House of Dextra: Cross-embodied Co-design for Dexterous Hands

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Fay, Kehlani, Djapri, Darin Anthony, Zorin, Anya, Clinton, James, Lahib, Ali El, Su, Hao, Tolley, Michael T., Yi, Sha, Wang, Xiaolong
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917419257167872
author Fay, Kehlani
Djapri, Darin Anthony
Zorin, Anya
Clinton, James
Lahib, Ali El
Su, Hao
Tolley, Michael T.
Yi, Sha
Wang, Xiaolong
author_facet Fay, Kehlani
Djapri, Darin Anthony
Zorin, Anya
Clinton, James
Lahib, Ali El
Su, Hao
Tolley, Michael T.
Yi, Sha
Wang, Xiaolong
contents Dexterous manipulation is limited by both control and design, without consensus as to what makes manipulators best for performing dexterous tasks. This raises a fundamental challenge: how should we design and control robot manipulators that are optimized for dexterity? We present a co-design framework that learns task-specific hand morphology and complementary dexterous control policies. The framework supports 1) an expansive morphology search space including joint, finger, and palm generation, 2) scalable evaluation across the wide design space via morphology-conditioned cross-embodied control, and 3) real-world fabrication with accessible components. We evaluate the approach across multiple dexterous tasks, including in-hand rotation with simulation and real deployment. Our framework enables an end-to-end pipeline that can design, train, fabricate, and deploy a new robotic hand in under 24 hours. The full framework will be open-sourced and available on our website: https://an-axolotl.github.io/HouseofDextra/ .
format Preprint
id arxiv_https___arxiv_org_abs_2512_03743
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle House of Dextra: Cross-embodied Co-design for Dexterous Hands
Fay, Kehlani
Djapri, Darin Anthony
Zorin, Anya
Clinton, James
Lahib, Ali El
Su, Hao
Tolley, Michael T.
Yi, Sha
Wang, Xiaolong
Robotics
Machine Learning
Dexterous manipulation is limited by both control and design, without consensus as to what makes manipulators best for performing dexterous tasks. This raises a fundamental challenge: how should we design and control robot manipulators that are optimized for dexterity? We present a co-design framework that learns task-specific hand morphology and complementary dexterous control policies. The framework supports 1) an expansive morphology search space including joint, finger, and palm generation, 2) scalable evaluation across the wide design space via morphology-conditioned cross-embodied control, and 3) real-world fabrication with accessible components. We evaluate the approach across multiple dexterous tasks, including in-hand rotation with simulation and real deployment. Our framework enables an end-to-end pipeline that can design, train, fabricate, and deploy a new robotic hand in under 24 hours. The full framework will be open-sourced and available on our website: https://an-axolotl.github.io/HouseofDextra/ .
title House of Dextra: Cross-embodied Co-design for Dexterous Hands
topic Robotics
Machine Learning
url https://arxiv.org/abs/2512.03743