# Untangling Dense Knots by Learning Task-Relevant Keypoints

arXiv: https://arxiv.org/abs/2011.04999 · HTML: https://mlsanity.com/paper/2011.04999/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Nov 2020.

## Built on by (most cited first, 21 of 21)

- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [Multi-Stage Cable Routing through Hierarchical Imitation Learning](https://mlsanity.com/paper/2307.08927/) (2023)
- [SpeedFolding: Learning Efficient Bimanual Folding of Garments](https://mlsanity.com/paper/2208.10552/) (2022)
- [Stabilize to Act: Learning to Coordinate for Bimanual Manipulation](https://mlsanity.com/paper/2309.01087/) (2023)
- [VoxAct-B: Voxel-Based Acting and Stabilizing Policy for Bimanual Manipulation](https://mlsanity.com/paper/2407.04152/) (2024)
- [Learning Bimanual Scooping Policies for Food Acquisition](https://mlsanity.com/paper/2211.14652/) (2022)
- [ScrewMimic: Bimanual Imitation from Human Videos with Screw Space Projection](https://mlsanity.com/paper/2405.03666/) (2024)
- [SimNet: Enabling Robust Unknown Object Manipulation from Pure Synthetic Data via Stereo](https://mlsanity.com/paper/2106.16118/) (2021)
- [AnyBimanual: Transferring Unimanual Policy for General Bimanual Manipulation](https://mlsanity.com/paper/2412.06779/) (2024)
- [Untangling Dense Non-Planar Knots by Learning Manipulation Features and Recovery Policies](https://mlsanity.com/paper/2107.08942/) (2021)
- [A Bayesian Treatment of Real-to-Sim for Deformable Object Manipulation](https://mlsanity.com/paper/2112.05068/) (2021)
- [Autonomously Untangling Long Cables](https://mlsanity.com/paper/2207.07813/) (2022)
- [Deep Reinforcement Learning Based on Local GNN for Goal-conditioned Deformable Object Rearranging](https://mlsanity.com/paper/2302.10446/) (2023)
- [Disentangling Dense Multi-Cable Knots](https://mlsanity.com/paper/2106.02252/) (2021)
- [Learning Sequential Acquisition Policies for Robot-Assisted Feeding](https://mlsanity.com/paper/2309.05197/) (2023)
- [SGTM 2.0: Autonomously Untangling Long Cables using Interactive Perception](https://mlsanity.com/paper/2209.13706/) (2022)
- [All You Need is LUV: Unsupervised Collection of Labeled Images using Invisible UV Fluorescent Indicators](https://mlsanity.com/paper/2203.04566/) (2022)
- [Graph-Transporter: A Graph-based Learning Method for Goal-Conditioned Deformable Object Rearranging Task](https://mlsanity.com/paper/2302.10445/) (2023)
- [Deformable Cluster Manipulation via Whole-Arm Policy Learning](https://mlsanity.com/paper/2507.17085/) (2025)
- [DexKnot: Generalizable Visuomotor Policy Learning for Dexterous Bag-Knotting Manipulation](https://mlsanity.com/paper/2603.07136/) (2026)
- [Learning Sim-Grounded Policies for Bimanual Rope Manipulation from Human Teleoperation Data](https://mlsanity.com/paper/2605.16043/) (2026)

## Abstract (arXiv)

Untangling ropes, wires, and cables is a challenging task for robots due to the high-dimensional configuration space, visual homogeneity, self-occlusions, and complex dynamics. We consider dense (tight) knots that lack space between self-intersections and present an iterative approach that uses learned geometric structure in configurations. We instantiate this into an algorithm, HULK: Hierarchical Untangling from Learned Keypoints, which combines learning-based perception with a geometric planner into a policy that guides a bilateral robot to untangle knots. To evaluate the policy, we perform experiments both in a novel simulation environment modelling cables with varied knot types and textures and in a physical system using the da Vinci surgical robot. We find that HULK is able to untangle cables with dense figure-eight and overhand knots and generalize to varied textures and appearances. We compare two variants of HULK to three baselines and observe that HULK achieves 43.3% higher success rates on a physical system compared to the next best baseline. HULK successfully untangles a cable from a dense initial configuration containing up to two overhand and figure-eight knots in 97.9% of 378 simulation experiments with an average of 12.1 actions per trial. In physical experiments, HULK achieves 61.7% untangling success, averaging 8.48 actions per trial. Supplementary material, code, and videos can be found at https://tinyurl.com/y3a88ycu.