Vision based techniques for rotorcraft low altitude flightAn overview of research in obstacle detection at NASA Ames Research Center is presented. The research applies techniques from computer vision to automation of rotorcraft navigation. The development of a methodology for detecting the range to obstacles based on the maximum utilization of passive sensors is emphasized. The development of a flight and image data base for verification of vision-based algorithms, and a passive ranging methodology tailored to the needs of helicopter flight are discussed. Preliminary results indicate that it is possible to obtain adequate range estimates except at regions close to the FOE. Closer to the FOE, the error in range increases since the magnitude of the disparity gets smaller, resulting in a low SNR.
Sridhar, Banavar (NASA Ames Research Center Moffett Field, CA, United States)
Suorsa, Ray (NASA Ames Research Center Moffett Field, CA, United States)
Smith, Philip (NASA Ames Research Center Moffett Field, CA, United States)
August 16, 2013
January 1, 1991
Publication: In: Intelligent robotics; Proceedings of the International Symposium, Bangalore, India, Jan. 2-5, 1991 (A93-49350 21-63)