Analysis of Virtual Agent Communities by Means of AI Techniques and Visualization

  • David Kadleček
  • David Řehoř
  • Pavel Nahodil
  • Pavel Slavík
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2792)

Abstract

The analysis of virtual agent communities combining advanced methods of visualization with traditional AI techniques is presented in this paper. As Multi-Agent Systems (MAS) are becoming more complex and thereby harder to analyze and assess, traditional methods of analysis are becoming insufficient. The main purpose of this research is to develop new methods improving the efficiency of analysis. Extensive assessment on different levels is necessary to analyze MAS and our approach covers from single agent to overall properties of a community. The most substantial merits of this research lie in the utilization of sensitivity analysis together with clustering methods. A simulated artificial life environment with intelligent agents has been used as test bed. We have selected this domain because our long-term goal is to model life as it could be so as to understand life as we know it. However our proposed techniques can be used across multiple domains with intelligent virtual agents.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Allbeck, J., Badler, N.: Toward Representing Agent Behaviors Modified by Personality and Emotion. In: Workshop on Embodied Conversational Agents, Bologna, Italy (2002)Google Scholar
  2. 2.
    Arkin, R.C.: Behavior based robotics. The MIT Press, Mass (1999)Google Scholar
  3. 3.
    Beneš, B., Espinosa, E.: Modeling Virtual Ecosystems With Proactive Guidance of Agents. To appear In IEEE Computer Animation and Social Agents, New Bruncwick, USA (2003)Google Scholar
  4. 4.
    Havre, S., Hetzler, E., et al.: ThemeRiver: Visualizing Thematic Changes in Large Document Collections. IEEE Trans. on Visualization and Computer Graphics 8(1) (2002)Google Scholar
  5. 5.
    Inselberg, A., Dimsdale, B.: Parallel Coordinates: A Tool for Visualizing Multivariate Relations. Plenum Publishing Corporation, New York (1991)Google Scholar
  6. 6.
    Kadleček, D., Nahodil, P.: CZAR: Hybrid Agent Architecture for Animal-Like Robots. In: Proc. European Symp. on Intelligent Technologies, Albufeira, Portugal, pp. 141-152 (2002)Google Scholar
  7. 7.
    Kadleček, D., Nahodil, P.: New hybrid architecture in artificial life simulation. In: Kelemen, J., Sosík, P. (eds.) ECAL 2001. LNCS (LNAI), vol. 2159, pp. 143–146. Springer, Heidelberg (2001)CrossRefGoogle Scholar
  8. 8.
    Leslie, M., Moore, B., Ray, K.: Statistical methods for sensitivity and performance analysis in computer experiments. In: Proc. Winter Simulation Conf., Phoenix, USA, pp. 486-491 (1999)Google Scholar
  9. 9.
    Nahodil, P., Petrus, M.: Behavior Coordination in Multi Robot Group. In: Proc. of IASTED International Conf. MIC 2002, Innsbruck, pp. 464–468. ACTA Press, Calgary (2002)Google Scholar
  10. 10.
    Řehoř, D., Kadleček, D., Slavík, P., Nahodil, P.: VAT – an approach to Multi-Agent Sestem Vizualization. To appear in IASTED International Conference on Visualization, Imaging and Image Processing, Benalmadena, Spain (2003)Google Scholar
  11. 11.
    Schroeder, M., Noy, P.: Multiagent visualisation based on multivariate data. In: Proc. 5th International Conf. on Autonomous Agents, Montreal, Canada, pp. 85-91 (2001)Google Scholar
  12. 12.
    Szekely, P., Rogers, C., Abrams, M.: Interfaces for understanding multi-agent behavior. In: Proc. ACM Conf. on Intelligent User Interfaces, Santa Fe New, Mexico, USA, pp. 161–166 (2001)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • David Kadleček
    • 1
  • David Řehoř
    • 2
  • Pavel Nahodil
    • 1
  • Pavel Slavík
    • 2
  1. 1.Dept. of CyberneticsCzech Technical University in PraguePrague 2Czech Republic
  2. 2.Dept. of Computer Science and EngineeringFEE, Czech Technical University in PraguePrague 2Czech Republic

Personalised recommendations