robotics

S2A2: Audio-Visual Imitation Learning for Manipulation Tasks Using Acoustic Spatial Information

arXiv:2607.26047

summary

The paper presents acoustic-aware manipulation tasks where robots use sound cues to locate and identify objects, and introduces the S2A2 multimodal imitation learning framework that combines visual and acoustic spatial/spectral information for robot manipulation.

Abstract

Acoustic information provides rich cues about object location, material properties, and changes caused by contact or motion. This paper introduces a new set of acoustic-aware manipulation tasks for imitation learning, in which robots must use auditory cues to determine manipulation targets. These tasks require sound source localization and identification for active exploration in robotic manipulation. Also, we propose a multimodal imitation learning framework, Spatial-Spectral Audio Action (S2A2), that integrates visual features with acoustic spatial and acoustic signal information for the acoustic-aware manipulation tasks. We implemented S2A2 models that integrates policies such as ACT, Diffusion Policy, VQ-BeT, and , into our framework. Simulation experiments showed that the proposed method is the most effective for tasks requiring both position and timbre. Furthermore, real-robot experiments confirm the applicability of the proposed tasks and framework to real-world manipulation.

Project page: https://azuma413.github.io/projects/s2a2

Topics & keywords

#acoustic perception#imitation learning#robotic manipulation#multimodal learning#sound source localizationacoustic spatial informationS2A2visual-audio integrationdiffusion policyACTVQ-BeTauditory cues
S2A2: Audio-Visual Imitation Learning for Manipulation Tasks Using Acoustic Spatial Information · wovepaper