Affordance-Based intention recognition in virtual spatial environments

  • Authors:
  • Michal Sindlar;John-Jules Meyer

  • Affiliations:
  • University of Utrecht, The Netherlands;University of Utrecht, The Netherlands

  • Venue:
  • PRIMA'10 Proceedings of the 13th international conference on Principles and Practice of Multi-Agent Systems
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

In applications for entertainment or training, behavior of characters often takes place in virtual environments with spatial dimensions that incorporate both agents and objects. Situated virtual characters can employ knowledge of their environment in reasoning about the goals or intentions of other characters, virtual or human, contributing to their believability. This paper presents lightweight techniques to that extent, utilizing object affordances and observed behavior of characters to define an observer reasoning about that behavior. In case an observer reasons about the behavior of virtual autonomous agents, knowledge of behavior-producing rules can also be employed. This is formalized by extending earlier work on mental state abduction of BDI-based agents with the techniques presented here. The presentation is technical, illustrated with practical examples.