SOURCE-LINKED INTELLIGENCE

Steerable and Reactive Grasping Through Modular Design with a Three-Point Interface

arXiv · AI, language, vision and robotics · article · Sep 21, 2026 · UTC

Dexterous grasping requires deciding where to grasp, reaching the target, and maintaining stable contact. We connect these stages through a compact three-point interface that separates global geometric reasoning from local contact control. Given object geometry and optional language commands, our framework samples contact triples from a precomputed grasp-affordance heatmap. A model-based reactive controller tracks the object, avoids collisions, and guides the hand toward the selected contacts. In the final centimeters, a Reinforcement Learning (RL) policy uses proprioceptive feedback to refine

Read original source ↗ Open in workspace

recordType

paper

region

Global