Active reorientation in an active granular system powered by toy vibrobots

While emergent collective motion in systems of self-propelled objects is remarkable, one proclivity of active agents, particularly animals, is often overlooked when building models: collision avoidance. We present here an active granular experiment of disks propelled by toy vibrobots, which integrates an “active reorientation” behavior in analogy to collision avoidance in animals. The system demonstrates local flocking, wherein particle velocities locally align. Inspired by this experiment, we develop a computational model of self-propelled disks with an active reorientation mechanism. This simple numerical model exhibits rich phase behaviors: a disordered state, a flocking state and a clustering state under different parametric conditions. We find a notable suppression of aggregation in regions of parameter space corresponding to strong collective motion. Clusters develop quickly in this region, but are metastable, and collapse once the flocking state is achieved. Our experiment and model demonstrate the importance of active reorientation on emergent behaviors in systems of self-propelled agents, and illustrate the profound interplay between different emergent phases of active matter.