Love robotics? Fill out the form below to stay
abreast of the latest news, research, and business
analysis in key areas of the fast-changing
robotics industry
Subscribe to Robotics
Trends Insights

Sponsored Links

Advertise with Robotics Trends
[ view all ]
Design and Development
Bookmark and Share
STORY TOOLBOX Print this story  |   Email to a friend  |   RSS feeds
Self-Assembling Roombots Morph into Furniture
It's the love child of Transformers and Lego.
By Judith Pfeffer - Filed May 22, 2014

More Design and Development stories
Lego meets Transformers. Inventors in Switzerland have devised modular robotics called Roombots – motorized half-spheres with retractable grippers – that transform into various forms of furniture at will. Chairs, benches tables – it's not merely form follows function, but furniture facilitating the future.

The furniture changes shape and function depending on the need and desire of the person controlling what's going on, according to Auke Jan Ijspeert of the Biorobotics Laboratory of  Ecole Polytechnique Federale de Lausanne, Switzerland) or EPFL for short.

He is an associate professor at the EPFL and head of the Biorobotics Laboratory with  a "diplome d'ingenieur" in physics from the EPFL, and a Ph.D. in artificial intelligence from the University of Edinburgh.

One obvious use for the invention, he says, would be "assistive technology," or being able to command the ball-like entities to create a table at a specified height and then to bring that table over to the user. But there are other applications down the line, even artistic ones, he says in the equally well-crafted two-minute video.

"The Roombot is controlled by machine learning algorithms modeled off animals' neural networks for movement, according to a research paper from the EPFL team," observes Motherboard

The research was co-­funded by the National Centre for Competence in Research Robotics and was documented by ScienceDirect of Elsevier, a top provider of information solutions for science, health, and technology professionals.

Ijspeert's team includes post-doctoral researchers Alexander Sprowitz and Rico MOeckel, and Ph.D. candidates Massimo Vespignani and Stephane Bonardi.

The system has not been perfected. "The movement of the modules when they're grouped needs to be accelerated and optimized, and the algorithms defining the sequences of motor actions involved in composing the various structures need to be improved," notes MediaCom, the information arm of EFPL.


Bookmark and Share
STORY TOOLBOX Print this story  |   Email to a friend  |   RSS feeds
Now you can follow Robotics Trends and
Robotics Trends Business Review on Facebook