Modeling Driving Behavior With Dynamic Bayesian Networks And Estimate Of Mental State

Although number of accidents cased by human error is still rising, most safety technologies have been focusing on vehicle components rather than human itself. Driver monitoring systems are indispensable technologies to prevent human-caused traffic accidents. Those accidents are frequently triggered by drivers’ unusual mental states such as drowsiness or fatigue. In this paper, we aimed to detect “hastiness” on drivers by using a biological signal. The detecting result was employed as a parameter for modeling driving behavior along with Dynamic Bayesian Networks. The recognition result of driving behavior in this study demonstrated that adopting drivers’ mental states is effective for accurate modeling.

Language

  • English

Media Info

  • Media Type: CD-ROM
  • Features: Figures; Photos; References;
  • Pagination: 7p
  • Monograph Title: ITS Connections: Saving Time. Saving Lives

Subject/Index Terms

Filing Info

  • Accession Number: 01144552
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Nov 10 2009 10:54AM