to this plurality of known abnormal operating states can be based on the transformation of standard data corresponding to normal operating states. For example, for the determination of the known abnormal operating zone corresponding to the abnormal bad calibration operating state, the vector of performance indicator values produced by the tracking system in the tracking calibration phase can be transformed by reducing the values of the performance indicators. tracking parameter values. Disturbances of tracking calibrations can be performed for each sensor used in the tracking system.
- step 105 the operating abnormality of the tracking system can be explained if the operating characteristic determined in step 103 is representative of an operating abnormality of the tracking system.
- the explanation can take different forms to explain why the operating state of the tracking system is not normal.
- step 105 can include determining at least one performance indicator corresponding to the operating abnormality and the determination of the values of the performance indicators corresponding to normal operation.
- step 105 can be based on a method of counter-factual explanation which consists in identifying the list of performance indicators explaining that the operating state is not normal and in identifying the minimum modification of the performance indicators of the identified list which would make it possible to return the operating characteristic of the system determined in step 103 to a normal operating state.
- the first term tx (77 (y) - c) 2 takes into account the fact that the new instance y is of class c indicating a normal operating state and the second term d (x, y) takes into account the fact that the new instance y is as close as possible to the vector x.
- the distance metric between the vectors x and y can be chosen from a group comprising the Euclidean distance and the distance associated with the L1 standard.
- At least one measure of the quality of service of the tracking system for the current situation can be determined from the values of performance indicators representing the current situation determined at step 101.
- the measurement of the quality of service can be carried out at the global level on the multi-sensor tracking or at a local level for a particular sensor.
- a measurement of the quality of service can correspond to a multi-sensor tracking corresponding to the quality of the overall tracking for a plurality of sensors or to a single-sensor tracking corresponding to the quality of service for a given sensor.
- the normality class can be used as an attribute of the quality of service model, in addition to the vector of performance indicators.
- a plurality of quality of service measures can be determined, the plurality of service quality measures comprising a measure of the overall multi-sensor tracking quality, and a measure of the quality. single-sensor tracking for each type of sensor (radar, WAM, ADS-B).
- a quality of service model can be determined for multi-sensor tracking, a quality of service model can be determined for each radar or group of radars, a quality of service model can be determined for the WAM information, and a QoS model can be determined for the ADS-B information, different performance metrics being used for each QoS model.
- the metrics of the ESASSP standard can be used while also considering the mandatory requirements and recommended by the ESASSP standard.
- a plurality of performance indicator and requirements may be used including a value of the 'range bias' indicator below 100m, a value of the 'azimuth bias' indicator. below 0.1, a value of the indicator 'standard deviation on the range' below 70m, a value of the indicator 'standard deviation in azimuth' below 0.1, a value of the indicator 'delays maximum on a report of a target 'below 2 seconds, a value of the indicator' false leads ratio 'below 0.1%, and a value of the indicator' probability of detection 'above 70 %.
- a plurality of indicators and mandatory requirements of the ESASSP standard can be used including a value of the indicator 'horizontal RMS error in position' less than 350m in ER and less than 150m in TMA, a value of the 'processing time' indicator less than 1 second in 'Data Driven' mode and less than 1 second plus the output period in 'periodic delayed' mode and less than 0.5 second for 'periodically predicted period' mode, a value of the indicator 'probability of detection of the position' greater than 97%, and a value of the indicator 'false leads ratio' less than 0.1%.
- a set A of performance indicators may correspond to statistics on events for which the values of the performance indicators include at least one value may not be calculable over the current time window.
- the vector x N ⁇ A designates the vector comprising the values of the performance indicators on the set N ⁇ A.
- the determination of the measurement of the quality of service in the presence of missing values of performance indicators can be carried out according to a first approach which consists in supplementing the vector x N ⁇ A with the missing values for the indicators of set A which are the more unfavorable for the vector x and to determine the measure of the quality of service Q (X N ⁇ A> Z A ) from the vectors x N ⁇ A and from the vector z A.
- the determination of the measurement of the quality of service in the presence of missing values of performance indicators can be carried out according to a second approach which is based on the a priori probability on the missing values and the 'evaluation of the expectation of the measurement of the quality of service Q ⁇ X N ⁇ A> Z A ) according to the probability on z A.
