Home /Research /Self-organised recruitment in a heteregeneous swarm
SWARM

Self-organised recruitment in a heteregeneous swarm

Carlo Pinciroli, Rehan O’Grady, Anders Lyhne Christensen, Marco Dorigo

Year
2009
Citations
10

Abstract

Abstract — In this paper, we present a heterogeneous recruitment system which allows aerial robots to recruit groups of wheeled robots. The system is novel because it is self-organized, is based only on simple probabilistic rules and relies only on local communication. Our approach is inspired by the aggregation behavior of cockroaches. The system allows aerial robots to recruit wheeled robot groups of different sizes in parallel. Although governed by probabilistic rules, we show that our system stabilises to a steady state distribution of wheeled robots that corresponds to the desired sizes of recruited groups. The system can also handle perturbations as robots enter and leave the system during execution—robots are redistributed accordingly and the system will once again stabilise. The system also displays sensible behaviour when there are not enough wheeled robots to satisfy the recruitment quotas. We conduct experiments based on physically embodied simulation of the robots to demonstrate these properties of the system, and to explore the scalability of our system. I.

Keywords

RobotSwarm roboticsProbabilistic logicScalabilityComputer scienceSwarm behaviourMobile robotState (computer science)Distributed computingAnt robotics

Related papers

Browse all SWARM papers