The European Future Technologies Conference and Exhibition 2011 Building Simple Formations in Large Societies of Tiny Mobile Robots

Abstract We envision that in the future lots of tasks that are handled nowadays by few but complex mobile robots will be performed byhuge swarms of tiny robots. This way the system becomes more robust against single failures and cheaper, since the robots can bevery simple. As a prerequisite, the robots have to be able to form given formations in order to organise their work. We review newstrategies that perform the task of building a simple line with robots that are limited to see only their direct surrounding. As energyis the main limiting factor, we analyse how the two main energy consumers (energy for sensing and energy for motion) are affectedby our algorithms.© Selection and peer-review under responsibility of FET11 conference organizers and published by Elsevier B.V. 1. Introduction Nowadays more and more tasks are handled by autonomic mobile robotic systems. So far, most of these systemsconsist of very complex robots that are expensive and error-prone. We envision a development towards large systemsof tiny, simple, cheap robots. Such systems of robots cannot be centrally controlled or optimised. Instead, they haveto work in a distributed fashion, where each robots bases its decision on very limited information about its localenvironment. The challenge arising in this context is to take decisions such that the local strategies of the robots resultin a globally desired behaviour of the whole system. If such strategies can be designed, they have several advantagesover solutions based on few complex robots: the cost will decrease because of the simplicity of the robots; the systemswill be much more robust against failures, because the solution does not depend critically on single entities; strategiesdesigned for such systems will often work well also in dynamic environments (e.g., if the tiny robots are moved bywind or mobile obstacles change the environment). In order to be able to solve complex tasks, as a prerequisite, thetiny robots need to be able to organise themselves into simple formations. One of the simplest formations one can thinkof is a line. Therefore, in the remainder of this work, we will present approaches how to form lines under the majorconstraint of limited information and we will review recent results.