Imitation learning from human demonstrations is a powerful framework to teach robots new skills. However, the performance of the learned policies is bottlenecked by the quality, scale, and variety of the demonstration data. In this paper, we aim to lower the barrier to collecting large and high-quality human demonstration data by proposing GELLO, a general framework for building low-cost and intuitive teleoperation systems for robotic manipulation. Given a target robot arm, we build a GELLO controller that has the same kinematic structure as the target arm, leveraging 3D-printed parts and off-the-shelf motors. GELLO is easy to build and intuitive to use. Through an extensive user study, we show that GELLO enables more reliable and efficient demonstration collection compared to commonly used teleoperation devices in the imitation learning literature such as VR controllers and 3D spacemouses. We further demonstrate the capabilities of GELLO for performing complex bi-manual and contact-rich manipulation tasks. To make GELLO accessible to everyone, we have designed and built GELLO systems for 3 commonly used robotic arms: Franka, UR5, and xArm. All software and hardware are open-sourced and can be found on our website: https://wuphilipp.github.io/gello/.

翻译：摘要：通过人类示教进行模仿学习是赋予机器人新技能的有效框架。然而，学习策略的性能受限于示教数据的质量、规模和多样性。本文旨在通过提出GELLO——一种用于构建低成本、直观的机器人操作遥操作系统的通用框架——降低收集大规模、高质量人类示教数据的门槛。针对目标机器人臂，我们利用3D打印部件和现成电机，构建了具有与目标臂相同运动学结构的GELLO控制器。GELLO易于构建且使用直观。通过广泛的用户研究，我们证明，与模仿学习文献中常用的遥操作设备（如VR控制器和3D空间鼠标）相比，GELLO能实现更可靠、高效的示教数据收集。我们进一步展示了GELLO在执行复杂双臂及高接触性操作任务中的能力。为让GELLO普及，我们针对三种常用机器人臂（Franka、UR5和xArm）设计并构建了GELLO系统。所有软硬件均开源，详见网站：https://wuphilipp.github.io/gello/。