About this Event
150 Western Ave, Allston, MA 02134
Letizia Gionfrida (King’s College London)
Robots operating alongside people need to do more than reproduce behaviours learned from fixed datasets: they need to understand, learn from, and adapt to the humans with whom they interact. In this talk, I will explore how humans can be placed within the robot learning loop, from sensing and modelling human movement and behaviour to learning controllers that adapt to individual users, changing capabilities, and physical interactions. I will discuss approaches for learning personalised assistance, extracting low-dimensional structure from human motion, and enabling robots to adapt their actions during shared physical tasks. Through examples, I will consider how sensing, learning, control, and robot design can be jointly developed around the human. Ultimately, this work asks how we can move from robots that simply learn from humans towards robots that continuously learn and adapt with them.
Dr. Letizia Gionfrida is a Senior Lecturer in Computer and Robot Vision in the Department of Informatics at King’s College London, where she leads the Vision in Human Robotics Lab and is the Head of the Robotics, Autonomous Systems and Planning (RASP) Group. Her research sits at the intersection of computer and robot vision, human movement understanding, biomechanics, robot learning and human–robot interaction. Her work investigates how information about human movement and behaviour can be sensed and modelled, and how these representations can be translated into adaptive robotic systems. Before joining King’s, she was a Postdoctoral Research Fellow at Harvard John A. Paulson School Of Engineering And Applied Sciences working on computer vision for wearable robotics, following a PhD in Bioengineering from Imperial College London. She holds a Royal Academy of Engineering Research Fellowship and is the recipient of a Philip Leverhulme Prize.
Event Details
See Who Is Interested
0 people are interested in this event