An analysis of coordination in Multi-Robot Systems

days, an important research area within Robotics and Artificial Intelligence and a growing number of systems have been recently presented in the literature. In this paper we present an analysis of the most relevant works on M R S by specifically examining their cooperative aspects. In particular, we propose a new taxonomy for a fine and precise analysis of the recent works on M R S and we describe some approaches which we consider representative in our taxonomy. We finally discuss the outcome of our analysis and try to highlight future trends of the research on MRS. keywords : Multi Robot System, Cooperation, Coordination. 1 I n t r o d u c t i o n Multi-Robot Systems (MRS) have been proposed in the last decade in a variety of settings and frameworks, based on rather different goals and successfully applied in many application domains. Special attention has been given to MRS developed to operate in a dynamic environment, where uncertainty and unforeseen changes can happen due to the presence of robots and other agents that are external to the MRS itself. A significant body of work on MRS has been originated from motivations that are essentially of engineering nature, where MRS are designed and realized in order to improve the effectiveness, performance and robustness of a robotic system. Another significant development of MRS stems from the studies on biological systems or complex models arising in cognitive science co.om cs for [6]) th s work we are more interested in the system engineering aspects of the MRS, we consider to some extent also a few biologically inspired approaches. The main motivation for studying MRS from an engineering point of view is that they can improve the effectiveness of a robotic system either from the viewpoint of the performance in accomplishing certain tasks, or in the robustness and reliability of the system, which can be increased by modularization [1]. In fact, MRS are useful not only when the robots can accomplish different functions, but also when they have the same capabilities • 0 7 8 0 3 7 9 5 2 7 / 0 3 / $ 1 7 . 0 0 (~) 2 0 0 3 I n n n . [10]. Moreover, even when a single robot can achieve the given task, the possibility of deploying a team of robots can improve the performance of the overall system. The technical solutions that are being developed for MRS have several common features, but the discipline is still relatively young and a common framework for the work on MRS is difficult to identify. Moreover, a MRS cannot be simply regarded as a generalization of the single robot case and the proposed approaches need to be more precisely characterized in terms of assumptions about the environment and in terms of the interhal system organization [18, 19]. Neither a MRS can be simply considered as a special case of a Multi-Agent System, because of the issues arising when dealing with a physical environment, such as uncertainty and incompleteness of the information acquired from the environment. Interesting contributions towards a classification of the work on MRS are the surveys by Cao et al. [6] and by Dudek et al. [10]. In [6] several dimensions for characterizing a MRS are proposed, while in [10] a classification of MRS, that is more focused on the communication and computation aspects of MRS is presented. In [32] an introduction to the field of MAS and MRS is presented along with a conceptual framework to organize the possible systems while the research topics in the MRS field are discussed in [23]. In [18, 20] the problems concerning the analysis and synthesis of intelligent group behaviors in MRS are addressed in detail and definitions for key concepts related to MRS coordination are given, thus characterizing precisely several imporrant aspects of the problem. Finally a chapter of [1] is devoted to social behavior for MRS, where several interesting characteristics and fundamental problems for MRS behaving in a society are pointed out. Although cooperation and coordination are central in many works on MRS and they are taken into account in the above cited surveys, a detailed analysis that specifically looks at these aspects can be interesting in at least two respects. First, the complexity of the systems and of the application domains requires more and more sophisticated forms of coordination. Consequently, an analysis of the proposals that appeared in the recent literature, can help the designer of MRS to choose the approach to coordination that is best suited for the application at hand. Second, following the coordination perspective it