MODELING THE GROUND PLANE TRANSFORMATION FOR REAL-TIME OBSTACLE DETECTION

Ground plane obstacle detection is a general problem for mobile robots . A number of different approaches have been investigated in previous research; however none is suitable for use with an active stereo visio n system. In this paper, we present a realtime approach to obstacle de tection for an active stereo vision system based on plane transformati on. By modeling the ground plane transformation between a stereo image pair captured by a common elevation stereo head, then identifying the related camera intrinsic parameters, we demonstrate that the ground p lane transformation can be computed in real-time using the identified parameters and the head feedback state. Hence it is possible to use it to detect ground plane obstacles in real-time. We also show that this formulation of ground plane obstacle detection unifies previous appro aches based on predicted ground plane disparities or on a predicted gr ound plane transformation for stereo cameras with fixed geometry. (C) 1998 Academic Press.