ManiWAV: Learning Robot Manipulation from In-the-Wild Audio-Visual Data
An 'ear-in-hand' gripper records audio with vision during human demos in the wild; contact sound then helps policies when vision alone is ambiguous.“手上长耳朵”：夹爪在野外人类演示时同步录制音频与视觉；当视觉信息不足时，接触声帮助策略完成接触密集操作。
BibTeX
@misc{liu2024maniwav,
title = {{ManiWAV: Learning Robot Manipulation from In-the-Wild Audio-Visual Data}},
author = {Zeyi Liu and Cheng Chi and Eric Cousineau and Naveen Kuppuswamy and Benjamin Burchfiel and Shuran Song},
year = {2024},
eprint = {2406.19464},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2406.19464},
}Tactile & Visuotactile Sensing触觉与视触觉感知Manipulation & Dexterous Hands操作与灵巧手Dexterous hand灵巧手audiohuman-demonstrationsrealcontact-rich