Robust Video/Ultrasonic Fusion-Based Estimation for Automotive Applications

In this paper the authors improve object tracking with robust estimation using either stationary or non-stationary cameras for applications in collision avoidance for vehicles utilizing Intelligent Transportation Systems (ITS)-developed technologies. By employing a robust extended Kalman filter (REKF) as well as increasing spatial diversity from multiple cameras, uncertainties in object recognition can be reduced. While optical flow based segmentation techniques for object segmentation provide similar results, the use of normal flow based segmentation provides a faster response time. It is also shown that the presented system provides better results when subjected to large uncertainties in the visual data. By using ultrasound cameras fused with standard video capture and applying REKF as opposed to only extended Kalman filtration (EKF), results were improved substantially.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/00189545
• Authors:
  • Pathirana, Pubudu N
  • Lim, Allan E K
  • Savkin, Andrey V
  • Hodgson, Peter D
• Publication Date: 2007-7

Language

• English

Media Info

• Media Type: Print
• Features: Figures; Photos; References; Tables;
• Pagination: pp 1631-1639
• Serial:
  • IEEE Transactions on Vehicular Technology
  • Volume: 56
  • Issue Number: 4
  • Publisher: Institute of Electrical and Electronics Engineers (IEEE)
  • ISSN: 0018-9545
  • Serial URL: http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=25

Subject/Index Terms

• TRT Terms: Crash avoidance systems; Dynamic models; Kalman filtering; Mathematical models; Optical scanners; Simulation; Video imaging detectors
• Subject Areas: Highways; Vehicles and Equipment; I91: Vehicle Design and Safety;

Filing Info

• Accession Number: 01056125
• Record Type: Publication
• Source Agency: UC Berkeley Transportation Library
• Files: BTRIS, TRIS
• Created Date: Aug 7 2007 12:55PM