RoboTurk: A Crowdsourcing Platform for Robotic Skill Learning through Imitation
Ajay Mandlekar‡, Yuke Zhu, Animesh Garg, Jonathan Booher, Max Spero, Albert Tung, Julian Gao, John Emmons, Anchit Gupta, Emre Orbay, Silvio Savarese, Li Fei-Fei Department of Computer Science, Stanford University
Abstract
Imitation Learning has empowered recent advances in learning robotic manipulation tasks by addressing shortcomings of Reinforcement Learning such as exploration and reward specification. However, research in this area has been limited to modest-sized datasets due to the difficulty of collecting large quantities of task demonstrations through existing mechanisms. This work introduces RoboTurk to address this challenge. RoboTurk is a crowdsourcing platform for high quality 6-DoF trajectory based teleoperation through the use of widely available mobile devices (e.g. iPhone). We evaluate RoboTurk on three manipulation tasks of varying timescales (15-120s) and observe that our user interface is statistically similar to special purpose hardware such as virtual reality controllers in terms of task completion times. Furthermore, we observe that poor network conditions, such as low bandwidth and high delay links, do not substantially affect the remote users’ ability to perform task demonstrations successfully on RoboTurk. Lastly, we demonstrate the efficacy of RoboTurk through the collection of a pilot dataset; using RoboTurk, we collected 137.5 hours of manipulation data from remote worker
中文速览
机器人操作任务通常需要大量高质量的人工示范数据，但现有采集方式要么太慢（手把手引导）、要么门槛太高（需要专用VR硬件），导致数据规模始终上不去。为此，研究者开发了RoboTurk——一个众包遥操作平台，让用户只需一部iPhone和一个网页浏览器，就能通过手机姿态追踪（ARKit）远程控制仿真机械臂，机器人模拟运行在云端并将实时视频回传到用户浏览器。用户研究表明，手机界面的任务完成时间与专用VR控制器在统计上没有显著差异，明显优于3D鼠标和键盘；即便在低带宽、高延迟的恶劣网络环境下，用户也能顺利完成示范任务。借助这一平台，团队仅用约22小时系统使用时长就从众包工作者处收集了137.5小时、超过2200条成功操作示范，并验证了更大规模的示范数据能显著提升稀疏奖励下策略学习的一致性和最终性能。RoboTurk的意义在于，它第一次让"为机器人学习大规模收集高质量轨迹数据"变得像标注图片一样可规模化、低门槛，为推动机器人模仿学习迈向大数据时代提供了关键基础设施。
原文 arXiv:1811.02790;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1811.02790v1