Robocup Symposium Talk 2

Tuesday, July 19, 16:40-17:30 (International Conference Hall)

Developmental Robotics and Robot Development

Giulio Sandini

LIRA-Lab, University of Genova, Italy

Abstract
In this talk I will stress two points: one scientific, the other methodological. The scientific point is the role of embodiment in the study of cognitive development and the implementation of cognitive robots. The methodological aspects are aimed at emphasizing the need to join forces to design and build robots that can be used as "bodies" to study cognition and implement cognitive systems. Human cognition is based on our abilities to use the hands (and to some extent our all body) to manipulate objects and communicate through gestures. Human ability to "invent" new uses of our hands, for example, to use new tools, grasp in unusual ways familiar objects, discover new uses of objects, resembles our ability to express new concepts with words and it is, in our view, a distinctive feature of human cognition. For this reason, from the methodological point of view, we believe that any attempt to implement "human-like" cognitive abilities in artificial systems requires the use of "bodies" with dexterous manipulation abilities. The problem with this is that "hands and arms" are complex mechatronic devices and only few centres in the world with a sufficient mix of mechanical, electronic and computer science knowledge can afford to design and build such complex systems. In many cases, most of the effort is devoted to the development of the robot hardware, leaving the study of cognitive development to a later stage. What we believe is that cognition poses such fundamental problems that without a major effort in joining forces it will progress very slowly in the next 10-15 years. One of the goals of this talk will be to stress the need for (and the specifications of) a cognitive research robot that can be used by research centres worldwide to incrementally build and share our knowledge about embodied cognition.

References

• Sandini, G., G. Metta, and J. Konczak. Human Sensori-motor Development and Artificial Systems. in International Symposium on Artificial Intelligence, Robotics and Intellectual Human Activity Support(AIR&IHAS '97). 1997. RIKEN - Japan.
• Metta, G., G. Sandini, and J. Konczak, A Developmental Approach to Visually-Guided Reaching in Artificial Systems. Neural Networks, 1999. 12(10): p. 1413-1427
• Atkeson, C. G., Hale, J. G., Pollick, F., Riley, M., Kotosaka, S., Schaal, S., et al. (2000).Using Humanoid Robots to Study Human Behavior. IEEE Intelligent Systems, 46-56.
• Metta, G., & Fitzpatrick, P. (2003). Early Integration of Vision and Manipulation. Adaptive Behavior, 11(2), 109-128.
• Sandini, G., G. Metta, and D. Vernon. RobotCub: An Open Framework for Research in Embodied Cognition. in IEEE-RAS/RJS International Conference on Humanoid Robotics. 2004. Santa Monica, CA: IEEE.
• Asada, M., et al., Cognitive Developmental Robotics As a New Paradigm for the Design of Humanoid Robots. Robotics and Autonomous Systems, 2001. 37: p. 185-193.
• Kozima, H., Infanoid: A Babybot that Explores the Social Environment, in Socially Intelligent Agents: Creating Relationships with Computers and Robots, K. Dautenhan, et al., Editors. 2002, Kluwer Academic Publishers: Amsterdam. p. 157-164.
• Weng, J., et al., Autonomous mental development by robots and animals. Science, 2001. 291: p. 599-600.

Acknowledgements
Research described here has been supported by the projects MIRROR and ADAPT of the EU and will be carried out within the project RobotCub (www.robotcub.org)

Biography
Education

Academic Appointments

Back