Transferring Grasp Configurations using Active Learning and Local Replanning
This addresses the challenge of robotic grasping for novel objects, though it is incremental as it builds on prior grasp transfer methods with specific assumptions.
The paper tackles the problem of transferring grasp configurations from example objects to novel objects with similar topology and shape by performing 3D segmentation, active learning for grasp space computation, bijective contact mapping, and local replanning, resulting in a general approach applicable to various object representations and robotic hands.
We present a new approach to transfer grasp configurations from prior example objects to novel objects. We assume the novel and example objects have the same topology and similar shapes. We perform 3D segmentation on these objects using geometric and semantic shape characteristics. We compute a grasp space for each part of the example object using active learning. We build bijective contact mapping between these model parts and compute the corresponding grasps for novel objects. Finally, we assemble the individual parts and use local replanning to adjust grasp configurations while maintaining its stability and physical constraints. Our approach is general, can handle all kind of objects represented using mesh or point cloud and a variety of robotic hands.