# Dextrous Tactile In-Hand Manipulation Using a Modular Reinforcement Learning Architecture

arXiv: https://arxiv.org/abs/2303.04705 · HTML: https://mlsanity.com/paper/2303.04705/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Mar 2023.

## Built on by (most cited first, 20 of 20)

- [DexCap: Scalable and Portable Mocap Data Collection System for Dexterous Manipulation](https://mlsanity.com/paper/2403.07788/) (2024)
- [General In-Hand Object Rotation with Vision and Touch](https://mlsanity.com/paper/2309.09979/) (2023)
- [DexGraspVLA: A Vision-Language-Action Framework Towards General Dexterous Grasping](https://mlsanity.com/paper/2502.20900/) (2025)
- [Rotating without Seeing: Towards In-hand Dexterity through Touch](https://mlsanity.com/paper/2303.10880/) (2023)
- [Sequential Dexterity: Chaining Dexterous Policies for Long-Horizon Manipulation](https://mlsanity.com/paper/2309.00987/) (2023)
- [Twisting Lids Off with Two Hands](https://mlsanity.com/paper/2403.02338/) (2024)
- [Lessons from Learning to Spin "Pens"](https://mlsanity.com/paper/2407.18902/) (2024)
- [AnyRotate: Gravity-Invariant In-Hand Object Rotation with Sim-to-Real Touch](https://mlsanity.com/paper/2405.07391/) (2024)
- [CordViP: Correspondence-based Visuomotor Policy for Dexterous Manipulation in Real-World](https://mlsanity.com/paper/2502.08449/) (2025)
- [METIS: Multi-Source Egocentric Training for Integrated Dexterous Vision-Language-Action Model](https://mlsanity.com/paper/2511.17366/) (2025)
- [Estimator-Coupled Reinforcement Learning for Robust Purely Tactile In-Hand Manipulation](https://mlsanity.com/paper/2311.04060/) (2023)
- [Vegetable Peeling: A Case Study in Constrained Dexterous Manipulation](https://mlsanity.com/paper/2407.07884/) (2024)
- [ViTaS: Visual Tactile Soft Fusion Contrastive Learning for Visuomotor Learning](https://mlsanity.com/paper/2602.11643/) (2026)
- [R×R: Rapid eXploration for Reinforcement Learning via Sampling-based Reset Distributions and Imitation Pre-training](https://mlsanity.com/paper/2401.15484/) (2024)
- [VTAO-BiManip: Masked Visual-Tactile-Action Pre-training with Object Understanding for Bimanual Dexterous Manipulation](https://mlsanity.com/paper/2501.03606/) (2025)
- [Composing Dextrous Grasping and In-hand Manipulation via Scoring with a Reinforcement Learning Critic](https://mlsanity.com/paper/2505.13253/) (2025)
- [Learning Time-Optimal and Speed-Adjustable Tactile In-Hand Manipulation](https://mlsanity.com/paper/2411.13148/) (2024)
- [Robust In-Hand Manipulation via Priors in Reinforcement Learning and Mechanical Design](https://mlsanity.com/paper/2607.12105/) (2026)
- [The Role of Tactile Sensing for Learning Reach and Grasp](https://mlsanity.com/paper/2502.20367/) (2025)
- [Estimate, Don't Imitate: Reusing Differentiable State-Based Policies for Visuomotor Control](https://mlsanity.com/paper/2609.34018/) (2026)

## Abstract (arXiv)

Dextrous in-hand manipulation with a multi-fingered robotic hand is a challenging task, esp. when performed with the hand oriented upside down, demanding permanent force-closure, and when no external sensors are used. For the task of reorienting an object to a given goal orientation (vs. infinitely spinning it around an axis), the lack of external sensors is an additional fundamental challenge as the state of the object has to be estimated all the time, e.g., to detect when the goal is reached. In this paper, we show that the task of reorienting a cube to any of the 24 possible goal orientations in a π/2-raster using the torque-controlled DLR-Hand II is possible. The task is learned in simulation using a modular deep reinforcement learning architecture: the actual policy has only a small observation time window of 0.5s but gets the cube state as an explicit input which is estimated via a deep differentiable particle filter trained on data generated by running the policy. In simulation, we reach a success rate of 92% while applying significant domain randomization. Via zero-shot Sim2Real-transfer on the real robotic system, all 24 goal orientations can be reached with a high success rate.