Deception and Risk Aware Dynamic Routing

Abstract

This report addresses the question of how routing algorithms should be modified when they are applied to routing agents, such as unmanned vehicles, in contested environments. Our primary motivation is prior work at the Naval Postgraduate School that showed how optimization-based routing that does not account for the presence of an observing adversary can result in predictable paths from which operational intent can be inferred. We propose the use of a randomized routing strategy based on the solution of a two-player game and evaluate its effectiveness on data derived from a multi-thread experiment. We also propose methods for dynamic routing in contested environments. Our work highlights the trade-off that needs to be made between efficiency and predictability in route planning and can potentially inform the development of new adversary-aware routing algorithms for unmanned systems.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 21, 2023
Accession Number
AD1224311

Entities

People

  • Jefferson Huang
  • Ruriko Yoshida

Organizations

  • Naval Postgraduate School

Tags

Fields of Study

  • Computer science

Readers

  • Computer Networking
  • Distributed Systems and Data Platform Development
  • Game Theory.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - DoD AI Strategy
  • AI & ML - Machine Learning Algorithms
  • Autonomy
  • Autonomy - UAVs