A Robust Aggregation Method for Quasi-blind Robots in an Active Environment

Abstract

Abstract. The aggregation of a swarm of autonomous agents into compact clusters is often a required behavior of multi-agent systems. In the case where no central control or coordination exists, this problem is known as the Decentralized Gathering. This presents a first step in the application of a bio-inspired aggregation scheme to aggregate robots whose abilities are very restricted: the only way they can communicate is through an active environment (stigmergy) and the only information they can receive is the local detection of the waves produced by other robots. The active environment obeys a cellular automaton rule and is simulated with a projection of light on the robots. Keywords:DecentralizedGathering,AgentAggregation,SwarmRobotics, Reaction-diffusion-chemotaxis