Robot Genome Project
Event details
| Date | 22.09.2015 |
| Hour | 15:00 |
| Speaker | Prof. Fumiya Iida, University of Cambridge (UK) |
| Location | |
| Category | Conferences - Seminars |
An approach of artificial evolution to morphologically improve our physical robots
The concepts and sprits of embodied artificial intelligence have been around for the last decades and demonstrated the far reaching implications for our understanding of the nature of autonomy and adaptability through the countless case studies. It is however not fully clarified methodologies with which the different aspects of embodiment can be coherently and consistently integrated into one entity of autonomous agent. In the last few years, we have been developing an approach in which embodied agents could be studied in the level of genetic encoding, rather than that of behaviors. The framework consists of a mother robot (a robotic manipulator) that is capable of handling various materials and robotic modules assembled into a variety of child robots and evaluating the behavioral performances of child robots in order to progressively improve designs over generations. Unlike the conventional evolutionary robotics research, our approach does not use mathematical modeling in the evolutionary process, we are able to explore how embodied system-environment interactions influence the dynamics of physical robot evolution by observing the mother robot “inventing” surprising solutions to the given task. By introducing our recent experiments of building over 700 individual physical robots, we will discuss the implications and challenges of our project.
Bio: Fumiya Iida received his bachelor and master degrees in mechanical engineering at Tokyo University of Science (Japan, 1999), and Dr. sc. nat. in Informatics at University of Zurich (2006). In 2004 and 2005, he was also engaged in biomechanics research of human locomotion at Locomotion Laboratory, University of Jena (Germany). From 2006 to 2009, he worked as a postdoctoral associate at the Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology in USA. In 2006, he awarded the Fellowship for Prospective Researchers from the Swiss National Science Foundation, and in 2009, he was appointed as a Swiss National Science Foundation Professor for bio-inspired robotics at ETH Zurich. His research interests include biologically inspired robotics, embodied artificial intelligence, and biomechanics, and he has been involved in a number of research projects related to dynamic legged locomotion, navigation of autonomous robots, and human-machine interactions. He has so far published over forty publications in major robotics journals and conferences, and edited two books. Currently he serves on the editorial board of the Journal of Intelligent & Robotic Systems and Frontiers in Neuroscience (Neurorobotics), and as a program committee member for international conferences and workshops. In addition, he has organized a few seminal meetings such as the International Conference of Morphological Computation and International Seminar of Embodied Artificial Intelligence.
The concepts and sprits of embodied artificial intelligence have been around for the last decades and demonstrated the far reaching implications for our understanding of the nature of autonomy and adaptability through the countless case studies. It is however not fully clarified methodologies with which the different aspects of embodiment can be coherently and consistently integrated into one entity of autonomous agent. In the last few years, we have been developing an approach in which embodied agents could be studied in the level of genetic encoding, rather than that of behaviors. The framework consists of a mother robot (a robotic manipulator) that is capable of handling various materials and robotic modules assembled into a variety of child robots and evaluating the behavioral performances of child robots in order to progressively improve designs over generations. Unlike the conventional evolutionary robotics research, our approach does not use mathematical modeling in the evolutionary process, we are able to explore how embodied system-environment interactions influence the dynamics of physical robot evolution by observing the mother robot “inventing” surprising solutions to the given task. By introducing our recent experiments of building over 700 individual physical robots, we will discuss the implications and challenges of our project.
Bio: Fumiya Iida received his bachelor and master degrees in mechanical engineering at Tokyo University of Science (Japan, 1999), and Dr. sc. nat. in Informatics at University of Zurich (2006). In 2004 and 2005, he was also engaged in biomechanics research of human locomotion at Locomotion Laboratory, University of Jena (Germany). From 2006 to 2009, he worked as a postdoctoral associate at the Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology in USA. In 2006, he awarded the Fellowship for Prospective Researchers from the Swiss National Science Foundation, and in 2009, he was appointed as a Swiss National Science Foundation Professor for bio-inspired robotics at ETH Zurich. His research interests include biologically inspired robotics, embodied artificial intelligence, and biomechanics, and he has been involved in a number of research projects related to dynamic legged locomotion, navigation of autonomous robots, and human-machine interactions. He has so far published over forty publications in major robotics journals and conferences, and edited two books. Currently he serves on the editorial board of the Journal of Intelligent & Robotic Systems and Frontiers in Neuroscience (Neurorobotics), and as a program committee member for international conferences and workshops. In addition, he has organized a few seminal meetings such as the International Conference of Morphological Computation and International Seminar of Embodied Artificial Intelligence.
Links
Practical information
- Informed public
- Free
Organizer
- Auke Ijspeert
Contact
- Auke Ijspeert