Dr. Sylvain CalinonThe NewScientist article Humanoid robot finds learning child’s play said
Could a robot learn its tasks as easily and as naturally as a child?
Sylvain Calinon and his colleagues at the Swiss Federal Institute of Technology in Lausanne think so. They have created software that allows them to teach a humanoid robot new tasks – such as how to move chess pieces – simply by guiding its limbs through the necessary motions. The 60-centimeter-tall robot, called HOAP-3, is made by Fujitsu.
In humans, parents “mould” children in this way, when teaching them how to hold a pen, for example. “We are taking insights from studies in developmental psychology and humans’ capabilities at transferring skills,” says Calinon.
Sylvain Calinon, Ph.D. is
Research Scientist at the Department of Advanced Robotics,
Italian Institute of Technology (IIT),
where he is working on the
development of robot learning by imitation approaches to handle
human-robot collaborative working scenarios and the development of
systems to handle versatile human-robot interactions.
Sylvain is winner of the 2008 Robotdalen Scientific Award, winner of the 2008 ABB Award, earned first prize in the Grand Challenge on Human-Robot Interaction at the International Conference on Robotics and Automation 2008, and won the Best Paper Award of the 2007 IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), Jeju, Korea. His paper was titled Active Teaching in Robot Programming by Demonstration.
He authored the book Robot Programming by Demonstration Robot Programming by Demonstration and coauthored the papers Robot Programming by Demonstration, Learning of Gestures by Imitation in a Humanoid Robot, Statistical Learning by Imitation of Competing Constraints in Joint Space and Task Space, Dynamical System Modulation for Robot Learning via Kinesthetic Demonstrations, Reinforcement Learning for Imitating Constrained Reaching Movements, What is the Teacher’s Role in Robot Programming by Demonstration? – Toward Benchmarks for Improved Learning, and On Learning, Representing and Generalizing a Task in a Humanoid Robot. Read the full list of his publications!
Sylvain earned his B.Sc. in Microengineering at EPFL, Switzerland in 2001. He earned his M.Sc. in Microengineering, specialization in robotics at EPFL, Switzerland in 2003 with the thesis PDA Interface for Humanoid Robots using Speech and Vision Processing. He earned his Ph.D. in robot programming by demonstration at the Learning Algorithms and Systems Laboratory (LASA), Ecole Polytechnique Federale de Lausanne (EPFL), Switzerland in 2007 with the thesis Continuous Extraction of Task Constraints in a Robot Programming by Demonstration Framework.
Read Robot whips up a snack… Visit his site Programming by Demonstration. Watch The chief cook robot (learning to prepare a cheese’n’ham omelet).