Real-Time Detection of In-ﬂight Aircraft Damage
Lawrence Livermore National Laboratory
Herbert K. H. Lee
University of California, Santa Cruz
NASA Ames Research Center
Abstract: When there is damage to an aircraft, it is critical to be able to quickly
detect and diagnose the problem so that the pilot can attempt to maintain control of
the aircraft and land it safely. We develop methodology for real-time classiﬁcation of
ﬂight trajectories to be able to distinguish between an undamaged aircraft and ﬁve dif-
ferent damage scenarios. Principal components analysis allows a lower-dimensional
representation of multi-dimensional trajectory information in time. Random Forests
provide a computationally efﬁcient approach with sufﬁcient accuracy to be able to
detect and classify the different scenarios in real-time. We demonstrate our approach
by classifying realizations of a 45 degree bank angle generated from the Generic
Transport Model ﬂight simulator in collaboration with NASA.
Keywords: Ensemble learning; Sliding window; Aviation safety.
This research was conducted at NASA Ames Research Center. Reference herein to
any speciﬁc commercial product, process, or service by trade name, trademark, manufactur-
er, or otherwise, does not constitute or imply its endorsement by the United States Govern-
ment. This work was performed under the auspices of the U.S. Department of Energy by
Lawrence Livermore National Laboratory under contract DE-AC52-07NA27344. Lawrence
Livermore National Security, LLC. Author completed this research as part of an M.S. in the
AMS department at the University California, Santa Cruz.
Corresponding Author’s Address: B. Blair, Lawrence Livermore National Labora-
tory, 7000 East Ave., Livermore, CA 94550, email: email@example.com.
Journal of Classiﬁcation 3 (2017)4: - 494 513
Published online: 2