Humans can toss casually, but the action of throwing things is actually very sophisticated. The robot ” TossingBot ” that can learn the motion and physics of throwing random things using machine learning.
Predictive throwing is not easy – even for humans. Grip, posture, mass, air resistance, friction, aerodynamics, and countless other variables can affect the trajectory of an object.
Tossingbot has been developed by Andy Zeng, Student researcher at Google Robotics, in collaboration with Princeton University, Columbia University and Massachusetts Institute of Technology.
Because it is possible to move things faster than moving and placing one by one, throwing objects is a very efficient way of moving objects. If the robot arm can throw things properly, it may be useful in many situations such as picking at a delivery site, in a factory, or at a disaster site.
Learn more: http://tossingbot.cs.princeton.edu/
