04/27/07
Embodied Artificial Intelligence: Connecting Body, Brain and Environment
Presented by Fumiya Iida

Traditionally, in robotics, artificial intelligence and neuroscience, there has been a focus on the study of the control or the neural system itself. Recently there has been an increasing interest in the notion of embodiment not only in robotics and artificial intelligence, but also in the neurosciences, psychology and philosophy. In this talk, I introduce the notions of embodied artificial intelligence, and demonstrate how it can be exploited on the one hand for designing intelligent, adaptive robotic systems, and on the other hand for understanding natural systems. In the second half of my talk, we discuss how the notions of embodied artificial intelligence can actually improve the designs and behavioral performance of robotic systems. With the case studies of a number of locomotion robots that we developed during the last few years, I will demonstrate how rapid dynamic behavior (e.g. dancing, walking, hopping, running and swimming) can be achieved by using "cheap" control architectures. We will then discuss how the dynamics of embodied systems influence the motor control through learning processes.

Fumiya Iida received his MSc. of mechanical engineering at Tokyo University of Science (Japan), and Ph.D. at the Artificial Intelligence Laboratory of University of Zurich (Switzerland, 2005). From 2004 to 2005, he was also engaged in biomechanics of human locomotion at Locomotion Laboratory, University of Jena (Germany). He is currently a postdoctoral fellow at the Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology. He has been involved in a number of research projects related to embodied artificial intelligence and robotics, and his research interest includes biologically inspired robotics, navigation of autonomous robots, and dynamic legged locomotion.