Function: Controls the robot from behind, with the robot directly mimicking human movements.
Function: Controls the robot from the front, with the robot mimicking human movements like a mirror (right hand controls the robot's left hand and vice versa).
Use Case: Suitable for scenarios where the operator controls the robot from behind.
Use Case: Suitable for scenarios where the operator controls the robot from the front.
Learning from demonstrations has shown to be an effective approach to robotic manipulation, especially with the recently collected large-scale robot data with teleoperation systems. Building an efficient teleoperation system across diverse robot platforms has become more crucial than ever. However, there is a notable lack of cost-effective and user-friendly teleoperation systems for different end-effectors, e.g., anthropomorphic robot hands and grippers, that can operate across multiple platforms. To address this issue, we develop ACE, a cross-platform visual-exoskeleton system for low-cost dexterous teleoperation. Our system utilizes a hand-facing camera to capture 3D hand poses and an exoskeleton mounted on a portable base, enabling accurate real-time capture of both finger and wrist poses. Compared to previous systems, which often require hardware customization according to different robots, our single system can generalize to humanoid hands, arm-hands, arm-gripper, and quadruped-gripper systems with high-precision teleoperation. This enables imitation learning for complex manipulation tasks on diverse platforms.
We would like to express our gratitude to Xuanbin Peng, Chengjing Yuan, Mazeyu Ji, Chenhao Lu, Haichuan Che, Jiyue Zhu, and Rizhao Qiu for their invaluable assistance with the experiments, insightful discussions, and continuous support.
@article{yang2024ace,
title={ACE: A Cross-platfrom Visual-Exoskeletons for Low-Cost Dexterous Teleoperation},
author={Yang, Shiqi and Liu, Minghuan and Qin, Yuzhe and Runyu, Ding and Li Jialong and Cheng, Xuxin and Yang, Ruihan and Yi, Sha and Wang, Xiaolong},
journal={arXiv preprint arXiv:240},
year={2024}
}