Off-Road Obstacle Detection with Robust Parametric Modelling of the Ground Stereo Geometry

S. Kodagoda, G. Dong, C.H. Yan, and S.H. Ong (Singapore)


Autonomous vehicles, Stereo vision, Scene analysis,Geometric modeling, Hough transforms, Piecewise linearapproximation


Autonomous navigation in off-road environments presents many challenges in contrast to the more conventional, urban environments. Unstructured surroundings, non-uniform visual cues and lack of prior knowledge about the scene complicate the design of even basic functionalities such as obstacle detection. This paper presents a stereo vision based ground geometry modeling and obstacle detection algorithm that is well suited for cross-country navigation. Our mathematical analysis shows that the “v-disparity” method is inadequate for accurate terrain modeling under vehicle pose variations; to compensate for this shortcoming, we propose a novel extension to the original algorithm. As the preliminary step of this extension, lateral gradient of the ground disparity is estimated using histogram analysis. This information is subsequently propagated to a modified “v-disparity” algorithm that models the longitudinal terrain disparity variation. The effectiveness of this two-phase ground modeling technique for obstacle detection is demonstrated with empirical results.

Important Links:

Go Back