Flocking and spreading dynamics in populations of self-propelled agents

TL;DR

This paper introduces a model where mobile agents' movement and information spread influence each other, leading to enhanced flocking, complex structures, and lowered epidemic thresholds, bridging active matter physics and agent-based modeling.

Contribution

It presents a novel framework integrating agent motion and SIS epidemic dynamics, revealing new flocking behaviors and phase transitions driven by information spreading.

Findings

Flocking and spreading are mutually enhanced by feedback.

Complex spatial structures emerge and can be controlled by velocity.

Low velocities lead to dense swarms reducing epidemic thresholds.

Abstract

Populations of self-propelled mobile agents - animal groups, robot swarms or crowds of people - that exchange information with their surrounding, host fascinating cooperative behaviors. While in many situations of interest the agents motion is driven by the transmission of information (e.g. the presence of an approaching predator) from neighboring peers, previous modeling efforts have focused on situations where agents either sit on static networks, or move independently of the information spreading across the population. Here, we introduce a reference model to tackle this current lack of general framework. We consider mobile agents which align their direction of motion (based on the Kuramoto dynamics) and carry an internal state governed by the Susceptible-Infected-Susceptible (SIS) epidemic process, characterizing the spread of information in the population, and affecting the way agents move in space. We show that the feedback between the agents motion and information spreading is responsible for (i) the enhancement of both flocking and information spreading, (ii) the emergence of complex spatial structures, or swarms, which can be controlled by the velocity of the agents. The SIS dynamics is able to drive a flocking phase transition even in the absence of explicit velocity-alignment, recovering a behavior reminiscent of Vicsek-like systems but featuring macro-phase separation rather than micro-phase separation. We show that the formation of dense swarms at low velocities reduces the epidemic threshold of information spreading with respect to the mean field limit. By bridging together soft active matter physics and agent based modeling of complex systems, we shed light upon a general positive feedback mechanism that crucially affects the collective behavior of mobile agents, providing a reference framework to study realistic situations where this mechanism is at play.