Keynote 2: Understanding and Modeling Object Affordances with Cognitive Robotics

Jose� Santos-Victor (Instituto Superior T�cnico - IST, Lisbon. Portugal), Abstract: I will give an overview of recent research projects combining artificial vision and humanoid robotics to help understand certain functions of human cognition. Part of this work comprises modeling human behavior in an interdisciplinary research involving engineering, computational neuroscience, and developmental psychology. I will discuss the concept of object affordances, and how it can be modeled and used to endow humanoid robots with higher cognitive functionalities, namely action recognition, goal acquisition and social learning. I will describe our different research approaches for modeling and using object affordances and different ways to extend that framework for language grounding and understanding tools. During the talk, I will provide examples of the use of humanoid robots (with our first platform, Baltazar, the iCub, and our social robot Vizzy) as testbeds to study human cognition, learning and sensorimotor coordination and affordances, while offering engineers with new approaches to build artificial systems.