Seminar on Computational Learning and Adaptation

 Unsupervised Anomaly Detection for Rocket Propulsion Health Monitoring

Dr. Mark Schwabacher
NASA Ames Research Center
Moffett Field, CA

Abstract:

This talk describes the results of applying four machine-learning-based unsupervised anomaly detection algorithms --- Orca, GritBot, the Inductive Monitoring System (IMS), and one-class Support Vector Machines (SVM) --- to data from two rocket propulsion testbeds. The first testbed uses historical data from the Space Shuttle Main Engine. The second testbed uses data from an experimental rocket engine test stand located at NASA Stennis Space Center. The talk describes nine anomalies detected by the four algorithms. The four algorithms use four different definitions of anomalousness. Orca uses a nearest-neighbor approach, defining a point to be an anomaly if its nearest neighbors in feature space are far away from it. IMS clusters the training data, and then uses the distance to the nearest cluster as its measure of anomalousness. GritBot learns rules from the training data, and then classifies points as anomalous if they violate these rules. One-class SVMs map the data into a high-dimensional space in which all of the normal points are on one side of a hyperplane, and then classify points on the other side of the hyperplane as anomalous. Because of these different definitions of anomalousness, different algorithms detect different anomalies. We therefore conclude that it is useful to use multiple algorithms.
---------------------------------

This talk presents joint work with Nikunj Oza, Ashok Srivastava, and Rodney Martin (NASA ARC) and Bryan Matthews (Perot Systems Inc)

Return to the seminar schedule