Fifth International Conference on Computer Vision (ICCV'95)
Weakly-calibrated stereo perception for rover navigation
Massachusetts Institute of Technology, Cambridge, Massachusetts
June 20-June 23
ISBN: 0-8186-7042-8
L. Robert, Inst. Nat. de Recherche en Inf. et Autom., Sophia Antipolis, France
M. Buffa, Inst. Nat. de Recherche en Inf. et Autom., Sophia Antipolis, France
M. Hebert, Inst. Nat. de Recherche en Inf. et Autom., Sophia Antipolis, France
Presents a vision system for autonomous navigation based on stereo perception without 3D reconstruction. This approach uses weakly calibrated stereo images, i.e. images for which only the epipolar geometry is known. The vision system first rectifies the images, matches selected points between the two images, and then computes the relative elevation of the points relative to a reference plane as well as the images of their projections on this plane. We have integrated this vision module into a complete navigation system. In this system, the relative elevation is used as a shape indicator in order to compute appropriate steering directions everytime a new stereo pair is processed. We have conducted initial experiments in unstructured, outdoor environments with an wheeled rover.
Index Terms:
stereo image processing; computerised navigation; navigation; image matching; mobile robots; path planning; robot vision; calibration; weakly-calibrated stereo perception; vision system; autonomous navigation; stereo perception; rover navigation; weakly calibrated stereo images; epipolar geometry; image rectification; point matching; relative point elevation; reference plane; projection images; vision module; complete navigation system; shape indicator; appropriate steering directions; unstructured outdoor environments; wheeled rover
Citation:
L. Robert, M. Buffa, M. Hebert, "Weakly-calibrated stereo perception for rover navigation," iccv, pp.46, Fifth International Conference on Computer Vision (ICCV'95), 1995