David Sumpter
Department of Mathematics
Uppsala University
The motion of animal groups, such as bird flocks, locust swarms and fish schools, provide beautiful examples of non-linear systems wher many interacting individuals producing complex patterns. Recently a class of models known as self-propelled particle models has been proposed to explain such flocking. In these models stochastic 'particles' align with their nearest neighbours and collective motion 'emeges' without leadership or central control. I describe models and experiments on homing pigeons and locusts which illustrate this approach and show how it can be usefully applied. I will also discuss the problem of giving a macroscopic description to systems with large numbers of interacting individuals. In particular I will describe some prelimenary work on the use of the equation free approach to estimating the long term distribution of macroscopic variables, such as flock alignment.