How to debug a swarm using non-equilibrium statistical physics

Cooperation is vital for the survival of a swarm. No single bird is faster than a jet plane, and no single fish is faster than a speed boat --- humans beat individual animals in air, land, and sea. But, when animals cooperate and swarm, they beat us since biblical times. The science of swarm cooperation contains many open questions, awaiting the discovery of new principles in disordered, far-from-equilibrium, many-body physics. Moreover, since multicellular organisms are, in fact, a cooperative heterogeneous swarm, studying swarm cooperation may help us understand living matter. A powerful approach to study swarms experimentally is to design them bottom-up. This requires us to manufacture active-particles in large numbers, as when it comes to swarms --- more is different. To-date artificial swarms fail to reproduce the fluidity observed in natural swarms, both on the micro-scale and the macro-scale. In my talk, I will describe two projects aiming to bridge this gap by looking at swarm cooperation through the lens of non-equilibrium statistical physics. In the first part, I will describe a novel approach to make artificial micro-swimmers that remain active, even at high densities. The swimmers display a turbulent flow structure (seen in living fluids), and cooperate to perform a greater task (that can be mapped to an athermal phase separation). In the second part,  I will describe a collection of ``smart'' particles that are not only active, but can also locally compute --- a robotic swarm. Treating the individual robots as aligning active-agents, I will show how a small morphological change leads to vastly different collective ability to perform different tasks. When robots are allowed to collide, their collective dynamics fluidizes, allowing us to link their hydrodynamic properties with the swarm's ability to cooperatively learn.