MULTISENSORY FUSION AND NEURAL NETWORKS METHODOLOGY: APPLICATION TO THE ACTIVE SECURITY IN DRIVING BEHAVIOR

This work has as its objective the development of an automobile copilot that informs the pilot of the driving situation in real time. Information from sensors on board the vehicle are used and processed in a multisensory fusion system and learned in a neural networks (NN) system. A new methodology called principal component analyze (PCA) and the NN are presented. The methodology is tested in different situations to characterize the normal or abnormal driving situations. One of the problems of the NN is the difficulty in learning the construction step. This is due to the large number of variables that is resolved by the multisensory fusion, which intends to simplify the working space dimension and then construct the NN using a new algorithm known as baricentric correction procedure. Three different experiments to validate the methodology are proposed to study driver behavior: 1) driver arrives at a cross road, 2) driver in an oval circuit, and 3) driver at stop. In each case the driver is in normal or abnormal situations. A significant advantage of the method lies in the adjustable selection of principal variables. The methodology is used in the driving copilot application and can be used in various industrial process monitoring applications as well.

  • Supplemental Notes:
    • Five volumes of papers and one volume of abstracts comprise the published set of conference materials.
  • Corporate Authors:

    VERTIS

    TORANOMOM 34 MORI BUILDING 1-25-5
    TORANOMON, MINATOKU, TOKYO 105  Japan 
  • Authors:
    • Hernandez-Gress, N
    • Esteve, D
  • Conference:
  • Publication Date: 1995-11

Language

  • English

Media Info

  • Pagination: p. 1146

Subject/Index Terms

Filing Info

  • Accession Number: 00722052
  • Record Type: Publication
  • Report/Paper Numbers: Volume 3
  • Files: TRIS
  • Created Date: Jun 25 1996 12:00AM