engineering research theme 3 retreat...iii preface welcome to this special edition focused on human...

“Human Technologies”

November 2016

Engineering Research Theme 3 Retreat

Enabling Technologies in Sport and Health

Edited by

Daniel A. James Zia Javanbakht Andreas Öchsner

Human Technologies Engineering Applications in Sport and Health

Hosted by

Griffith University

SABEL Labs

Venue

Sheraton Grand Mirage Resort, Gold Coast

71 Sea World Drive, Main Beach, Gold Coast QLD 4217 Australia

Editors

A/Prof Daniel A. James, Queensland Sports Technology Cluster

Zia Javanbakht, Griffith University

Professor Andreas Öchsner, Griffith University

Scientific Committee

A/Prof Daniel A. James, Queensland Sports Technology Cluster

Professor Andreas Öchsner, Griffith University

ISBN-13: 978-0-9875232-1-1

Published by:

Queensland Sports Technology Cluster, Griffith University

Queensland, Australia

© Copyright Australian Sports Technologies Network, 2013, the authors and their institutions.

Printed in Australia by Griffith University

i

Table of Contents

Preface ................................................................................................................................................ iii

Innovation strategies for research technologists engaged in sports and health applications

Daniel A. James ................................................................................................................................. 1

Research in the Triangle of the School of Engineering, the School of Medical Sciences, and the Gold

Coast Hospitals

Andreas Öchsner .............................................................................................................................. 3

On the preliminary design of an instrumented ball for sports applications

Hugo G. Espinosa, Andreas Ӧchsner, Ali Mirnajafizadeh, and Daniel A. James .............................. 5

Non-destructive detection of defects in composites - technology and applications

Huaizhong Li, Wayne Hall, and Andreas Öchsner ............................................................................ 7

Platform Technologies and Visualisation Tools for Sensors in Sport and Health Applications

Raymond Leadbetter, and Daniel A. James ...................................................................................... 9

Optimising for power in wearables

David Rowlands, Mitchell McCarthy, and Daniel James ................................................................ 11

Microwave Imaging for Biomedical Applications

Andrew Seagar ............................................................................................................................... 13

Novel Robust Sensor for Intraoral Bite Force Measurement in Human Subjects

Jarred Fastier-Wooller, Hoang-Phuong Phan,

Toan Dinh, Tuan-Khoa Nguyen, and Dzung Viet Dao .................................................................... 17

Assessing the impact of integrating IMU wearable technology in an elite netball training

environment for performance analysis

Jonathan Shepherd, Christine Voge, David Rowlands, and Daniel James ..................................... 19

Designers Don’t Text - They Visualise: Integrating Coding into the University Design Curriculum

James Novak, and Jennifer Loy ....................................................................................................... 21

Modelling resistance exercise lifting technique with inertial sensors:

Knee extension during deadlifts

Sam Gleadhill, Daniel James, and James Lee ................................................................................. 23

Progress in Metal Additive Manufacturing towards a Wide Implementation in Dentistry

Leonhard Hitzler, Frank Alifui-Segbaya, Wayne Hall, and Andreas Öchsner ................................. 27

Using Inverse Finite Element Simulation to Characterise Soft Tissues by MSC Marc/Mentat

Zia Javanbakht, and Andreas Öchsner ........................................................................................... 31

Fused deposition modeling of 3D parts using acrylonitrile butadiene styrene (ABS) polymer

Maksym Rybachuk, Charlène Alice Mauger, Thomas Fiedler, and Andreas Öchsner ................... 33

iii

Preface

Welcome to this special edition focused on Human Technologies. For the last decade our university

has been engaged in biomedical and sports engineering research activities. The former had been

something of a curiosity and the latter the source of some amusement until this new field began to

transform the way sport is played.

Today our engineering school encompasses a wide variety of disciplines that contribute to biomedical

and sports engineering research activities. We recognise these as Human Technologies as there are

many similarities between the two application areas. Engaging in human technologies requires a

mind-set shift for the research tyro and experienced alike requiring often that it become a holistic

human centric activity, rather than something to be done in the lab that we might try and fit to a human

benefit.

This special edition is a snap shot of recent and emerging research activities, and encompasses

strategies to engage with industry, the medical community featuring work on sensor and devices,

modelling/simulation and experimenting. It features activities from SABEL Labs, gCORE, School of

Engineering, School of Medical Sciences and some of the hospitals in the vicinity of our campuses.

We hope you enjoy reading this as much as we did writing it.

December 2016 Daniel James and Andreas Öchsner

Engineering Research Theme 3 Retreat Enabling Technologies in Sport and Health

Innovation strategies for research technologists engaged in sports and health

applications

Daniel A. James

SABEL Labs, Griffith University, Australia

[email protected]

Keywords: Innovation, Platform Technologies, Strategy.

Extended Abstract. The challenges of bringing research from out of the laboratory to have impact

in the application areas of sports and health are many. The least of which is that bench based

experiments are not nearly so predictable as working with humans. In this case its useful to take a

step back from our daily activities, that of pursuing our very interesting lines of enquiry to ask

ourselves some rhetorical questions. How do we reach out? and are there processes that can be helpful

for this? Happily there are and innovation is more than just a buzzword, there is actually considerable

body of work behind this. It is said that innovation is creativity plus implementation. Creativity is

something we do very well in the research context, implementation is where specific demand-pull

from industry can be very helpful in providing direction.

Some very useful approaches come from the innovation and particularly the startup literature [1],

where run lean and moving with agility are important. Also its helpful to think about who our

customers might be. Of course using the language of ‘customers’ is somewhat strange to use in a

research context, where granting bodies and journals might be our only contact with the world.

