Grasp-Taxonomy Conditioning as a Mode Switch for Dexterous Grasping of Climbing Holds
cs.RO, cs.AI
Submitted: 2026-11-09
Updated: 2026-11-09
Comments: 6 pages, 4 figures, 2 tables
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
The gist: Dexterous grasping in robotics relies on distinct hand configurations depending on object geometry.
Terminology
Abstract
Dexterous grasping in robotics relies on distinct hand configurations depending on object geometry. We introduce BetaGrasp, a 3D diffusion policy that conditions a DP3-style point-cloud policy on a four-class grasp label (crimp, sloper, pinch, and jug), running on a Franka arm with a 16-DoF LEAP hand. We also introduce a climbing-hold benchmark: a spring testbed whose ratchet gives a slip-force reading in newtons. In our analysis the policy's point-cloud encoder did not separate the four holds, so the scan alone gave it little to choose a grasp from. Without the label (a DP3-style baseline), simulated rollouts from the robot's reset pose all end in a jug grasp, whatever the hold (25% correct grasp type, versus 94% for BetaGrasp). The label works as a switch for which grasp the policy performs. On 80 paired hardware trials, BetaGrasp raises average slip force from 9.6 N to 14.3 N, and grasps slipping fully off the hold drop from 44% of trials to 10%.
Sources
- Denoising Diffusion Probabilistic Models
- 3D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations
- Dexonomy: Synthesizing All Dexterous Grasp Types in a Grasp Taxonomy
- Grasp as You Say: Language-guided Dexterous Grasp Generation
- Denoising Diffusion Implicit Models
- PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
- Generalizable Humanoid Manipulation with 3D Diffusion Policies
- GenDP: 3D Semantic Fields for Category-Level Generalizable Diffusion Policy
- DexGraspNet: A Large-Scale Robotic Dexterous Grasp Dataset for General Objects Based on Simulation
- UniDexGrasp: Universal Robotic Dexterous Grasping via Learning Diverse Proposal Generation and Goal-Conditioned Policy
- DexCap: Scalable and Portable Mocap Data Collection System for Dexterous Manipulation
- AnyTeleop: A General Vision-Based Dexterous Robot Arm-Hand Teleoperation System
- Open-TeleVision: Teleoperation with Immersive Active Visual Feedback
- Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware
- OpenVR: Teleoperation for Manipulation
- LEAP Hand: Low-Cost, Efficient, and Anthropomorphic Hand for Robot Learning
- Dexterous Functional Pre-Grasp Manipulation with Diffusion Policy
- Cross-Embodiment Dexterous Grasping with Reinforcement Learning
- OmniDexVLG: Learning Dexterous Grasp Generation from Vision Language Model-Guided Grasp Semantics, Taxonomy and Functional Affordance
- DexGraspVLA: A Vision-Language-Action Framework Towards General Dexterous Grasping
Related papers
- Rethinking Actions: Continuity-Aware Priors for World-Model Planning
- MOSAIC: Modular Obstacle and Spatial-Aware Framework for Manipulation In Cluttered Scenes
- SAGAS: Semantic-Aware Graph-Assisted Stitching for Offline Temporal Logic Planning
- DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
- LaST0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning
- CIALLO: Learning Shared Intent Representations for Cross-Embodiment Manipulation