Learning Visible Connectivity Dynamics for Cloth Smoothing
This work addresses the problem of robotic cloth manipulation for robotics, offering a novel approach that improves performance and enables real-world application, though it builds incrementally on existing particle-based and connectivity inference ideas.
The paper tackles the challenge of robotic cloth manipulation by learning a particle-based dynamics model from partial point cloud observations, inferring visible connectivity to overcome partial observability. The method outperforms previous state-of-the-art approaches in simulation and demonstrates successful zero-shot sim-to-real transfer on a Franka arm, smoothing various cloth types from crumpled configurations.
Robotic manipulation of cloth remains challenging for robotics due to the complex dynamics of the cloth, lack of a low-dimensional state representation, and self-occlusions. In contrast to previous model-based approaches that learn a pixel-based dynamics model or a compressed latent vector dynamics, we propose to learn a particle-based dynamics model from a partial point cloud observation. To overcome the challenges of partial observability, we infer which visible points are connected on the underlying cloth mesh. We then learn a dynamics model over this visible connectivity graph. Compared to previous learning-based approaches, our model poses strong inductive bias with its particle based representation for learning the underlying cloth physics; it is invariant to visual features; and the predictions can be more easily visualized. We show that our method greatly outperforms previous state-of-the-art model-based and model-free reinforcement learning methods in simulation. Furthermore, we demonstrate zero-shot sim-to-real transfer where we deploy the model trained in simulation on a Franka arm and show that the model can successfully smooth different types of cloth from crumpled configurations. Videos can be found on our project website.