Distributed Mobile Multi Sensor Systems for Localization and Cooperated Scene Recognition

Since several years, the automotive research activities deal with the task of localization of the own vehicle as well as the detection of obstacles in the vehicle environment. A high precision of such systems is only reachable by using multiple sensors for localization and obstacle detection. An optimal combination of sensors is not only dependent on compatibility/complementarities according to sensor principles and according to the kind of attributes that are detected but also dependent on the several sensor positions relative to the scene they are observing together. The combination of data coming from several vehicles is a challenging task in Sensor Data Fusion because the sensor positions of the cooperating vehicles are changing over time which increases the advantages reachable over time. The paper describes a system that uses the new upcoming communication facilities (WLAN, UMTS) to transmit the local obstacle detection result to a cooperating vehicle equipped with the same obstacle detection system and combines the own and the cooperating vehicle detections to a common scene recognition result. As the possibility of fusing the detection results is fundamentally based on a precise localization an approach is described that combines GPS with maps and an image based detection of the road. The paper addresses especially the precision of the data fusion results that are reached by the cooperation.

  • Corporate Authors:

    ITS America

    1100 17th Street, NW, 12th Floor
    Washington, DC  United States  20036
  • Authors:
    • Cramer, Heiko
    • Scheunert, Ullrich
    • Fardi, Basel
    • Wanielik, Gerd
  • Conference:
  • Publication Date: 2005


  • English

Media Info

  • Media Type: Print
  • Features: CD-ROM; Figures; References;
  • Pagination: 12p
  • Monograph Title: Proceedings of the 12th World Congress on Intelligent Transport Systems

Subject/Index Terms

Filing Info

  • Accession Number: 01016262
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Dec 16 2005 9:35AM