Cornell University, along with its partners at UC Berkeley, Brown University, and Stanford University, is developing a "robot brain" that can teach other robots how to do household chores, among other things. The undertaking, aptly called Robo Brain, has earned the support of many large organizations, including the National Science Foundation, Office of Naval Research Science and Technology, National Robotics Initiative, Google, Microsoft, Qualcomm, and the Alfred P. Sloan Foundation.
"Robo Brain is a large-scale computational system that learns from publicly available Internet resources, computer simulations, and real-life robot trials," the group says in the Robo Brain website. "It accumulates everything robotics into a comprehensive and interconnected knowledge base. Applications include prototyping for robotics research, household robots, and self-driving cars."
Basically,what the computer scientists are trying to achieve is to have a central system containing the data for robotic "skills" that can be used in or by other robots. As explained by Engadget, companies or robot developers can access Robo Brain online and download the robotic skill that they need for their machines.
"If a robot encounters a situation it hasn't seen before, it can query Robo Brain in the cloud," robot learning and project lead Ashutosh Saxena said in a statement and quoted by CBS News.
The developers of the system are tackling multiple challenges in robotics, namely: machine learning; large-scale data processing; perception; artificial intelligence and reasoning systems; robotics and automations; embodiment; and language and dialogue.
According to the CBS News report, Robo Brain is now processing 100 million manuals, 120,000 YouTube videos, and one billion photographs.
