Date and time: Monday, June 9th, 3-4pm

Location: room 4-231

Speaker: Dana Kulic, University of Tokyo

Title: Robots in Human Environments: Safety, Perception and Learning

Abstract: In order to work safely and cooperatively with humans in unstructured environments, robots will need to perceive and understand the environment and the humans in it, operate in a safe manner, and learn useful skills from humans who are not trained as robot operators. The first part of this talk will present strategies that ensure the safety of the human participant through planning and control, based on explicit measures of danger during interaction. Next, the ability of the robot to perceive its environment, and specifically the human behavior and reaction to robot movements, is considered. Human attention and affect data (extracted from physical markers such as head rotation and physiological monitoring) is used to improve safety and the user’s perception of safety during interaction. In the second part of the talk, recent work
on robot learning for humanoid robots will be presented. Learning behavior and motion primitives from observation is a key skill for humanoid robots, enabling humanoids to take advantage of their similar body structure to humans. The goal of this research is to develop robots which can learn during continuous co-location and interaction with humans in the human environment. An algorithm for on-line, incremental learning of whole body motion primitives by observation of human motion will be described. As new motion patterns are observed, they are stochastically segmented and incrementally grouped together using local clustering based on relative distances in the model space. Over time, the algorithm incrementally forms a tree structure representing the motion space, with specialized motions at the tree leaves, and generalized motions closer to the root. The talk will conclude with an overview of preliminary experimental results and a discussion of future research directions.

Bio: Dana Kulic received the combined B. A. Sc. and M. Eng. degree in electro-mechanical engineering, and the Ph. D. degree in mechanical engineering from the University of British Columbia, Canada, in 1998 and 2005, respectively. From 1997 to 2002, she worked as a systems engineer designing fuel cell systems with Ballard Power Systems, and developing operational control software for the CanadaArm II at MacDonald Dettwiler. She is currently a JSPS post-doctoral fellow at the Nakamura-Yamane Laboratory in the Department of Mechano-Informatics at the University of Tokyo, Japan. Her research interests include human-robot interaction, robot learning, humanoid robotics and mechatronics.