Robots Learn to Perform Tasks by “Watching” YouTube Videos

DARPA program advances robots’ ability to sense visual information and turn it into action

Robots can learn to recognize objects and patterns fairly well, but to interpret and be able to act on visual input is much more difficult. Researchers at the University of Maryland, funded by DARPA's Mathematics of Sensing, Exploitation and Execution (MSEE) program, recently developed a system that enabled robots to process visual data from a series of "how to" cooking videos on YouTube. Based on what was shown on a video, robots were able to recognize, grab and manipulate the correct kitchen utensil or object and perform the demonstrated task with high accuracy—without additional human input or programming.

"The MSEE program initially focused on sensing, which involves perception and understanding of what's happening in a visual scene, not simply recognizing and identifying objects," said Reza Ghanadan, program manager in DARPA's Defense Sciences Offices. "We've now taken the next step to execution, where a robot processes visual cues through a manipulation action-grammar module and translates them into actions."

Another significant advance to come out of the research is the robots' ability to accumulate and share knowledge with others. Current sensor systems typically view the world anew in each moment, without the ability to apply prior knowledge.

"This system allows robots to continuously build on previous learning—such as types of objects and grasps associated with them—which could have a huge impact on teaching and training," Ghanadan said. "Instead of the long and expensive process of programming code to teach robots to do tasks, this research opens the potential for robots to learn much faster, at much lower cost and, to the extent they are authorized to do so, share that knowledge with other robots. This learning-based approach is a significant step towards developing technologies that could have benefits in areas such as military repair and logistics."

The DARPA-funded researchers presented their work today at the 29th meeting of the Association for the Advancement of Artificial Intelligence. The University of Maryland paper is available here:

Featured Product

US Digital - E4T Miniature Optical Kit Encoder

US Digital - E4T Miniature Optical Kit Encoder

US Digital is pleased to announce the launch of the E4T, their latest series of miniature high performance optical encoders. The E4T series delivers a marked performance increase over similar encoder models and designed to be an enhanced replacement for the E4P encoder series. The E4T utilizes state of the art transmissive optical sensing technology, and incorporates US Digital's own proprietary OptoASIC. Assembly of the E4T is simple and efficient and retains the previous E4P's form factor. Key features of the new E4T include: • Mechanically and Electrically Interchangeable with E4P • Improved Quadrature Signal Strength • 100 kHz Frequency Response • Transmissive Optical Design • Collet Style Push on Optical Disk Design (Patent Pending) • Simple & Efficient Assembly Process As with all of our products the E4T is designed and manufactured in their Vancouver, Washington USA facility and is available for purchase as of December 2014.