You are here

Developing a Cloud Robotics Infrastructure for a Humanoid Robotic Teaching Assistant

Abstract

Robotics is an area of intense interest currently, with humanoid robots emerging that can mimic the behaviour of humans much more accurately than ever before. This provides an opportunity to conduct research on how robots might help individuals with robotic assistance and skill development in areas as diverse as business, information technology, and education.

As the 21st century moves on, consumer grade, pre-built anthropomorphic robots designed to complete human-like tasks are becoming more common. For instance, Nao (pronounced now) is an autonomous, programmable humanoid robot developed by Aldebaran Robotics, a French company headquartered in Paris. Nao robots have been used for research and education purposes in numerous academic institutions worldwide and, as of 2015, over 7,000 Nao units are in use in more than 50 countries, according to Aldebaran (see https://www.ald.softbankrobotics.com/).

Whilst many projects with Nao are conducted locally, the field of Cloud Robotics is growing and could be applied to Nao (and eventually to other robots) for robotic assistance purposes. Cloud computing technologies enable robot systems to be endowed with powerful capability whilst reducing costs by offloading processing to computers in the cloud. Humans can also delegate tasks to robots remotely through networks, and Artificial Intelligence in the cloud can be applied to enhance the autonomous actions of the robot. This provides an excellent platform to develop a cloud-driven, autonomous robotic assistance model applicable in a variety of different areas.

For instance, with the recent announcement by the Queensland State Government that robotics should become core in all classrooms in 2016 (see http://advancingeducation.qld.gov.au), consideration needs to be given on how this technology might fit into the classroom. One option is that robotics be taught as part of the curriculum, but what if robotics could be used more directly to support teaching activities in the classroom? Specifically, what if a robot could be used to provide direct assistance to the teacher, serving as both a proxy for students as well as a teacher’s aide? This proposed research project will look to answer these research questions and other questions relating to the use of a robot as a semi-autonomous robotic assistant driven by a cloud infrastructure.

This paper begins this work by describing a prototype system demonstrating robot network connection functionality and a cloud based architecture for robotic control. This infrastructure can be built on to enable artificial intelligence interactions with robots in a variety of settings, such as assistant positions in corporations, or within an educational setting. Through experimental testing, this infrastructure will be assessed for its affect in the classroom and on the use of a robot as a teaching assistant.

The proposed outcome of the work is a greater understanding of how robotics can be used to support a general set of classroom activities, beyond those conducted in the traditional computer science classroom.