Robotics and the human brain - nothing like the movies?

Cyborg man
Professor Danica Kragic
Professor Danica Kragic
Research in robotics and computer science is set to advance as we develop our understanding of the human neurological system and its processes in more detail. The lessons available from the supercomputers in our heads can provide inspiration for new technology. Although it may appear that science fiction is veering alarmingly close to reality, science fact is much more complex and the way we deal with increased robotic capabilities will be more subtle and gradual than the depictions on the silver screen sometimes seem to suggest, argue experts in the field.

15 to 20 years ago nobody could understand why they would want to carry a phone with them all the time; now, we can hardly imagine life without mobile phones. Once you get used to a product and see that there are additional functionalities you can add to it, it’s much easier to accept those extra functionalities because you are already used to the product.
Professor Danica Kragic
It was announced on 28th January this year that The Human Brain Project (HBP) would be one of the European Commission’s Future and emerging Technologies (FET) Flagship projects. Lasting for a period of ten years and aggregating the efforts of no fewer than 87 research institutes, the project’s broad research areas will be tackling neuroscience, medicine and – more tangentially – computing. As well as being part of the means to the end of understanding the brain, more sophisticated computing and brain-inspired technologies are some of the project’s goals.

Professor Danica Kragic is Vice-Dean at the School of Computer Science and Communication at KTH Royal Institute of Technology, Sweden. She is also Director of the Centre for Autonomous Systems and head of the Computer Vision and Active Perception Lab. In an interview with ScienceOmega.com, Professor Kragic discussed the ways in which the Human Brain Project and other programmes of research are contributing to the advent of consumer robotics and advancing the debate on robot ethics.

On the topic of the project more generally, the science of understanding the human brain is an important endeavour in itself. For the ‘Future Computing’ branch in particular, the hope of those involved and of observers is that a better understanding of the brain will allow us to design and build new systems that, although they may not directly imitate it, are inspired by it. According to Professor Kragic, this will help solve some of the difficult problems involved in trying to build machines that can simulate human movement and interact with their environment.

"We still cannot build machines that exactly emulate the human body," she said. "Even if we built an artificial brain, it’s not likely that we would couple it to a body that is biological. The ‘body’ that we would want to control is that of a robot – which is different from a human body both in terms of shape and capabilities.

"I think this is why it is important to have an understanding of brain processes rather than copying them. I expect that the project will give some insights into these processes and also provide some basis, in the area of robotics, for inspiration to build machines that can perform tasks elegantly and efficiently, as humans do."

There are many obstacles to consumer-grade robots becoming ubiquitous from Professor Kragic’s point of view, chief among which is the very fact that robots are not yet in widespread use, as this limits opportunities to study their usefulness. Although some basic capabilities have been developed with which robots can be built that are devoted and able to fulfil certain tasks, these are not necessarily suited to regular users who struggle to understand why anyone would want a robot to dust the floor, for example, when a person can do it more efficiently.

"15 to 20 years ago nobody could understand why they would want to carry a phone with them all the time; now, we can hardly imagine life without mobile phones," Professor Kragic noted. "Once you get used to a product and see that there are additional functionalities you can add to it, it’s much easier to accept those extra functionalities because you are already used to the product."

Many challenges will have to be overcome if we are all to become used to robots and the idea of having them in our homes, a transition that Professor Kragic describes as going from something that dusts the floor to something that can multitask, is interactive, and may even have ‘emotional capacities’ as part of its logical processes. This is perhaps where the big divide between robots and other products lies, and where concerns begin to arise.

"There is a big difference between consumer robotics and mobile phones, because we want robots to interact with their environment and interact with the user," said Professor Kragic. "A phone does not need to do that – it is used passively – but robots are more active. When this activity comes into play, safety issues are raised."

In order to build systems that are safe, developers need to know what the failure modes are and which areas need to be focussed on. To find out – to understand what might go wrong and how the quality of the products can be improved – they will need to break out of the lab and tests with (mostly) skilled users with the aim of conducting more research on the interaction perspective with ‘real’ users.

While we have been talking about robots for only 60 years or so, the effort to understand the human brain has been ongoing for centuries. Professor Kragic was keen to stress that it takes time, when constructing a complex new product, to improve it. She clarified the point by drawing a comparison between robots and cars, which are still being made better, faster, safer, and cheaper. Indeed, a robot is a much more complex machine than a car and is used in many more contexts.

"I think we feel like we have been talking about robots for a very long time because we are used to seeing robots in movies," remarked Professor Kragic. "We are impressed by robots in films; we are afraid of them, love them and hate them by turns. We have built a vision of what a robot is and what it should be able to do based on these portrayals, but they do not represent reality. It’s easy for us to think that we understand what robots and robotics are all about because it seems so easy in the cinema. Research in the field is not all that easy."

The fact that we are now much closer to having systems that think like humans, look like humans and interact with humans will necessarily engender some manner of change in the way we talk about robots. Despite Rex, the bionic man on display until 11th March at the Science Museum, being designed to highlight the human physiological functions that can be replicated by existing technologies rather than being a fully functional humanoid, we seem to be realising all of a sudden that the possibility of creating androids is just around the corner.

"I think that some important questions are being raised now," Professor Kragic went on. "For example, should robots look like humans? Should robots have rights? Just because we built a machine, should we be able to turn it off?"   

Any product can be used or misused, as the professor pointed out, but robots have the potential to help us work less, and to carry out tasks that we consider boring or even dangerous. She believes that fears of robots somehow ‘taking over’ are largely unwarranted.

"You don’t need to have something that is embodied, i.e. that has a body that moves, for it to take over. You could potentially have a virus in a system that completely destroys a whole computer network and does many strange things to society. In the same way we have to protect all computer systems, we will be able to provide codes that make robots safe so that they do not endanger humans and so on. However, a car can be a very dangerous thing depending on who is using it, and the same can be said for any system."


Professor Kragic discussed the question What is the ultimate destiny of man and robot? with Alastair Reynolds and other scientists in a Crosstalks web broadcast on 28th February.

COMMENTS


(NOT DISPLAYED)




YOUR COMMENT WILL BE APPROVED BY A MODERATOR
HTML CODE IS NOT PERMITTED.
RELATED CONTENT
  • New study explores the emergence of the 'Twitter tribe'
  • ALMA unveiled: starburst galaxies and distant water discovered
  • Plight of the bumblebee: science meets art in ecological opera
  • Caffeine buzz sees bees return to certain flowers
This is one reason why responsible caregivers need to be allowed to have these creatures - to stop them from becoming extinct; albeit in captivity only.


Commented Cheryl on
Mass lizard extinction looms as climate change continues

publicservice.co.uk Ltd, Ebenezer House, Ryecroft, Newcastle-under-Lyme, Staffordshire ST5 2UB
Tel: +44 (0)1782 630200, Fax: +44 (0)1782 625533, www.publicservice.co.uk
Registered in England and Wales  Co. Reg No. 4521155   Vat Reg No. 902 1814 62