Improving Efficiency of Leading a Flock in Ad Hoc Teamwork Settings

Designing robots that can influence their teammates to behave in a particular manner is desirable — especially if the robots are able to do this without prior coordination or explicit communication. In this paper, we examine an aspect of the problem of leading teammates in an ad hoc teamwork setting. Specifically, we consider how to best utilize ad hoc agents to lead a flock to a desired orientation. Specifically, we extend upon recent work [4] by presenting a more efficient search methodology for determining how the ad hoc agents should orient themselves to optimally influence a flock.