Robots that develop emotions in interaction with humans

Aug 12, 2010
Robots that develop emotions in interaction with humans

The first prototype robots capable of developing emotions as they interact with their human caregivers and expressing a whole range of emotions have been finalised by researchers.

The first prototype robots capable of developing emotions as they interact with their human caregivers and expressing a whole range of emotions have been finalised by researchers.

Led by Dr. Lola Cañamero at the University of Hertfordshire, and in collaboration with a consortium of universities and robotic companies across Europe, these robots differ from others in the way that they form attachments, interact and express emotion through bodily expression.

Developed as part of the interdisciplinary project FEELIX GROWING (Feel, Interact, eXpress: a Global approach to development with Interdisciplinary Grounding), funded by the European Commission and coordinated by Dr. Cañamero, the robots have been developed so that they learn to interact with and respond to humans in a similar way as children learn to do it, and use the same types of expressive and behavioural cues that babies use to learn to interact socially and emotionally with others.

The robots have been created through modelling the early attachment process that human and chimpanzee infants undergo with their caregivers when they develop a preference for a primary caregiver.

They are programmed to learn to adapt to the actions and mood of their human caregivers, and to become particularly attached to an individual who interacts with the in a way that is particularly suited to its personality profile and learning needs. The more they interact, and are given the appropriate feedback and level of engagement from the human caregiver, the stronger the bond developed and the amount learned.

The robots are capable of expressing anger, fear, sadness, happiness, excitement and pride and will demonstrate very visible distress if the fails to provide them comfort when confronted by a stressful situation that they cannot cope with or to interact with them when they need it.

"This behaviour is modelled on what a young child does," said Dr Cañamero. “This is also very similar to the way chimpanzees and other non-human primates develop affective bonds with their caregivers.”

This is the first time that early attachment models of human and non-human primates have been used to program robots that develop emotions in interaction with humans.

“We are working on non-verbal cues and the emotions are revealed through physical postures, gestures and movements of the body rather than facial or verbal expression,” Dr Cañamero added.

The researchers led by Dr. Cañamero at the University of Hertfordshire are now extending the prototype further and adapting it as part of the EU project ALIZ-E, which will develop robots that learn to be carer/companion for diabetic children in hospital settings.

Within this project, coordinated by Dr Tony Belpaeme of the University of Plymouth, the Hertfordshire group will lead research related to the emotions and non-linguistic behaviour of the robots. The future robot companions will combine non-linguistic and linguistic communication to interact with the children and become increasingly adapted to their individual profiles in order to support both, therapeutic aspects of their treatment and their social and wellbeing.

Explore further: Robots lending a helping hand to build planes

Related Stories

Robots to help children to form relationships

May 29, 2007

A project which is using robots to help children with developmental or cognitive impairments to interact more effectively has just started at the University of Hertfordshire.

Androids might soon become science fact

May 09, 2006

New Zealand and European scientists say it's time to send R2-D2 back to science fiction land -- and get ready to greet androids that think and act like humans.

Developing long-term relations with robots

Apr 13, 2008

Scientists at Queen Mary, University of London are leading an international project which is set to advance the relationship between robots and humans, as part of new European project called LIREC - Living with Robots and ...

Is my robot happy to see me?

Oct 19, 2009

(PhysOrg.com) -- People are social creatures. Robots... not so much. When we think of robots, we think of cold, metallic computers without emotion. If science fiction has taught us anything, though, it's that ...

Piecing together the next generation of cognitive robots

May 05, 2008

Building robots with anything akin to human intelligence remains a far off vision, but European researchers are making progress on piecing together a new generation of machines that are more aware of their ...

Recommended for you

Robots lending a helping hand to build planes

Aug 26, 2014

Trying to squeeze into small enclosed areas, carrying out highly repetitive tasks, retiring with back injuries even while your expertise is needed: these everyday realities of working in aviation construction ...

C2D2 fighting corrosion

Aug 22, 2014

Bridges become an infrastructure problem as they get older, as de-icing salt and carbon dioxide gradually destroy the reinforced concrete. A new robot can now check the condition of these structures, even ...

Meet the "swarmies"- robotics' answer to bugs

Aug 22, 2014

(Phys.org) —A small band of NASA engineers and interns is about to begin testing a group of robots and related software that will show whether it's possible for autonomous machines to scurry about an alien ...

