The team demonstrated its system by commissioning a two-finger robot, RiceGrip, to transform malleable foam into a desired shape much like you could make sushi. It used a depth camera and object recognition to identify the foam, and used the model to introduce the foam as a dynamic graph of malleable materials. While it already had an idea of how the particles would react, it would adjust its model if the "sushi" behaved in a way that it did not expect.
It's still early, and scientists want to improve their approach They're using partially observable situations (for example, knowing how a stack of boxes will fall.) They also want them to work directly with images, and if that happens, it could But Making Breakthroughs for Robots It's easier to manipulate virtually any type of object, even if liquids or soft solids could make it difficult to pinpoint results in advance.While robots can not replace sushi chefs in the foreseeable future, MIT's learning method offers the prospect much more realistic.