Getting our name our there might involve some thought for marketing, which includes talking to

people and generating good word of mouth [2] and in the world of sports, some thought to branding

is also important [3]. Its also useful in some senses to think of our partnerships with other researchers,

sports organisations and private industry as business to business (B2B) customers. The next step is

building a kind of intimacy with them, which is important to understanding their needs and the

demand pull, it's a fairly common strategy these days [3]. In our laboratory, where our specialty is

wearable sensors [4] we realized that setting up the building blocks for each project where often the

same and there were great efficiency gains (operational efficiency) to be made here. By identifying

this it made life easier for student researchers and reduced our response time to trying out an idea or

meeting a partner’s needs. Thus we developed a series of platform technologies using our sensor

expertise [5] around a hardware and software framework [6], built with the need for flexibility in

mind and thus could be applied to a variety of application [7]. In fact the tool soon became quite

popular within our networks and so was a way to rapidly gain scale through developing partnerships

around the technologies, where our core domain expertise in wearables was helpful in working in a

transdisciplinary way with specialist sports and health applications.

With the research environment constantly under challenge, the adoption of these approaches around

a core area of expertise can help protect and expand areas of research strength. Together with

judicious use of strategies used in technology industries [8] it can help meet the emerging agendas of

research and innovation together with the requirements to work more closely with external

organisations and businesses.


References

[1] Blank, S., & Dorf, B. (2012). The startup owner's manual: The step-by-step guide for building a

great company.

[2] Merrilees, B. (2007). A theory of brand-led SME new venture development. Qualitative Market

Research: An International Journal, 10(4), 403-415.

[3] Michael Levens (2012), Marketing: Defined, Explained, Applied (2e). Prentice Hall, Pearson

[4] James, D. A., & Petrone, N. (2016). Sensors and Wearable Technologies in Sport: Technologies,

Trends and Approaches for Implementation. SpringerBriefs in applied sciences and technology.

[5] Cutmore, T. R., & James, D. A. (2007). Sensors and sensor systems for psychophysiological

monitoring: A review of current trends. Journal of Psychophysiology, 21(1), 51-71.

[6] Espinosa, H. G., Lee, J., & James, D. A. (2015). The inertial sensor: A base platform for wider

adoption in sports science applications. Journal of Fitness Research, 4.

[7] McNab, T., James, D. A., & Rowlands, D. (2011). iPhone sensor platforms: Applications to sports

monitoring. Procedia Engineering, 13, 507-512.

[8] Byers, T. H., Dorf, R. C., & Nelson, A. J. (2011). Technology ventures: from idea to enterprise.

New York: McGraw-Hill.


Research in the Triangle of the School of Engineering, the School of Medical

Sciences, and the Gold Coast Hospitals

Andreas Öchsner

Griffith School of Engineering, Griffith University (Gold Coast Campus), Parklands Drive

Southport Queensland 4214, Australia

[email protected]

Keywords: Griffith University, Health and Knowledge Precinct, Gold Coast University Hospital, Gold Coast

Private Hospital, Gold Coast Commonwealth Games Village, Interdisciplinary Research.

Extended Abstract. Griffith University Gold Coast Campus is growing in the next few years in a

unique infrastructure, the so-called Gold Coast Health and Knowledge Precinct, see Fig. 1. Residing

after 2018 in the former Gold Coast Commonwealth Games Village, this project will assemble in the

closed neighborhood the teaching and research facilities of Griffith University and two new and

modern hospitals to strengthen and coordinate research, for example, in the biomedical area. It should

be not forgotten that this precinct is situated vey close to Surfers Paradise and the famous Gold Coast

beaches, easy to access via the new Gold Coast Light Rail.

Figure 1. Stakeholders of the Gold Coast Health and Knowledge Precinct.

This unique infrastructure and a research focus in the biomedical and bioscience area opens new

research opportunities for the School of Engineering, especially for the Disciplines of Mechanical,

Electrical/Electronic as well as Mechatronic Engineering. The following list summarizes a few

projects, which were already designed in this new context of multidisciplinary research:

3D printing of artificial bone scaffolds based on the polymer printing facilities in the School

of Engineering [1,2], experimental testing and numerical simulation [3,4,5] under the

consideration of damage evolution [6].

Finite element characterization of biological materials with depth-sensing indentation [7].


Numerical simulation and experimental investigations of trabecular bone [8,9]; in cooperation

with University Technology Malaysia.

Application of selective laser melting in dentistry [10]; in cooperation with the School of

Dentistry and Oral Health (DOH).

Fracture resistance of lithium disilicate crowns in simulated oral environment [11]; in

cooperation with DOH.

Systematic pilot study on radial head fracture; in cooperation with Gold Coast University

Hospital (GCUH).

Design of a monitoring system for club foot correction; in cooperation with GCUH.

Bite strength in human facial morphology [12]; in cooperation with DOH and Griffith

Criminology Institute.

References

[1] G. Hughes, A. Öchsner, Design, Manufacture and testing of three-dimensional scaffolds, in: A.

Öchsner, H. Altenbach (Eds.), Applications of computational tools in biosciences and medical

engineering, Springer, Cham, 2015, pp. 133-179.

[2] J. Harbusch-Hecking, A. Öchsner, Simulating the mechanical properties of three-dimensional

printed artificial bone scaffolds, Materialwiss. Werkst. 47, 5-6 (2016) 549-563.

[3] A. Öchsner, Computational statics and dynamics: An introduction based on the finite element

method, Springer, Singapore, 2016.

[4] A. Öchsner, M. Öchsner, The finite element analysis program MSC Marc/Mentat, Springer,

Singapore, 2016.

