Emergent coordinate systems for microrobot swarms built on topological defects

Our ability to make cell-sized robotic devices has dramatically improved in recent decades, suggesting future applications of ‘microbot’ swarms which collaborate to manipulate biological systems or additively manufacture larger objects in a massively parallel way. However, a major barrier to realizing this vision is coordination – how should individuals be designed so that a group of millions performs complex, goal-driven behavior?

In this numerical study, we focus on a key enabling question: how can agents agree upon a collective coordinate system while only interacting at short range and without individual identities? We solve this problem by leveraging the properties of topological defects to act as a long-lived and stable reference point from which a coordinate system can be built. In simulations of thousands of self-propelled agents, we demonstrate how communication between individuals on several independent channels results in a set of coupled fields which can form a stable coordinate system. Crucially, this information can be used to inform individual motion and enable large collectives to spontaneously adopt complex group geometries without external control or pre-defined leaders. This work highlights the usefulness of describing control algorithms for many-body systems as the interaction of coupled fields, where each field is represented by a variable spread across the memories of all individuals.