Phototactic Movement of Battery-Powered and Self-Charging Robot Swarms

Abstract

In biology, phototaxing is one of the many forms of directed motions that organisms can exhibit. If stimulated by light, these organisms can move towards or away from the source. In this paper we present an algorithm that is reliably capable of moving a swarm of virtual robots away from a light source, all while the individual virtual robots are incapable of determining the direction and origin of the light source. The individual agents cannot generate phototactic behavior by themselves, instead this is realized through the overall, emergent behavior of the swarm. This paper explores an algorithm to produce phototactic movement in swarms of virtual robots. The algorithm is developed so that robots in a swarm of any size can execute it in a decentralized fashion, without the need for any globalized control or communication between them. We explore two different sets of virtual robots, one with self-charging and one with battery-powered characteristics. In a rigorous testing and evaluation we find that the swarms behavior and performance with regards to moving away from a light source varies strongly.