User comments : 27

Adjust slider to filter visible comments by rank

Display comments: newest first

cmn
4 / 5 (7) Aug 12, 2010
This robot doesn't develop emotions anymore than your Sims characters develop emotions. Saying that the robot "expresses anger, fear, sadness, and happiness" would indicate that it actually has feelings. It doesn't. What it's doing is just a fancy method of reinforcement learning, based on the caregiver's visual or physical queues.

http://en.wikiped...learning

I'm afraid saying it develops emotions is just another marketing ploy.
danman5000
5 / 5 (2) Aug 12, 2010
"That one's called anger. You ever simulate anger before?"
toyo
3.7 / 5 (6) Aug 12, 2010
Yay!
Now we're spending money to produce nuisance robots, demanding robots, complaining robots et al.
A Great way to spend research money!!
Congratulations for producing robots who: "...will demonstrate very visible distress if the caregiver fails to provide them comfort when confronted by a stressful situation that they cannot cope with or to interact with them when they need it."
For what purpose?
For the purpose of: "... carer/companion for diabetic children in hospital settings."

Huh?
Run that one past me again?
I must be missing something here...
kits
not rated yet Aug 12, 2010
hope we end up creating more DATAS than LORES since in few decades we might have the likes of both...
RobertKLR
1 / 5 (1) Aug 12, 2010
"The robots are capable of expressing anger, fear, sadness, happiness, excitement ..."

BS. The robot is executing a slick and deceiving program. The "emotions" are an illusion.
Modernmystic
1.5 / 5 (2) Aug 12, 2010
FTR: I'm not saying that they've made robots with emotions.

That said I'll also state for the record I'm a HUGE technophile. I love everything from bic lighters to Saturn rockets. However, AI scares the shit out of me and if/when they do develop intelligent robots with emotions it will basically be the end of human civilization. Whether or not the end is violent or without bloodshed...it'll still be the end.
DamienS
3 / 5 (4) Aug 12, 2010
Another pointless 'research' program in a long line of many where someone thinks it would be cute to mimic some aspect of human behaviour in a robot. It's a gimmick of course. If they're really serious, they should concentrate on building a massively connected neural simulator (with sensory inputs) that can learn and be brought up (programmed) like a child, then you'd get all of these emergent behaviours popping out on their own, rather then having to be hand coded in. The Blue Brain project is a promising start.
http://en.wikiped..._Project
HealingMindN
2 / 5 (3) Aug 12, 2010
Therefore, it's a mime robot created for purposes of entertainment and companionship. Paw parrots are cheaper. If it must be a bot, then perfect the farting/burping battle bot. They're loads of laughs.
Auxon
1.7 / 5 (3) Aug 12, 2010
What a waste of time and money. Focus on making stupid robots to do all our chores then more people will have free time to spend with kids in the hospital, and it will be real emotions and caring.
Kedas
3.5 / 5 (4) Aug 13, 2010
The project will probably be more important to find out how people react to it, because believe it or not we are also machines and it only a matter of time before we won't see anymore why one machine is more important than the other.
rab96
1 / 5 (3) Aug 13, 2010
@Kedas
If you think we are machines then you should be able to explain the difference between a living machine and a dead machine.
CHollman82
5 / 5 (3) Aug 13, 2010
I'm not saying these robots experience emotion but...

At the end of the day, our emotions are nothing but the product of biochemical signals in our brains... It is a pretty arbitrary line you must draw to claim that one thing that expresses emotions actually feels them and another thing that expresses emotions does not, at least with our current understanding.

Just something to keep in mind as we progress within the field of AI...
Kedas
2.3 / 5 (3) Aug 13, 2010
@Kedas
If you think we are machines then you should be able to explain the difference between a living machine and a dead machine.

You have working machines and you have broken machines, the working ones that are learning/complex enough we call living the other ones are dead or broken.
DamienS
3 / 5 (3) Aug 13, 2010
@Kedas
If you think we are machines then you should be able to explain the difference between a living machine and a dead machine.

Software.
Arikin
1.5 / 5 (2) Aug 13, 2010
quote:interact with the children and become increasingly adapted to their individual profiles in order to support both, therapeutic aspects of their treatment and their social and emotional wellbeing.

This implies a controlled response to help development of the child. Which of the many child development theories do they plan on using??

For posters who say this is a waste of time think of it this way. While reading this you have no clue as to the emotional context behind the text.

The faked emotions help convey more of the message being given. Conversely, if the robot can learn to read the user's or child's body language it will receive more information to make better decisions.

For example:
1. You barely move when you tell the robot to do something = Task is not high priority, higher priority tasks can be done first.
2. You wave your fist at the robot while yelling the command = robot immediately stops current task and does what you say.
MarkyMark
3 / 5 (2) Aug 14, 2010
Have to say great programming! But remember "simulating" Emotions is like an Aircraft Simulator...

