DexNet 4.0 in IEEE Spectrum
Ken Goldberg's team of is getting some more well deserved press for Dex-Net 4.0, this time from the Institute of Electrical and Electronics Engineers' Spectrum magazine. IEEE Spectrum's article "Dex-Net 4.0 Enables 'Ambidextrous' Robots to Choose Best Gripper for the Job" summarizes the recent paper on Dex-Net published in Science Robotics. “Deep Learning of Ambidextrous Robot Grasping Policies for Universal Picking" was co-authored by Jeffrey Mahler, Matthew Matl, Vishal Satish, Mike Danielczuk, Bill DeRose, Stephen McKinley, and Ken Goldberg.
From the article:
The new and exciting bit about this latest version of Dex-Net is that it’s able to successfully grasp 95 percent of unseen objects at a rate of 300 per hour, thanks to some added ambidexterity that lets the robot dynamically choose between two different kinds of grippers.
Read the rest of the article here.