Future domestic robotic assistants will be complex hybrid consumer products, integrated into human living spaces to perform daily tasks with us. While robotics technologies have seen a rapid advance in the last decade, state-of-the-art robotic systems are still challenged by frequent failures, especially for long-term autonomy. This is a potential barrier against acceptability and trustworthiness of these future systems. Through a socio-technical multidisciplinary approach, this project aims to develop an understanding of robotic failures and explore how people perceive and resolve robot failures as an effort to design human-robot interactions that can enhance consumer experience of future robotic systems.
This project will explore the potential of using a robotic arm to aid humans with the task of unloading a dishwasher in the home environment to facilitate the future uptake of the technology and aid inclusion/accessibility and independence of users in a range of settings.
Project team: Ayse Kucukyilmaz, Eike Schneiders, Bernd Stahl, Horia Maior, Murray Goulden, Liz Dowthwaite, Pablo Lopez-Custodio, Simon Castle-Green
Project partner: Beko (Dr Muhammad Chughtai)
Project start date 1 April 2023 – 30 September 2024
Introduction blog, mid, final.
Research into this area will continue through the UK-RAS Networks Robotic and Autonomous Systems project ‘Principles of Learning from Unstructured Human-Robot Interactions‘.
Outputs:
Paper “In-the-Wild Failures in a Long-Term HRI Deployment”
Paper “A Taxonomy of Domestic Robot Failure Outcomes: Understanding the impact of failure on trustworthiness of domestic robots”, TAS Hub Symposium, Sept 2024, Texas Austin.
This project sits within Horizon’s Consumer Products Campaign
Horizon Digital Economy Research, University of
Nottingham Innovation Park, Triumph Road, Nottingham,
NG7 2TU
horizon@nottingham.ac.uk
