Abstract
This paper proposes and demonstrates the viability of a novel approach to control multi-robot systems. This approach is based on computing with membranes (P systems) that is a branch of molecular computing inspired by the structure and function of biological cells. There are many variants to the original P system model that have been introduced and theoretically studied over the last two decades. Most of them were proved to be computationally universal. Many applications for P systems have been proposed including the recent use of numerical P systems to control a single robot. This paper makes several contributions to the area of P systems applied to robot control. First, it approaches the problem of controlling multi-robot systems containing dozens of robots with very limited capabilities. Second, it proposes the use of a symbolic approach based on XP colonies. Third, this paper presents a detailed comparison of the proposed solution to a standard one based on finite state machines, which highlights the advantages and disadvantages of the proposed solution. Finally, in addition to using physically realistic simulations, this paper demonstrates the feasibility of the proposed approach on real robots. The source code is available from an online repository, which makes all the experiments fully reproducible. Online demonstration videos illustrate all of the experiments.
Get full access to this article
View all access options for this article.
