Multitiered representation for autonomous agents

Wasson, Glenn; Martin, Worthy N.

doi:10.1117/12.335687

Cited by 2 publications

(1 citation statement)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They may be interested in the most detailed resolution level to make an order or a lower resolution level to command a team. In hierarchical autonomous agent modeling, agents may jointly execute multiple layers in order to utilize the capabilities of each layer [62], for example, to use the deliberative layer to arrange the next behavior to complete a task, and to use the reactive layer to perceive and react to the environment. When talking about representation of models, resolution is obviously a term that needs to be defined.…”

Section: Multi-resolution Modelingmentioning

confidence: 99%

Autonomous agents for distributed virtual worlds

Wang¹

View full text Add to dashboard Cite

Autonomous agents and multi-agent systems (MASs) provide a valuable tool for handling increasing software complexity and developing efficient applications. Large scale distributed simulation is a technology for linking simulation components at possibly geographically different locations to create a common virtual world. It is attractive to integrate these two technologies together, in order to facilitate rapid and accurate decision making in simulation applications, and to increase reusability and interoperability of agent-based simulation components. These are the reasons for our investigation of integrating agents into distributed simulations. There are many applications such as battlefield simulations, interactive games, etc that provide the motivation for this integration. In this project, we develop a generic framework for autonomous agents in distributed simulations, where the behaviors of certain entities in the simulation are described by means of agents. The novelty of our approach comes from the way in which agents are integrated into a distributed simulation. We adopt the Foundation for Intelligent Physical Agents (FIPA) standard and choose a FIPA-compliant agent platform JADE to provide general agent-specific services. Components are linked using the Runtime Infrastructure (RTI) of the High Level Architecture (HLA), which is an IEEE standard for modeling and simulation. Thus we achieve a general integration architecture which is standard-compliant from the perspectives of not only agents but also distributed simulations. To build a seamless integration of agents and distributed simulations, we develop a special kind of federate, called a gateway federate. Gateway federates enable agents to perceive information from the environment in which they are embedded and make responses to it. A prototype system has been implemented to demonstrate the feasibility of this approach. Agents are enabled to communicate with one another via the HLA RTI using a special kind of agent, called a mailbox agent, that ensures causality and i

show abstract

Section: Multi-resolution Modelingmentioning

confidence: 99%