Many robotic systems can exhibit self‐organized collective behaviors similar to those of natural swarms. Some of these systems achieve coordination by enhancing individual intelligence or enabling explicit communication. In contrast, others consist of highly simplified modules capable of only basic sensing and physical interactions. Indeed, the former type faces challenges in manufacturing, modeling, and controlling the swarms, while the latter is usually incapable of complex collective tasks. Here, we present a diatom‐inspired robotic colony that explores group coordination through limited physical interactions. The modules in the colony, BarBots, possess neither individual mobility nor explicit communication capabilities, yet they can exhibit collective behaviors by sliding along their neighbors. We model the relationship between individual and group patterns, enabling the completion of designed collective behaviors. BarBots can also automatically assign actuation phases across a colony using simple light‐sensing abilities, promoting light‐stimulated self‐organization and object manipulation. These findings indicate the feasibility of tuning groups of simple robotic individuals for robust planned motions and collective behaviors.
Hu et al. (Sun,) studied this question.