Naval Imagery Infrastructure Revitalization (NIIR) - IUPUI

Abstract

Targeting and tracking are key technologies in EO/IR imagers. These are used for situational awareness, combat identification, and engagement of targets with kinetic and directed energy weapons. Artificial Intelligence targeting and tracking systems developed with machine learning are showing promise toward increasing systems effectiveness while reducing the burden on the analyst or warfighter. These algorithms need to be trained with vast amounts of real world data. This data may not be available, or only available for limited scenarios. This is where virtual data can be used to extend the real world data, increase robustness, and be used to test black-box trackers for accuracy and robustness. In co-ordination with Dr. Craig Cameron at Navy/CRANE, the proposed research goals are to: (1) develop measures of performance (MOP) for ML and AIbased trackers, (2) develop evaluation measures of performance (MOP) for the effectiveness of synthetic virtual scene generation, and (3) develop adversarial examples from a few common AI/ML trackers, in order to improve and understand the tracker robustness.

Document Details

Document Type
DoD Grant Award
Publication Date
Jul 20, 2020
Source ID
N000142012674

Entities

People

  • Lauren Christopher

Organizations

  • Indiana University
  • Office of Naval Research
  • United States Navy

Tags

Fields of Study

  • Computer science

Readers

  • Neural Network Machine Learning.
  • Sensor Fusion and Tracking Systems.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • AI & ML - Neural Networks
  • Directed Energy