[5] A. Öchsner, M. Merkel, One-dimensional finite elements – an introduction to the FE method,

Springer Verlag, Berlin, 2013.

[6] A. Öchsner, Continuum damage and fracture mechanics, Springer, Singapore, 2016.

[7] Z. Javanbakht, A. Öchsner, Advanced finite element simulation with MSC Marc: application of

user subroutines, Springer, Cham, 2017.

[8] A. Syahrom, M.R.A. Kadir, M.N. Harun, A. Öchsner, Permeability study of cancellous bone and

its idealised structures, Med. Eng. Phys. 37 (2015) 77-86.

[9] S.J. Fatihhi, M.N. Harun, M.R.A. Kadir, J. Abdullah, T. Kamarul, A. Öchsner, A. Syahrom,

Uniaxial and multiaxial fatigue life prediction of the trabecular bone based on physiological loading:

a comparative study, Ann. Biomed. Eng. 43 (2015) 2487-2502.

[10] L. Hitzler, C. Janousch, J. Schanz, M. Merkel, F. Mack, A. Öchsner, Non-destructive evaluation

of AlSi10Mg prismatic samples generated by selective laser melting: Influence of manufacturing

conditions, Materialwiss. Werkst. 47, 5-6 (2016) 564-581.

[11] N. Nawafleh, F. Mack, A. Öchsner, Masticatory loading and oral environment simulation in

testing lithium disilicate restorations: a structured review, in: A. Öchsner, H. Altenbach (Eds.),

Applications of computational tools in biosciences and medical engineering, Springer, Cham, 2015,

pp. 189-215.

[12] J. Fastier-Wooller, H.-P. Phan, T. Dinh, T.-K. Nguyen, A. Cameron, A. Öchsner, D.V. Dao,

Novel low-cost sensor for human bite force measurement, Sensors 16, 8 (2016) 1244 (10 pages).


On the preliminary design of an instrumented ball for sports applications

Hugo G. Espinosa1,3,a*, Andreas Ӧchsner2,b, Ali Mirnajafizadeh2,c

and Daniel A. James1,3,d, 1School of Engineering, Nathan Campus, Griffith University, Brisbane, Queensland, Australia

2School of Engineering, Gold Coast Campus, Griffith University, Brisbane, Queensland, Australia

3SABEL Labs, Griffith University, Brisbane, Queensland, Australia

[email protected], [email protected], [email protected], dsabellabs.com

Keywords: cricket, inertial sensors, bowling, bat, smart ball, impact forces, shock absorption.

Extended Abstract. An instrumented hard ball that can measure aspects of athlete’s/player’s

performance including flight time, rotational velocity (spin), distance, velocity and impact forces, is

being developed. The Smart Ball will allow coaches to work with quantifiable outcomes from the

flight of delivery for each bowler. In the case of a spin bowler, different strategies for a bowler’s

action can be assessed in terms of how many revolutions per second, and therefore ability to spin the

ball, have been imparted onto the ball as it leaves the bowler’s hand. The technology transmits a

wireless signal and/or stores the characteristics of the ball’s flight. The characteristics of the ball are

not altered. The instrumented ball uses a series of mini inertial sensors, such as tri-axial

accelerometers, gyroscopes and magnetometers that are encased in a novel impact resistant housing.

The sensors allow for the determination of quantifiable measures of ball flight, impact and usage. An

in-house sensor platform based on SABEL Sense [1, 2] is being used on the instrumented ball. The

platform consists of wearable sensors that collect data from digital MEMS (Microelectromechanical

systems) inertial sensors. The accelerometer is capable of measuring acceleration forces of ±10g in

three perpendicular directions (g being the gravitational force). The platform contains wireless

connectivity (2.4 GHz) for real-time data streaming.

The impact force between a cricket ball and a bat is extremely high, so an impact proof system,

together with a shock damping structure are required for the protection of the electronics. The average

force acting on the ball during the hit is approximately 10 kN for a velocity of 25 m/s [3]. The

challenge lies in the development of a shock protection system, which is being developed through the

assessment of different materials such as foams, cork, polycarbonates, nylon, and novel structures

based on the trabecular bone structure of the body to protect the electronics from very large shocks,

yet maintain the structural integrity and feel of the ball to players. These materials are being tested

using a drop test tower.

The dynamic impact characteristics will be determined experimentally through the coefficient of

restitution [4, 5] and contact time. In addition, a Finite Element model will be used to determine the

contact properties associated with impact between cricket balls and bats [6].

The final prototype will result in a technology with the potential for large impact on sporting

performance coaching. Due to its portability, the ball kinematics will be determined in an outdoor

environment.


References

[1] H. G. Espinosa, J. Lee, D. A. James, The inertial sensor: A base platform for wider adoption

in sports science applications, Journal of Fitness Research, 4, 1 (2015) 13-20.

[2] N. B. Nordsborg, H. G. Espinosa, D. V. Thiel, Estimating energy expenditure during front

crawl swimming using accelerometers, Procedia Engineering, 72 (2014) 132-137.

[3] N. Cheng, M. Takla, A. Subic, Development of an FE model of a cricket ball, Procedia

Engineering, 13 (2011) 238-245.

[4] F. Colombo, K. Seibert, H. G. Espinosa, D. V. Thiel, Novel methodology for measuring the

coefficient of restitution from various types of balls and surfaces, Procedia Engineering, 147

(2016) 872-877.

[5] H. G. Espinosa, K. Seibert, F. Colombo, D. V. Thiel, Measuring the court pace rating of tennis

courts using low-cost portable devices, Symposium on Sports and Human Dynamics (SHD),

Yamagata, Japan, Nov. 2016.

