Europe and Japan Join Forces to Map Out Future of Intelligent Robots
Innovations Report (07/29/08) Lau, Thomas

Making robots that can truly learn and adapt to unexpected situations like humans, or be able to move with the fluidity and grace of animals, has been a difficult challenge. However, significant progress has been made over the last few years, and Europe and Japan are both set for a push toward a new generation of intelligent machines. A conference jointly organized by the European Science Foundation (ESF) and the Japan Society for the Promotion of Science (JSPS) brought together researchers actively working in the fields of cognitive science and robotics. Professor Florentin Worgotter from Gottingen University in Germany suggested that with a greater understanding of how animals coordinate their movements, the principles involved in natural movement, such as mechanics, neuronal feedback and instantaneous adaptability, could be applied to robotics. Shuuji Kajita from the Japanese research group AIST demonstrated biped robots that use new walking techniques based on the Zer o-Moment Point principle, which is designed to ensure that a top-heavy system such as a humanoid robot can walk without losing its balance or placing too much stress on its points of contact. Another major focus of the ESF/JSPS conference was enabling robots to be adaptable and learn from their mistakes. University of Tokyo professor Yasuo Kuniyoshi said that traditional artificial intelligence techniques have not succeeded in making adaptable robots. He said current techniques break down events that a robot has not been programmed to expect into smaller parts in an attempt to analyze them. The problem is that the robot has no context in which to decide how to act. Alternative approaches involve imposing constraints on the robot's interactions that allow more intelligent behavior to emerge.
Web Link