Fault Tolerant Coordination of Robot Teams

Abstract

Coordinating interaction among robot teams in order to maintain a formation is difficult. The objective of our research is to develop fault tolerant methods of accomplishing this task that will perform well in real world environments. We are examining this problem from a distributed computing perspective and have experimented with symbolic machine learning as a method for addressing uncertainties in communication among distributed, autonomous robots. Our intent is to develop solutions that have a broader applicability to distributed computing environments and applications. Introduction Coordinating a robot team in the pursuit of a common goal is difficult. Elements of this problem can be equated to that of communication and coordination in distributed computing applications. In distributed computing applications, issues such as maintaining global state and determining causality among a series of events are fraught with difficulty, even under controlled conditions (Chow and Johnson 19...