Collective Behavior of Intelligent Active Brownian Particles in the Presence of a Static Obstacle

Abstract

Using computer simulations, we investigate the collective behavior of two-dimensional active Brownian particles (ABPs) with excluded-volume interactions in the bulk and in the presence of a single static circular obstacle. Perception-mediated interactions are introduced through a vision-based steering mechanism that enables each particle to reorient its propulsion direction according to the instantaneous positions of neighboring particles within a prescribed vision cone. In the bulk, these intelligent active Brownian particles (iABPs) exhibit distinct collective states, including aggregated clusters, worm-like chains, worm-aggregate coexistence, and dilute gas phases. We characterize these states using a shape anisotropy parameter and construct the corresponding phase diagrams. The presence of a static obstacle, which interacts with the particles through purely repulsive forces, qualitatively alters their collective behavior. In contrast to conventional ABPs, whose isotropic accumulation around the obstacle increases with activity, iABPs exhibit the opposite trend, with boundary accumulation decreasing as the activity increases. We further identify empirical scaling relations that describe particle accumulation and cluster formation at the obstacle boundary. Consequently, the relative effective diffusion coefficient of iABPs displays a nonmonotonic dependence on self-propulsion speed, whereas that of conventional ABPs decreases monotonically with increasing activity. In addition, we show that the residence times of iABPs are orders of magnitude shorter than those of conventional ABPs, indicating that perception-mediated interactions can be useful for controlling the organization and transport of active particles in complex environments.