An embeddable testbed for insurgent and terrorist agent theories: InsurgiSim

Abstract

Many simulators today contain traditional opponents and lack an asymmetric insurgent style adversary. InsurgiSim prototypes an embeddable testbed containing a threat network of agents that one can easily configure and deploy for training and analysis purposes. The insurgent network was constructed inside a socio-cognitive agent framework (FactionSim-PMFserv) that includes: (a) a synthesis of best-of-breed models of personality, culture, values, emotions, stress, social relations, mobilization, as well as (b) an IDE for authoring and managing reusable archetypes and their task-sets (Section 2). Agents and markups in this library are not scripted, and act to follow their values and fulfill their needs. So it's desirable to profile the agents (eg, faction leaders, cell logisticians, followers, bomb maker, financier, recruiter, etc.) as faithfully to the real world as possible. Doing this will improve the utility of InsurgiSim for studying what may be driving the insurgent agents in a given area of operation as Section 3 explains. InsurgiSim's bridge is an HLA federate and can be embedded to drive all or some of the insurgent agents in a 3rd party simulator. Three such examples are summarized in Section 4. The paper closes with next steps to improve InsurgiSim's capabilities and utility. Comments Postprint version. Published in Intelligent Decision Technologies, November 2008. Publisher URL: http://iospress.metapress.com/content/120767/ Author(s) Barry G. Silverman, Aline Normoyle, Praveen Kannan, Richard Pater, Deepthi Chandrasekaran, and Gnana K. Bharathy This journal article is available at ScholarlyCommons: http://repository.upenn.edu/ese_papers/456 Galley Proof 24/10/2008; 13:38 File: idt37.tex; BOKCTP/sx p. 1 Intelligent Decision Technologies 2 (2008) 1–11 1 IOS Press An embeddable testbed for insurgent and terrorist agent theories: InsurgiSim Barry G. Silverman∗, Aline Normoyle, Praveen Kannan, Richard Pater, Deepthi Chandrasekaran and Gnana Bharathy Electrical and Systems Engineering, University of Pennsylvania, Philadelphia, PA 19104-6315, USA Abstract. Many simulators today contain traditional opponents and lack an asymmetric insurgent style adversary. InsurgiSim Many simulators today contain traditional opponents and lack an asymmetric insurgent style adversary. InsurgiSim prototypes an embeddable testbed containing a threat network of agents that one can easily configure and deploy for training and analysis purposes. The insurgent network was constructed inside a socio-cognitive agent framework (FactionSim-PMFserv) that includes: (a) a synthesis of best-of-breed models of personality, culture, values, emotions, stress, social relations, mobilization, as well as (b) an IDE for authoring and managing reusable archetypes and their task-sets (Section 2). Agents and markups in this library are not scripted, and act to follow their values and fulfill their needs. So it’s desirable to profile the agents (eg, faction leaders, cell logisticians, followers, bomb maker, financier, recruiter, etc.) as faithfully to the real world as possible. Doing this will improve the utility of InsurgiSim for studying what may be driving the insurgent agents in a given area of operation as Section 3 explains. InsurgiSim’s bridge is an HLA federate and can be embedded to drive all or some of the insurgent agents in a 3rd party simulator. Three such examples are summarized in Section 4. The paper closes with next steps to improve InsurgiSim’s capabilities and utility.

Extracted Key Phrases

8 Figures and Tables

Cite this paper

@article{Silverman2008AnET, title={An embeddable testbed for insurgent and terrorist agent theories: InsurgiSim}, author={Barry G. Silverman and Aline Normoyle and Praveen Kannan and Richard Pater and Deepthi Chandrasekaran and Gnana Bharathy}, journal={Intelligent Decision Technologies}, year={2008}, volume={2}, pages={193-203} }