ARTIFICIAL NEURAL NETWORK APPLICATIONS IN NAVAL WARFARE

Authors
  1. Jacobson, J.Z.
  2. Laplante, M.
  3. Mogford, R.H.
Corporate Authors
Government Consulting Group, Ottawa ONT (CAN);Operational Research and Analysis Establishment, Ottawa ONT (CAN) Directorate of Mathematics and Statistics
Abstract
The Operational Research and Analysis Establishment commissioned a study on the application of artificial neural networks or parallel distributed processing models of information storage to specific tasks in naval warfare. In a task of a naval air defence, networks were trained with data from simulated engagements under thousands of different conditions. After training, networks were able to rapidly choose the correct fire control doctrine in untrained situations perfectly where shoot-double-salvo (or shoot-look-shoot) were preferable doctrines, but made a few errors where shoot-single-salvo would be preferred. A second task involved network discrimination of passive SONAR signals. In that case, networks trained on specific contacts were easily and quickly able to identify new signals from the same targets and to identify those signals from simulated background sound.
Date of publication
15 Mar 1988
Number of Pages
148
DSTKIM No
89-01593
CANDIS No
59523
Format(s):
Hardcopy;Originator's fiche received by DSIS

Permanent link

Document 1 of 1

Date modified: