The world-wide aviation system is one of the most complex dynamical systems ever developed and is generating data at an extremely rapid rate. Most modern commercial aircraft record several hundred flight parameters including information from the guidance, navigation, and control systems, the avionics and propulsion systems, and the pilot inputs into the aircraft. These parameters may be continuous measurements or binary or categorical measurements recorded in one second intervals for the duration of the flight. Currently, most approaches to aviation safety are reactive, meaning that they are designed to react to an aviation safety incident or accident. Here, we discuss a novel approach based on the theory of multiple kernel learning to detect potential safety anomalies in very large data bases of discrete and continuous data from world-wide operations of commercial fleets. We pose a general anomaly detection problem which includes both discrete and continuous data streams, where we assume that the discrete streams have a causal influence on the continuous streams. We also assume that atypical sequence of events in the discrete streams can lead to off-nominal system performance. We discuss the application domain, novel algorithms, and also briefly discuss results on synthetic and real-world data sets. Our algorithm uncovers operationally significant events in high dimensional data streams in the aviation industry which are not detectable using state of the art methods.

Sample report in support of "Discovering Anomalous Aviation Safety Events Using Scalable Data Mining Algorithms" manuscript.

The world-wide aviation system is one of the most complex dynamical systems ever developed and is generating data at an extremely rapid rate. Most modern commercial aircraft record several hundred flight parameters including information from the guidance, navigation, and control systems, the avionics and propulsion systems, and the pilot inputs into the aircraft. These parameters may be continuous measurements or binary or categorical measurements recorded in one second intervals for the duration of the flight. Currently, most approaches to aviation safety are reactive, meaning that they are designed to react to an aviation safety incident or accident. In this paper, we discuss a novel approach based on the theory of multiple kernel learning to detect potential safety anomalies in very large data bases of discrete and continuous data from world-wide operations of commercial fleets. We pose a general anomaly detection problem which includes both discrete and continuous data streams, where we assume that the discrete streams have a causal influence on the continuous streams. We also assume that atypical sequences of events in the discrete streams can lead to off-nominal system performance. We discuss the application domain, novel algorithms, and also discuss results on real-world data sets. Our algorithm uncovers operationally significant events in high dimensional data streams in the aviation industry which are not detectable using state of the art methods.

Modern aircraft are producing data at an unprecedented rate with hundreds of parameters being recorded on a second by second basis. The data can be used for studying the condition of the hardware systems of the aircraft and also for studying the complex interactions between the pilot and the aircraft. NASA is developing novel data mining algorithms to detect precursors to aviation safety incidents from these data sources. This talk will cover the theoretical aspects of the algorithms and practical aspects of implementing these techniques to study one of the most complex dynamical systems in the world: the national airspace.

The National Airspace System (NAS) is an ever changing and complex engineering system. As the Next Generation Air Transportation System (NextGen) is developed, there will be an increased emphasis on safety and operational and environmental efficiency. Current operations in the NAS are monitored using a variety of data sources, including data from flight recorders, radar track data, weather data, and other massive data collection systems. Although numerous technologies exist to monitor the frequency of known but undesirable behaviors in the NAS, there are currently few methods that can analyze the large repositories to discover new and previously unknown events in the NAS. Having a tool to discover events that have implications for safety or incidents of operational importance, increases the awareness of such scenarios in the community and helps to broaden the overall safety of the NAS, whereas only monitoring the frequency of known events can only provide mitigations for already established problems. This paper discusses a novel approach for discovering operationally significant events in the NAS that are currently not monitored and have potential safety and/or efficiency implications using radar-track data. This paper will discuss the discovery algorithm and describe in detail some flights of interest with comments from subject matter experts who are familiar with the operations in the airspace that was studied.

