Robots are not part of some science fiction future, they’re here now and being used in ever more interesting and adventurous ways. Julianne Evans discusses the ways robots are changing human care in New Zealand with Elizabeth Broadbent and Bruce MacDonald.
Originally meeting at a seminar, they realised they shared a consuming interest in how robots and humans can work together to make people’s lives better.
Both have degrees in electrical and electronic engineering from Canterbury University; Elizabeth later moving into the area of health psychology, while Bruce continued in engineering, although he always remained interested in artificial intelligence and machine learning.
Elizabeth leads the methodology; designing the research and carrying out studies of how robots and people can most fruitfully interact, particularly in the areas of health and education. Bruce designs the robotic software in partnership with his team at CARES, an interdisciplinary group of 30 staff across the faculties of health, engineering, and science, based at Newmarket Campus.
Their most recent study involved a trial using robots at home to support patients with chronic, obstructive, pulmonary disease (COPD). It found that the iRobiQ robots, who have expressive faces and look a bit like Star Wars character R2-D2, helped patients to regularly take their medication and do exercises as well as offering them companionship.
The study randomly allocated COPD patients to one of two groups, with 25 people receiving an iRobiQ robot in their homes for four months as well as standard care, while 29 received standard care alone.
The robots were programmed to measure heart rate, breathlessness and quality of life weekly; remind patients to take medication and do pulmonary rehabilitation exercises; provide education about COPD and data to respiratory clinicians. The study was funded by a Health Research Council partnership innovation grant and was in collaboration with the respiratory team at Counties Manukau DHB.
It found the group with robots were significantly more consistent in taking their long-acting inhalers than the control group, and also significantly increased their rehabilitation exercises compared to the control group, though there were no significant differences in quality of life.
Of the 25 patients who had a robot, 19 liked having it; saying their friends and family were interested in it, they felt the robot had a “personality” and was good company. Many gave it a name.
However, six people didn’t like them, three of whom returned theirs early, saying they were good at managing their medication and exercises on their own. The other three said they were unnerved by having a robot in their home.
“The research is helpful as it suggests that a homecare robot can improve adherence to medication and increase exercise,” says Elizabeth. However, she says more research is needed with a larger sample size to further investigate effects on hospitalisations after improvements are made to the robots.
“Robots can never be substitutes for people”, says Bruce. “They are assistants to help out in the absence of a person. They’re about empowering people. They will only help you if you need help. Robots can help dementia patients, for example, whose memory problems mean they can’t remember when to do things or if they’ve already done them.”
Another of Elizabeth and Bruce’s recent studies involved robots in schools, collaboration with Tufts University in Boston, from where student Danielle Feerst travelled to help collect data. Elizabeth visited Boston on a Fulbright scholarship in 2017 to study companion robots.
“To date, schools have mainly used robotic kits to teach children how to build and program robots” says Elizabeth. “We were interested in how students and teachers would respond to a companion robot. Would it be seen as useful? Would it make students more interested in learning about science and maths?”
In all, 207 students and 22 teachers from preschool to high school participated in 30-minute sessions with two popular companion robots; Paro, a fluffy, white soft toy created to resemble a baby harp seal and iRobiQ, a more traditional robot used in early childhood education in Korea.
Elizabeth chose schools in the Central Plateau and Buller regions because rural schools typically have fewer resources. The sessions started with a brief demonstration of the robots and then children and teachers could spend 10 minutes engaging with them.
Overall the students enjoyed them, with 84 percent saying they would like to have Paro at school and 80 percent said they would like to have iRobiQ. Girls were more enthusiastic about the robots in general than boys.
“As we need to motivate more students to go into science and technology careers, especially girls, these results are promising,” says Elizabeth. Elizabeth says both teachers and children saw Paro, who can open and close its eyes, move its tail and make a baby seal noise, as a pet and a comfort.
“Given the prevalence of anxiety and depression in school-age children, robots may be a useful tool in school-based approaches to promote mental health. However, further research is needed to test this,” she says.
Robotics is multi-disciplinary, connecting experts from different subject areas and bringing a high level of expertise to all parts of the research being done at CARES.
“We’ve focused on putting robots in real world conditions, unlike many researchers who mainly study robots in the laboratory, which gives us a much richer source of new knowledge,” says Bruce.
Elizabeth Broadbent is an Associate Professor in Psychological Medicine at the University of Auckland.
Bruce MacDonald is a Professor in Electrical and Computer Engineering at the University of Auckland.
Associate Professor Elizabeth Broadbent (School of Medicine) and Professor Bruce MacDonald (School of Engineering) have been collaborating in the field of leading edge robotics for more than a decade.
This article was originally published in the April edition of UniNEWS.