PlaNet-Pick: Effective Cloth Flattening Based on Latent Dynamic Planning
Why do Recurrent State Space Models such as PlaNet fail at cloth manipulation tasks? Recent work has attributed this to the blurry reconstruction of the observation, which makes it difficult to plan directly in the latent space. This project explores the reasons behind this by applying PlaNet in the pick-and-place cloth-flattening domain.
Keywords
Robotics, Cloth-like Deformable Object, Reinforcement Learning, Predictive Processing
Staff
[Kasim Terzic]{kt54} and Abudureyimu Halite