- the determination of the measurement of the quality of service in the presence of missing values of performance indicators can be carried out according to a third approach which is based on the determination of a new function of quality of service.
- the function ⁇ (.) can be a general function or a monotonic and normalized function, or a function using a Choquet integral.
- the Choquet integral is an aggregation function O m .
- R n ® R having as parameter a vector m: 2 N ® [0,1] ⁇
- m (5) represents the importance of the criteria S.
- Y t ® R is the utility (normalization) function on the indicator t, Y t designating the set of values that the indicator t can take. This aggregation function F is monotonic and normalized.
- the sum can be replaced by an integral when the indicators take continuous values.
- the Choquet integral makes it possible to model the criteria which interact with each other. A particular case consists in limiting oneself to interactions only between pairs of criteria.
- the expression of the Choquet integral - then called the 2-additive Choquet integral - is given by with x> i denoting the importance of the criterion t, and / ⁇ ; ⁇ denoting the level of interaction between the criteria i and j.
- the new function Q_ A can be written in the form: Q- A A), where F_ A denotes a Choquet integral
- the aggregation function can be organized in a hierarchical manner.
- the hierarchical decomposition of the aggregation function can be composed of several Choquet integrals.
- the restriction operators can thus be applied to each integral of Choquet.
- a tree structure of normalization and aggregation criteria can be determined for each quality of service model.
- Mandatory and recommended requirements can be separated and piece-affine utility functions and a Choquet integral can be used for aggregations.
- the tracking system can be used for the tracking of at least one aircraft in an airspace distributed in a plurality of spatial zones, the evaluation of the tracking quality being able to be carried out on each spatial area separately in order to identify in which area the quality of service is degraded or an operational problem has occurred.
- at least one measure of the quality of service of the tracking system can be determined in step 107 in association with each of the plurality of spatial areas.
- the determination of quality of service measurements over a plurality of spatial areas can be based on a uniform tiling of the airspace in a plurality of cells defined by the ESASSP standard, the cells being grouped together. so that the air traffic is homogeneous group of cells to another.
- the determination of the groups of cells can be based on a method comprising the steps of:
- the division of the sets of cells can be carried out dynamically on each calculation of a measure of the quality of service, the number of measures present in each cell changing over time.
- an explanation of the degradation of the quality of service of the tracking system can be determined if the at least one measure of the quality of service determined in step 107 is representative of a degradation of the quality of service, for example if at least one measure of the quality of service is below a predefined quality of service threshold.
- the influence index of the indicator te ⁇ 1,. .., n ⁇ can be determined from partial influence indices denoted ô * 'y ° pt'T'Q (i), each partial influence index ô * ' y ° pt'T'Q (i ) being determined from the permuted vectors obtained by applying a permutation p selected in the set p (T) of compatible permutations in the tree T such that (p (1), ..., 7r (/ c) ⁇ .
- the influence index of the indicator te ⁇ 1, ..., n ⁇ can thus be determined as being the average of the indicator indices partial over all permutations p of the set p (T) such that
- the influence index of an indicator ie ⁇ 1, ..., n ⁇ can be determined by first determining the missing values.
- the calculations performed for determining the influence indices of the performance indicators can use sum-compensated algorithms rather than the standard summation operators.
- each performance indicator can be associated with a set of executable evaluation processes (set of reflex sheets) making it possible to assess the state of a subsystem of the tracking system and of '' identify the origin of potential operating anomalies in the tracking system.
- step 111 the evaluation processes (in the form of reflex sheets for identifying the root causes) associated with the performance indicators corresponding to an operating abnormality and / or to a degradation of the quality of service. can be determined and performed by the tracking system maintenance operator. More precisely, step 111 can comprise the determination or the identification, and the execution of the most relevant reflex sheets associated with the performance indicators explaining the abnormal operation of the tracking system to the current situation and the reflex sheets. the most relevant associated with the performance indicators explaining the degradation of the quality of service of the tracking system to the current situation.
- step 111 may consist of identifying and executing the reflex sheets associated with the performance indicators corresponding to values for the current situation which are significantly different from the values for the counterfactual example considered.
