OPEN TEACH: A Versatile Teleoperation System for Robotic Manipulation

Aadhithya Iyer1, Zhuoran Peng1, Yinlong Dai1, Irmak Guzey1, Siddhant Haldar1, Soumith Chintala2, Lerrel Pinto1
1New York University   2Meta
description arXiv
code Code

Abstract

Open-sourced, user-friendly tools form the bedrock of scientific advancement across disciplines. The widespread adoption of data-driven learning has led to remarkable progress in multi-fingered dexterity, bimanual manipulation, and applica- tions ranging from logistics to home robotics. However, existing data collection platforms are often proprietary, costly, or tailored to specific robotic morphologies. We present OPEN TEACH, a new teleoperation system leveraging VR headsets to immerse users in mixed reality for intuitive robot control. Built on the affordable Meta Quest 3, which costs $500, OPEN TEACH enables real- time control of various robots, including multi-fingered hands, bimanual arms, and mobile manipulators, through an easy-to- use app. Using natural hand gestures and movements, users can manipulate robots at up to 90Hz with smooth visual feedback and interface widgets offering closeup environment views. We demonstrate the versatility of OPEN TEACH across 38 tasks on different robots. A comprehensive user study indicates significant improvement in teleoperation capability over the AnyTeleop framework. Further experiments exhibit that the collected data is compatible with policy learning on 10 dexterous and contact- rich manipulation tasks. Currently supporting Franka, xArm, Jaco, Allegro, and Hello Stretch platforms, OPEN TEACH is fully open-sourced to promote broader adoption.



Hardware

OPEN TEACH utilizes the Quest 3 VR headset for teleoperation, providing low latency and high-frequency visual feedback. The Quest 3 is chosen for its affordability and accessibility. Below we provide videos of the full-color passthrough provided by Quest 3, and the user interface of the Open Teach app. As shown, we implement a screen within the VR app that displays the feed from a camera positioned within the scene. This enables the user to view a robotic scene both through their eyes, via visual passthrough, as well as realtime streams from the robot’s camera. We emphasize the exceptional clarity of the Quest 3 visual passthrough, enabling users to perceive tasks in a 3D environment rather than solely on a 2D screen. Furthermore, the high-frequency, low-latency teleoperation facilitated by Open Teach allows users to observe real-time movements of the robot, enabling immediate corrections for any errors in its pose.