Simulation and Multi-agent Environment for Aircraft Learning

  • Authors:
  • Guy Gouardères;Anton Minko;Luc Richard

  • Affiliations:
  • -;-;-

  • Venue:
  • AIMSA '00 Proceedings of the 9th International Conference on Artificial Intelligence: Methodology, Systems, and Applications
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents the earlier results of the CMOS project prototype including an embedded multi-agents ITS (Intelligent Tutoring Systems) aimed to help efficiently the learner who faces troubleshooting maintenance tasks. This environment gives responses dedicated to aeronautical training sessions according to a three-step principle: first to «introduce», second to «convince» and, finally, to get to do. We emphasize two main characteristics: a real-time full simulation of the technical domain, which works with a tutoring multiagent architecture, ASITS. ASITS is supplied with reactive and cognitive agents to track the learner's performance, to detect inherent negative effects (the learner's "cognitive gaps"), and as a feedback issue, to identify some deficiencies that current training simulator lacks. Therefore, as the measuring of gap values with quantitative rules keeps sometimes hazardous, the concept of simulation has to be extended to a Qualitative Simulation approach.