AI and robotics experts from the National Robotarium, based at Heriot-Watt University, have started work on a multi-user conversational robot for healthcare.
Matt Davis
Part of a multimillion-pound collaborative project involving experts from eight European and Asian institutions, SPRING (Socially Pertinent Robots in Gerontological Healthcare) is the first research project to be announced by the National Robotarium.
SPRING, a four-year project funded by Horizon2020, will develop Socially Assistive Robots (SARs) with the capacity to perform multi-person interactions and open domain social conversation for the first time in a healthcare setting. The work builds on the success of Heriot-Watt University’s Amazon Alexa Prize conversational AI system 'Alana’.
The project will focus on supporting elderly patients by carefully coupling scientific findings and user-focussed technological developments to bring social robots into gerontological healthcare.
Over the past five years, social robots have been introduced into many public spaces ranging from museums and malls to hospitals and retirement homes. The robots have been able to provide both information and entertainment, but the technology has faced challenges.
Limitations include the fact that both the hardware and supporting software is often designed for reactive, single-user interactions, leading to limited one-on-one conversations. As a result, the robots typically wait for commands or questions based on a limited set of scripted actions.
Professor Oliver Lemon, Heriot-Watt University, said: “Research shows that the careful use of robots in group settings can have a positive impact on health, such as decreased stress and loneliness, and improved mood and sociability.
“Healthcare practitioners have been supportive of the use of robots during the non-medical phases of time in hospital because social robots can help explain complex concepts to patients with limited medical knowledge.
“Social robot technology is of interest for elder care because robot companionship has the long-term potential to better connect people with each other. Social robots could improve both psychological well-being and the relationship between patients and hospital professionals.
"While overcoming the limitations of current social robots raises numerous scientific and technological challenges, it has the potential to create tremendous social impact and economic value. The National Robotarium’s focus on creating societal benefits is ideally aligned to addressing such challenges.
"This type of technology is touch-free and hands-free so will be in great demand in the future as it will reduce the risk and spread of infection."
SPRING will develop new research into conversational AI, computer vision, machine learning and human-robot interaction, alongside human behaviour analysis and sensorimotor robot control.
The work will focus on helping social robots to understand various individuals and group situations and take appropriate decisions such as identifying patients that have been waiting alone for a long time or who might be anxious. The social robots will ultimately engage in face-to-face conversation with patients, their family members, staff members, and with whole groups of people.
SPRING’s partners include: Inria Grenoble (coordinator), Università degli Studi di Trento, Czech Technical University Prague, Heriot-Watt University Edinburgh, Bar-Ilan University Tel Aviv, ERM Automatismes Industriels Carpentras, PAL Robotics Barcelona, and Hôpital Broca Paris.