[6] T. Y. Pang, A. Subic, M. Takla, Finite element analysis of impact between cricket ball and

cantilever beam, Procedia Engineering, 13 (2011) 258-264.


Non-destructive detection of defects in composites - technology and applications

Huaizhong Li1, a *, Wayne Hall1,b and Andreas Öchsner1,c 1Griffith School of Engineering, Gold Coast campus, Griffith University, QLD 4222, Australia

a [email protected]; [email protected]; b [email protected]; c [email protected]

Keywords: Composite structures; Defects; Non-destructive detection; Vibration.

Extended Abstract. Advanced composite materials such as fiber reinforced plastics offer distinct

properties including high strength and low weight, good vibrational damping, low coefficient of

thermal expansion, fatigue strength, resistance to corrosion, oxidation and wear. They are now widely

used in aircraft, aerospace, automobile, marine, and sports equipment sectors. However, defects in

composite structures can occur during materials processing, component manufacture, or in-service

use, which make the composite structures susceptible to reduction in performance [1]. Many defect

types may exist in composite structures, for example, cracks, delamination and debond, impact

damage, embedded foreign objects or contamination, voids [2]. Among these, delamination and

debond are considered as the most commonly observed failure modes in composite structures [1].

The presence of defects can affect the strength of composites, reduce the buckling load capacity and

the stiffness substantially, which in turn will affect the structural performance and cause failure of

composites [3]. It is important to be able to discover the defects in an early stage to avoid potential

catastrophic consequence. Since damage is often hidden within the structure, with little to no surface

indication, non-destructive testing approaches must be evaluated and employed. There are numerous

non-destructive testing (NDT) techniques that have been developed for the inspection and detection

of defects in composite structures without damaging the component. Examples of available non-

destructive defect detection methods are ultrasonic testing, radiography, interferometry, X-ray,

thermography, microwave, eddy current and acoustic emission [2,4]. Different principles are

employed in each of these techniques to detect the defects in the materials. Considering the diversity

of the shape, size, physical and material properties of the component being tested, different NDT

approaches may have different advantages and limitations.

Ultrasonic testing is a very efficient method for materials quality inspection. It uses high frequency

sound signals beyond human hearing (> 20 kHz) to evaluate the depth and size of the nonconformity

by analyzing the measured change of sound amplitude as the sound passes through the material.

There are two basic ultrasonic detection approaches. The pulse-echo (A-scan) detection is by

analyzing the reflected sound signal, while the through transmission (C-scan) method is by analyzing

the transmitted signals through the material [2,5]. Ultrasonic inspection is very sensitive to commonly

found defects in composites such as delamination and debond. However, it is a local detection

approach, which may take significant amount of time due to the limited scanning rate. Radiography

involves penetrating the object with short wavelength electromagnetic radiation [4]. Computed

tomography (CT) is a new and precise radiography NDT method. It uses a scanning head that rotates

around the object and an X-ray beam passes through the object and produces a set of slices, which

are reconstructed to the three-dimensional data array [6]. The industrial CT can achieve a very high

spatial resolution of lower than 1 m, which allows for identification of even single fibre breakage in

the fibre-reinforced composite structure [6]. The main challenge for using the CT method is on the

interpretation of measurement results caused by the measurement noise and artefacts. The very high

cost of inspections is also a major concern [6]. In thermography testing, a composite structure is

excited by an external heating source. The differences in a temperature distribution on its surface are

observed through infrared (IR) imaging recorded by an IR camera to detect defects within the

component [4,6]. The principle behind is that the existence of defects will disturb the heat flow and

hence can be detected [4]. The aforementioned approaches can be classified as local-damage

detection methods which are usually time consuming and costly.

mailto:[email protected]





For large and more complex structures, there is often a demand for quick detecting of damage

throughout the whole structure, which requires global-damage detection methods. Global vibration-

based structural damage detection has been proposed in [7,8]. It is believed that defects within the

structure can cause a change in the structural dynamic properties in terms of the frequency response

function (FRF) and modal parameters of the structural system. Therefore, detection of the change of

the structural modal parameters including natural frequencies, modal damping, and modal shapes can

be used as a signal of early damage occurrence within the composite component [9]. The current

vibration-based method is model-dependent which means that the accuracy of the model affects the

effectiveness of the method significantly. The current study aims to develop a hybrid method for rapid

and reliable detection of defects in composite structures, which will use a vibration based approach

for a rapid global detection to check if a defect exists, and then followed by a local detection method

to find out the exact location and type of the defect. The developed method will be applied to detect

the damages in composite sports equipment such as a hokey stick.

References

[1] K. Senthil, A. Arockiarajan, R. Palaninathan, B. Santhosh, K.M. Usha, Defects in composite

structures : Its effects and prediction methods – A comprehensive review, Compos. Struct.

106 (2013) 139–149. doi:10.1016/j.compstruct.2013.06.008.

[2] R.B. Heslehurst, Defects and Damage in Composite Materials and Structures, CRC Press,

2014.

[3] X. Yin, D.A. Hutchins, Composites : Part B Non-destructive evaluation of composite

materials using a capacitive imaging technique, Compos. Part B. 43 (2012) 1282–1292.

doi:10.1016/j.compositesb.2011.10.018.

[4] M. Jolly, A. Prabhakar, B. Sturzu, K. Hollstein, R. Singh, S. Thomas, et al., Review of Non-

destructive Testing ( NDT ) Techniques and their applicability to thick walled composites,

Procedia CIRP. 38 (2015) 129–136. doi:10.1016/j.procir.2015.07.043.

[5] T.D. Orazio, M. Leo, A. Distante, C. Guaragnella, V. Pianese, G. Cavaccini, Automatic

ultrasonic inspection for internal defect detection in composite materials, NDT&E Int. 41

(2008) 145–154. doi:10.1016/j.ndteint.2007.08.001.

