Towards Single-Operator Control of Tightly-Coordinated Robot Teams
Gal A. Kaminka, Yehuda Elmaliach · Adaptive Agents and Multi-Agents Systems · 2004
There is growing recognition that many applications of robots will require a human operator to direct multiple robots that collaborate to achieve the operator’s goals. However, the bulk of existing work in this area assume that robots are independent of each other and thus key challenges and opportunities are ignored in monitoring and operating tightly-coordinating teams. This paper takes steps to address these open issues. First, we introduce a graphical display that explicitly shows the coordination in the team, in terms of the robots’ state with respect to each other (socially-attentive monitoring). As a result, the operator can easily detect coordination failures, even before these cause overall failure in the task. We also take advantage of the robots’ teamwork, to allow the robots to actively assist the operator in maintaining her control of the team (Distributed Call-Request Resolution). We evaluate Distributed Call-Request Resolution techniques in several multi-robot task. Preliminary results show that they can lead to significant improvements in task completion time, and number of coordination failures during execution.