Decentralized task-oriented local group generation for robot swarms

This paper addresses an adjustable group generation problem for heterogeneous robot groups performing cooperative tasks simultaneously. As its solution approach, we propose a decentralized task-oriented group generation scheme, which is composed of group consensus and self-adjustment algorithms. The group consensus algorithm enables robots to select leader robots and generate individual groups based on the leaders. Through the self-adjustment algorithm, to meet assigned task conditions, the leader attempts to recruit more members or any robots are dismissed against the group. By doing this, robots can reach consensus on self-organizing themselves according to the task conditions. Extensive simulations are performed to verify that the proposed scheme effects an adjustable self-organization.