The iCub IROS '10 workshop
Giorgio Metta(1), Angelo Cangelosi(2), Giulio Sandini(3)
(1) Robotics, Brain and Cognitive Science Department, Italian Institute of Technology (IIT), Via Morego, 30 16163 Genova, Italy. Ph: +39 010 71781411, email: giorgio.metta_(at)_iit.it
(2) Centre for Robotics and Neural Systems, University of Plymouth, Drake Circus, Plymouth, Devon PL4 8AA, United Kingdom. Ph: +44 1752 586217, email: acangelosi_(at)_plymouth.ac.uk
(3) Robotics, Brain and Cognitive Science Department, Italian Institute of Technology (IIT), Via Morego, 30 16163 Genova, Italy. Ph: +39 010 71781416, email: giulio.sandini_(at)_iit.it
When and where
The workshop is held at IROS 2010 in Taipei on October 18th, 2010. This is a full day workshop (number W02-MoF in the conference program).
What is the iCub
The iCub is a humanoid robot shaped as a child of about four years of age available as an open systems platform following the GPL license. The iCub was originally designed by a consortium of 11 partners, with background ranging from engineering to neurophysiology and developmental psychology, within the RobotCub Integrated Project funded by European Commission through its Cognitive Systems and Robotics Unit . The iCub can crawl on all fours and sit up. Its hands allow dexterous manipulation and its head and eyes are fully articulated. It has visual, vestibular, auditory, and haptic sensory capabilities.
To facilitate the adoption of iCub as an open research platform, the RobotCub project launched a call for research proposals in the area of embodied cognition which resulted in the selection of seven winners that received a complete iCub free-of-charge. More recently other groups joined this initiative by building their own iCubs, e.g. the ITALK, ImClever, Roboskin, CHRIS, Poeticon EU funded projects to name a few.
This workshop is the first cohesive presentation of iCub-related activities ranging from software development, control, vision, cognition and hardware, sensors and so forth.
Human intelligence develops through the interaction with the environment and it is shaped profoundly by the interactions with other human beings. These concepts altered in a fundamental way the relation between fields such as neuroscience, information science, psychology, and robotics which are now learning from each other and together are building a new interdisciplinary science. As such, the study of artificial cognition and intelligence is dependent on the use of physical bodies, and, ultimately, on the use of humanoid robots. Humanoids are essential tools in the study of human intelligence: since the development of cognitive skills is function of both neural and physical activity, it is important to simulate human action when investigating learning, adaptation, anticipation, reasoning, memory, communication, and language to name a few.
The RobotCub humanoid platform represents an important opportunity to move this research agenda forward: through open collaboration, on the common theme of embodied cognition, enabled by a shared humanoid platform – the iCub –, and supported by a community capable of delivering on this vision in the long run. The availability of a common platform, such as the iCub, allows benchmarking and testing theories by comparing results obtained using different method and, more importantly, it enables the accumulation of results into a common “repository” shared among scientists. This paves the way to a more scientific approach to intelligence for robots by enabling researchers worldwide to access the same experimental facilities.
Our goal is to present a broad overview of these activities which have now reached a consistent critical mass.
- To show and demonstrate the iCub humanoid platform, including a vision of the hardware and software
- To showcase the strength of collaborative research using a common platform
- To show that open source development can help in building a critical mass of researchers in the important topic of robot intelligence
List of topics
- Introduction to the iCub hardware and software platform
- Cognitive systems research on humanoid robots
- Robot control
- Manipulation and motor affordances
- Speech and language
- Human-robot interaction
List of speakers
- Giorgio Metta, Italian Institute of Technology (Italy)
- The iCub project: www.icub.org
- Lorenzo Natale, Italian Institute of Technology (Italy)
- YARP middleware for the iCub
- Angelo Cangelosi, University of Plymouth (United Kingdom)
- Teaching the iCub to speak: the Modi experiment
- Peter Ford Dominey, INSERM (France)
- Towards Multi-Partner Integration on iCub: A Case Study from the CHRIS project
- Brenna Argall & Aude Billard, EPFL (Switzerland)
- Demonstration Learning and Adaptation with the iCub: Perturbation Recovery and Tactile Corrections
- Philippe Bidaud, Joseph Salini & Vincent Padois, EPMC (France)
- iCub@Paris: Whole Body control and a focus on the MACSi project
- Sebastien Gay, Sarah Degallier, Ludovic Righetti & Auke Ijspeert, EPFL (Switzerland)
- A bio-inspired robotic controller for discrete and rhythmic movements
- Jochen Steil & Britta Wrede, Univ. of Bielefeld (Germany)
- Neural Dynamic Motion Primitives for iCub: a contribution from the AMARSi project
- Lydia Majure, UI Urbana Champaign (USA)
- Integrating Language and Motor Function on a Humanoid Robot
- Stefano Nolfi, CNR (Italy)
- Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation
- Yiannis Aloimonos, Univ. of Maryland (USA) & Katerina Pastra, ILSP, Athens (Greece)
- The Poeticon project & the praxicon
- Alexandre Bernardino & Jose’ Santos-Victor, IST Lisbon (Portugal)
- Object Detection and Tracking in the iCub
- Nikos Tsagarakis, Italian Institute of Technology (Italy)
- On the Mechatronic Implementations of the AMARSI Compliant Humanoid Platform
- Gustavo Medrano-Cerda, Martin Brown & Houman Dallali, University of Manchester (UK)
- A hybrid control approach for walking robots
- Michael Beetz, TUM (Germany)
- iCub @ TUM: Playing with legos
- Jean-Christophe Baillie, Gostai (France)
- Programming the iCub with Urbi Open Source and YARP 
This workshop has a broad audience by showcasing cutting edge humanoid robotics. In particular, it can be informative for researchers interested in joining the iCub community or more in general interested in open source development.
