Can we trust robots?

Sci-fi movies haven’t done robots’ reputations any favours. Humans still don’t seem to have overcome the fear that the robots we build to serve us may one day turn against us.

To address this irrational fear, researchers from the University of Hertfordshire are involved in the Trustworthy Robotic Assistants project, which will explore how robots can participate in sophisticated interactions with humans in an increasingly safe and trustworthy manner.

Robots are increasingly being used as active ‘helpers’ in situations where humans require assistance to undertake certain tasks, such as being home companions for older people or personal care robots to help patients during their recovery. For this to be viable and productive, the humans involved must have complete faith in robot behaviour.

“People need to be able to trust robots that they come into contact with,” said Professor Kerstin Dautenhahn from the University of Hertfordshire’s Adaptive Systems Research Group.

“As part of this new project, our research team will focus on safety issues and trustworthy behaviour in the application of robots as home companions. This research will be carried out in our Robot House where we can observe these behaviours in a realistic environment.”

The research will address issues concerning the barriers between robots and humans that have hampered the development of human-robot interactions, and will look at not only whether the robot makes safe moves but whether it knowingly or deliberately makes unsafe moves.

The Trustworthy Robotic Assistants project involves teams from the University of Hertfordshire’s Adaptive Systems Research Group, the University of Liverpool’s Centre for Autonomous Systems Technology and the Bristol Robotics Laboratory, as well as industrial partners including the British Automation and Robot Association (BARA) and RU Robots Limited.