Selecting stable image features for robot localization using stereo

Abstract
To navigate and recognize where it is, a mobile robot must be able to identify its current location. In an un- known initial position, a robot needs to refer to its environment to determine its location in an external coordinate system. Even with a known initial position, drift in odometry causes the estimated position to de- viate from the correct position, requiring correction. We show how to nd landmarks without models. We use dense stereo data from our mobile robot's trinocular system to discover image regions that will be stable over widely diering viewpoints. We nd image brightness \corners" in images and select those that do not straddle depth discontinuities in the stereo depth data. Selecting corners only in regions of nearly pla- nar stereo data results in landmarks that can be seen in images taken from dierent viewpoints.

This publication has 14 references indexed in Scilit: