Aircraft anomaly detection using performance models trained on fleet data

Gorinevsky, D.; Matthews, Bryan; Martin, Roscoe C.

doi:10.1109/cidu.2012.6382196

Cited by 32 publications

(29 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [15] the authors proposed an approach based on a specially designed linear regression model to describe the aerodynamic forces acting on an aircraft. The constructed model accounts for the flight-to-flight and aircraft-to-aircraft variability, which enables the fitting of a single model to the entire dataset.…”

Section: Related Workmentioning

confidence: 99%

Semi-Markov Switching Vector Autoregressive Model-Based Anomaly Detection in Aviation Systems

Melnyk

Banerjee

Matthews

et al. 2016

Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Self Cite

View full text Add to dashboard Cite

In this work we consider the problem of anomaly detection in heterogeneous, multivariate, variablelength time series datasets. Our focus is on the aviation safety domain, where data objects are flights and time series are sensor readings and pilot switches. In this context the goal is to detect anomalous flight segments, due to mechanical, environmental, or human factors in order to identifying operationally significant events and provide insights into the flight operations and highlight otherwise unavailable potential safety risks and precursors to accidents. For this purpose, we propose a framework which represents each flight using a semi-Markov switching vector autoregressive (SMS-VAR) model. Detection of anomalies is then based on measuring dissimilarities between the model's prediction and data observation. The framework is scalable, due to the inherent parallel nature of most computations, and can be used to perform online anomaly detection. Extensive experimental results on simulated and real datasets illustrate that the framework can detect various types of anomalies along with the key parameters involved.

show abstract

Section: Related Workmentioning

confidence: 99%

Semi-Markov Switching Vector Autoregressive Model-Based Anomaly Detection in Aviation Systems

Melnyk

Banerjee

Matthews

et al. 2016

Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Self Cite

View full text Add to dashboard Cite

show abstract

“…1 A comprehensive archive of safety data is available through the FAA's Aviation Safety Information Analysis and Sharing (ASIAS) 2 repository, which is being mined by researchers to identify anomalous situations for further analysis to detect emerging issues, potentially leading to policy or procedure changes. 3,4,5 Once discovered, these precursors can also be probed in real-time as a component of real-time safety assessment. Similarly, the FAA uses data mining techniques to monitor operational parameters that define Key Performance Indicators (KPI) throughout the NAS to identify issues and modify operations before issues become hazards; 6 EUROCONTROL's Automatic Safety Monitoring Tool (ASMT) serves a comparable purpose in Europe.…”

Section: Introductionmentioning

confidence: 99%

Initial Demonstration of the Real-Time Safety Monitoring Framework for the National Airspace System Using Flight Data

Roychoudhury

Daigle

Goebel

et al. 2016

16th AIAA Aviation Technology, Integration, and Operations Conference

View full text Add to dashboard Cite

As new operational paradigms and additional aircraft are being introduced into the National Airspace System (NAS), maintaining safety in such a rapidly growing environment becomes more challenging. It is therefore desirable to have an automated framework to provide an overview of the current safety of the airspace at different levels of granularity, as well an understanding of how the state of the safety will evolve into the future given the anticipated flight plans, weather forecast, predicted health of assets in the airspace, and so on. Towards this end, as part of our earlier work, we formulated the Real-Time Safety Monitoring (RTSM) framework for monitoring and predicting the state of safety and to predict unsafe events. In our previous work, the RTSM framework was demonstrated in simulation on three different constructed scenarios. In this paper, we further develop the framework and demonstrate it on real flight data from multiple data sources. Specifically, the flight data is obtained through the Shadow Mode Assessment using Realistic Technologies for the National Airspace System (SMART-NAS) Testbed that serves as a central point of collection, integration, and access of information from these different data sources. By testing and evaluating using real-world scenarios, we may accelerate the acceptance of the RTSM framework towards deployment. In this paper we demonstrate the framework's capability to not only estimate the state of safety in the NAS, but predict the time and location of unsafe events such as a loss of separation between two aircraft, or an aircraft encountering convective weather. The experimental results highlight the capability of the approach, and the kind of information that can be provided to operators to improve their situational awareness in the context of safety.

show abstract

“…Motivated by a wide variety of applications ranging from fraud detection to aviation safety management through the health monitoring of complex net-works, data center infrastructure management or food risk analysis, unsupervised anomaly detection is now the subject of much attention in the data science literature, see e.g. Gorinevsky et al (2012); T. Fawcett (1997); Viswanathan et al (2012). In frequently encountered practical situations and from the viewpoint embraced in this paper, anomalies coincide with rare measurements that are extremes, i.e.…”

Section: Introductionmentioning

confidence: 99%

A multivariate extreme value theory approach to anomaly clustering and visualization

et al. 2019

View full text Add to dashboard Cite

In a wide variety of situations, anomalies in the behaviour of a complex system, whose health is monitored through the observation of a random vector X = (X1, . . . , X d ) valued in R d , correspond to the simultaneous occurrence of extreme values for certain subgroups α ⊂ {1, . . . , d} of variables Xj. Under the heavy-tail assumption, which is precisely appropriate for modeling these phenomena, statistical methods relying on multivariate extreme value theory have been developed in the past few years for identifying such events/subgroups. This paper exploits this approach much further by means of a novel mixture model that permits to describe the distribution of extremal observations and where the anomaly type α is viewed as a latent variable. One may then take advantage of the model by assigning to any extreme point a posterior probability for each anomaly type α, defining implicitly a similarity measure between anomalies. It is explained at length how the latter permits to cluster extreme observations and obtain an informative planar representation of anomalies using standard graph-mining tools. The relevance and usefulness of the clustering and 2-d visual display thus designed is illustrated on simulated datasets and on real observations as well, in the aeronautics application domain.

show abstract

Aircraft anomaly detection using performance models trained on fleet data

Cited by 32 publications

References 8 publications

Semi-Markov Switching Vector Autoregressive Model-Based Anomaly Detection in Aviation Systems

Semi-Markov Switching Vector Autoregressive Model-Based Anomaly Detection in Aviation Systems

Initial Demonstration of the Real-Time Safety Monitoring Framework for the National Airspace System Using Flight Data

A multivariate extreme value theory approach to anomaly clustering and visualization

Contact Info

Product

Resources

About