[6] A. Katunin, K. Dragan, M. Dziendzikowski, Damage identification in aircraft composite

structures : A case study using various non-destructive testing techniques, Compos. Struct.

127 (2015) 1–9. doi:10.1016/j.compstruct.2015.02.080.

[7] Y. ZOU, L. TONG, G.P. STEVEN, Vibration-Based Model-Dependent Damage

(delamination) Identification and Health Monitoring for Composite Structures — A Review,

J. Sound Vib. 230 (2000) 357–378. doi:10.1006/jsvi.1999.2624.

[8] Y.J. Yan, L. Cheng, Z.Y. Wu, L.H. Yam, Development in vibration-based structural damage

detection technique, Mech. Syst. Signal Process. 21 (2007) 2198–2211.

doi:10.1016/j.ymssp.2006.10.002.

[9] S.S. Kessler, S.M. Spearing, M.J. Atalla, C.E.S. Cesnik, C. Soutis, Damage detection in

composite materials using frequency response methods, Compos. Part B. 33 (2002) 87–95.

Figure 1. Experimental

setup of vibration-based

defect detection


Platform Technologies and Visualisation Tools for Sensors in Sport and Health

Applications

Raymond Leadbetter1, a * and Daniel James1, 2 1 SABEL Labs, School of Engineering, Griffith University, Brisbane, Queensland, Australia

2 Centre of Excellence for Applied Sports Science Research, Queensland Academy of Sport, Brisbane

Queensland, Australia

[email protected]

Keywords: Inertial sensors, MATLAB, Human monitoring, 9DOF, visualisation

Introduction: Inertial sensors have been a key component in human movement studies conducted in

the lab [1]. An in-house inertial sensor system was created over several iterative development stages

[2]. The in-house system provides the flexibility and customisation required by the lab.

A large variety of trials are performed in the lab, consisting of high intensity tasks such as cricket

monitoring [3] and running gait; medium intensity activities such as swimming [2]; and low intensity

daily tasks such as walking or lifting [4]. The sensor system hardware must be flexible enough to

capture the wide range of expected intensity, while still maintaining a useful resolution for analysis.

The software controlling the sensor system is required to be robust, user-friendly, and highly

customizable. A key requirement of the software system is to provide easy to use video and inertial

sensor synchronisation. This aids in data analysis by providing a reference to key events and patterns.

Results: The variety of activities requiring monitoring led to the development of sensor hardware

with a dynamically adjustable measurement range. The primary inertial measurement unit (IMU)

used by the sensor device (MPU-9150) has an adjustable accelerometer scale (2/4/8/16g) and an

adjustable gyroscope scale (250/500/1000/2000deg/s). This measurement range meets most of the

system usage cases. High intensity tasks required the development of a modular daughter board with

sensors capable of measuring up to 200g and 6000deg/s. The desired measurement range is set in

software to meet the user’s specific application.

Discussion: The software was developed to perform two primary purposes. First is sensor control,

using a wireless link from a PC the user can interface to all sensor devices. The sensor interface

provides device status (battery life, logging state, previous sessions) and device control, allowing for

sensor configuration changes, data download, and network control. The second primary purpose is

data analysis, which often starts with viewing synchronised video and sensor data. The

synchronisation process was streamlined by using a synchronisation LED combined with wirelessly

connected sensors. The synchronisation LED is filmed when a sync marker command is sent from

the software, as shown in Figure 1. This provides a reference frame in the video and the system marks

the location in sensor data. This allows for semi-automated synchronisation, as the user must only

input the correct video frame.


Figure 1 Demonstration of a synchronisation command from the sensor system.

Once the data is synchronised the analysis tool [6] can be used to mark locations of key events and

patterns. The software can also be used to apply certain data processing techniques, including filtering

and sensor orientation estimations [5]. This assists the user in visualising the inertial sensor data and

the effect of movements on the device.

Conclusion: The developed inertial sensor system has helped service the lab by providing a flexible

measurement device to suit a wide range of applications. Future development will continue to refine

the system based off user demands and technology advancements. Development is progressing into

the cloud by integrating the system with a recently developed cloud-based data storage system. This

provides a searchable database with data tagging and indexing, with the ability for data distribution

and sharing between lab members.

References:

[1] Espinosa, H.G., Lee Jim, James, D.A. (2015). The inertial sensor: A base platform for wider

adoption in sports science applications. Journal of Fitness Research, 4, 1, 13-20.

[2] James, D.A., Leadbetter, R.I., Neeli, M.R., Burkett, B.J., Thiel, D.V. & Lee, J.B. (2011). An

integrated swimming monitoring system for the biomechanical analysis of swimming strokes. Sports

Technology, 4(3-4), 141-150.

[3] W Spratford, M Portus, A Wixted, R Leadbetter, DA James (2015). Peak outward acceleration

and ball release in cricket. Journal of sports sciences 33 (7), 754-760.

[4] S Gleadhill, JB Lee, D James (2016). The development and validation of using inertial sensors

to monitor postural change in resistance exercise. Journal of biomechanics 49 (7), 1259-1263.

[5] T Wada, RI Leadbetter, DD Rowlands, DA James (2016). Developing an AHRS tool for a

wearable sensor to obtain attitude and heading information. Journal of Fitness Research 5 (2).

[6] R Leadbetter, D James (2016). Platform technologies and Visual analytics for inertial sensors.

Journal of Fitness Research 5 (2).


