Arrow Research search
Back to AAMAS

AAMAS 2018

Benchmark Framework for Virtual Students' Behaviours

Conference Paper Socially Interactive Agents Track Extended Abstracts Autonomous Agents and Multiagent Systems

Abstract

This paper demonstrates the integration and evaluation of different atmosphere models into Virtual Reality (VR) training for teacher education. We developed three behaviour models to simulate different levels of class discipline. We evaluated their performances using a combination of objective and subjective measurements. Our initial results suggest that the more believable and distinguishable classroom atmospheres are produced by creating more consistent behaviours across virtual students. Our results confirm the importance of similar behaviours to elicit a particular atmosphere.

Authors

Keywords

  • Virtual Training
  • Agents Behaviour Generation
  • Benchmarking

Context

Venue
International Conference on Autonomous Agents and Multiagent Systems
Archive span
2002-2026
Indexed papers
8043
Paper id
248741904276387201
v2026.09.13