New Challenges in AI for Military Simulation: Are Multilevel Heterogeneous Models the Solution?

Between a Rock and a Hard Place: Cognitive Science Principles Meet AI-Hard Problems

Volume

Papers from the 2006 AAAI Spring Symposium

Contents

As our nations military forces face new challenges in the field as the result of trying to defeat unconventional adversaries, so it goes with our military modeling and simulation community, who are attempting to build decision-support tools for our command staff. This paper will address two extraordinarily large issues which the modeling and simulation community seeks to make progress on, and how multilevel heterogeneous models (MHM) may play a major part in moving forward. The first of the issues concerns modeling our adversaries, who we conceive of as being an organization: from nation-states to terror cells, as systems with interdependent political, military, economic, social, infrastructural, and information-based components. Secondly, we face the more daunting problem of developing cognitive models of the human elements within these systems: from leaders of countries to the common citizen and everything in between. However, there seem to be parallels between these problems in that they both are readily able to be modeled by as multilevel heterogeneous systems. In this paper, I argue that while both problems are able to be modeled as MHMs, the real issue lies in providing strategies for integrating among the parts of the MHMs.