ORCA: A Platform for Open-Source Dexterity Research

2026-06-12

AI Summary

This work addresses the longstanding challenge in dexterous manipulation research—the absence of a unified, open-source hardware-software platform compatible with mainstream robot learning ecosystems—by introducing ORCA, the first open-source research stack tailored for dexterous manipulation. ORCA integrates low-level hand control, high-fidelity simulation, teleoperation via consumer-grade VR devices, and gesture retargeting, while natively supporting popular robot learning frameworks such as Lerobot. It establishes an end-to-end workflow spanning expert demonstration collection, policy training, and deployment evaluation, thereby standardizing dexterous manipulation within established robot learning paradigms. By offering a reproducible, modular, and accessible full-stack solution, ORCA significantly lowers the barrier to entry and has been fully open-sourced to foster community-wide advancement.

Abstract

Robotics manipulation research increasingly focuses on two-finger parallel grippers for their effectiveness, affordability, and ease of teleoperation. Grippers are nonetheless limited by their form factor, often requiring bimanual setups even for simple reorientation tasks. Anthropomorphic hands are a more natural platform for dexterous robot learning -- closer to the human hand, and capable of learning from human video -- yet they remain hard to use in learning research: even where open and accessible hand hardware exists, the software for control, simulation, teleoperation, and retargeting is scattered in one-off code bases, and largely disconnected from the robot-learning ecosystem. In this work, we introduce the \orca~learning stack, an open-source research stack for dexterity as a first-class robot learning domain. Our \orca~stack unifies low-level control, simulation, teleoperation from a range of consumer platforms, and hand retargeting, behind a single interface, and integrates natively with popular robot-learning frameworks such as \lerobot, so dexterous hand researchers can leverage the same data, training, and evaluation pipelines used for non-dexterous robot learning. We demonstrate a complete end-to-end workflow, collecting expert demonstrations of an in-hand reorientation task by teleoperation with a consumer-grade VR headset, training an autonomous policy with \lerobot, and evaluating the learned policy in a fully reproducible and observable setup. We open-source the entire stack as a shared, reproducible foundation for dexterous-manipulation research.

Problem

• dexterous manipulation
• anthropomorphic hands
• robot learning
• teleoperation
• open-source platform

Innovation

• dexterous manipulation
• open-source platform
• teleoperation
• hand retargeting
• robot learning