S Russo
Development of a practical electrical tomography system for flexible contact sensing applications
Russo, S
Authors
Contributors
S Nefti-Meziani S.Nefti-Meziani@salford.ac.uk
Supervisor
Abstract
Tactile sensing is seeing an increase in potential applications, such as in humanoid and industrial robots; health care systems and medical instrumentation; prosthetic devices; and in the context of human-machine interaction. However, these applications require the integration of tactile sensors over various objects with different surface shapes. This emphasises the need of developing sensors which are flexible in contrast with the common rigid type. Moreover, flexible sensing research is considered to be in its infancy. Many technological and system issues are still open, mainly: conformability; scalability; system integration; high system cost; sensor size; and power consumption.
In light of the above, this thesis is concerned with the development of a flexible fabric-based contact sensor system. This is done through an interdisciplinary approach whereby electronics, system engineering, electrical tomography, and machine learning have been considered. This results in a practical flexible sensor that is capable of accurately detecting contact locations with high temporal resolution; and requires low power consumption.
The sensor is based on the principle of electrical tomography. This is essential since this technique allows us to eliminate electrodes and wiring from within the sensing area, confining them to the periphery of the sensor. This improves flexibility all while eliminating electrode fatigue and deterioration due to repeated loading.
We start by developing an electrical tomography sensor system. This comprises of a piezoresistive flexible fabric material, a data acquisition card, and a custom printed circuit board for managing both current injection and data collection. We show that current injection and voltage measurement protocols respond differently to different positions of the input contact region of interest, consequently affecting the overall performance of the tomography sensor system. Then, an approach for classifying contact location over the sensor is presented. This is done using supervised machine learning, namely discriminant analysis. Accurate touch location identification is achieved, along with an increase in the detection speed and sensor versatility. Finally, the sensor is placed over different surfaces in order to show and validate its efficiency.
The main finding of this work is that electrical tomography flexible sensor systems present a very promising technology, and can be practically and effectively used for developing inexpensive and durable flexible sensors for tactile applications. The main advantage of this approach is the complete absence of wires in the internal area of the sensor. This allows the sensor to be placed over surfaces with different shapes without losing its functionality.
The sensor's applicability can be further improved by using machine learning strategies due to their ability of empirical learning and extracting meaningful tactile information.
The research work in this thesis was motivated by the problems faced by industrial partners which were part of the sustainable manufacturing and advanced robotics training network in Europe (SMART-e).
Citation
Russo, S. (in press). Development of a practical electrical tomography system for flexible contact sensing applications. (Thesis). University of Salford
Thesis Type | Thesis |
---|---|
Acceptance Date | Jun 29, 2018 |
Deposit Date | Sep 25, 2018 |
Publicly Available Date | Jun 29, 2020 |
Related Public URLs | https://usir.salford.ac.uk/view/authors/58139.html |
Additional Information | Funders : People Programme (Marie Curie Actions) of the European Union’s Seventh Framework Programme FP7/2007-2013/ Projects : SMART-e Grant Number: 608022 |
Files
Thesis Stefania Russo corrected.pdf
(31.2 Mb)
PDF
You might also like
Geometry model for marker-based localisation
(2020)
Thesis
A comparative review on mobile robot path planning : classical or meta-heuristic methods?
(2020)
Journal Article
Toward Kinecting cognition by behaviour recognition-based deep learning and big data
(2020)
Journal Article
Downloadable Citations
About USIR
Administrator e-mail: library-research@salford.ac.uk
This application uses the following open-source libraries:
SheetJS Community Edition
Apache License Version 2.0 (http://www.apache.org/licenses/)
PDF.js
Apache License Version 2.0 (http://www.apache.org/licenses/)
Font Awesome
SIL OFL 1.1 (http://scripts.sil.org/OFL)
MIT License (http://opensource.org/licenses/mit-license.html)
CC BY 3.0 ( http://creativecommons.org/licenses/by/3.0/)
Powered by Worktribe © 2024
Advanced Search