Driver models based on lateral dynamics for adaptation of assistance systems

The continuous demand to increase the road safety induces that the driver in the vehicle development and research being taken increasingly in the focus. An important focus of research is to determine the driver's ability to respond to situations appropriately and thus to drive the vehicle safely in traffic. At the Institute of Automotive Engineering (IAE) of the Technical University of Braunschweig, a driver model has been developed, which makes it possible to identify the specific driver characteristic in the form of an individual "fingerprint". By means of the knowledge of the "fingerprint" conclusions can be drawn on: the driving style, the driver's age/ driver control behaviour and the driver performance. Within the context of this paper, the relevant quantities are derived by means of a driver model that serves the adaptation of driver assistance systems. Furthermore, the existing correlations between the driver's performance and the driver control behaviour will be analysed as part of the driver's fingerprint. For this purpose, relevant vehicle and track camera signals are extracted from the CAN bus and used for the driver modelling. Genetic algorithm is used for the identification and optimization of the so-called "time-shifted driver model" control parameters. The brief presentation of a methodology used for calculating the individual driver condition parameter sets in real time on the test vehicle constitutes the conclusion of this article.

Language

  • English

Media Info

  • Pagination: pp 509-513
  • Monograph Title: FAST-zero'15: 3rd international symposium on future active safety technology toward zero traffic accidents: September 9-11, 2015 Gothenburg, Sweden: proceedings

Subject/Index Terms

Filing Info

  • Accession Number: 01602330
  • Record Type: Publication
  • Source Agency: Swedish National Road and Transport Research Institute (VTI)
  • Files: ITRD, VTI
  • Created Date: Jun 20 2016 1:29PM