UBSoft: A Simulation Platform for Robotic Skill Learning in Unbounded Soft Environments

Chunru Lin, Jugang Fan, Yian Wang, Zeyuan Yang, Zhehuan Chen, Lixing Fang, Tsun-Hsuan Wang, Zhou Xian, Chuang Gan
Proceedings of The 8th Conference on Robot Learning, PMLR 270:4879-4894, 2025.

Abstract

It is desired to equip robots with the capability of interacting with various soft materials as they are ubiquitous in the real world. While physics simulations are one of the predominant methods for data collection and robot training, simulating soft materials presents considerable challenges. Specifically, it is significantly more costly than simulating rigid objects in terms of simulation speed and storage requirements. These limitations typically restrict the scope of studies on soft materials to small and bounded areas, thereby hindering the learning of skills in broader spaces. To address this issue, we introduce UBSoft, a new simulation platform designed to support unbounded soft environments for robot skill acquisition. Our platform utilizes spatially adaptive resolution scales, where simulation resolution dynamically adjusts based on proximity to active robotic agents. Our framework markedly reduces the demand for extensive storage space and computation costs required for large-scale scenarios involving soft materials. We also establish a set of benchmark tasks in our platform, including both locomotion and manipulation tasks, and conduct experiments to evaluate the efficacy of various reinforcement learning algorithms and trajectory optimization techniques, both gradient-based and sampling-based. Preliminary results indicate that sampling-based trajectory optimization generally achieves better results for obtaining one trajectory to solve the task. Additionally, we conduct experiments in real-world environments to demonstrate that advancements made in our UBSoft simulator could translate to improved robot interactions with large-scale soft material. More videos can be found at https://ubsoft24.github.io.

Cite this Paper

BibTeX
@InProceedings{pmlr-v270-lin25b, title = {UBSoft: A Simulation Platform for Robotic Skill Learning in Unbounded Soft Environments}, author = {Lin, Chunru and Fan, Jugang and Wang, Yian and Yang, Zeyuan and Chen, Zhehuan and Fang, Lixing and Wang, Tsun-Hsuan and Xian, Zhou and Gan, Chuang}, booktitle = {Proceedings of The 8th Conference on Robot Learning}, pages = {4879--4894}, year = {2025}, editor = {Agrawal, Pulkit and Kroemer, Oliver and Burgard, Wolfram}, volume = {270}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v270/main/assets/lin25b/lin25b.pdf}, url = {https://proceedings.mlr.press/v270/lin25b.html}, abstract = {It is desired to equip robots with the capability of interacting with various soft materials as they are ubiquitous in the real world. While physics simulations are one of the predominant methods for data collection and robot training, simulating soft materials presents considerable challenges. Specifically, it is significantly more costly than simulating rigid objects in terms of simulation speed and storage requirements. These limitations typically restrict the scope of studies on soft materials to small and bounded areas, thereby hindering the learning of skills in broader spaces. To address this issue, we introduce UBSoft, a new simulation platform designed to support unbounded soft environments for robot skill acquisition. Our platform utilizes spatially adaptive resolution scales, where simulation resolution dynamically adjusts based on proximity to active robotic agents. Our framework markedly reduces the demand for extensive storage space and computation costs required for large-scale scenarios involving soft materials. We also establish a set of benchmark tasks in our platform, including both locomotion and manipulation tasks, and conduct experiments to evaluate the efficacy of various reinforcement learning algorithms and trajectory optimization techniques, both gradient-based and sampling-based. Preliminary results indicate that sampling-based trajectory optimization generally achieves better results for obtaining one trajectory to solve the task. Additionally, we conduct experiments in real-world environments to demonstrate that advancements made in our UBSoft simulator could translate to improved robot interactions with large-scale soft material. More videos can be found at https://ubsoft24.github.io.} }
Endnote
%0 Conference Paper %T UBSoft: A Simulation Platform for Robotic Skill Learning in Unbounded Soft Environments %A Chunru Lin %A Jugang Fan %A Yian Wang %A Zeyuan Yang %A Zhehuan Chen %A Lixing Fang %A Tsun-Hsuan Wang %A Zhou Xian %A Chuang Gan %B Proceedings of The 8th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Pulkit Agrawal %E Oliver Kroemer %E Wolfram Burgard %F pmlr-v270-lin25b %I PMLR %P 4879--4894 %U https://proceedings.mlr.press/v270/lin25b.html %V 270 %X It is desired to equip robots with the capability of interacting with various soft materials as they are ubiquitous in the real world. While physics simulations are one of the predominant methods for data collection and robot training, simulating soft materials presents considerable challenges. Specifically, it is significantly more costly than simulating rigid objects in terms of simulation speed and storage requirements. These limitations typically restrict the scope of studies on soft materials to small and bounded areas, thereby hindering the learning of skills in broader spaces. To address this issue, we introduce UBSoft, a new simulation platform designed to support unbounded soft environments for robot skill acquisition. Our platform utilizes spatially adaptive resolution scales, where simulation resolution dynamically adjusts based on proximity to active robotic agents. Our framework markedly reduces the demand for extensive storage space and computation costs required for large-scale scenarios involving soft materials. We also establish a set of benchmark tasks in our platform, including both locomotion and manipulation tasks, and conduct experiments to evaluate the efficacy of various reinforcement learning algorithms and trajectory optimization techniques, both gradient-based and sampling-based. Preliminary results indicate that sampling-based trajectory optimization generally achieves better results for obtaining one trajectory to solve the task. Additionally, we conduct experiments in real-world environments to demonstrate that advancements made in our UBSoft simulator could translate to improved robot interactions with large-scale soft material. More videos can be found at https://ubsoft24.github.io.
APA
Lin, C., Fan, J., Wang, Y., Yang, Z., Chen, Z., Fang, L., Wang, T., Xian, Z. & Gan, C.. (2025). UBSoft: A Simulation Platform for Robotic Skill Learning in Unbounded Soft Environments. Proceedings of The 8th Conference on Robot Learning, in Proceedings of Machine Learning Research 270:4879-4894 Available from https://proceedings.mlr.press/v270/lin25b.html.

Related Material