Obstacle Detection for Unmanned Ground Vehicles: A Progress Report

  • Larry Matthies
  • Alonzo Kelly
  • Todd Litwin
  • Greg Tharp


To detect obstacles during off-road autonomous navigation, unmanned ground vehicles (UGV’s) must sense terrain geometry and composition (ie. terrain type) under day, night, and low-visibility conditions. To sense terrain geometry, we have developed a real-time stereo vision system that uses a Datacube MV-200 and a 68040 CPU board to produce 256 × 45-pixel range images in about 0.6 seconds/frame. To sense terrain type, we are using the same computing hardware with red and near-infrared imagery to classify 256 × 240-pixel images as a rate of 10 frames/second. This paper reviews the rationale behind the choice of these sensors, describes their recent evolution and on-going development, and summarizes their use in demonstrations of autonomous UGV navigation over the past five years. This work has been the first to show that stereo vision can be practical for autonomous UGV navigation, and is now the first to show a real-time terrain classification system with very low computing requirements.


Stereo Vision Autonomous Navigation Obstacle Detection Unmanned Ground Vehicle Terrain Type 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    L. H. Matthies. Stereo vision for planetary rovers: stochastic modeling to near real-time implementation. International Journal of Computer Vision, 8(1):71 -91, July 1992.Google Scholar
  2. [2]
    L. H. Matthies and P. Grandjean. Stochastic performance modeling and evaluation of obstacle detectability with imaging range sensors. IEEE Transactions on Robotics and Automation, 10 (6): 783 - 791, December 1994.CrossRefGoogle Scholar
  3. [3]
    M. Drumheller and T. Poggio. On parallel stereo. In Proc. IEEE Conf. on Robotics and Automation, pages 1439-1448. IEEE, 1986.Google Scholar
  4. [4]
    P. Fua. A parallel stereo algorithm that produces dense depth maps and preserves image features. Technical Report No. 1369, INRIA, January 1991.Google Scholar
  5. [5]
    D. H. Ballard and C. M. Brown. Computer Vision. Prentice-Hall, Englewood Cliffs, NJ, 1982.Google Scholar
  6. [6]
    M. Daily et al. Autonomous cross-country navigation with the ALV. In Proc. IEEE Int’l Conf on Robotics and Automation, pages 718 - 726. IEEE Computer Society, April 1988.Google Scholar
  7. [7]
    P. Grandjean and L. Matthies. Perception control for obstacle detection by a cross-country rover. In Proc. IEEE Int’l Conf Robotics and Automation, May 1993.Google Scholar
  8. [8]
    C. Elachi. Introduction to the Physics and Techniques of Remote Sensing. John Wiley and Sons, New York, NY, 1987.Google Scholar
  9. [9]
    A. Kelly, A. Stentz, and M. Hebert. Terrain map building for fast navigation on rugged outdoor terrain. In Proceedings SPIE Conference on Mobile Robots. SPIE, November 1992.Google Scholar
  10. [10]
    D. Langer, J. K. Rosenblatt, and M. Hebert. An integrated system for autonomous off-road navigation. In Proc. 1994 IEEE Int’l Conf. on Robotics and Automation, pages 414 - 419, May 1994.Google Scholar

Copyright information

© Springer-Verlag London Limited 1996

Authors and Affiliations

  • Larry Matthies
    • 1
  • Alonzo Kelly
    • 1
  • Todd Litwin
    • 1
  • Greg Tharp
    • 1
  1. 1.Jet Propulsion LaboratoryPasadenaUSA

Personalised recommendations