Researchers at the Georgia Institute of Technology (Georgia Tech) are working on making the process of a robot autonomously grasping an object easier.
Their new method involves eliminating the manual positioning of objects by humans, and using a friendly, interactive interface that handles everything with only one or two clicks.
In addition to the conventional, full manual approach and the point-and click grasping approach, the Georgia Tech researchers also implemented a middle-ground "constrained positioning" method, which intelligently limits the amount of degrees of freedom that a user needs to position. Instead, the user needs to select only a grasp point, approach angle, and grasp depth. When these factors are combined, they create a spectrum of options for teleoperated grasping.
As the autonomy of these grasping approaches improves, the system makes increasing use of scene information, making it easy to scale up and deploy to new environments without training.
From IEEE Spectrum
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA