Abstract

Swarm technology will enable reconfigurable robotic tasks that require high resilience and boundless scaling, especially in vast, unstructured, and dynamic environments. A robot swarm can be defined as a decentralized multi-robot system that, through only local interactions, can exhibit a collective behavior that facilitates the realization of tasks well beyond the reach of an individual. However, in contrast with the control of individuals, control laws for robot swarms are incredibly fragile against imperfections. In general, tiny imperfections in sensing and actuation are amplified throughout large numbers of robots and rapidly erode and make unpredictable the overall performance of the robot swarm. Notwithstanding, imperfections can result in surprising complex emergent behaviors such as intricate trajectory patterns of mobile robot swarms.

This talk will explore how to unleash and ally with imperfections in sensing and actuation instead of suppressing their ``damaging’’ effects as a formal strategy to control emergent behaviors of robot swarms.