Bimanual Deformable Bag Manipulation Using a Structure-of-Interest Based Neural Dynamics Model

The manipulation of deformable objects by robotic systems presents significant challenges due to their complex dynamics and infinite-dimensional configuration spaces. This article introduces a novel approach to deformable object manipulation (DOM) by emphasizing the introduction and manipulation of structures of interest (SOIs) in deformable fabric bags. We propose a bimanual manipulation framework that leverages a graph neural network-based neural dynamics model to succinctly represent and predict the behavior of these SOIs. Our approach involves global particle sampling process to construct a particle representation from partial point clouds of the SOIs and learning the neural dynamics model that effectively captures the essential deformations of the SOIs for fabric bags. By integrating this neural dynamics model with model predictive control, we enable robotic manipulators to perform precise and stable manipulation tasks focused on the SOIs. We validate our new framework through various experiments that demonstrate its efficacy in manipulating deformable bags and T-shirts. Our contributions not only address the complexities inherent in DOM, but also provide new perspectives and methodologies for enhancing robotic interactions with deformable materials by concentrating on their critical structural elements.