Evolution of Discrete Opinions on Human-Swarm Interaction in Influence Networks

The aim of this paper is to explore social dynamics in conjunction with robot swarms dynamics to reach opinion consensus in a discussions series. First, it is presented the DeGroot-Friedkin model for the self-confidence dynamics of individuals and how it is updated to discussions on a sequence of topics, reaching a consensus in each case. Also, it is proposed an adaptation to a gossip-based algorithm in order to handle the consensus of opinions limited to a finite set of possible values. After each issue discussion, the interactions weights are updated. Then it is considered the case where some agents are replaced by robotic agents and it is analyzed how this impact the discussion outcome compared to the previous case. It is concluded that these agents with static self-confidence make others less confident and consequently have a greater influence on the discussion over a sequence of topics outputs.