AT 2018 Invited Speaker
TITLE: Shared Decision Making in Human-Robot Teams
The future of robotics lies in developing safe and trustworthy methodologies for people and robots to collaborate. Whether operating in the office, classroom, home, hospital or disaster scenario, robots and humans need to be able to support each other in mixed-initiative interactions and take advantage of the unique capabilities of each species. Our research investigates ways in which robots can collect data from their environment, analyse the data, and then use that knowledge to make decisions in partnership with humans. This talk will focus on the use of computational argumentation and argumentation-based dialogue to facilitate shared decision making in human-robot teams. Experimental results obtained through studies with physical robots and human subjects will be highlighted.
Dr. Elizabeth Sklar is a Professor of Robotics in the Department of Informatics at King's College London, UK, where she is head of the Centre for Robotics Research (CoRe). She worked as a Software Engineer in industry for more than 10 years, designing and building real-time interactive systems, before holding academic positions in the US at Columbia University and the City University of New York. She was a US-UK Fulbright Scholar in 2013-14, a Senior Research Fellow at University of Liverpool and moved to King's College London in 2015. Her research investigates interaction in multi-robot and human-robot systems. She has recently contributed to shared decision making by applying computational argumentation-based dialogue to human-agent and human-robot systems. Sklar's work has been funded by the US National Science Foundation, US Department of Education, US Army Research Lab, and UK research councils (EPSRC and ESRC). She has published over 150 papers in refereed conferences, workshops and journals and has edited two books. She is a Founder Trustee of the RoboCup Federation, former member of the Board of Directors for the International Foundation for Autonomous Agents & MultiAgent Systems (IFAAMAS) and on the Editorial Board for the Journal of Autonomous Agents & MultiAgent Systems (JAAMAS).
EUMAS 2018 Invited Speaker
TITLE: Trustworthy Autonomy
Truly autonomous systems, such as (future) robots or vehicles, will make their own decisions about what to do without necessarily consulting humans. How can we trust such systems if we do not know what, and crucially why, these systems will make their decisions? In this talk, I will describe how the use of agents at the core of these systems can form the basis for trustwortiness. Yet we need more: at least transparency and explainability, but ideally verifiability. Once we have autonomous systems based on a hybrid agent architecture that supports verifiability, then we can strongly analyse these systems and move towards not only trustworthiness, but ethical and responsible behaviour. I will provide some examples from our current and previous projects examining the deployment of hybrid agent architectures in both vehicles (road and air) and robots (domestic and industrial) and assessing safety, ethics, fault tolerance, etc.
This work, and the examples described, are derived from projects funded in the UK by EPSRC: http://cgi.csc.liv.ac.uk/~matt/AVLab
Dr. Michael Fisher is a Professor in the Department of Computer Science at the University of Liverpool, and Director of the University's Centre for Autonomous Systems Technology. He is a Fellow of both the BCS and the IET, member of the BSI AMT/10 committee on Robotics and the IEEE P7009 Standard for Fail-Safe Design of Autonomous Systems, and lead the UK Network on the Verification and Validation of Autonomous Systems.