Sponsored by:
and
News:
15-October-2007: The workshop program is available from today!
15-October-2007: Registered participants will receive a hard copy of the workshop proceedings (Supported by NXP Research).
16-August-2007: Depending on the amount of high quality papers, selected workshop papers might be considered for a Robotics & Autonomous Systems journal special issue.
8-August-2007: Due to problems with IROS support the paper submission deadline is extended till 24th of August and the workshop date is changed to 2-November!
Call for papers:
(click here for PDF version)
Invited speakers:
Laurent Itti, University of Southern California,USA
http://ilab.usc.edu/itti/
Dieter Fox, University of Washington, USA
http://www.cs.washington.edu/homes/fox/
Antonio Torralba, MIT CSAIL, Cambridge, USA.
http://web.mit.edu/torralba/www/
Description:
One of the basic robot skills is goal-directed navigation, including localization, path planning and path following. Traditionally mapping and localization has focused on metric environment representations. The environment models were typically described in terms of simple geometric features, such as points, lines and planes. However, it is hard to communicate with a robot in terms of these types of geometric primitives. For natural communication the robot should have an appropriate space representation enabling communication in terms of semantic concepts such as objects, rooms etc.
The aim of the workshop is to bring together researchers that work on space representations appropriate for humans-robot interaction and on development of algorithms for learning various semantic concepts/categories from the robot sensor data.
Suggested topics include, but are not limited to the following areas:
- space representations appropriate for communication with humans including semantic concepts (e.g. objects, rooms,..)
- applications and analysis of the results from cognitive science about the human spatial concepts
- space representations suited to cover cognitive requirements of learning, knowledge acquisition and contextual control
- representations of spatial relations between objects suitable for human-robot interaction