Pint of Robotics: In-situ learning from human interactions in public spaces, Francesco Del Duchetto, University of Lincoln & Saga Robotics

Speaker: Francesco Del Duchetto, University of Lincoln & Saga Robotics

Title: In-situ learning from human interactions in public spaces

When: Wednesday 27th January, 1700-1800

Where: Zoom (free but registration required) https://universityofleeds.zoom.us/meeting/register/tZAkce6tqTwuGtLrw3EckaI_KAxVmUnZandw

Abstract: In a future where robots will ubiquitously be among us, in public spaces and our homes, we want them to be able to learn from their mistakes and to adapt from the interactions with us. In this talk, I will describe what are the challenges of deploying autonomous robots in such scenarios and how we are tackling some of them. In particular, I will show a couple of ways robots can obtain feedback data from humans, either by directly asking the users (explicit feedback) or by observing them (implicit feedback), to understand how well they are performing their operations. The feedback data collected from interactions is used, in our work, to improve the robot future behaviour and to adapt it over time.

Short Bio: Francesco Del Duchetto is a PhD student at the Lincoln Centre for Autonomous Systems (L-CAS) at the University of Lincoln and a Research Associate at SAGA Robotics. He is interested in studying the methodologies that allow robots to operate in the real-world among humans in long-term scenarios. In practice, he works in the intersection of the fields of Human-Robot Interaction, Navigation, Learning from Demonstrations, and Reinforcement Learning. He graduated cum laude from Sapienza University of Rome with an MSc in “Artificial Intelligence and Robotics” and from the University of Bologna with a BSc in “Computer Science and Engineering”.