Please use this identifier to cite or link to this item:
For citation please use:
Main Title: Passive and Active Acoustic Sensing for Soft Pneumatic Actuators - Code and Data
Author(s): Wall, Vincent
Zöller, Gabriel
Brock, Oliver
Other Contributor(s): Hebecker, Marius
Type: Generic Research Data
Language Code: en
Abstract: This data and code accompanies the paper "Passive and Active Acoustic Sensing for Soft Pneumatic Actuators" [1]. Abstract: We propose a sensorization method for soft pneumatic actuators that uses an embedded microphone and speaker to measure relevant actuator states. The physical state of the actuator influences the modulation of sound as it travels through the structure. Using simple machine learning, we create a computational sensor that infers the current state from sound recordings. We demonstrate the acoustic sensor on a PneuFlex actuator and use it to measure the contact location, contact force, object material and actuator inflation. We show that the sensor is reliable (the classification rate for six contact locations is 93%), precise (spatial resolution below 4mm), and robust against common disturbances like background noise. Finally, we compare different sounds and learning methods and achieve best results with 20ms of white noise and a support vector classifier as the sensor model. [1] Vincent Wall, Gabriel Zöller, and Oliver Brock. "Passive and Active Acoustic Sensing for Soft Pneumatic Actuators." (in preparation)
Issue Date: Dec-2020
Date Available: 19-Feb-2021
DDC Class: 005 Computerprogrammierung, Programme, Daten
Subject(s): acoustic sensing
soft sensors
force and tactile sensing
pneumatic actuators
soft material robotics
Sponsor/Funder: EC/H2020/645599/EU/Soft-bodied intelligence for Manipulation/SoMa
DFG, EXC 2002/1, 390523135, Science of Intelligence (SCIoI)
DFG, SPP 2100, 359715917, Soft Material Robotic Systems
Appears in Collections:FG Robotics » Research Data

Files in This Item:
Format: Markdown | Size: 2.78 kB

Python code used to record, convert, and evaluate all experimental data shown in the manuscript.

Format: ZIP Archive | Size: 1.56 MB

Converted data; All data sets converted into spectra and corresponding labels.

Format: ZIP Archive | Size: 2.82 GB

Description of the data sets and which experiment from the paper they correspond to.

Format: Microsoft Excel | Size: 11.5 kB

Plots for each of the evaluations

Format: ZIP Archive | Size: 12.38 MB

Raw sound files. (Description of the data sets given in 'depositOnce_data_sets.xls')

Format: ZIP Archive | Size: 1.29 GB

Results of the evaluations; Contains trained sensor models as well as the prediction scores and confusion matrices. The plots are generated from these files.

Format: ZIP Archive | Size: 3.37 GB

Version History
Version Item Date Summary
2 10.14279/depositonce-11059.2 2021-02-18 10:58:41.975 Added additional experimental data and corresponding evaluations.
1 10.14279/depositonce-11059 2020-12-15 14:34:10.0
Item Export Bar

Items in DepositOnce are protected by copyright, with all rights reserved, unless otherwise indicated.