Optimising for power in wearables

David Rowlands, Mitchell McCarthy, Daniel James Sport and Biomedical Engineering Laboratories (SABEL), Griffith University, Nathan Campus, 170 Kessels Road, Nathan QLD, Australia, 4111

*Correspondence: [email protected]; Tel.: +61-737-355-084

Keywords: Wearable, IMU, Inertial Sensors, Optimisation, Power

Introduction: Recently the popularity and application of wearable technologies has greatly

increased in the medical, sporting and consumer space [1]. These wearables are typically used for

activity classification and monitoring of sporting actions [1,2]. There are a number of features that

are common to all Wearable technologies:

They must be small and light weight so that they do not affect the activity being monitoring.

They have a sensor or sensors to monitor an activity. Typically these are inertial sensors.

They have sufficient storage to collect the data from the activity being.

They have a communication system to transmit the data to a remote device such as USB,

Bluetooth etc

They have a powerful enough microcontroller with appropriate firmware.

They have a large enough battery to allow sufficient time between recharging.

The issue of the time between charging Wearable devices is signifiant since it can affect the

duration of activity measurement and whether or not the processing is performed on the device or

off the device. Processing off the device will limit the ability for Real Time Monitoring but

processing on the device will limit the battery lifetime. This means that the hardware and the

software will both affect the time between battery recharging since both affect the power drain of

the battery. This paper will discuss what can be done to optimise both the hardware and the

software to reduce the power usage. Since a typical use of Wearable technology requires the use of

classifiers, then a case study showing the effects of the varied classifiers is also presented.

Hardware and Software: There are a number of procedures that can be performed to reduce the

hardware power drain :

Reduce the active time of the processor [3]. This means that it is best to run the processor

intermittently (burst mode) rather than continuously.

Reduce the clock frequency [3]. The higher clock frequency means that there is more

switching so that more power is used by the transistors in the microcontroller.

Use the low power modes of the microcontroller [3]. Most microcontrollers have a low

power mode that will draw A of current hence prolonging the battery charge.

Disable any unused subsystems. The microcontroller is built for general use and contains

many subsystems such as I2C, SPI, ADC, UART, GPIO etc. By turning off the subsystems

that are not in use then they will not draw any power hence reducing the power drain. This is

akin to switching off the lights in rooms that are not being used.

Minimise the usage of the radios (Bluetooth, wifi etc). In order to transmit these will draw

more power.

There are a number of procedures that can be performed to reduce the software power drain:

Write the code to take advantage of power modes and turn off unused subsystems.

Write the code to use interrupts rather than polling. This allows the microcontroller sleep

modes to be used.


Design the algorithms to reduce the computational complexity. The higher the

computational complexity the longer time the processor is running hence more switching is

occurring which uses more power.

Case Study: Commonly used classification algorithms were examined for their suitability for

Wearables with the aim to determine their effect on the power usage [4]. The supervised classifiers

had to distinguish between walking, running, stair ascending, and stair descending from data

gathered by an IMU mounted on the foot [4]. The time taken to execute the classifier algorithm was

taken as the metric since this is related to the power usage [4]. This analysis was performed on a

typical Desktop PC using the MATLAB numerical language. Figure 1 shows the timing results for

the different classifiers.

Figure 1 The time taken for the 1000 runs of the different classifiers in milliseconds [5].

Discussion and Conclusion: This paper noted that a desirable aspect for Wearables is to have a

long time between battery recharging. The power drain on the battery is related to both the

hardware and the software so they must both be taken into account when designing a Wearable

device. This paper gave a number of suggestions that can be used to reduce the effect on the battery

drain from the hardware and the software. A key aspect of the software is to carefully design the

algorithms used.

A case study was presented for the comparison of typical activity classifier techniques for a foot

mounted sensor. Table 1 gives the time taken for the typically used classifier algorithm types. The

results are as expected where the least computationally complex algorithms use less time to

compute and the more computationally complex algorithms use more time to compute. Therefore

for the desired accuracy, the least computationally complex algorithm should be chosen to reduce

the effect on the time between battery recharges.

References: [1] David D Rowlands, L Laakso, T McNab, Daniel A James, “ Cloud based activity monitoring system for health

and sport”, The 2012 International Joint Conference on Neural Networks (IJCNN), pp. 1–5, 2012.

[2] Jonathon Neville, David Rowlands, Andrew Wixted, Daniel James, “ Determining over ground running speed

using inertial sensors” Procedia Engineering., pp. 487-492, 2011.

[3] Murat Ozkan, “Fundamentals of Low Power Design,” http://electronicdesign.com/boards/fundamentals-low-

power-design, accessed Nov 2016.

[4] M. McCArthy, D.A.JAmes, J.B.Lee, T.Wada, D.Rowlands, “Effect of Machine Learning Techniques Upon

Wearable Devices,” SHD2016 Yamagata Japan., , no. 16-40, pp. A30, Nov. 2016.


Microwave Imaging for Biomedical Applications

Andrew Seagar

Griffith University School of Engineering, Gold Coast, Queensland, Australia

[email protected]

Keywords: Cauchy integral, electromagnetic radiation, inverse scattering, microwave imaging

Extended Abstract. The application of electromagnetic radiation in its various forms as a means of

obtaining useful diagnostic information in a non-invasive fashion is exploited from the lowest

frequencies with impedance imaging [1] to the highest frequencies with X-ray computed tomography

(CT) [2]. In imaging it is often believed that the resolution of the images produced is limited by the

size of the wavelength of the radiation. That may be the case for some techniques but it is by no means

a fundamental limit. For example, in the case of impedance imaging the wavelength of the radiation

is many times the size of the object under scrutiny but useful images showing structure detailed within

such objects are indeed obtained [3]. However the resolution is not as good as for X-ray CT where

higher frequencies are used. In spite of somewhat lower resolution, imaging at the lower

electromagnetic frequencies is attractive firstly because the photon energy is below the threshold for

ionization, thereby eliminating the adverse health affects of ionizing radiation, and secondly because

the cost of the technology is typically much lower. With ever improving technology both in terms of

the analogue components and circuitry available for data acquisition, and in terms of the

computational hardware available for image reconstruction there has of recent times been a

resurgence in interest for exploiting the intermediate frequencies occupying the microwave range of

the electromagnetic spectrum [4]. There is widespread effort to develop effective microwave imaging

systems for application to diagnosis of breast cancer [5], greenstick fractures [6], intracranial bleeding

[7] and osteoporosis [8].