- step 111 may include a sub-step consisting in comparing the results obtained on the different quality of service models comprising the model used for multi-sensor tracking and the models used for each type. sensor for single-sensor tracking.
- Step 111 may further comprise a sub-step consisting in identifying the indicators having the greatest influence on the degradation of the quality of service among the indicators of the different models compared, and in selecting a number of indicators, this number d 'indicators being determined either by retaining the indicators which are associated with the p largest influence indices or by applying a clustering algorithm to the distribution of the influence indices and by selecting the indicators from the first class.
- the selection of performance indicators by applying the clustering algorithm advantageously makes it possible to select performance indicators dynamically according to the distribution of values.
- step 111 may consist in collecting all the levels of influence of each indicator by including the various quality of service models, in classifying the indicators according to the sum of the levels of influence that he has, and to carry out the reflex sheets associated with these indicators in the order of the classification of the indicators.
- the invention also provides a device for supervising an air traffic tracking system, in a current situation of the tracking system defined by a position and a speed vector relating to less than one aircraft, the position and the vector. speed being determined from data from one or more sensors, the device being configured to determine performance indicator values representing the current situation, characterized in that the device is configured for: - determining an operating characteristic of the tracking system for said current situation as a function of said values of performance indicators;
- said operating characteristic is representative of an operating abnormality of the tracking system, determining at least one performance indicator corresponding to the operating abnormality, and determining values of at least one performance indicator corresponding to normal operation ;
- At least one measure of the quality of service is representative of a degradation of the quality of service, determine at least one performance indicator corresponding to the degradation of the quality of service;
- the invention further provides a computer program product comprising code instructions for performing the process steps when said program is executed on a computer.
- the embodiments of the invention can be implemented by various means, for example by hardware (“hardware”), software, or a combination thereof.
- Computer 20 may include various compute, storage, and communications units configured to interact with each other through a data and address port 29, comprising a processor 21, one or more storage peripherals 23, an input / output interface (I / O) 25 and a Human-Machine interface (HMI) 27.
- processor 21 processor 21, one or more storage peripherals 23, an input / output interface (I / O) 25 and a Human-Machine interface (HMI) 27.
- I / O input / output interface
- HMI Human-Machine interface
- the processor 21 can include one or more selected devices: microprocessors, microcontrollers, digital signal processors, microcomputers, central processing units, programmable gate networks, programmable logic devices, machines state defined, logic circuits, analog circuits, digital circuits, or any other device used to manipulate signals (analog or digital) based on operating instructions stored in memory.
- the memory may include a single device or a plurality of memory devices, including but not limited to read-only memory (ROM), random access memory (RAM), volatile memory, non-volatile memory, static random access memory (SRAM), dynamic random access memory (DRAM), flash memory, cache memory or any other device capable of storing information.
- Mass memory can include data storage devices such as a hard drive, optical disc, magnetic tape drive, volatile or non-volatile solid state circuitry, or any other device capable of storing information.
- a database may reside on the mass memory storage device and may be used to collect and organize data used by the various systems and modules described herein.
- the processor 21 can operate under the control of an operating system which resides in the memory.
- the operating system can manage the computer resources in such a way that the program code of the computer, integrated in the form of one or more software applications;
- routines executed to implement the embodiments of the invention may be referred to herein as “computer program code” or simply “program code”.
- Program code typically includes computer readable instructions that reside at various times in various memory and storage devices in a computer and which, when read and executed by one or more processors in a computer, cause the computer to perform the operations necessary to run the operations and / or the elements specific to the various aspects of the embodiments of the invention.
