1 Hit in 2.7 sec

DiffSRL: Learning Dynamical State Representation for Deformable Object Manipulation with Differentiable Simulator [article]

Sirui Chen, Yunhao Liu, Jialong Li, Shang Wen Yao, Tingxiang Fan, Jia Pan
2022 arXiv   pre-print
We propose DiffSRL, a dynamic state representation learning pipeline utilizing differentiable simulation that can embed complex dynamics models as part of the end-to-end training.  ...  However, current dynamic state representation learning methods scale poorly on complex dynamic systems such as deformable objects, and cannot directly embed well defined simulation function into the training  ...  To improve the performance of the state representation learning on deformable objects, we propose a new pipeline, DiffSRL, that utilizes a differentiable simulator to encode dynamic and constraint-related  ... 
arXiv:2110.12352v2 fatcat:rtdu2v2nznf23alsgznc7vvw4e