Skip directly to content

Adaptable robots learn to work together

NSF Award:

EFRI-ARESCI:Controlling the Autonomously Reconfiguring Factory  (Massachusetts Institute of Technology)

State:
Congressional Districts:
Research Areas:

A team of researchers at the Massachusetts Institute of Technology has developed a new programming language and two robotic prototypes that can build and disassemble objects without human intervention.

While robots are already highly valued in manufacturing, one of the biggest factors limiting their use is their inability to adapt to new and unique situations. Creating robots that can act on their own without human intervention will open up new processes to automation, making it easier and more cost effective than ever to meet manufacturing demands.

The robotic systems developed as a part of this research were designed to work as a team, alternatively delivering materials and assembling the objects. Additionally, if one goes offline or is unable to continue working, the other can quickly take over the task, working independently to complete it without additional intervention.

The robots can also be programmed to act according to certain preferences; for example, to execute slow, but reliable actions over actions that are faster but more prone to error.

 

Image

  • two robots work together to reconfigure two structures
A wheeled robot and a climbing robot reconfigure two structures.
Daniela Rus, MIT

Recent Award Highlights

soft polymers with hollow centers coated with a stiffer polymer can change shape on demand

Morphing surfaces optimize aerodynamics

Surfaces change texture from smooth to dimpled and back again

Research Areas: Engineering Locations: Massachusetts
illuminated by a beam of blue light, a neuron fires an electrical spike

Mapping brain function with light

Optogenetic tools enable precise control of neuronal activity

Research Areas: Engineering Locations: Massachusetts