A layered modeling and simulation architecture for agent-based system development

doi:10.1109/5.910855

A layered modeling and simulation architecture for agent-based system development

2001-01-01

41

PRE

AI

H

Hessam S. Sarjoughian

DOI：10.1109/5.910855

Original request for help

Save

Abstract

En 中文

This paper proposes a layered architectural framework to support agent-based system development in a collaborative, multidisciplinary engineering setting. This architecture is viewed from two distinct perspectives. First, the environment must enable agent-based modeling and simulation. Second, it should support concurrent (team-oriented) engineering. The main focus is on the proposed layered architecture delineating various needs of an agent-based system, thus supporting incremental specification, design, implementation, and testing. In our discussions, we distinguish between performing agents and simulated agents. The former refers to agents as they are performing their tasks in real-world settings. The latter refers to agents that have their behavior simulated in a virtual environment. In these terms, the proposed framework is intended to form the basis for environments that support development of agents, in both performance and simulation modes, as well as in hybrid combination (both performing and simulated agents interacting at the same time). The proposed framework is a step toward realization of agent-based systems under the umbrella of the simulation-based acquisition (SBA) initiative of the U.S. Department of Defense.

Keywords:

agent architecture requirements

agent modeling and simulation

DEVS

distributed agent simulation

high-level architecture

A layered modeling and simulation architecture for agent-based system development

Abstract

AI Summary

Journal

Organization