ABSTRACT
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.
Supplemental Material
- I. C. Statler, "The Aviation System Monitoring and Modeling (ASMM) Project: A Documentation of its History and Accomplishments: 1999--2005". 2007,Technical Report: NASA/TP-2007-214556.Google Scholar
- F. R. Bach, G. R. G. Lanckriet and M. I. Jordan, "Multiple kernel learning, conic duality, and the SMO algorithm". 2004,International Conference on Machine Learning. Google ScholarDigital Library
- S. D. Bay and M. Schwabacher, "Mining Distance-Based Outliers in Near Linear Time with Randomization and a Simple Pruning Rule". 2003,Proceedings of The Ninth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Google ScholarDigital Library
- National Transportation Safety Board, "Loss of Control and Impact with Pacific Ocean Alaska Airlines Flight 261", 2002,ulr: http://www.ntsb.gov/publictn/2002/AAR0201.pdf.Google Scholar
- S. Budalakoti and A. N. Srivastava and M. E. Otey, "Anomaly Detection and Diagnosis Algorithms for Discrete Symbol Sequences with Applications to Airline Safety", 2008,Vol. 39, No. 1, IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, pages 101--113. Google ScholarDigital Library
- C. J. C. Burges, "A tutorial on support vector machines for pattern recognition", 1998,Vol. 2, DMKD, pages 121--167. Google ScholarDigital Library
- V. Chandola, A. Banerjee and V. Kumar, "Anomaly Detection: A Survey", 2009,ACM Computing Surveys. Google ScholarDigital Library
- O. Chapelle and P. Haffner, "Support Vector Machines for Histogram-Based Classification", 1999,Vol. 10, No. 5, IEEE Transactions on Neural Networks, pages 1055--1064. Google ScholarDigital Library
- S. Das, B. L. Matthews, K. Bhaduri, N. C. Oza and A. N. Srivastava, "Detecting Anomalies in Multivariate Data Sets with Switching Sequences and Continuous Streams", 2009,NIPS 2009 Workshop: Understanding Multiple Kernel Learning Methods.Google Scholar
- Z. Harchaoui and F. R. Bach, "Image classification with Segmentation Graph Kernels", 2007,Computer Vision and Pattern Recognition, pages 1--8.Google Scholar
- J. W. Hunt and T. G. Szymanski, "A Fast Algorithm for Computing Longest Common Subsequences", 1977,Vol. 20, No. 5, Communications of the ACM, pages 350--353. Google ScholarDigital Library
- D. L. Iverson, R. Martin, M. Schwabacher, L. Spirkovska, W. Taylor, R. Mackey, and J. P. Castle "General Purpose Data-Driven System Monitoring for Space Operations", 2009, Proceedings of the AIAA Infotech@Aerospace Conference.Google Scholar
- H. Kashima, K. Tsuda and A. Inokuchi, "Kernels for Graphs", 2004, Vol. 39, No. 1, Kernel Methods in Computational Biology, pages 101--113.Google Scholar
- G. R. G. Lanckriet, N. Cristianini, L. El Ghaoui, P. Bartlett and M. I. Jordan, "Learning the Kernel Matrix with Semidefinite Programming", 2004, Vol. 5, Journal of Machine Learning Research, pages 27--72. Google ScholarDigital Library
- P. Patel and E. Keogh and J. Lin and S. Lonardi, "Mining Motifs in Massive Time Series Databases", 2002,International Conference on Data Mining. Google ScholarDigital Library
- B. Scholkopf, J. C. Platt, J. Shawe-taylor, A. J. Smola and R. C. Williamson, "Estimating the Support of a High-Dimensional Distribution". 2001, Vol. 13, No. 7, Neural Computation, issn 0899--7667,MIT Press,Cambridge, MA, USA, pages 1443--1471. Google ScholarDigital Library
- B. Scholkopf and A. Smola, "Learning with Kernels", 2002,MIT Press, Cambridge, MA, USA.Google Scholar
- D. M. J. Tax and R. P. W. Duin, "Support Vector Domain Description", 1999,Vol. 20, Pattern Recognition Letters, pages 1191--1199. Google ScholarDigital Library
- Aviation Safety Team, "Statistical Summary of Commercial Jet Airplane Accidents", 2007, Boeing Commercial Airplanes.Google Scholar
- H. Zhang, A. C. Berg, M. Maire and J. Malik, "SVM-KNN: Discriminative Nearest Neighbor Classification for Visual Category Recognition", 2006, Computer Vision and Pattern Recognition, pages 2126--2136.. Google ScholarDigital Library
Index Terms
- Multiple kernel learning for heterogeneous anomaly detection: algorithm and aviation safety case study
Recommendations
Finding Precursors to Anomalous Drop in Airspeed During a Flight's Takeoff
KDD '17: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data MiningAerodynamic stall based loss of control in flight is a major cause of fatal flight accidents. In a typical takeoff, a flight's airspeed continues to increase as it gains altitude. However, in some cases, the airspeed may drop immediately after takeoff ...
Anomaly Detection on ADS-B Flight Data Using Machine Learning Techniques
Computational Collective IntelligenceAbstractWith the rapid increase in the number of flights all over the world, the management and control of flight operations has become difficult in recent years. Moreover, the expectations for the aviation sector indicate that this increase will continue ...
Anomaly Detection of Aerospace Facilities Using Ganomaly
ICMSSP '20: Proceedings of the 2020 5th International Conference on Multimedia Systems and Signal ProcessingIn the field of aerospace, the abnormal detection of data is of great significance. The rapid and effective detection of abnormal parameters is key to find potential failures of spacecraft. Traditional methods of anomaly detection need much manual ...
Comments