Preliminary Development for Vision-based Human Augmented Mapping: Human Robot Interaction and Human Following

Abstract

Human augmented mapping is a semi-autonomous map building approach for mobile robots. Among many prerequisites for human augmented mapping, human robot interaction and human following are the most fundamental ones, since the main concept is a guided tour. In this paper, we only employ a stereo camera and assume that the user has no background knowledge on robotics so that the robot should provide an easy interaction method. For that, we suggest an intuitive HRI method using virtual buttons and vision-based human following using Haar-like visual features. We implemented these preliminary developments on a real mobile robot. The experimental results show that our system is applicable to human augmented mapping using a stereo camera particularly for naive users.