Marcus J. Huber, ORINCON Corporation, Edmund H. Durfee, University of Michigan, USA
Plan recognition remains a largely unexplored paradigm for facilitating coordination. In this paper, we begin to explore domain, task, and agent characteristics which impact upon the utility of using plan recognition for coordinating multiple agents and, in particular, collections of agents organized into competing teams. Agents in our research are supplied plan recognition capabilities in the form of specially instantiated belief networks. These networks, called plan recognition networks (PRSn), are automatically constructed from plan models by the ASPRN system. Our initial experiments demonstrate that plan recognition can provide effective coordination, both for cooperating with team agents as well as for competing against non-team agents. We have also begun to explore the factors involved with PRN-based coordination relative to communication-based coordination and we describe results of some early experiments.