Generating self-organizing collective behavior using separation dynamics from experimental data

Abstract

We propose a simple dynamical model of collective behavior that attempts to follow a rule of local neighbor interactions, which was abstracted from experimental flight data of pigeon flocks. This rule is consistent with the previous hypothesis of the basic mechanisms affecting collective motion: short range repulsion to avoid collisions, longer range attraction to keep the group together, and velocity alignment to maintain the same navigational direction. The local interactions of our model consist of using naive neighbor estimates: essentially assuming that nearest neighbors will move with the same velocity as in the previous time interval. The dynamics of our model try to follow the experimental rule by using a simple adjustment mechanism with respect to the naive neighbor estimates. From our simulations, we show that by changing the initial conditions or the number of individuals involved in the interactions, the model is capable of exhibiting a wide range of realistic behaviors. Our study emphasizes the importance of using experimental data for making better models of complex systems, and that this should contribute to a better understanding of nonlinear collective dynamics.