Teaching robots to handle a knife on soft objects is easier said than done due to the complexity and predictability of tissue under manipulation. Artūras Straižys, Michael Burke, and Subramanian Ramamoorthy have come up with a novel approach to handle such problem. They used a “binary medium classifier trained using joint torque measurements, and a closed loop control law that relies on an error signal compactly encoded in the decision boundary of the classifier.”
A robot was used to cut a grapefruit with precision by following the boundary between grapefruit pulp and peel. The above video shows it in action.