Path Formation and Goal Search in Swarm Robotics
Nouyan Shervin

Abstract:

In this work, we present a swarm robotic approach to exploration and navigation. Taking inspiration from swarm intelligence methods, we address the problem of solving complex tasks with the group of robots while using simple control strategies for an individual robot. In particular, our approach consists in visually connected robotic chains, where neighbouring members of a chain can perceive each other with a camera. A chain of robots can be used to establish a path between different locations, in this way allowing other robots to exploit the chain to navigate along the formed path. We present the results of two series of experiments. While in the first one we analyse the general capabilities of chain formation, in the second one the robots have to find a goal location and establish a path towards it starting from a home location. Three chain formation strategies are tested, differing in the degree of movement allowed to the robots which are aggregated into a chain.