Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Yongxu, Yi, Weiyun, Kong, Xinhao, Li, Wanting
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2503.21257
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909555043074048
author Wang, Yongxu
Yi, Weiyun
Kong, Xinhao
Li, Wanting
author_facet Wang, Yongxu
Yi, Weiyun
Kong, Xinhao
Li, Wanting
contents With the rapid development of embodied intelligence, leveraging large-scale human data for high-level imitation learning on humanoid robots has become a focal point of interest in both academia and industry. However, applying humanoid robots to precision operation domains remains challenging due to the complexities they face in perception and control processes, the long-standing physical differences in morphology and actuation mechanisms between humanoid robots and humans, and the lack of task-relevant features obtained from egocentric vision. To address the issue of covariate shift in imitation learning, this paper proposes an imitation learning algorithm tailored for humanoid robots. By focusing on the primary task objectives, filtering out background information, and incorporating channel feature fusion with spatial attention mechanisms, the proposed algorithm suppresses environmental disturbances and utilizes a dynamic weight update strategy to significantly improve the success rate of humanoid robots in accomplishing target tasks. Experimental results demonstrate that the proposed method exhibits robustness and scalability across various typical task scenarios, providing new ideas and approaches for autonomous learning and control in humanoid robots. The project will be open-sourced on GitHub.
format Preprint
id arxiv_https___arxiv_org_abs_2503_21257
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle OminiAdapt: Learning Cross-Task Invariance for Robust and Environment-Aware Robotic Manipulation
Wang, Yongxu
Yi, Weiyun
Kong, Xinhao
Li, Wanting
Robotics
Artificial Intelligence
With the rapid development of embodied intelligence, leveraging large-scale human data for high-level imitation learning on humanoid robots has become a focal point of interest in both academia and industry. However, applying humanoid robots to precision operation domains remains challenging due to the complexities they face in perception and control processes, the long-standing physical differences in morphology and actuation mechanisms between humanoid robots and humans, and the lack of task-relevant features obtained from egocentric vision. To address the issue of covariate shift in imitation learning, this paper proposes an imitation learning algorithm tailored for humanoid robots. By focusing on the primary task objectives, filtering out background information, and incorporating channel feature fusion with spatial attention mechanisms, the proposed algorithm suppresses environmental disturbances and utilizes a dynamic weight update strategy to significantly improve the success rate of humanoid robots in accomplishing target tasks. Experimental results demonstrate that the proposed method exhibits robustness and scalability across various typical task scenarios, providing new ideas and approaches for autonomous learning and control in humanoid robots. The project will be open-sourced on GitHub.
title OminiAdapt: Learning Cross-Task Invariance for Robust and Environment-Aware Robotic Manipulation
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2503.21257