Target Discrimination with Neural Networks,

Abstract

The feasibility of discriminating the warhead from an intentionally segmented exoatmospheric threat missile is demonstrated by applying the time delay neural network (TDNN) and the adaptive time delay neural network (ATNN). Exoatmospheric threats are especially difficult to distinguish using currently available techniques because all threat segments follow the same trajectory. Thus, classification must be done using infrared sensors that record the signal over time. Results have demonstrated that the trained neural networks were able to successfully identify warheads from other missile parts on a variety of simulated scenarios, including differing angles and tumbling. The network with adaptive time delays (the ATNN) performs highly complex mapping on a limited set of training data and achieves better generalization to overall trends of situations compared to the TDNN, which includes time delays but adapts only its weights. The ATNN was trained on additive noisy data, and it is shown that the ATNN possesses robustness to environment variations.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 01, 1996
Accession Number
ADA319959

Entities

People

  • Cheryl Resh
  • Daw-tung Lin
  • Judith Dayhoff

Organizations

  • University of Maryland

Tags

Communities of Interest

  • Weapons Technologies

DTIC Thesaurus Topics

  • Additives (Chemicals)
  • Classification
  • Cooperation
  • Detectors
  • Discrimination
  • Environment
  • Infrared Detectors
  • Neural Networks
  • Segmented
  • Target Discrimination
  • Training
  • Trajectories
  • Tumbling

Readers

  • Brain and Cognitive Science; Experimental Psychology; Cognitive Neuroscience
  • Electrochemical Surface Science
  • Missile Defense Systems.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • AI & ML - Neural Networks