Designing Resilient Networked Swarms: A New Evolutionary Dynamic Approach that Introduces Dynamics and Variability to Population Game

Abstract

Our research will lead to new methods to characterize the self organizing capabilities of large swarms when each agent makes decisions and revises its strategies according to simple bounded rationality rules (protocols) inspired on evolutionary biology principles. The high level guidance is a payoff mechanism that continually ascribes a reward to each possible strategy in terms of the situational awareness and the overall configuration of the swarm. Considering that the protocols will favor higher reward strategies, the payoff mechanism, if appropriately chosen, can lead the swarms towards desirable configurations. Unlike existing work, we will consider that the payoff mechanism, which needs to broadcast the reward structure to at least some members of the swarm, may be adversely affected by delays, variability and other detrimental effects such as jamming and information transmission outages. Our technical approach is mostly inspired on so called passivity based methods that were initially developed to certify the resilience and stability of feedback interconnections of dynamical systems.

Document Details

Document Type
DoD Grant Award
Publication Date
Jan 14, 2022
Source ID
FA95501910315

Entities

People

  • Nuno C. Martins

Organizations

  • Air Force Office of Scientific Research
  • United States Air Force
  • University of Maryland

Tags

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Systems Analysis and Design
  • Theoretical Analysis.