The Fingertip Manipulability Assessment of Tendon-Driven Multi-Fingered Hands

Li, Junnan; Ganguly, Amartya; Figueredo, Luis F. C.; Haddadin, Sami

Abstract

The ability of robotic fingers to exert force and exhibit motion is vital for achieving dexterity in manipulation tasks. To evaluate dexterous capabilities in terms of both features, i.e., quantifying finger performance that facilitates task planning and design optimization, we introduce the Fingertip Manipulability (FtM) metric. The FtM is a comprehensive assessment tool linked to finger parameters rather than specific task requirements, e.g., wrench information, contact-points, among others. It takes into account the entire voxelized fingertip workspace of all fingers, filling a gap in providing a global representation during the design and deployment phase of tendon-driven robotic hands. It composes the assessment map of a multi-fingered hand that enables real-time performance monitoring and planning for dexterous tendon-driven hands. To illustrate the practical application of this metric, we showcase its assessment of the Shadow Hand, demonstrating its characteristics in optimizing poses for a multi-finger grasping scenario.

Citation

Li, J., Ganguly, A., Figueredo, L. F. C., & Haddadin, S. (2024). The Fingertip Manipulability Assessment of Tendon-Driven Multi-Fingered Hands. IEEE Robotics and Automation Letters, 9(3), 2726-2733. https://doi.org/10.1109/lra.2024.3360816

• Journal Article Type: Article
• Acceptance Date: Aug 30, 2024
• Online Publication Date: Jan 31, 2024
• Publication Date: 2024-03
• Deposit Date: Mar 21, 2025
• Publicly Available Date: Mar 25, 2025
• Journal: IEEE Robotics and Automation Letters
• Electronic ISSN: 2377-3766
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• Volume: 9
• Issue: 3
• Pages: 2726-2733
• DOI: https://doi.org/10.1109/lra.2024.3360816
• Keywords: Artificial Intelligence, Control and Optimization, Computer Science Applications, Computer Vision and Pattern Recognition, Mechanical Engineering, Human-Computer Interaction, Biomedical Engineering, Control and Systems Engineering
• Public URL: https://nottingham-repository.worktribe.com/output/31880890
• Publisher URL: https://ieeexplore.ieee.org/document/10417117