From a theoretical viewpoint most methods used to reconstruct images from microwave radiation

either transmitted through or scattered from an object of interest require a model of the way the

electromagnetic radiation interacts with the material of which the object is composed. A popular

mathematical model are the equations developed by Maxwell [9] as a description of most classical

electromagnetic phenomena. Maxwell’s equations are suited for calculating directly the radiation

transmitted or scattered when the material properties are already known, but are not in a form suitable

to make the inverse calculation so easily. One approach is to make a first estimate of the material

properties then to refine the estimate iteratively by repeatedly calculating the corresponding

transmitted or scattered radiation and making adjustments to the estimate as necessary according to

how well the calculations match the actual measured data. For this approach to be successful it is

necessary to adopt a technique for calculating the scattered radiation which is highly efficient.

Calculating the scattered radiation from Maxwell’s equations is usually treated in terms of a set of

four first order partial differential equations, often converted for simplicity into a second order

ordinary differential equation. It has however been shown recently [10] that it is possible to take

another path and convert the four first order partial differential equations into a single first order

ordinary differential equation. The solution of first a single first order ordinary differential equation

is of course much easier and more efficient that the solution of either a second order ordinary

differential equation or a set of four first order partial equations. For the application of microwave

imaging to biomedical and other applications where efficiency is paramount there is a strong

imperative for adopting the newer approach.


The newer approach is implemented in the framework of Clifford algebra [11]. Although unfamiliar

to many, ultimately this is easier than the traditional approach of using vector calculus [12]. The new

1corresponding author approach leads to a solution in terms of the Cauchy integral [13] in its

multidimensional form [14], rather than as conventionally [15] in terms of integrals involving Green’s

functions. This approach has already been properly validated in one-dimension for materials of all

types [16], and in two dimensions initial successes have been reported [17]. Although the new method

is similar to the traditional approach in that they both solve a linear system of equations, the method

is sufficiently different to offer an implementation which avoids calculating most of the elements of

the corresponding matrix. Not calculating most matrix elements is a huge saving in term of

computational effort. Rather than inverting a matrix the method, known as the Clifford-Cauchy-Dirac

(CCD) technique [10], finds the solution iteratively in as few as twenty iterations using projections

defined from the Cauchy integral, onto convex sets [18]. The conventional method simply cannot do

that because it is not based on the Cauchy integral, and for that reason cannot achieve the same

computational efficiency.

Based on its computational efficiency the CCD technique is ideally suited for the purpose of

reconstructing diagnostic biomedical images from microwave radiation.

References

[1] A.D. Seagar D.C. Barber and B.H. Brown. Theoretical limits to sensitivity and resolution in

impedance imaging. Clinical Physics and Physiological Measurement, 8(A):13–31, November

1987.

[2] G. N. Hounsfield. Computerized transverse axial scanning (tomography): Part 1. description of

system. British Journal of Radiography, 46(552):1016–1022, 1973.

[3] Inez Frerichs and Marcelo B P Amato et al. Chest electrical impedance tomography examination,

data analysis, terminology, clinical use and recommendations: Consensus statement of the

translational eit development study group. Thorax, 0:1–11, October 2016.

[4] Paul M. Meaney. Microwave imaging and emerging applications. International Journal of

Biomedical Imaging, 2012:1–2, 2012.

[5] N. R. Epstein, P. M. Meaney, and K. D. Paulsen. 3d parallel-detection microwave tomography

for clinical breast imaging. Review of Scientific Instruments, 85(12):1–12, December 2014.

[6] Giuseppe Ruvio and Antonio Cuccaro et al. Microwave bone imaging: A preliminary scanning

system for proof-ofconcept. Healthcare Technology Letters, 3(3):218–221, September 2016.

[7] Stefan Candefjord and Johan et al. Winges. Microwave technology for detecting traumatic

intracranial bleedings: Tests on phantom of subdural hematoma and numerical simulations.

Medical & Biological Engineering & Computing, pages 1–12, October 2016.

[8] T Zhou, PM Meaney, MJ Pallone, S Geimer, and KD Paulsen. Microwave tomographic imaging

for osteoporosis screening: A pilot clinical study. In Engineering in Medicine and Biology Society

(EMBC 2010), Annual International Conference of the IEEE, pages 1218–1221, Buenos Aires,

Argentina, August31–04 2010.

[9] James C. Maxwell. A Treatise on Electricity and Magnetism, volume 2. Clarendon Press, Oxford,

UK, 1873.

[10] Andrew Seagar. Application of Geometric Algebra to Electromagnetic Scattering – the Clifford-

Cauchy-Dirac Technique. Springer, Singapore, 2016.

[11] William K. Clifford. Applications of Grassmann’s extensive algebra. American Journal of

Mathematics, 1(4):350–358, 1878.

[12] Edwin B. Wilson. Vector Analysis: A Text-Book for the Use of Students of Mathematics and

Physics. Charles Scribner’s Sons, New York, US-NY, 1901.


