Feelix Growing is developing software empowering robots that can learn when a person is sad, happy or angry.
The learning part is achieved through the use of artificial neural networks, which are well suited to the varied and changing inputs that ‘perceptive’ robots would be exposed to.
Using cameras and sensors, the very simple robots being built by the researchers – using mostly off-the-shelf parts – can detect different parameters, such as a person's facial expressions, voice, and proximity to determine emotional state.
The technology pulls together research in robotics, adaptive systems, developmental and comparative psychology, neuroscience and ethology, which is all about human behaviour.
Are you feeling ok?
Much like a human child, the robot learns from experience how to respond to emotions displayed by people around it.
If someone shows fear or cries out in pain, the robot may learn to change its behaviour to appear less threatening, backing away if necessary. If someone cries out in happiness, it may even detect the difference, and one day fine-tune its responses to individuals.
“It's mostly behavioural and contact feedback,” project coordinator Dr Lola Canamero is quoted as saying in a BBC News story on Feelix Growing. “Tactile feedback and emotional feedback through positive reinforcement, such as kind words, nice behaviour or helping the robot do something if it is stuck,” she said.
The three-year, Sixth Framework Programme project involves six countries and 25 specialists who are building demonstration robots as proof of concept.
One demo follows the researchers around like a young bear cub might its mother, learning from experience when to trail behind or stick close to her. A robot face is also in development which can express different 'emotions'.
The main idea is, by being more in tune with human emotions, giving the impression of empathy, the robots should be more readily accepted by the people they may one day serve.
Not exactly I Robot
Robots that can adapt to people's behaviours are needed if machines are to play a part in society, such as helping the sick, the elderly, people with autism or house-bound people, working as domestic helpers, or just for entertainment, according to Canamero.
The work is still well shy of an I Robot scenario with emotionally complex machines taking matters into their own hands, but the empathy empowering software being developed by Feelix Growing is a big step forward for robotics.
And gauging by the attention the project has garnered in leading press, such as the BBC, Wired and engadget, and most recently in a report on Euronews, Feelix Growing is maturing very well.
Christian Nielsen | alfa
CubeSats prove their worth for scientific missions
17.04.2019 | American Physical Society
Largest, fastest array of microscopic 'traffic cops' for optical communications
12.04.2019 | University of California - Berkeley
A stellar flare 10 times more powerful than anything seen on our sun has burst from an ultracool star almost the same size as Jupiter
A localization phenomenon boosts the accuracy of solving quantum many-body problems with quantum computers which are otherwise challenging for conventional computers. This brings such digital quantum simulation within reach on quantum devices available today.
Quantum computers promise to solve certain computational problems exponentially faster than any classical machine. “A particularly promising application is the...
The technology could revolutionize how information travels through data centers and artificial intelligence networks
Engineers at the University of California, Berkeley have built a new photonic switch that can control the direction of light passing through optical fibers...
Physicists observe how electron-hole pairs drift apart at ultrafast speed, but still remain strongly bound.
Modern electronics relies on ultrafast charge motion on ever shorter length scales. Physicists from Regensburg and Gothenburg have now succeeded in resolving a...
Engineers create novel optical devices, including a moth eye-inspired omnidirectional microwave antenna
A team of engineers at Tufts University has developed a series of 3D printed metamaterials with unique microwave or optical properties that go beyond what is...
17.04.2019 | Event News
15.04.2019 | Event News
09.04.2019 | Event News
18.04.2019 | Life Sciences
18.04.2019 | Physics and Astronomy
18.04.2019 | Life Sciences