Automatic mapping of dynamic office environments

Abstract

We present a robot, called InductoBeast, that greets a new office building by learning the floorplan automatically, with minimal human intervention and a priori knowledge. Our robot architecture is unique because it combines both abductive and inductive mapping methods to solve this problem. We present experimental results spanning three office environments. We hope the breadth of these results helps to establish a performance benchmark against which robust and adaptive navigator robots of the future may be measured.