A 3D World Model Builder with a Mobile Robot

摘要

This article describes a system to incrementally build a world model with a mobile robot in an unknown environment. The model is, for the moment, segment based. A trinocular stereo system is used to build a local map about the environment. A global map is obtained by integrating a sequence of stereo frames taken when the robot navigates in the environment. The emphasis of this article is on the representation of the uncer tainty of 3D segments from stereo and on the integration of segments from multiple views. The proposed representation is simple and very convenient to characterize the uncertainty of segment. A Kalman filter is used to merge matched line seg ments. An important characteristic of our integration strategy is that a segment observed by the stereo system corresponds only to one part of the segment in space, so the union of the different observations gives a better estimate on the segment in space. We have succeeded in integrating 35 stereo frames taken in our robot room.