From Power to Precision: Learning Fine-grained Dexterity for Multi-fingered Robotic Hands

2025-11-17

AI Summary

Multi-finger robotic hands struggle to simultaneously achieve robust power grasping and dexterous manipulation. This paper proposes a hardware-control co-optimization framework: (1) designing lightweight, parameterized fingertip geometries; and (2) constructing a learnable contact-surface model enabling dynamic grasp-mode switching and reducing fine manipulation to coordinated thumb-index motion. The method integrates large-scale differentiable neural physics-based simulation optimization, sim-to-real reinforcement learning, and real-to-real validation. It achieves an 82.5% zero-shot sim-to-real transfer success rate on unseen objects and a 93.3% success rate in real-world bread-pinching tasks—significantly enhancing dexterity without compromising grasp stability. Key innovations include contact-model-driven co-optimization of morphology and control, and an adaptive grasp-mode switching mechanism.

Technology Category

• Intelligent Robots: Manipulation
• Search and Optimization: Sampling/Simulation-based Search
• Computer Vision: Low Level & Physics-based Vision

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: The sharing economy

Abstract

Human grasps can be roughly categorized into two types: power grasps and precision grasps. Precision grasping enables tool use and is believed to have influenced human evolution. Today's multi-fingered robotic hands are effective in power grasps, but for tasks requiring precision, parallel grippers are still more widely adopted. This contrast highlights a key limitation in current robotic hand design: the difficulty of achieving both stable power grasps and precise, fine-grained manipulation within a single, versatile system. In this work, we bridge this gap by jointly optimizing the control and hardware design of a multi-fingered dexterous hand, enabling both power and precision manipulation. Rather than redesigning the entire hand, we introduce a lightweight fingertip geometry modification, represent it as a contact plane, and jointly optimize its parameters along with the corresponding control. Our control strategy dynamically switches between power and precision manipulation and simplifies precision control into parallel thumb-index motions, which proves robust for sim-to-real transfer. On the design side, we leverage large-scale simulation to optimize the fingertip geometry using a differentiable neural-physics surrogate model. We validate our approach through extensive experiments in both sim-to-real and real-to-real settings. Our method achieves an 82.5% zero-shot success rate on unseen objects in sim-to-real precision grasping, and a 93.3% success rate in challenging real-world tasks involving bread pinching. These results demonstrate that our co-design framework can significantly enhance the fine-grained manipulation ability of multi-fingered hands without reducing their ability for power grasps. Our project page is at https://jianglongye.com/power-to-precision

Problem

• Enabling multi-fingered robotic hands to perform both power and precision grasps
• Overcoming the limitation of current robotic hands in fine-grained manipulation
• Jointly optimizing control and hardware design for versatile grasping capabilities

Innovation

• Jointly optimizing control and hardware design
• Lightweight fingertip geometry modification with contact plane
• Dynamic switching between power and precision manipulation