Northeastern University professor Robert Platt says recent advances in machine learning, big data, and robot perception have put the robotics field on the brink of a major breakthrough in the ability of robots to perform fine motor tasks and function in controlled environments.
Platt notes robot dexterity and artificial intelligence have been held back by Moravek's Paradox, which dictates that what is hard for humans is relatively easy for robots, and what is easy for humans is nearly impossible for robots.
However, autonomous robots are poised to make a significant leap forward in their ability to manipulate unfamiliar objects.
Platt and his team have trained a robot to find, grab, and remove unfamiliar objects from a pile of clutter with 93% accuracy. The researchers accomplished this using reinforcement learning, in which the robot learns through trial and error.
Platt also says advances in depth perception enable robots to identify individual objects in a crowded field.
From Northeastern University News
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found