Leader election algorithms for static swarms

Biologically Inspired Cognitive Architectures. 2015. No. 12. P. 54–64.

Valery Karpov, Irina Karpova

Work solutions are proposed for problems of leader definition and role distribution in homogeneous groups of robots. It was shown that transition from a swarm to a collective of robots with hierarchical organisation is possible using exclusively local interaction. The local re-voting algorithm is central to the procedure for choice of leader while distribution of roles can be achieved by a wave method. The basis for this approach is the static swarm model characterised by the absence of a set control centre; it represents the network fixed at some time interval as a set of locally interacting agents. A task of cooperative hunting by distributed mobile robots based on local interaction was considered. Two strategies were used for the hunting task solution: individual hunting and pack-hunting. Simulation results showed that symbiosis of leader election and role distribution procedures has advantages over the individual strategy.

Keywords: процедура голосования, swarm robotics, roles distribution, роевая робототехника, распределение ролей, voting procedure, local interaction, выбор лидера группы, локальное взаимодействие, group leader selection, pack-hunting, групповая охота