IEEE Robotics & Automation Magazine - June 2011 - 20

FROM THE GUEST EDITORS *

A Special Issue Toward a WWW for Robots
van de Molengraft (M.j.G.v.d.Molengraft@tue.nl), Michael Beetz, and Toshio Fukuda
By Rene

he autonomous robotics community is starting to tackle
more complex manipulation
activities, such as doing household chores, in more realistic operating
environments such as human working
and living environments. One of the
biggest challenges in such applications
is the open endedness of the task
domains and the enormous amount of
knowledge needed to achieve success
for the reliable task. Luckily, almost
every robot has access to the World
Wide Web (WWW), the world's largest
knowledge and information source.
Digital Object Identifier 10.1109/MRA.2011.941631
Date of publication: 14 June 2011

IEEE ROBOTICS & AUTOMATION MAGAZINE

JUNE 2011

While up to now robots have only
made limited use of the Web as an
information and knowledge source, the
opportunities are obvious and manifold.
Robots can use search engines such as
Google images, video, or three-dimensional (3-D) warehouse. They can also
ask for help from humans using tools
such as the Amazon Mechanical Turk.
Robots could perhaps form communities to build up large common knowledge bases, much in the same spirit as
people work on Wikipedia. Or, robots
could reproduce the situations they cannot handle in Web games and observe
how the players deal with the situations
and what the consequences might be.
The articles in this issue aim at exploring the route toward a WWW for robots.
Some of the challenging research topics
that are addressed are as follows: dealing
with the large amount of potentially useful
information on the Internet, learning from
and reasoning with this knowledge at a
semantic level, the recognition of objects
and actions and their interaction, and the
distributed nature of robotic systems.
Mozos et al. address the problem of
exploiting the structure in today's designed workplace interiors as an example
for how future object model Web databases can be used by service robots to
add semantics to their sensors' readings
and to build models of their environment. Their method automatically learns
a vocabulary of object parts from CAD
models downloaded from the Web,
where the method is evaluated on real
and synthetic scans of indoor scenes.
Blake et al. introduce both developmental and operational paradigms,
whereas robots can be outfitted with
Web-oriented software interfaces that give
them access to universally standard Web
resources. A case study is performed to
investigate and demonstrate the conversion of traditional robotic data exchange
for communicating with Web services.

Ciocarlie et al. present three data sets
related to mobile manipulation in human
environments. The data sets share a common framework, both in software and
hardware, and are compared and contrasted from multiple points of view,
including data collection tools, annotation methodology, and applications.
Tenorth et al. propose an approach
that allows robots to make use of information from the Web, such as instructions to perform everyday tasks or
descriptions of properties and the appearance of objects. The authors propose techniques to translate the information from
human-readable form-instructions in
natural language, pictures, and 3-D models of objects-into representations the
robot can use, formally described knowledge, task descriptions, and object models
that can be used for recognition.
Waibel et al. present the design
and first implementation of a system
for sharing knowledge between robots,
named RoboEarth. Demonstrators show
how prior knowledge from RoboEarth
can be reused by different robots and
how reasoning can allow a robot to execute a task that was not explicitly
planned at the design time.
Roggen et al. discuss the use of
methods for recognition of human
activities that are commonly used in
wearable computing for the robotics
domain. A WWW for robots will benefit from such activity recognition
approaches in labeling large robot-generated activity data sets, enabling activity-aware human-robot interaction,
and opening the way to self-learning
autonomous robots capable of monitoring their own proficiency at a task.
We sincerely hope that the contents
of this issue will be both informative
and inspiring for the active researchers
in the field as well as for the interested
readers from the IEEE Robotics and
Automation Society community.

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - June 2011