RoboAcoustics@ICRA 2025

1st Workshop on Acoustic Sensing and Representations for Robotics

IEEE International Conference on Robotics and Automation (ICRA) 2025 - Workshop

May 19, 2025, Atlanta, GA, USA

Location: Room 316

Email: roboacoustics@googlegroups.com

Zoom: Link

Abstract

Acoustic sensing is rapidly emerging as a critical area of research, offering new possibilities for robust robotic perception, localization, and interaction in complex environments. This workshop aims to bring together researchers and practitioners from various domains and disciplines to discuss the latest advancements, challenges, and future directions in acoustic sensing for robotics.

Acoustic sensors, including microphones, sonar, and ultrasonic devices, provide unique advantages in scenarios where vision-based perception may be limited by adversarial lighting conditions, occlusions, or environmental complexity. Acoustic sensors may endow robots with heightened situational awareness, allow them to reason about interactions with the environment and enable sound-based feedback mechanisms that can react to voice commands or environment noise. However, acoustic signals also present unique open challenges, such as susceptibility to noise, multi-path propagation effects, the need for real-time processing in dynamic environments, and extracting acoustic features or learning representations that are robust to context-dependent distortions. Addressing these challenges is crucial for unlocking the full potential of acoustic sensing in robotics.

Speakers

Oliver

Brock

Technical University

Berlin

Shabnam Ghaffarzadegan

Bosch Center for AI

Lerrel

Pinto

New York University

Naomi

Fitter

Oregon State University

Shuran

Song

Stanford University

Kazuhiro

Nakadai

Institute of Science Tokyo

Ruohan Gao

University of Maryland

Topics of Interest

Partially supported by Computer & Robot Vision TC, the workshop will cover a wide range of topics, including but not limited to:

Signal processing techniques to extract acoustic signal features from raw sensor readings
Auditory signal design and communication with robots (e.g., integration with LLMs) for effective human-robot interaction and haptics
Representation learning for acoustic signals, studying its impact on robotic perception and decision-making
Foundation models that can utilize acoustic signals for generalizable scene understanding across diverse tasks
Sensor fusion techniques that combine acoustic data with visual, tactile, or other sensory inputs for enhanced situational awareness
Generative models or simulation frameworks for generating synthetic acoustic datasets
Data-efficient learning approaches for acoustic perception in resource-constrained robotic systems
Applications of acoustic sensing for manipulation, navigation, human-robot interaction, environmental monitoring, and other robotics domains
Challenges in designing and deploying acoustic sensors in real-world robotic systems
Preliminary studies demonstrating practical implementations of acoustic sensing for robotics
Simulation and synthetic data generation for audio and vibrotactile signals
Design processes for embedding acoustic sensors on different robot embodiments (e.g., arms, end-effectors, mobile robots)

Schedule

9:05 - 9:10: Opening Remarks

9:10 - 10:10: Speaker Session 1

9:10 - 9:40: Ruohan Gao

9:40 - 10:10: Lerrel Pinto

10:10-11:00: Coffee Break and Poster Session 1

11:00 - 12:00: Speaker Session 2

11:00 - 11:30: Shabnam Ghaffarzadegan

11:30 - 12:00: Oliver Brock

12:00 - 13:30: Lunch

13:30 - 14:45: Speaker Session 3

13:45 - 14:15: Kazuhiro Nakadai

14:15 - 14:45: Naomi Fitter

14:45 - 15:00: Paper Lightning Talks

15:00 - 16:00: Break and Posters

16:00 - 16:30: Speaker Session 4 -

16:00 - 16:30: Shuran Song

16:30 - 17:00 [30 min]: Panel discussion/Round Table

Panelists: Lerrel Pinto, Shabnam Ghaffarzadegan, Kazuhiro Nakadai, Naomi Fitter, Shuran Song, and Ruohan Gao.

17:00 - 17:05 [5 min]: Closing remarks + Awards

Accepted Papers

WildFusion: Multimodal Implicit 3D Reconstructions in the Wild [Paper]
Yanbaihui Liu, Boyuan Chen

Visual-Audio Representation Enables a Customer-Driven Domain Adaptation for Audio-Controlled Robots [Paper]
Peixin Chang, Shuijing Liu, Tianchen Ji, Neeloy Chakraborty, Kaiwen Hong, Katherine Rose Driggs-Campbell

Self-Mixing Laser Interferometry: In Search of an Ambient Noise Resilient Alternative to Acoustic Sensing [Paper]
Remko Proesmans, Thomas Lips, Francis Wyffels

SonicBoom: Contact Localization Using Array of Microphones [Paper]
Moonyoung Lee, Uksang Yoo, Jean Oh, Jeffrey Ichnowski, George Kantor, Oliver Kroemer

Bayesian inference and neural estimation of acoustic wave propagation [Paper]
Yongchao Huang, Yuhang He, Hong Ge

Acoustic Representation Learning: A β-VAE Approach for Anomaly Detection in Underwater Robotic Laser Cutting [Paper]
Sungmoon Joo

Acoustic Sensing for Universal Jamming Grippers [Paper]
Lion Weber, Theodor Wienert, Martin Splettstößer, Alexander Koenig, Oliver Brock

Acoustic Neural 3D Reconstruction Under Pose Drift [Paper]
Tianxiang Lin, Mohamad Qadri, Kevin Zhang, Adithya Pediredla, Christopher Metzler, Michael Kaess

AcousTac: Tactile Sensing with Acoustic Resonance for Electronics-Free Soft Structures [Paper]
Monica Li, Tae Myung Huh, Jadesola Aderibigbe, Hannah Stuart

SonarSplat: Novel View Synthesis and De-Streaking of Imaging Sonar via Gaussian Splatting [Paper]
Advaith Venkatramanan Sethuraman, Max Rucker, Onur Bagoren, Pou-Chun Kung, Nibarkavi Naresh Babu Amutha, Katherine A. Skinner

Active Acoustic Sensing: Capturing "EMG"-like Signals from Objects for Contact-Rich Robotic Manipulation [Paper]
Shihan Lu, Minho Heo, Heather Culbertson

Connectivity-Aware Coordination of AUV Fleets in Dynamic Underwater Environments [Paper]
Prajjwal Dutta, Jnaneshwar Das, Xi Yu