Learning visual-based deformable object rearrangement with local graph neural networks
Authors:
- Yuhong DengYDYuhong Deng
- Xueqian WangXWXueqian Wang
- Lipeng ChenLCLipeng Chen
Preprint
Abstract
Goal-conditioned rearrangement of deformable objects (e.g. straightening a rope and folding a cloth) is one of the most common deformable manipulation tasks, where the robot needs to rearrange a deformable object into a prescribed goal configuration with only visual observations. These tasks are typically confronted with two main challenges: the high dimensionality of deformable configuration space and the underlying complexity, nonlinearity and uncertainty inherent in deformable dynamics. To address these challenges, we propose a novel representation strategy that can efficiently model the deformable object states with a set of keypoints and their interactions. We further propose local-graph neural network (GNN), a light local GNN learning to jointly model the deformable rearrangement dynamics and infer the optimal manipulation actions (e.g. pick and place) by constructing and updating two dynamic graphs. Both simulated and real experiments have been conducted to demonstrate that the proposed dynamic graph representation shows superior expressiveness in modeling deformable rearrangement dynamics. Our method reaches much higher success rates on a variety of deformable rearrangement tasks (96.3% on average) than state-of-the-art method in simulation experiments. Besides, our method is much more lighter and has a 60% shorter inference time than state-of-the-art methods. We also demonstrate that our method performs well in the multi-task learning scenario and can be transferred to real-world applications with an average success rate of 95% by solely fine tuning a keypoint detector.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Robotic dexterous manipulation: from tele-operation to autonomous learning and adaptive controlComplex & Intelligent Systems, 2022
- Offline-Online Learning of Deformation Model for Cable Manipulation With Graph Neural NetworksIEEE Robotics and Automation Letters, 2022
- Challenges and Outlook in Robotic Manipulation of Deformable ObjectsIEEE Robotics & Automation Magazine, 2022
- GraspVDN: scene-oriented grasp estimation by learning vector representations of graspsComplex & Intelligent Systems, 2021
- Modeling, learning, perception, and control methods for deformable object manipulationScience Robotics, 2021
- Dynamic Manipulation of Deformable Objects With Implicit IntegrationIEEE Robotics and Automation Letters, 2021
- Track deformable objects from point clouds with structure preserved registrationThe International Journal of Robotics Research, 2019
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding2019
- A geometric approach to robotic laundry foldingThe International Journal of Robotics Research, 2011