SPARK Hand: Scooping-Pinching Adaptive Robotic Hand with Kempe Mechanism for Vertical Passive Grasp in Environmental Constraints

2025-04-29

AI Summary

Conventional grippers struggle to stably grasp thin or flat objects in resource-constrained environments (e.g., desktop-edge settings). To address this, we propose SPARK Hand—a dual-finger adaptive robotic hand enabling vertical passive grasping. Its core innovation is a passive mode-switching mechanism that integrates Kempe linkages with parallelogram structures, enabling autonomous, actuation-free transitions between parallel pinch and scooping modes. Coupled with a multi-link architecture, embedded elastic elements, and guided motion constraints, SPARK Hand achieves dual adaptation—geometric conformity and mechanical compliance. Experimental results demonstrate significantly improved grasping success rates and stability for irregularly shaped and ultra-thin objects. The design proves effective and practical under dynamic spatial constraints, offering a lightweight, low-power solution suitable for edge-deployable robotic manipulation.

Technology Category

• Intelligent Robots: Manipulation
• Computer Vision: Multi-modal Vision
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments

Abstract

This paper presents the SPARK finger, an innovative passive adaptive robotic finger capable of executing both parallel pinching and scooping grasps. The SPARK finger incorporates a multi-link mechanism with Kempe linkages to achieve a vertical linear fingertip trajectory. Furthermore, a parallelogram linkage ensures the fingertip maintains a fixed orientation relative to the base, facilitating precise and stable manipulation. By integrating these mechanisms with elastic elements, the design enables effective interaction with surfaces, such as tabletops, to handle challenging objects. The finger employs a passive switching mechanism that facilitates seamless transitions between pinching and scooping modes, adapting automatically to various object shapes and environmental constraints without additional actuators. To demonstrate its versatility, the SPARK Hand, equipped with two SPARK fingers, has been developed. This system exhibits enhanced grasping performance and stability for objects of diverse sizes and shapes, particularly thin and flat objects that are traditionally challenging for conventional grippers. Experimental results validate the effectiveness of the SPARK design, highlighting its potential for robotic manipulation in constrained and dynamic environments.

Problem

• Develops adaptive robotic hand for versatile grasping modes
• Achieves stable vertical grasp with passive switching mechanism
• Enhances manipulation of thin, flat objects in constrained environments

Innovation

• Kempe linkages enable vertical linear fingertip trajectory
• Parallelogram linkage maintains fixed fingertip orientation
• Passive switching mechanism adapts between grasping modes