- the instructions of a program, readable by computer, for carrying out the operations of the embodiments of the invention can be, for example, the assembly language, or else a source code or an object code written in combination with one or several programming languages.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Traffic Control Systems (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
Description
Claims
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020227024205A KR20220113509A (ko) | 2019-12-20 | 2020-12-11 | 교통 관제 시스템을 감독하기 위한 방법 및 디바이스 |
AU2020408906A AU2020408906A1 (en) | 2019-12-20 | 2020-12-11 | Method and device for supervising a traffic control system |
US17/787,557 US20220406199A1 (en) | 2019-12-20 | 2020-12-11 | Method and device for supervising a traffic control system |
EP20821013.8A EP4078409A1 (fr) | 2019-12-20 | 2020-12-11 | Procédé et dispositif de supervision d'un système de pistage |
BR112022012137A BR112022012137A2 (pt) | 2019-12-20 | 2020-12-11 | Método e dispositivo para supervisionar um sistema de controle de tráfego |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FRFR1915153 | 2019-12-20 | ||
FR1915153A FR3105544B1 (fr) | 2019-12-20 | 2019-12-20 | Procede et dispositif de supervision d'un systeme de pistage |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021122397A1 true WO2021122397A1 (fr) | 2021-06-24 |
Family
ID=70613973
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2020/085841 WO2021122397A1 (fr) | 2019-12-20 | 2020-12-11 | Procédé et dispositif de supervision d'un système de pistage |
Country Status (7)
Country | Link |
---|---|
US (1) | US20220406199A1 (fr) |
EP (1) | EP4078409A1 (fr) |
KR (1) | KR20220113509A (fr) |
AU (1) | AU2020408906A1 (fr) |
BR (1) | BR112022012137A2 (fr) |
FR (1) | FR3105544B1 (fr) |
WO (1) | WO2021122397A1 (fr) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113643571A (zh) * | 2021-10-18 | 2021-11-12 | 中国电子科技集团公司第二十八研究所 | 一种基于航班正常性目标的空域网络优化方法 |
-
2019
- 2019-12-20 FR FR1915153A patent/FR3105544B1/fr active Active
-
2020
- 2020-12-11 BR BR112022012137A patent/BR112022012137A2/pt unknown
- 2020-12-11 EP EP20821013.8A patent/EP4078409A1/fr active Pending
- 2020-12-11 WO PCT/EP2020/085841 patent/WO2021122397A1/fr unknown
- 2020-12-11 US US17/787,557 patent/US20220406199A1/en active Pending
- 2020-12-11 KR KR1020227024205A patent/KR20220113509A/ko unknown
- 2020-12-11 AU AU2020408906A patent/AU2020408906A1/en active Pending
Non-Patent Citations (6)
Title |
---|
B. ABICHOU ET AL: "Choquet integral capacities-based data fusion for system health monitoring", IFAC THE 2012 IFAC WORKSHOP ON AUTOMATIC CONTROL IN OFFSHORE OIL AND GAS PRODUCTION, vol. 45, August 2012 (2012-08-01), Red Hook, NY, pages 31 - 36, XP055729067, ISSN: 1474-6670, ISBN: 978-1-123-47890-7, DOI: 10.3182/20120829-3-MX-2028.00260 * |
CUSTINNE BERNARD ET AL: "Surveillance chain performance assessment against ESASSP", 2014 TYRRHENIAN INTERNATIONAL WORKSHOP ON DIGITAL COMMUNICATIONS - ENHANCED SURVEILLANCE OF AIRCRAFT AND VEHICLES (TIWDC/ESAV), IEEE, 15 September 2014 (2014-09-15), pages 12 - 16, XP032677352, DOI: 10.1109/TIWDC-ESAV.2014.6945440 * |
EUROCONTROL SPÉCIFICATION FOR ATM SURVEILLANCE SYSTEM PERFORMANCE, vol. 1, March 2012 (2012-03-01), ISBN: ISBN : 978-287497-022-1 |
TONG LI ET AL: "Risk Management based on Fuzzy Measure and Integral: an Application to Air Traffic Control Management", OPERATIONS RESEARCH AND MANAGEMENT SCIENCE, April 2014 (2014-04-01), pages 153 - 157, XP055728736, Retrieved from the Internet <URL:http://m.wdfxw.net/goDownFiles.aspx?key=35161455> [retrieved on 20200908], DOI: 10.1016/j.eswa.2016.03.045 * |
YAVUZ OZDEMIR ET AL: "Aircraft selection using fuzzy ANP and the generalized choquet integral method: The Turkish airlines case", JOURNAL OF INTELLIGENT AND FUZZY SYSTEMS, vol. 31, no. 1, 13 June 2016 (2016-06-13), NL, pages 589 - 600, XP055728738, ISSN: 1064-1246, DOI: 10.3233/IFS-162172 * |
