Simulating the collective dynamics of active matter

Abstract

This thesis presents a performant, modular implementation of the Vicsek Model of active matter, developed to aid the study of phase transitions driven by control parameters. To check the validity of the implementation, selected results from papers by Yu et al. and Puzzo et al. have been replicated. Simulations of 2D noise sweeps reproduced stationary susceptibility scaling, supporting a first-order noise-driven phase transition, and the 3D Short-Time Dynamics protocol successfully identified the lower spinodal from disordered initial conditions. The results obtained support the claim that the noise-driven transition in the Vicsek Model is first-order, discontinuous. The validated package establishes a robust, open-source foundation for active matter research of the Vicsek model. Future work will focus on better optimization and Reactant-based automatic differentiation.