Advertisement
Advertisement
Professor Nadia Thalmann (right) shaking hands with Nadine, a robot made in her likeness at a Singapore university.

Meet your newest friend, nurse and carer: a robot who reads your moods and emotions

A social robot equipped with personality, a humanoid face and a good memory could soon be, say, a coach for your autistic son, a minder for your diabetic daughter, a carer for your dad with dementia or a companion for your elderly aunt who lives alone

At Nanyang Technological University in Singapore, a new receptionist on campus has been turning heads. Nadine, with soft skin and flowing brunette hair, is always ready with a friendly smile and handshake. She looks you in the eye when talking, and the next time you meet her, she’s sure to remember you and your previous chat with her.

Her appearance might fool you but Nadine is actually a robot – the world’s most human-like one – and a doppelganger of her creator, Professor Nadia Thalmann, director of the university’s Institute for Media Innovation.

Thalmann says social robots like Nadine – equipped with personality, moods, emotions and a good memory – could become personal coaches or companions, especially for people with special health or medical needs.

“As countries worldwide face challenges of an ageing population, social robots can be one solution to address the shrinking workforce, become personal companions for children and the elderly at home, and even serve as a platform for health care services in the future,” says Thalmann, who has pioneered research into virtual humans over the past 30 years.

“For autistic children, for example, we can control the emotional degrees of the robot to make it attractive and adapted to children with special needs. For the elderly with dementia, social robots can help them maintain social skills and to feel less alone.”

Nadine is powered by intelligent software similar to Apple’s Siri or Microsoft’s Cortana. Unlike conventional robots, social robots can adapt to each individual, recognise particular actions and features of a specific person, and react appropriately, Thalmann explains.

Over the past four years, Thalmann and colleagues have been fostering cross-disciplinary research in social robotics technologies – involving engineering, computer science, linguistics, psychology and other fields – to transform a virtual human in a computer into a physical being that’s able to observe and interact with humans.

Edgar 1 (left) and Edgar 2, robots built at Nanyang Technological University in Singapore.

Our psychology when interacting with a social robot versus a character on a screen is starkly different, according to Dr Cory Kidd, who has worked in health care technology for nearly two decades. While getting his Masters and PhD at the Massachusetts Institute of Technology’s Media Lab, Kidd and colleagues researched this psychology of interaction, using six robots and more than 400 people in a series of studies over two years.

They found two main differences between people interacting with robots versus someone on a screen.

“People were more drawn to interacting with the robots, more engaged in the conversation. Part of that is novelty. But after that novelty wore off, people remained engaged with the thing for significantly longer than an identical-looking character on the screen having the same conversation,” explained Kidd during a recent TedxBerlin talk.

“And when trying to convey information through the robot, the information seemed more credible, more informative. The robot came across as more trustworthy.”

Kidd was previously based in Hong Kong for about five years, after starting up Intuitive Automata in 2008 in this city with two partners. Their main offering was Autom, a weight-loss coach robot created by Kidd at MIT.

In a trial involving 45 people at Boston University Medical Centre’s weight management clinic, participants were assigned either the robot, a computer with an identical touch screen the robot had, or a paper log, to help them with their weight-loss goals.

At the end of the two-month study, Kidd found that people assigned to the computer or paper log stopped using those tools after a couple of weeks.

Softbank Robotics’ emotion-reading robot Pepper is present at the Consumer Electronics Show in Las Vegas this week. Photo: Reuters
“The robots? Not only were people still using them, they dressed them up, with a hat or scarf. Every single person had named their robot, for example Casper or Mary. We found a creation of a relationship between the person and the robot.”

Unfortunately Autom didn’t take off in the market. In late 2013, Kidd moved to San Francisco and launched Catalia Health with a new social robot, the Mabu personal health care companion, which will become publicly available this year.

Mabu – short for “mabudachi”, Japanese for “best friend” – is designed to help patients with the myriad challenges of chronic disease management. She isn’t mobile, but can make eye contact while carrying on a conversation with someone and is capable of simple gestures with her head and eyes. She holds a tablet-like screen in front of her that she uses during conversations to convey additional information.

“The cost of the robot is less than having a home health nurse visit even once a week,” Kidd says. “As we get older, this is the kind of technology that’s going to help all of us live longer, healthier, happier lives.”

New research at the University of Auckland shows social robots in the homes of independent elderly rural residents may help lessen social isolation and help with medical care. In the study, five participants trialled the Korean-made iRobi in their homes for at least three months.

Some participants enjoyed hearing the robot talk and interacting with it. Others liked the robot reminding them to take their medicine on time.

One day we could have robots with knowledge of language and etiquette like C-3PO, the robot (above right) played by Anthony Daniels in the new film Star Wars: The Force Awakens, one scientist says. Photo: AP
“The results suggest the health care robots were feasible for use with a rural population and may have benefits for some patients in reducing the need for medical care, increasing quality of life, reminding patients to take medicine on time, and providing companionship,” says study senior author Dr Elizabeth Broadbent.

Other studies have also shown the effectiveness of social robots with children. One study by the University of Southern California, for example, found Nao robots to be effective in helping autistic children practise imitation behaviour that encouraged their autonomy.

The study’s lead researcher Maja Mataric hopes that, within a decade, children with autism might have their own personal robots to assist them with therapy, help prompt them through daily tasks, coach them through interactions with others, and encourage them to play with peers.

“The idea is to eventually give every child a personalised robot dedicated to providing motivation and praise and nudges toward more integration,” Mataricć says.

The 60cm tall, French-made Nao robot was also extensively used in the ALIZ-E project, an €8.3 million (HK$69.6 million) initiative funded by the European Commission which ran from 2010 to 2014, and involved a number of studies in health care and education settings.

In one study, the robot was found to help diabetic children accept the nature of their condition and become more confident about their futures. Among the tasks performed by the robot was assisting diabetic children in keeping a diary in which they record food intake, insulin injections and blood sugar levels. The study found youngsters provide more information if a robot has prompted them to do so. For those on return visits, the robot helped children – through a number of quizzes and games – to understand diabetes more, and absorb the huge amount of information they are given in a way they can understand.

ALIZ-E uses “Nao” robots – each standing about 60 cm tall – and equips them with specially designed speech recognition software through which they can interact with a child and deliver personalised responses.

Of all these social robots, however, none are nearly as human as Nadine. But with rapid technological advancements, it perhaps won’t be long till these lifelike droids become commonplace among us.

“In future,” says Thalmann, “these socially intelligent robots could be like C-3PO, the iconic golden droid from Star Wars, with knowledge of language and etiquette.”

Post