Learning to Grasp Unknown Objects Based on 3D Edge Information

In this work we re ne an initial grasping behavior based on 3D edge information by learning. Based on a set of autonomously generated evaluated grasps and relations between the semi-global 3D edges, a prediction function is learned that computes a likelihood for the success of a grasp using either an of me or an online learning scheme. Both methods are implemented using a hybrid arti cial neural network containing standard nodes with a sigmoid activation function and nodes with a radial basis function. We show that a signi cant performance improvement can be achieved.