Spontaneously ordered motion of self-propelled particles

Abstract

We study a biologically inspired, inherently non-equilibrium model consisting of self-propelled particles. In the model, particles move on a plane with a velocity of constant magnitude; they locally interact with their neighbors by choosing at each time step a velocity direction equal to the average direction of their neighbors. Thus, in the limit of vanishing velocities the model becomes analogous to a Monte-Carlo realization of the classical XY ferromagnet. We show by large-scale numerical simulations that, unlike in the equilibrium XY model, a long-range ordered phase characterized by non-vanishing net flow $\phi$ emerges in this system in a phase space domain bordered by a critical line along which the fluctuations of the order parameter diverge. The corresponding phase diagram as a function of two parameters, the amplitude of noise $\eta$ and the average density of the particles $\varrho$ is calculated and is found to have the form $\eta_c(\varrho)\sim \varrho^{1/2}$. We also find that $\phi$ scales as a function of the external bias $h$ (field or ``wind'') according to a power law $\phi\sim h^{0.9}$. In the ordered phase the system shows long-range correlated fluctuations and $1/f$ noise.