Visual Registration Method for a Low Cost Robot

Abstract

An autonomous mobile robot must face the correspondence or data association problem in order to carry out tasks like place recognition or unknown environment mapping. In order to put into correspondence two maps, most correspondence methods first extract early features from robot sensor data, then matches between features are searched and finally the transformation that relates the maps is estimated from such matches. However, finding explicit matches between features is a challenging and computationally expensive task. In this paper, we propose a new method to align obstacle maps without searching explicit matches between features. The maps are obtained from a stereo pair. Then, we use a vocabulary tree approach to identify putative corresponding maps followed by a Newton minimization algorithm to find the transformation that relates both maps. The proposed method is evaluated on a typical office dataset showing good performance.