Synthesis and Analysis of Non-Reactive Controllers for Multi-Robot Sequential Task Domains

In this paper we present a macroscopic model for the analysis of homogeneous task-directed multi-robot systems (MRS). The model is used to compute the probability that a given MRS will correctly execute a given sequential Markovian task. We consider distributed MRS composed of non-communicative robots that maintain a limited amount of non-transient internal state. The model shares a common formal framework with our past work on the development of systematic methods for the synthesis of MRS. As such, it can be used to improve design procedures by providing an analytical approach to the evaluation of design decisions. The unified nature of the modeling and synthesis methods are part of our ongoing work toward a general, comprehensive and principled MRS design methodology. We apply the model to the analysis of system performance in a multi-robot construction task domain. Our past work on synthesis methods provides robot controllers for this task domain; the model provides quantitative pr...

• Type: Conference
• Year: 2004
• Where: ISER
• Authors: Chris Jones, Maja J. Mataric
• Keywords: ISER 2004, Model, Robotics, Sequential Markovian Task, Synthesis Methods