As in not real hence the term simulation.
MaxwellsDemon
not rated yet Aug 14, 2010
@ DemianS
The Blue Brain project is a promising start.
http://en.wikiped..._Project

Now this is fascinating. I'm all for it, as long as we don't hand our launch codes over to this thing.

denschmitz
1.7 / 5 (3) Aug 14, 2010
Giving robots emotions: What could possibly go wrong?

It seems like none of these guys have ever read or seen The Terminator, The Matrix, Blade Runner, 2001: a Space Odyssey, Metropolis, The Hitchhiker's Guide, The Demon Seed, Saturn 3, Futurama, Battlestar Galactica, or any anime with a robot character.

I could go on and on, but geeze, what are these people thinking? Depending on your definition, emotions are what guide most animals. Should we be starting a Vegas pool on what scientist gets vacuumed to death by his Roomba because he's paying too much attention to his TiVo?
Modernmystic
2 / 5 (2) Aug 14, 2010
We wouldn't even have to do something as overt as handing over launch codes for it to be a problem. If it had any kind of wireless/wired connectivity to the outside it could be a huge problem.
MaxwellsDemon
2 / 5 (2) Aug 14, 2010
I was just kidding around, but you’re right – the idea of a fully functioning electronic form of higher consciousness scares the poop out of me.

And the ethical dilemmas are dizzying. Seriously, do we have the right to create a form of consciousness that, by its very nature and through no fault of its own, is basically a slave under 24/7 surveillance? I know that would drive [i]me[/i] insane…
DamienS
not rated yet Aug 14, 2010
Seriously, do we have the right to create a form of consciousness that, by its very nature and through no fault of its own, is basically a slave under 24/7 surveillance?

The same question can be turned around by substituting 'form of consciousness' to 'human baby'. Either way you're creating an uncertain future, at least with AI you have some chance in building in safeguards. In any case, such an AI won't just be created in one fell swoop, but will begin to appear organically (ahem) over time. The so called singularity is already a growth industry.
http://en.wikiped...gularity
http://en.wikiped..._Is_Near
kaypee
5 / 5 (2) Aug 14, 2010
Human emotions are also a clever illusion, but the complex buffet of emotions and feelings that create self-importance urge us to reject this. Actually, I think that the qualia we experience can only be experienced by other entities that have the same hardware functioning in the same way. Different hardware = different subjective experience, where the difference depends on HOW different the hardware is.
bottomlesssoul
not rated yet Aug 15, 2010
Wow, kaypee! You're right, it's chemistry, why not emulate it in hardware, we'll react the same. It's hardwired in either case.

I wonder if it will be programmable to probe mental illness? That would be cool. To really mimic primates it would need critical periods where circuitry becomes stable and mostly locked.

I'm really impressed, I think it's a whole new way to learn about ourselves.
bottomlesssoul
not rated yet Aug 15, 2010
@MaxwellsDemon
I was just kidding around, but you're right - the idea of a fully functioning electronic form of higher consciousness scares the poop out of me.

And the ethical dilemmas are dizzying. Seriously, do we have the right to create a form of consciousness that, by its very nature and through no fault of its own, is basically a slave under 24/7 surveillance? I know that would drive [i]me[/i] insane
.Yikes! You just described a modern city but with out the AI! Being a slave to 24/7 surveillance has been around for decades, ever heard of Stalin? It's self fulfilling fear. The only remedy is exercise what control you have over your own fear. Oh and pray to whatever you pray too or appeal too hope or whatever you do to find internal peace. Pray no one will be so stupid as to give their self control over to a person let alone a machine.
cmn
1 / 5 (1) Aug 16, 2010
I'm not saying these robots experience emotion but...

At the end of the day, our emotions are nothing but the product of biochemical signals in our brains... It is a pretty arbitrary line you must draw to claim that one thing that expresses emotions actually feels them and another thing that expresses emotions does not, at least with our current understanding.

Just something to keep in mind as we progress within the field of AI...


I'd say the line is quite clear, to anyone that has emotions...
rgwalther
not rated yet Aug 18, 2010
"The robots are capable of expressing anger, fear, sadness, happiness, excitement ..."

BS. The robot is executing a slick and deceiving program. The "emotions" are an illusion.

ArcainOne
not rated yet Aug 20, 2010
I feel like I just walked into the player haters ball for technology. "This is dumb robots will kill us! waist of money", "No its not robots are kewl, wave of the future!"

We cannot let fear keep us from advancing technology, and at the same time we must exercise caution and evaluate the possible repercussions of our actions. Movies such as "Terminator, The Matrix, Battlestar Galactica, etc" are excellent cautions to what "may" happen if we are not careful. We should use these ideas to improve our development, not hinder it.