Tentative schedule and agenda
|Time||Speaker and title||Notes|
|9.00-9.25||Giorgio Metta: Welcome and introduction to the iCub||--|
|9.25-9.50||Lorenzo Natale: YARP middleware for the iCub||--|
|9.50-10.15||Peter Ford Dominey: Towards Multi-Partner Integration on iCub: A Case Study from the CHRIS project||--|
|10.15-10.40||Angelo Cangelosi: Teaching the iCub to speak: the Modi experiment||--|
|11.10-11.35||Brenna Argall: Demonstration Learning and Adaptation with the iCub: Perturbation Recovery and Tactile Corrections||--|
|11.35-12.00||Philippe Bidaud: iCub@Paris: Whole Body control and a focus on the MACSi project||--|
|12.00-12.35||Jochen Steil: Neural Dynamic Motion Primitives for iCub: a contribution from the AMARSi project||--|
|14.00-14.25||Sebastien Gay: A bio-inspired robotic controller for discrete and rhythmic movements||--|
|14.25-14.50||Lydia Majure: Integrating Language and Motor Function on a Humanoid Robot||--|
|14.50-15.15||Stefano Nolfi: Co-development of linguistic and behavioural skills: compositional semantics and behaviour generalisation||--|
|15.15-15.40||Yiannis Aloimonos: The Poeticon project & the praxicon||--|
|16.10-16.35||Alexandre Bernardino: Object Detection and Tracking in the iCub||--|
|16.35-17.00||Nikos Tsagarakis: On the Mechatronic Implementations of the AMARSI Compliant Humanoid Platform||--|
|17.00-17.25||Gustavo Medrano-Cerda: A hybrid control approach for walking robots||--|
|17.25-17.50||Michael Beetz: iCub @ TUM: Playing with legos||--|
|17.50-18.30||Concluding remarks, discussion about a special issue, next workshop at Humanoids 2010|
- Borisyuk R., Kazanovich Y., Chik D., Tikhanoff V. & Cangelosi A. (2009). A neural model of selective attention and object segmentation in the visual scene: An approach based on partial synchronization and star-like architecture of connections. Neural Networks. 22, pp. 707-719.
- Fitzpatrick, P., Metta, G., Natale,L. Towards long lived genes. Robotics and Autonomous systems, 2008. 56(6): p. 29-45.
- Metta, G., Fitzpatrick, P., Natale, L. YARP: Yet Another Robot Platform. International Journal of Advanced Robotics Systems, special issue on Software Development and Integration in Robotics, 2006. 3(1): p. 43-48.
-  Nosengo N.. "Robotics: The bot that plays ball" Nature Vol 460, 1076-1078 (2009) doi:10.1038/4601076a.
- Tikhanoff V., Cangelosi A., Fitzpatrick P., Metta G., Natale L. & Nori F. (2008). An open-source simulator for cognitive robotics research: The prototype of the iCub humanoid robot simulator. R. Madhavan & E.R. Messina (Eds.), Proceedings of IEEE Workshop on Performance Metrics for Intelligent Systems Workshop (PerMIS’08). Washington, D.C.
|<youtube v="JRqdIFCIZd8" />||<youtube v="USCKIiwareU" />|
|<youtube v="sUErJodlPtQ" />||<youtube v="x9-u6OGo0zA" />|