SonicBoom: Contact Localization Using Array of Microphones

TL;DR

SonicBoom is a system that uses an array of contact microphones and machine learning to accurately localize contact points on a robot's end effector in cluttered, occluded environments, enabling better navigation and manipulation.

Contribution

It introduces a novel hardware and learning pipeline for contact localization using microphone arrays and feature engineering, addressing challenges of solid media localization.

Findings

Achieves 0.42cm localization error in distribution

Maintains 2.22cm error with novel objects and conditions

Enables practical haptic mapping in occluded environments

Abstract

In cluttered environments where visual sensors encounter heavy occlusion, such as in agricultural settings, tactile signals can provide crucial spatial information for the robot to locate rigid objects and maneuver around them. We introduce SonicBoom, a holistic hardware and learning pipeline that enables contact localization through an array of contact microphones. While conventional sound source localization methods effectively triangulate sources in air, localization through solid media with irregular geometry and structure presents challenges that are difficult to model analytically. We address this challenge through a feature engineering and learning based approach, autonomously collecting 18,000 robot interaction sound pairs to learn a mapping between acoustic signals and collision locations on the robot end effector link. By leveraging relative features between microphones, SonicBoom achieves localization errors of 0.42cm for in distribution interactions and maintains robust performance of 2.22cm error even with novel objects and contact conditions. We demonstrate the system's practical utility through haptic mapping of occluded branches in mock canopy settings, showing that acoustic based sensing can enable reliable robot navigation in visually challenging environments.