Giuseppe Averta – Associate Professor of Robot Learning

Human-Inspired robot learning

For robots to work in collaboration with humans, it is fundamental to have them move and act in a human-like fashion. We take inspiration from principles of human motor control to devise planning and control guidelines for robots.Read more

Egocentric Vision for human behavior understanding

We develop methods able to reason about human behavior collected in first-person videos, and to generate transferrable knowledge for intelligent manipulators.Read more

Scene understanding for robot manipulation and beyond

Our robots need to understand the environment in which they are operating, and be able to interact with it. We develop methods to find objects in the scene relevant for the task, and train robots to interact with them.Read more