Ken Goldberg's Dex-Net Research Spotlighted in Nature
How robots are grasping the art of gripping, a Nature spotlight article covering robots' capabilities to use their "hands," features BCNM's very own Ken Goldberg. According to Goldberg,“Grasping is the critical grand challenge right now." Engineers are struggling to create working hands in the field of robotics.
An excerpt from the article reads:
Goldberg uses machine learning to teach robots to grasp, too. But rather than gather data from real-world attempts, his Dex-Net software is trained virtually. “We can simulate millions of grasps very quickly,” he says. The software lets an industrial robot pick objects from a pile with a success rate of more than 90%, even if it hasn’t seen those objects before. It can also decide for itself whether to use a parallel-jaw gripper or suction tool for a particular object.
Dex-Net’s fourth incarnation will be presented in 2018. According to a metric being developed by Goldberg and roboticists around the world to aid reproducibility, known as mean picks per hour, the Dex-Net system is now among the fastest pickers around. It can achieve over 200 picks an hour — still behind human capacity, estimated at 400–600 picks an hour, but far ahead of the numbers achieved by the teams at the most recent Amazon Robotics Challenge (see ‘A measure of success’).
Goldberg is working to create the product that will do its best at assisting humans, but not as to put anyone out of a job. The development of capable robots with a successful rate of dexterity is steadily becoming reality.