Biography

Emre Ugur is an assistant professor in the Department of Computer Engineering Department, Bogazici University, Turkey. After receiving his PhD in Computer Engineering from the Middle East, he worked at ATR  Japan as a researcher (2009-2013), at the University of Innsbruck as a  senior researcher (2013-2016), and at Osaka University as specially appointed assistant professor (2015, 2016). He participated in several   EU funded projects, including Xperience, ROSSI, MACS and Swarm-bots,  and is currently PI of IMAGINE project supported by European Commission, Horizon 2020 Programme. He is interested in developmental and cognitive robotics, and intelligent and adaptive manipulation.

Abstract

IMAGINing Action Consequences in Robot Manipulation

Predicting the consequences of one’s own actions is an important requirement for intelligent control and decision making in both biological and artificial systems. Neurophysiological and behavioural data suggest that human brain benefits from internal forward models that continuously predict the outcomes of the generated motor commands for trajectory planning, movement control, and multi-step planning. In this talk, I will talk about robots learning consequences of their manipulation actions in different levels. First,  I will focus on discovering and predicting effect categories given objects and actions in an affordances-driven developmental framework. Next, I will present models that enable a robot to learn to predict the low-level spatiotemporal consequences of its manipulation actions from its own interaction experience on objects of various shapes. Finally, I will present our recent extension of propagation networks that enable the robot to predict the effects of its actions in scenes containing articulated multi-part multi-objects.