Reliable Multi-Agent Control in Failure-Prone Environments via Inhomogeneous Markov Chains

Abstract

This project developed protocols for the coordination of autonomous nodes (such as UAVs, ROVs, and other robotic platforms) operating in fault-prone environments characterized by noisy and time-varying communication, message losses, and persistent and unpredictable node failures. For problems of formation control, leader-following, cooperative estimation and learning, resource allocation, and others tasks fitting within a separable optimization framework, control strategies were developed with fast and reliable performance, even in simulated networks of thousands of nodes.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 04, 2019
Accession Number
AD1085638

Entities

People

  • Alex Olshevsky

Organizations

  • University of Illinois Urbana–Champaign

Tags

Communities of Interest

  • Autonomy
  • C4I
  • Energy and Power Technologies
  • Engineered Resilient Systems

DTIC Thesaurus Topics

  • Air Force
  • Air Force Research Laboratories
  • Algorithms
  • Computer Science
  • Convergence
  • Coordinate Systems
  • Environment
  • Grids
  • Illinois
  • Machine Learning
  • Markov Chains
  • Multiagent Systems
  • Neural Networks
  • Scientific Research
  • Supervised Machine Learning
  • Training
  • Trees (Data Structures)

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Computer Networking
  • Robotics and Automation.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - Machine Learning Algorithms
  • Autonomy
  • Autonomy - Autonomous System Control