Augmented Reality as a Medium for Human-Robot Interaction

Abstract

Quals talk:

During interactions between humans and robots, it may be difficult for humans to predict a robot's intentions. In recent years, augmented reality (AR) has been increasingly studied as a medium to facilitate human-robot interactions (HRI). AR allows visual augmentations to be rendered over the physical world enabling unique forms of communication, such as displaying future motion intent. In this talk, we will discuss a recently conducted HRI study that used AR to help participants identify failures in a pick-and-place task. We will also discuss an AR framework, DARA, designed to enable configuration changes without requiring the application to be rebuilt and to support visualizations for multiple distinct robotics platforms simultaneously. Additionally, we outline current and future work exploring how AR might be used to aid in interactive machine learning.