I am Tenured Senior Researcher at the Istituto Italiano di Tecnologia (IIT) in Genova. I am leading the Humanoid Sensing and Perception research group.
Today, robots find wide application in industry where they operate by relying on accurate a priori models of the environment. Programming robots is therefore expensive and time-consuming, their use limited to accurately specified applications, and in strict separation from human workers.
The goal of my research is to develop humanoid robots that can effectively work in unstructured environments, operating in close interaction and cooperation with humans. Robotics is an interdisciplinary field that draws from fields ranging from engineering to machine learning and computer vision. Technical difficulties are unfortunatly a big barrier preventing contributions from different disciplines to robotics. This is one of the reasons that motivated me to work on the development of an open humanoid platform like the iCub. My research focuses on the study of algorithms for perceiving the environment and the development of sensing technologies. Research demonstrates that perception in humans is a multimodal and active process: we constantly integrate information from all senses and exploit actions to simplify perception and structure sensory data to aid learning. Similarly, my strategy is to study algorithms that allow the robot to explore the environment and improve perception by leveraging on the integration of the available sensors (i.e. vision, touch and proprioception).
For more information see:
European Laboratory for Leanring and Intelligent Systems Genoa Unit
Human Guided Learning and Benchmarking of Robotic Heap Sorting
ITP funded by the RobMoSys EU project
ITP funded by the RobMoSys EU project
The iCub robot vision dataset.
An open source middleware for robotics.