Senior Researcher in Robotics
Oxford Robotics Institute, University of Oxford, UK

https://ori.ox.ac.uk/people/bruno-lacerda/

Modelling and Adapting to Human Behaviour in Shared Autonomy Systems

As the deployment of robot systems alongside humans increases, the ability to reason about human-robot interactions is becoming more crucial. Specifically, robots must be able to assess the capabilities of humans and adapt their behaviour accordingly. To achieve this, robots need principled ways to update their belief over the capabilities of the specific human they are interacting with, as they observe how that human acts. In this talk, I will focus on human modelling in shared autonomy systems, where humans can take control of robots at specific moments of operation. I will describe how framing this problem as a Bayes-adaptive Markov decision process, modelling humans as a set of possible transition functions and maintaining a belief over which human we are interacting with, enables the synthesis of adaptive policies that determine when to hand over control to humans, based on the internal robot model and on its belief on how well the human performs.

Bio: Bruno Lacerda is a Senior Researcher at the Oxford Robotics Institute, University of Oxford, UK. He received his Ph.D. in Electrical and Computing Engineering from the Instituto Superior Técnico, University of Lisbon, Portugal, in 2013. Between 2013 and 2017, he was a Research Fellow at the School of Computer Science, University of Birmingham, UK. His research focuses on the intersection of decision-making under uncertainty, formal methods, and mobile robotics. In particular, he is interested in using a combination of techniques from learning, planning and model checking to synthesise intelligent, robust and verifiable behaviour, for both single and multi-robot systems.