A Fast Self-Organizing Pattern Formation Method for Swarm Robots in Dynamic Multi-Region Environments

Abstract

Pattern formation of swarm robots is applied in many domains. While the environment is dynamic, it is difficult for swarm robots to form the correct pattern quickly. In particular, when there are multi regions that are not connected in the environment, the swarm robots need to adjust the spatial formation in time by self-organization. This paper provides a bio-inspired method for pattern formation of the swarm robots, which can be realized through three basic behaviors. On the other hand, swarm robots can be divided into some social organizations, which reduce the time of pattern formation especially for the multi regions. Experiments show that our method can form the pattern quickly in the dynamic multi regions.