A Framework for Studying Multi-Robot Task Allocation

Abstract

Although many multi-robot task allocation (MRTA) architectures can be found in the literature, relatively little has been said regarding the fundamental theoretical characteristics of the task allocation problem. We present a formal, but practical, framework for studying MRTA. In constructing our framework, we borrow from the Operations Research community and show that MRTA can be understood as an instance of the Optimal Assignment Problem. We use this framework to analyze several recently proposed approaches to MRTA, describing their fundamental characteristics in such a way that they can be objectively studied, compared, and evaluated. In so doing, we demonstrate the utility of such frameworks in formalizing robotics research, which we argue is vital to the development of the field.