Multi-agent model of biological swarming

Abstract

An agent-based approach is used to explain the formation of vortex swarms in biological systems. The dynamics of the multiagent system is described by 3N coupled equations, modeling for each agent its position, its velocity and its internal energy depot. The energy depot considers the conditions for active biological motion, such as energy take-up, metabolism, and energy conversion. The equation of motion results from a superposition of deterministic and stochastic terms (random noise). The deterministic part considers indirect interactions with other agents to describe local avoidance behavior, and external influences resulting from an attractive environmental potential. Stochastic computer simulations of the multi-agent system are shown in very good agreement with the behavior observed in Daphnia swarms.