ARTIFICIAL INTELLIGENCE APPLICATIONS FOR SOLVING COMBAT IDENTIFICATION PROBLEMS CONCERNING UNKNOWN UNKNOWNS
MacKinnon, Douglas J.
Johnson, Bonnie W.
Cook, Glenn R.
MetadataShow full item record
The Navy has identified a need to exploit the benefits of artificial intelligence, particularly in the realm of common tactical picture (CTP), combat identification (CID), and battle management aids. Of grave concern to the Navy is the CID of unknown unknowns or things that are not known to exist or can be easily tracked. Artificial Intelligence, and related machine learning, deep learning, and deep analytics tools, provides a technology that can be used to assist commanders in processing information to help determine these unknown unknowns. The limitations of current CID systems, coupled with the increased amount of the sensor data overloading current watch-stander's abilities to determine regularized patterns and anomalies, offers a technological opportunity to exploit and ease a human’s workload. Machine learning and other AI systems may be able to fill this gap, assisting in the determination of unknown-unknowns. Research conducted on machine learning and deep learning techniques determined possible applications for CID use in the surface Navy, with technology acquisition and integration as the limiting factors. Continued research concerning the integration of legacy systems and new technologies is necessary to realize the true potential of AI in CID unknown-unknown applications.
Approved for public release. distribution is unlimited
Showing items related by title, author, creator and subject.
Application of thin film photovoltaic CIGS cells to extend the endurance of small unmanned aerial systems Lai, Matthew D. (Monterey, California: Naval Postgraduate School, 2017-06);Vulnerability and unknowns are characteristic in all military missions. With an end goal to enhance situational awareness and provide the military unit an advantage, small unmanned aerial systems (SUAS) have been utilized ...
Van Bossuyt, Douglas L.; O’Halloran, Bryan M.; Arlitt, Ryan M. (IEEE, 2018);System of systems (SoS) failures can sometimes be traced to a system within the SoS behaving in unexpected ways. Due to their emergent complexity, these types of failures are notoriously challenging to foresee. This ...
Sheng Li; Yun Fu (2016);In this paper, we aim at learning robust and discriminative subspaces from noisy data. Subspace learning is widely used in extracting discriminative features for classifica- tion. However, when data are contaminated with ...