Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/36451
Title: All-elastomer sensor technology for three-axial contact force measurement
Authors: Wang, Fei
Advisors: Tao, Xiaoming (ITC)
Keywords: Detectors -- Materials.
Elastomers.
Issue Date: 2015
Publisher: The Hong Kong Polytechnic University
Abstract: With more upcoming aging societies, the need for surgery robots in hospitals, and the demand for automated assembly lines, reliable and soft tactile sensing technologies have become imperative for human-machine interactions in robotics, biomedicine and health-care, etc. During the past decades, substantial effort has been made to create a touch-sensitive artificial skin. Such endeavors produced diverse tactile sensors with distinctive traits analogous to or even outperforming that of the human skin. Nevertheless, it remains challenging to produce a tactile sensor with sufficiently high sensitivity, stretchability, softness and three-axial force mapping capability, especially in the low pressure regime (<10 kPa). This challenge is mostly due to the limited study on elastomeric conductive composites with sufficiently low time-dependency, an effective structural design, reliable and stretchable conductors compatible to elastomeric materials, and the processing techniques of relevant soft materials. To address these issues, a systematic research has been conducted in this thesis towards all-elastomer sensor technology for three-axial contact force measurement. Inspired by the skin of a toad, cell-structured pressure sensors were designed and fabricated by injecting corn oil between two polydimethylsiloxane (PDMS) layers and printing carbon/silicone strain gauges. When being evaluated on an electromechanically coupled testing device, the sensor demonstrated a high sensitivity of 0.643 kPa-1 in a low pressure regime (0.1-7 kPa) with the lowest hysteresis around 7.7%, which is on a par with reported flexible pressure sensors. But unlike flexible sensors, this sensor was exceptionally soft since it was mainly composed of elastomeric materials and liquid. Critical issues examined include hysteresis, liquid/PDMS compatibility and elastic conductive tracks. Previous research on hysteresis has attributed to time-dependency of the carbon/silicone composite. However, this study revealed a geometrical effect, that is, the hysteresis is related to the printing thickness and aspect ratio, of which optimized values were 50 µm and 3.2:1 respectively by using a central composite design method. Among the liquids considered, polyethylene glycol (PEG) 400 was more compatible with the PDMS used, achieving a shelf life of over 4 months.
To address the problems on stretchable conductors, a new kind of printable, stretchable and conductive silver/silicone composite has been explored, of which the conduction development mechanism was experimentally investigated with consideration of heating temperature and time. The silver/silicone mixture was thermally sintered within 20 seconds, achieving a conductivity as high as 4000 S/cm. Such a fast fabrication method is comparable with photonic and microwave sintering of conductive inks thereby potentially enabling roll-to-roll industrial production. When serpentine structures were employed, the printed connectors selected by a comparative study demonstrated a fatigue resistance of over 10 000 cycles at 40% maximum strain, which is among the best in reported printable connectors. Based on the above studies, a biomimetic all-elastomer sensor array for three-axial contact force measurement was fabricated by injecting PEG 400 between two PDMS layers with stencil printed carbon/silicone strain gauges and silver/silicone conductive interconnects. Being composed exclusively of elastomeric materials and liquid, the sensor array exhibited a stretchability and elastic modulus approaching those of a human skin, which marks substantial progress of tactile sensors in this regard. The sensor was tested on an electromechanically coupled setup with a pulley to redirect loading. A sensitivity of 0.097 N⁻¹ to normal forces below 1.62 N and an excellent sensitivity of 0.337 N⁻¹ to shear forces below 1.3 N (with a normal force at 0.4 N) have been achieved, exceeding the best reported flexible 3D force sensors, especially for small shear forces below 0.5 N. Such a soft shear force sensor can be used in gait analysis, slippage detection and dexterous manipulation of robots, etc, where friction measurement is of utmost importance. A geometric model on deformation of the sensor under compression further confirmed the arrangement of strain gauges on sensors. Moreover, another design of soft pressure sensor (SPS) as proposed in the preliminary study was concretized for smart protective clothing against impact loadings. Both pressure measuring range and sensitivity of the SPS are tunable. Dynamic calibration and evaluation of the sensors were carried out by using a drop-tower impact method. The SPS demonstrated a large workable pressure range (0-8 MPa), a high sensitivity (100% /MPa), an excellent repeatability (lowest non-repeatability ±2.4% from 0.8 to 8 MPa) as well as a good shelf stability (the lowest drift in 6 months was 2.6%), which is by far the best reported result for soft pressure sensors against impact loadings. Additional quasi-static compression tests revealed strain rate effect of the SPS. Numerical analysis based on Finite Element method illustrated that the cylinder height/diameter ratio of 0.5 would be the best in terms of both electrical response and shear resistance. Smart clothing equipped with such SPS will provide spatial and temporal pressure distributions on a human body during contact sports, traffic accidents or occupational accidents, thereby greatly shortening the detection time of rescue, reducing injury and saving more lives.
Description: PolyU Library Call No.: [THS] LG51 .H577P ITC 2015 WangF
xxiii, 209 pages :color illustrations
URI: http://hdl.handle.net/10397/36451
Rights: All rights reserved.
Appears in Collections:Thesis

Files in This Item:
File Description SizeFormat 
b28277636_link.htmFor PolyU Users208 BHTMLView/Open
b28277636_ira.pdfFor All Users (Non-printable)6.96 MBAdobe PDFView/Open
Show full item record

Page view(s)

108
Last Week
5
Last month
Checked on Sep 25, 2017

Download(s)

43
Checked on Sep 25, 2017

Google ScholarTM

Check



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