ExBody2: Advanced Expressive Humanoid Whole-Body Control

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ji, Mazeyu, Peng, Xuanbin, Liu, Fangchen, Li, Jialong, Yang, Ge, Cheng, Xuxin, Wang, Xiaolong
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913730835513344
author Ji, Mazeyu
Peng, Xuanbin
Liu, Fangchen
Li, Jialong
Yang, Ge
Cheng, Xuxin
Wang, Xiaolong
author_facet Ji, Mazeyu
Peng, Xuanbin
Liu, Fangchen
Li, Jialong
Yang, Ge
Cheng, Xuxin
Wang, Xiaolong
contents This paper tackles the challenge of enabling real-world humanoid robots to perform expressive and dynamic whole-body motions while maintaining overall stability and robustness. We propose Advanced Expressive Whole-Body Control (Exbody2), a method for producing whole-body tracking controllers that are trained on both human motion capture and simulated data and then transferred to the real world. We introduce a technique for decoupling the velocity tracking of the entire body from tracking body landmarks. We use a teacher policy to produce intermediate data that better conforms to the robot's kinematics and to automatically filter away infeasible whole-body motions. This two-step approach enabled us to produce a student policy that can be deployed on the robot that can walk, crouch, and dance. We also provide insight into the trade-off between versatility and the tracking performance on specific motions. We observed significant improvement of tracking performance after fine-tuning on a small amount of data, at the expense of the others.
format Preprint
id arxiv_https___arxiv_org_abs_2412_13196
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ExBody2: Advanced Expressive Humanoid Whole-Body Control
Ji, Mazeyu
Peng, Xuanbin
Liu, Fangchen
Li, Jialong
Yang, Ge
Cheng, Xuxin
Wang, Xiaolong
Robotics
Artificial Intelligence
Machine Learning
This paper tackles the challenge of enabling real-world humanoid robots to perform expressive and dynamic whole-body motions while maintaining overall stability and robustness. We propose Advanced Expressive Whole-Body Control (Exbody2), a method for producing whole-body tracking controllers that are trained on both human motion capture and simulated data and then transferred to the real world. We introduce a technique for decoupling the velocity tracking of the entire body from tracking body landmarks. We use a teacher policy to produce intermediate data that better conforms to the robot's kinematics and to automatically filter away infeasible whole-body motions. This two-step approach enabled us to produce a student policy that can be deployed on the robot that can walk, crouch, and dance. We also provide insight into the trade-off between versatility and the tracking performance on specific motions. We observed significant improvement of tracking performance after fine-tuning on a small amount of data, at the expense of the others.
title ExBody2: Advanced Expressive Humanoid Whole-Body Control
topic Robotics
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2412.13196