Predicting Cognitive Emitter Behavior with GAIL

Abstract

Through interactions with the surroundings, modern cognitive radar systems learn to optimize their decision-making process to intelligently select transmission waveforms and operating parameters. Because of their waveform agility and the ability to respond dynamically, cognitive radars are difficult to track and disrupt. This work aims to explore if Generative Adversarial Imitation Learning (GAIL) can be applied to capture, imitate, and predict the behavior of cognitive radars. We study the basic principles of GAIL, explore its existing applications, and research implementation of the approach for tracking the actions of self-driving cars. We conclude with the feasibility analysis of utilizing GAIL for predicting the behavior of cognitive radar systems.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 18, 2023
Accession Number
AD1190622

Entities

People

  • Anthony Tai

Organizations

  • Naval Surface Warfare Center

Tags

Communities of Interest

  • Materials and Manufacturing Processes
  • Sensors

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Artificial Intelligence Software
  • Computer Languages
  • Computer Vision
  • Detection
  • Image Processing
  • Information Processing
  • Information Systems
  • Machine Learning
  • Natural Language Processing
  • Neural Networks
  • Radar
  • Reinforcement Learning
  • Signal Processing
  • Supervised Machine Learning
  • Surface Warfare
  • Target Detection
  • Unmanned Vehicles

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Computational Modeling and Simulation
  • Radar Systems Engineering.