Unsupervised Neural Motion Retargeting for Humanoid Teleoperation

TL;DR

This paper introduces a GAN-based online motion retargeting method for humanoid teleoperation that eliminates the need for pairwise datasets, simplifying setup and improving accessibility.

Contribution

It presents a novel unsupervised neural approach for motion retargeting that reduces complexity and setup requirements compared to traditional methods.

Findings

Retargeted a variety of upper-body motions successfully.

Achieved end-effector error comparable to conventional methods.

Demonstrated practical usability with a pick-and-place task.

Abstract

This study proposes an approach to human-to-humanoid teleoperation using GAN-based online motion retargeting, which obviates the need for the construction of pairwise datasets to identify the relationship between the human and the humanoid kinematics. Consequently, it can be anticipated that our proposed teleoperation system will reduce the complexity and setup requirements typically associated with humanoid controllers, thereby facilitating the development of more accessible and intuitive teleoperation systems for users without robotics knowledge. The experiments demonstrated the efficacy of the proposed method in retargeting a range of upper-body human motions to humanoid, including a body jab motion and a basketball shoot motion. Moreover, the human-in-the-loop teleoperation performance was evaluated by measuring the end-effector position errors between the human and the retargeted humanoid motions. The results demonstrated that the error was comparable to those of conventional motion retargeting methods that require pairwise motion datasets. Finally, a box pick-and-place task was conducted to demonstrate the usability of the developed humanoid teleoperation system.