mlsanityData to 14 Oct 2026

Dexterous Manipulation Based on Prior Dexterous Grasp Pose Knowledge

arXiv 2412.15587 ↗First posted 20 Dec 2024cs.RO cs.LGcited by 2

On the map: Deep reinforcement learning methods → RL for robot locomotion and manipulation

Abstract (arXiv)

Dexterous manipulation has received considerable attention in recent research. Predominantly, existing studies have concentrated on reinforcement learning methods to address the substantial degrees of freedom in hand movements. Nonetheless, these methods typically suffer from low efficiency and accuracy. In this work, we introduce a novel reinforcement learning approach that leverages prior dexterous grasp pose knowledge to enhance both efficiency and accuracy. Unlike previous work, they always make the robotic hand go with a fixed dexterous grasp pose, We decouple the manipulation process into two distinct phases: initially, we generate a dexterous grasp pose targeting the functional part of the object; after that, we employ reinforcement learning to comprehensively explore the environment. Our findings suggest that the majority of learning time is expended in identifying the appropriate initial position and selecting the optimal manipulation viewpoint. Experimental results demonstrate significant improvements in learning efficiency and success rates across four distinct tasks.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods. Builds on 37 of its 46 references that are in the map. First posted on arXiv 20 Dec 2024.

Builds on

37 of 46 references.

References · 37

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  3. 2017Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrationscited by 715
  4. 2018Learning Dexterous In-Hand Manipulationcited by 687
  5. 2020SAPIEN: A SimulAted Part-based Interactive ENvironmentcited by 595
  6. 2021Bridge Data: Boosting Generalization of Robotic Skills with Cross-Domain Datasetscited by 250
  7. 2022AnyGrasp: Robust and Efficient Grasp Perception in Spatial and Temporal Domainscited by 208
  8. 2019Deep Dynamics Models for Learning Dexterous Manipulationcited by 206
  9. 2024DexCap: Scalable and Portable Mocap Data Collection System for Dexterous Manipulationcited by 160
  10. 2023UniDexGrasp: Universal Robotic Dexterous Grasping via Learning Diverse Proposal Generation and Goal-Conditioned Policycited by 124
  11. 2021Learning Generalizable Robotic Reward Functions from "In-The-Wild" Human Videoscited by 119
  12. 2017Grasp Pose Detection in Point Cloudscited by 111
25 more
  1. 2023UniDexGrasp++: Improving Dexterous Grasping Policy Learning via Geometry-aware Curriculum and Iterative Generalist-Specialist Learningcited by 107
  2. 2022VideoDex: Learning Dexterity from Internet Videoscited by 106
  3. 2019AVID: Learning Multi-Stage Tasks via Pixel-Level Translation of Human Videoscited by 105
  4. 2022Visual Dexterity: In-Hand Reorientation of Novel and Complex Object Shapescited by 101
  5. 2021Reset-Free Reinforcement Learning via Multi-Task Learning: Learning Dexterous Manipulation Behaviors without Human Interventioncited by 95
  6. 2021Coarse-to-Fine Imitation Learning: Robot Manipulation from a Single Demonstrationcited by 93
  7. 2022DexVIP: Learning Dexterous Grasping with Human Hand Pose Priors from Videocited by 89
  8. 2019ContactGrasp: Functional Multi-finger Grasp Synthesis from Contactcited by 87
  9. 2020Learning Dexterous Grasping with Object-Centric Visual Affordancescited by 85
  10. 2022DexPoint: Generalizable Point Cloud Reinforcement Learning for Sim-to-Real Dexterous Manipulationcited by 84
  11. 2023Rotating without Seeing: Towards In-hand Dexterity through Touchcited by 84
  12. 2020Reinforcement Learning with Videos: Combining Offline Observations with Interactioncited by 82
  13. 2022From One Hand to Multiple Hands: Imitation Learning for Dexterous Manipulation from Single-Camera Teleoperationcited by 82
  14. 2022Dexterous Imitation Made Easy: A Learning-Based Framework for Efficient Dexterous Manipulationcited by 78
  15. 2023Teach a Robot to FISH: Versatile Imitation from One Minute of Demonstrationscited by 75
  16. 2023DexArt: Benchmarking Generalizable Dexterous Manipulation with Articulated Objectscited by 66
  17. 2022Holo-Dex: Teaching Dexterity with Immersive Mixed Realitycited by 57
  18. 2022Learning Generalizable Dexterous Manipulation from Human Grasp Affordancecited by 51
  19. 2024Graspness Discovery in Clutters for Fast and Accurate Grasp Detectioncited by 48
  20. 2021Transferring Dexterous Manipulation from GPU Simulation to a Remote Real-World TriFingercited by 38
  21. 2020Physics-Based Dexterous Manipulations with Estimated Hand Poses and Residual Reinforcement Learningcited by 37
  22. 2019Learning Task-Oriented Grasping from Human Activity Datasetscited by 32
  23. 2023Sampling-based Exploration for Reinforcement Learning of Dexterous Manipulationcited by 24
  24. 2020Learning Dexterous Manipulation from Suboptimal Expertscited by 23
  25. 2022Learning Reward Functions for Robotic Manipulation by Observing Humanscited by 17

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Adaptive Vision-Language Grasping via Composable Foundation Priors and Generalizable Grasp Synthesiscited by 1
  2. 2026PhotoHOI: Synthesizing 3D Hand-Object Interactions from a Single RGB Photographcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Dexterous Manipulation through Imitation Learning: A Survey
  2. 2024Cross-Embodiment Dexterous Grasping with Reinforcement Learning
  3. 2026Learning Stable In-Grasp Manipulation in a Non-Dropping Action Space ↗
  4. 2025Dexterous Grasping with Real-World Robotic Reinforcement Learning
  5. 2025DemoGrasp: Universal Dexterous Grasping from a Single Demonstration
  6. 2024Benchmarking Reinforcement Learning Methods for Dexterous Robotic Manipulation with a Three-Fingered Gripper
  7. 2025Towards Human-level Dexterity via Robot Learning
  8. 2026LAMP: Latent Motion Prior-Guided Real-World Learning for Dexterous Hand Manipulation