Published at 12th AIAA Aviation Technology, Integration, and Operations (ATIO) Conference and 14th AIAA/ISSM 17 - 19 September 2012, Indianapolis, Indiana

For the purposes of this paper, the National Airspace System (NAS) encompasses the operations of all aircraft which are subject to air traffic control procedures. The NAS is a highly complex dynamic system that is sensitive to aeronautical decision-making and risk management skills. In order to ensure a healthy system with safe flights a systematic approach to anomaly detection is very important when evaluating a given set of circumstances and for determination of the best possible course of action. Given the fact that the NAS is a vast and loosely integrated network of systems, it requires improved safety assurance capabilities to maintain an extremely low accident rate under increasingly dense operating conditions. Data mining based tools and techniques are required to support and aid operators’ (such as pilots, management, or policy makers) overall decision-making capacity. Within the NAS, the ability to analyze fleetwide aircraft data autonomously is still considered a significantly challenging task. For our purposes a fleet is defined as a group of aircraft sharing generally compatible parameter lists. Here, in this effort, we aim at developing a system level analysis scheme. In this paper we address the capability for detection of fleetwide anomalies as they occur, which itself is an important initiative toward the safety of the real-world flight operations. The flight data recorders archive millions of data points with valuable information on flights everyday. The operational parameters consist of both continuous and discrete (binary & categorical) data from several critical subsystems and numerous complex procedures. In this paper, we discuss a system level anomaly detection approach based on the theory of kernel learning to detect potential safety anomalies in a very large data base of commercial aircraft. We also demonstrate that the proposed approach uncovers some operationally significant events due to environmental, mechanical, and human factors issues in high dimensional, multivariate Flight Operations Quality Assurance (FOQA) data. We present the results of our detection algorithms on real FOQA data from a regional carrier.

We discuss a statistical framework that underlies envelope detection schemes as well as dynamical models based on Hidden Markov Models (HMM) that can encompass both discrete and continuous sensor measurements for use in Integrated System Health Management (ISHM) applications. The HMM allows for the rapid assimilation, analysis, and discovery of system anomalies. We motivate our work with a discussion of an aviation problem where the identification of anomalous sequences is essential for safety reasons. The data in this application are discrete and continuous sensor measurements and can be dealt with seamlessly using the methods described here to discover anomalous flights. We specifically treat the problem of discovering anomalous features in the time series that may be hidden from the sensor suite and compare those methods to standard envelope detection methods on test data designed to accentuate the differences between the two methods. Identification of these hidden anomalies is crucial to building stable, reusable, and cost-efficient systems. We also discuss a data mining framework for the analysis and discovery of anomalies in high-dimensional time series of sensor measurements that would be found in an ISHM system. We conclude with recommendations that describe the tradeoffs in building an integrated scalable platform for robust anomaly detection in ISHM applications.

This paper presents a framework aimed at monitoring the behavior of aircraft in a given airspace. Trajectories that constitute typical operations are determined and learned using data driven methods. Standard procedures are used by air traffic controllers (ATC) to guide aircraft, ensure the safety of the airspace, and to maximize the runway occupancy. Even though standard procedures are used by ATC, the control of the aircraft remains with the pilots, leading to a large variability in the flight patterns observed. Two methods to identify typical operations and their variability from recorded radar tracks are presented. This knowledge base is then used to monitor the conformance of current operations against operations previously identified as typical. A tool called AirTrajectoryMiner is presented, aiming at monitoring the instantaneous health of the airspace, in real time. The airspace is “healthy” when all aircraft are flying according to the typical operations. A measure of complexity is introduced, measuring the conformance of current flight to typical flight patterns. When an aircraft does not conform, the complexity increases as more attention from ATC is required to ensure a safe separation between aircraft.

This paper presents a framework aimed at monitoring the behavior of aircraft in a given airspace. Trajectories that constitute typical operations are determined and learned using data-driven methods. Standard procedures are used by air traffic controllers (ATCs) to guide aircraft, ensure the safety of the airspace, and maximize runway occupancy. Even though standard procedures are used by ATCs, control of the aircraft remains with the pilots, leading to large variability in the flight patterns observed. Two methods for identifying typical operations and their variability from recorded radar tracks are presented. This knowledge base is then used to monitor the conformance of current operations against operations previously identified as typical. A tool called AirTrajectoryMiner is presented, aiming at monitoring the instantaneous health of the airspace, in real time. The airspace is “healthy” when all aircraft are flying according to typical operations. A measure of complexity is introduced, measuring the conformance of current flight to typical flight patterns. When an aircraft does not conform, the complexity increases as more attention from ATC is required to ensure safe separation between aircraft. IEEE Transactions on Intelligent Transportation Systems, Dec. 2011, 12(4), pp. 1511 - 1524.