A robot brain in the cloud: RoboEarth



I saw ‘Robot and Frank‘ this weekend, a charming little comedy about a robot who undergoes training as a lockpick and diamond thief. But the smart people at Roboearth figured out that training your robot to do something is probably a waste of time: millions of other people are probably trying to teach it the same things.

Maybe not lock picking skills and how to pull off a diamond heist, but at least making breakfast, doing dishes and ironing shirts. Says Markus Waibel, senior researcher at ETH Zurich, one of the people working on RoboEarth:

“Not all knowledge robots can learn is easily exchangeable. However, a fair amount of knowledge robots learn can be exchanged: for example, maps, CAD models of objects, and articulation models of doors and drawers have been successfully learnt and shared between different robots.”

“A particularly interesting area for learning are links between shared information, such as where is the fridge (map coordinates), what does it look like (object recognition model) and how do I open it (object articulation model).”

“Another are probabilities, such as given that I see a table, bed, and chair, where would I most likely find a pillow? Robots are well-suited for this type of learning not least because, unlike humans, they are capable of rapid, systematic, and accurate data collection. This capability provides unprecedented opportunities for obtaining consistent and comparable data sets as well as for performing large-scale systematic analysis and data mining.” (source, edited for brevity)

So, instead of training each robot individually, why not store all your robot’s knowledge into a robot wikipedia in the cloud? Or, if you want to be sinister about it, a Borg brain for your robot. That’s exactly what RoboEarth is, an open source  brain:

RoboEarth is a World Wide Web for robots: a giant network and database repository where robots can share information and learn from each other about their behavior and their environment.

Bringing a new meaning to the phrase “experience is the best teacher”, the goal of RoboEarth is to allow robotic systems to benefit from the experience of other robots, paving the way for rapid advances in machine cognition and behaviour, and ultimately, for more subtle and sophisticated human-machine interaction.

I remember reading that Google already uses a shared “brain” for its driverless cars. This is one of the reasons why they claim driverless cars will perform better than humans: because they can access a database of thousands (maybe millions) of cars who have experience with the very situation that your car is in – like ice on the road, or extreme mist, or an accident right in front of your car). Your car will have access to a database with similar situations and be able to react in an optimal way – whereas you might only experience those conditions a few times in your driving career and panic.

Just recently, RoboEarth introduced Rapyuta a cloud engine that allows robots to offload heavy computation into the cloud. It gives a good idea of how the process of shared knowledge works.

 

[RoboEarth][Photo: digitalshay, Flickr]

Powered by Facebook Comments

About the author

Raf Weverbergh

Editor of whiteboard. Raf Weverbergh was a magazine journalist whose work appeared in magazines like Rolling Stone, Playboy, Mail on Sunday, Publico and South China Morning Post. He is the co-founder of FINN, a corporate communications agency where he advises startups and multinationals on their PR and Mustr, the easiest media database for PR professionals. You can contact him on Twitter, Linkedin or Skype (rafweverbergh).

Related Posts