Filters








3 Hits in 1.5 sec

PlasticineLab: A Soft-Body Manipulation Benchmark with Differentiable Physics [article]

Zhiao Huang, Yuanming Hu, Tao Du, Siyuan Zhou, Hao Su, Joshua B. Tenenbaum, Chuang Gan
2021 arXiv   pre-print
We introduce a new differentiable physics benchmark called PasticineLab, which includes a diverse collection of soft body manipulation tasks.  ...  We expect that PlasticineLab will encourage the development of novel algorithms that combine differentiable physics and RL for more complex physics-based skill learning tasks.  ...  Yuanming Hu is partly supported by a Facebook research fellowship and an Adobe research fellowship.  ... 
arXiv:2104.03311v1 fatcat:tmp23xtv65d5niikoes3bcd22m

Contact Points Discovery for Soft-Body Manipulations with Differentiable Physics [article]

Sizhe Li, Zhiao Huang, Tao Du, Hao Su, Joshua B. Tenenbaum, Chuang Gan
2022 arXiv   pre-print
Differentiable physics has recently been shown as a powerful tool for solving soft-body manipulation tasks.  ...  To evaluate the effectiveness of our method, we introduce PlasticineLab-M that extends the existing differentiable physics benchmark PlasticineLab to seven new challenging multi-stage soft-body manipulation  ...  To extensively evaluate our method, we introduce PlasticineLab-M, a dataset that extends the existing differentiable physics benchmark PlasticineLab with seven new challenging multi-stage soft-body manipulation  ... 
arXiv:2205.02835v1 fatcat:mmy44gmryrdopmablhrwdlksbe

DiffSRL: Learning Dynamic-aware State Representation for Deformable Object Control with Differentiable Simulator [article]

Sirui Chen, Yunhao Liu, Jialong Li, Shang Wen Yao, Tingxiang Fan, Jia Pan
2021 arXiv   pre-print
We further establish a state representation learning benchmark on a soft-body simulation system, PlasticineLab, and our model demonstrates superior performance in terms of capturing long-term dynamics  ...  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.  ...  benchmark with differentiable simulation in PlasticineLab [6] .  ... 
arXiv:2110.12352v1 fatcat:tz4kixrqmreejlum4hmhrftbpi