A Network-Centric Formalism for Disturbance Rejection Design and Human Swarm Interaction
Abstract
The objective of the proposed research is to examine fundamental structural bounds on the disturbance rejection and human-swarm interaction properties of a network of diffusively interacting dynamic agents. These bounds are then employed to adaptively change or re-parameterize the network geometry in order to improve the disturbance resilience of the network while facilitating its manageability by human operator(s). The corresponding algorithms are examined in the context of centralized procedures where the network structure and its parameters are designed viasuitable optimization algorithms, or analyzed via the probabilistic method, as well as a game theoretic approach where each agent or a subset of agents, based on the available local information, update the network structure and parameters in order to improve the overall disturbance rejection properties of the network and its effectiveness for being managed by human operator(s). In this venue, we address a number of issues at the core of networked multiagent networks with the aim of highlighting the intricate relationship between network structure and agent dynamics, on one hand,and network disturbance rejection properties and external manageability, on the other.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 06, 2015
- Accession Number
- AD1001404
Entities
People
- Mehran Mesbahi
Organizations
- University of Washington