SpeedFolding: Learning Efficient Bimanual Folding of Garments
Open accessAbstract
Folding garments reliably and efficiently is a long standing challenge in robotic manipulation due to the complex dynamics and high dimensional configuration space of garments. An intuitive approach is to initially manipulate the garment to a canonical smooth configuration before folding. In this work, we develop SpeedFolding, a reliable and efficient bimanual system, which given user-defined instructions as folding lines, manipulates an initially crumpled garment to (1) a smoothed and (2) a folded configuration. Our primary contribution is a novel neural network architecture that is able to predict pairs of gripper poses to parameterize a diverse set of bimanual action primitives. After learning from 4300 human- annotated and self-supervised actions, the robot is able to fold garments from a random initial configuration in under 120 s on average with a success rate of 93 %. Real-world experiments show that the system is able to generalize to unseen garments of different color, shape, and stiffness. While prior work achieved 3–6 Folds Per Hour (FPH), SpeedFolding achieves 30–40 FPH. See https://pantor.github.io/speedfolding for code, videos, and datasets.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The number of bins needed to approximate a continuous action space grows exponentially with increasing dimensionality.
“However, the number of bins needed to approximate a continuous action space grows exponentially with increasing dimensionality.”
From Diffusion Policy: Visuomotor Policy Learning via Action Diffusion (2023) · cites this paper as Avigal et al. (2022)
Cites
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404 · see citing works
Learning Synergies Between Pushing and Grasping with Self-Supervised Deep Reinforcement Learning
Zeng, A., Song, S., Welker, S., Lee, J., Rodriguez, A., & Funkhouser, T. (2018) · 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2018.8593986 · Cited by 606 · see citing works
Maitin-Shepard, J., Cusumano-Towner, M., Lei, J., & Abbeel, P. (2010) · 2010 IEEE International Conference on Robotics and Automation · doi:10.1109/robot.2010.5509439 · Cited by 418 · see citing works
Seita, D., Florence, P., Tompson, J., Coumans, E., Sindhwani, V., Goldberg, K., & Zeng, A. (2021) · 2021 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra48506.2021.9561391 · Cited by 130 · see citing works
Self-citationFabricFlowNet: Bimanual Cloth Manipulation with a Flow-based Policy
Weng, T., Bajracharya, S., Wang, Y., Agrawal, K., & Held, D. (2021) · arXiv · doi:10.48550/arxiv.2111.05623 · Cited by 48 · see citing works
Learning Arbitrary-Goal Fabric Folding with One Hour of Real Robot Experience
Lee, R., Ward, D., Cosgun, A., Dasagi, V., Corke, P., & Leitner, J. (2020) · arXiv · doi:10.48550/arxiv.2010.03209 · Cited by 47 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Diffusion policy: Visuomotor policy learning via action diffusion
Chi, C., Xu, Z., Feng, S., Cousineau, E., Du, Y., Burchfiel, B., Tedrake, R., & Song, S. (2025) · The International Journal of Robotics Research · doi:10.1177/02783649241273668 · Cited by 613 · see citing works
“A popular approach to model multimodal action distributions while maintaining the simplicity of direction action mapping is convert the regression task into classi fi cation by discretizing the action space (Zeng et al., 2021; Wu et al., 2020; Avigal et al., 2022).”
InfluentialDiffusion Policy: Visuomotor Policy Learning via Action Diffusion
Chi, C., Feng, S., Du, Y., Xu, Z., Cousineau, E., Burchfiel, B., & Song, S. (2023) · Robotics: Science and Systems XIX · doi:10.15607/rss.2023.xix.026 · Cited by 530 · see citing works
Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
Chi, C., Xu, Z., Feng, S., Cousineau, E., Du, Y., Burchfiel, B., Tedrake, R., & Song, S. (2023) · arXiv · doi:10.48550/arxiv.2303.04137 · Cited by 12 · see citing works
“A popular approach to model multimodal action distributions while maintaining the simplicity of direction action mapping is convert the regression task into classi fi cation by discretizing the action space (Zeng et al., 2021; Wu et al., 2020; Avigal et al., 2022).”
InfluentialGoal-Conditioned Model Simplification for 1-D and 2-D Deformable Object Manipulation
Wang, S., Leonetti, M., & Dogar, M. (2025) · IEEE Transactions on Robotics · doi:10.1109/tro.2025.3577052 · Cited by 4 · see citing works
- Authors
- Avigal, Y., Berscheid, L., Asfour, T., Kröger, T., & Goldberg, K.
- Affiliations
- University of California, Berkeley (US)
- Karlsruhe Institute of Technology (DE)
- Venue
- 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
- Publisher
- IEEE
- Published
- 2022-10-23
- Pages
- 1-8
- ISSN
- 2153-0858 (ISSN-L)
- OpenAlex
- W4312610897
- Cited by
- 95
- References
- 6
- Citations by year
- 2026: 13 · 2025: 30 · 2024: 26 · 2023: 25 · 2022: 1
- FWCI
- 31.29
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Engineering › Control and Systems Engineering › Robot Manipulation and Learning
- Physical Sciences › Engineering › Biomedical Engineering › Soft Robotics and Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Keywords
- garment folding, bimanual manipulation, robotic manipulation, neural network architecture, self-supervised learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
