Emergent properties of nonreciprocal robotic active matter

Nonreciprocal interactions, in which particles influence one another asymmetrically, can give rise to unusual collective dynamics and mechanical properties, including odd elasticity and odd viscosity. Inspired by our earlier work in living chiral crystal [ Nature 2022 ],  We developed a programmable robotic active-matter platform, called MASBots, to investigate how these phenomena emerge and how different states of active matter are connected. [ preprint ]

​

By tuning the balance between hydrodynamic attraction, transverse interactions, and magnetic repulsion, we experimentally observe transitions between three distinct phases: an odd elastic solid, an odd viscous liquid, and a chiral active gas. Remarkably, the active gas remains spatially clustered through long-range hydrodynamic attraction and exhibits statistical signatures reminiscent of a two-dimensional self-gravitating vortex gas.

​

Through control of individual spinning speeds, interaction strengths, and spatially patterned activity, we can further tune the collective dynamics and mechanical responses, and induce behaviors such as phase separation and collective translation. This establishes robotic swarms as an experimental platform for exploring nonequilibrium physics and programmable states of matter, laying the groundwork for materials capable of adaptive and functional collective behavior.

​

Building on this foundation, we investigate how learning and memory can emerge from adaptive local interactions, and how information-driven organization enables increasingly sophisticated collective functions. Our long-term vision is intelligent matter, where collections of simple physical units can learn, adapt, and act collectively.

​

​