Autonomous systems must make their own decisions, often without direct human control. But can we be sure that these systems will always make the decisions we would want them to? And how can we deploy such systems if we are not clear where this responsibility lies? In this talk I will examine high-level decision-making in autonomous systems, the formal verification approaches we might apply to this decision making, and the impact this verification can have on safety, legality, ethics, and more.
Director of the cross-disciplinary Centre for Autonomous Systems Technology at the University of Liverpool (www.liv.ac.uk/cast); Leads the EPSRC Network on the Verification and Validation of Autonomous Systems (vavas.org); Member of British Standards Institution AMT/10 committee on Robotics (including standard on Robot Ethics); and Member of IEEE "Global Initiative for Ethical Considerations in Artificial Intelligence and Autonomous Systems" (including new P7009 standard on Fail-Safe Design of Autonomous Systems).
If you would like to meet Michael, please contact Amit Chopra (email@example.com)Add to my calendar