YING LI ET AL: "COMPREHENSIVE RISK ASSESSMENT BASED ON THE CHOQUET INTEGRAL", THE INTERNATIONAL JOURNAL OF ORGANIZATIONAL INNOVATION, vol. 11, no. 1, July 2018 (2018-07-01), pages 1 - 8, XP055728540 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113643571A (zh) * | 2021-10-18 | 2021-11-12 | 中国电子科技集团公司第二十八研究所 | 一种基于航班正常性目标的空域网络优化方法 |
CN113643571B (zh) * | 2021-10-18 | 2022-02-08 | 中国电子科技集团公司第二十八研究所 | 一种基于航班正常性目标的空域网络优化方法 |
US11756435B2 (en) | 2021-10-18 | 2023-09-12 | The 28Th Research Institute Of China Electronics Technology Group Corporation | Airspace network optimization method based on flight normality target |
Also Published As
Publication number | Publication date |
---|---|
FR3105544B1 (fr) | 2023-04-14 |
BR112022012137A2 (pt) | 2022-08-30 |
KR20220113509A (ko) | 2022-08-12 |
AU2020408906A1 (en) | 2022-07-14 |
EP4078409A1 (fr) | 2022-10-26 |
FR3105544A1 (fr) | 2021-06-25 |
US20220406199A1 (en) | 2022-12-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10956253B2 (en) | Anomaly detection in performance management | |
US11023325B2 (en) | Resolving and preventing computer system failures caused by changes to the installed software | |
EP2599039B1 (fr) | Procede et systeme d'analyse de donnees de vol enregistrees au cours d'un vol d'un avion | |
FR3052273A1 (fr) | Prediction de pannes dans un aeronef | |
US11675641B2 (en) | Failure prediction | |
US10387728B2 (en) | Mapping wind turbines and predicting wake effects using satellite imagery data | |
US10769866B2 (en) | Generating estimates of failure risk for a vehicular component | |
FR3062733A1 (fr) | Procede de surveillance d'un equipement de type actionneur electromecanique | |
US20200177468A1 (en) | Techniques for analyzing a network and increasing network availability | |
US20210005332A1 (en) | Systems and methods for generating trust metrics for sensor data | |
WO2021122397A1 (fr) | Procédé et dispositif de supervision d'un système de pistage | |
EP1792278A1 (fr) | Procede de detection et de pistage de cibles ponctuelles, dans un systeme de surveillance optronique | |
US11411811B2 (en) | Fault localization for cloud-native applications | |
US11294748B2 (en) | Identification of constituent events in an event storm in operations management | |
WO2023011618A1 (fr) | Prédiction de cause profonde d'une alerte à l'aide d'un réseau neuronal récurrent | |
US20230179501A1 (en) | Health index of a service | |
US20220167558A1 (en) | Locating picking points of anomalies | |
JPWO2020070792A1 (ja) | 船舶検出システム、方法およびプログラム | |
FR3079637A1 (fr) | Procede d'evaluation de la conformite d'un systeme de pistage a un ensemble d'exigences et dispositifs associes | |
US20240078467A1 (en) | Predicting the intent of a network operator for making config changes | |
US20230385702A1 (en) | Data fabric for intelligent weather data selection | |
WO2022222623A1 (fr) | Estimation d'événement composite par l'intermédiaire d'une logique temporelle | |
US20230376825A1 (en) | Adaptive retraining of an artificial intelligence model by detecting a data drift, a concept drift, and a model drift | |
US20220230077A1 (en) | Machine Learning Model Wildfire Prediction | |
US20230126323A1 (en) | Unsupervised data characterization utilizing drift |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20821013 Country of ref document: EP Kind code of ref document: A1 |
|
REG | Reference to national code |
Ref country code: BR Ref legal event code: B01A Ref document number: 112022012137 Country of ref document: BR |
|
ENP | Entry into the national phase |
Ref document number: 20227024205 Country of ref document: KR Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 2020408906 Country of ref document: AU Date of ref document: 20201211 Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2020821013 Country of ref document: EP Effective date: 20220720 |
|
ENP | Entry into the national phase |
Ref document number: 112022012137 Country of ref document: BR Kind code of ref document: A2 Effective date: 20220617 |