[13] Baron Augustin Cauchy. Résumé d’un mémoire sur la mécanique céleste et sur un nouveau

calcul appelé calcul des limites. In Bachelier, editor, Exercices d’Analyse et de Physique

Mathématique, volume 2, pages 41–98. Bachelier, Paris, France, 1841.

[14] Alan McIntosh. Clifford algebras and the higher-dimensional Cauchy integral. In Z. Ciesielski,

editor, Approximation and Function Spaces, number 22 in BCP series, pages 253–267. Banach

Center Publications, Warsaw, Poland, 1989.

[15] Walton C. Gibson. The Method of Moments in Electromagnetics. Chapman and Hall/CRC, Boca

Raton, US-FL, 2nd edition, 2015.

[16] Andrew Seagar and Ajalawit Chantaveerod. An EM scattering algorithm for all materials. In

International Conference on Electromagnetics in Advanced Applications (ICEAA’15), pages

256–259, Turin, Italy, September 7–11 2015.

[17] Andrew Seagar and Hugo Espinosa. A numerical comparison between EFIE/MoM and CCD

methods for EM scattering in two dimensions. In International Conference on Electromagnetics

in Advanced Applications (ICEAA’16), pages 328–330, Cairns, Australia, September 19–23

2016.

[18] C. Badea, S. Grivaux, and V.Müller. The rate of convergence in the method of alternating

projections. St. Petersburg Mathematical Journal, 23(3):413–434, 2012.

This page is intentionally left blank.


17

Novel Robust Sensor for Intraoral Bite Force Measurement in Human Subjects

Jarred Fastier-Wooller1,a, Hoang-Phuong Phan2,b, Toan Dinh2,c, Tuan-Khoa

Nguyen2,d, and Dzung Viet Dao1,2,e,* 1School of Engineering, Griffith University, Queensland 4215, Australia

2Queensland Micro- and Nanotechnology Centre, Griffith University, Queensland 4111, Australia

[email protected], [email protected], [email protected], [email protected], [email protected]

*corresponding author

Keywords: bite force; strain gauge; stainless steel; oral health.

Extended. Maximum Bite Force (MBF) measurements can be utilised by professional dentists to

assist in the diagnosis of a patient’s oral health. MBF measurements are also commonly used in the

design, rehabilitation, and evaluation of oral prosthetics and implants. Griffith University’s School of

Dentistry and School of Criminology and Criminal Justice submit a request for a 250 to 550 N

measurable range intraoral bite force sensor that is safe for use in both research and dental analysis.

Based on the conceptual design, a sensor prototype was designed and implemented using 420 stainless

steel and conventional milling process. The sensor was more compact and robust compared to the

acrylic prototype [1], and lower cost than silicon micro force sensor [2].

Sensor Design and Simulation

Structure of the sensor is shown in figure 1 (a), consisting of two metal pieces and a metallic strain

gauge bonded together using cyanoacrylate epoxy. Protective casings made from a Poly-Vinyl

Siloxane (PVS) addition silicone material commonly used in dentistry to make impression moulds of

a patient’s teeth, meaning the material is well tested and considered safe for intraoral use. The PVS

material is a robust silicone with excellent electrical resistive properties. When a bite force is applied

on the sensor, the upper plate will be deformed, leading to the change of resistance of the strain gauge

bonded to it.

Figure 1 (a) shows the sensors structure, with an overall size of 10 mm x 10 mm x 5 mm. Structural

simulation in Solidworks™ Simulation (2015) indicated that 440C stainless steel would be capable

of withstanding a total of 1300 N before plastically deforming. However, 420 stainless steel was

chosen due to its better availability in Australia. Using the same sensor design, 420 stainless steel

yields at approx. 300 N in simulations. The estimated measurement range of the sensor has thus been

reduced to below 23% of its initial 1300 N limit.

Fabrication and Characterisation Results

The sensor prototype was fabricated out of 420 stainless steel at Griffith University using a 3-axis

manual milling machine. The sensor went from a single piece of acrylic to layered stainless steel

plates that had to be held together with epoxy. Due to this, the application of the sensing element

(strain gauge) was made much easier than the acrylic prototype. The final prototype’s structure was

modified to reduce size and number of layers to two, while maintaining the benefit of being able to

easily bond the strain gauge to the sensor. The assembled sensor can be seen in figure 1 (b).


Calibration and testing is performed using an Instron universal testing machine in order to make

accurate and reliable comparisons between applied forces and sensor output. Calibration

measurements are performed using a Wheatstone bridge and a voltage amplifier (AD 623) connected

to an oscilloscope (MSO-X 3104A, Agilent Technologies, California, United States). A custom

designed circuit and software package capable of simultaneously reading, displaying, and saving the

data from the sensor in real-time on a connected computer was created for use in practical testing. A

Sparkfun HX711 and Arduino Pro Micro were used to provide a means to easily calibrate and tare

the sensor with software, allowing the circuit to produce readable force values without performing

manual conversions. Three human volunteers provided preliminary practical test data (GU Human

Ethics Protocol 2016/142), resulting in bite forces within the expected range derived from literature

(≤ 700 N) [1]. Linearity and repeatability of the sensor was improved over the acrylic prototype [1].

Despite the effective measurable force of the sensor being limited, the third prototype provide

exceptional results from the calibration process up to 300 N as seen in figure 1 (c).

Throughout the calibration and testing phase, it was noted that the sensor design would also be

suitable for other applications. Such applications include using the sensor in an array to perform

multi-point measurements (not limited to bite forces). The sensor could be effectively applied in

robotics due to its small size, high sensitivity, and modular functionality. An example of this would

be to use the sensor as haptic feedback to give robotics a highly effective and low-cost sense of touch.

(a)

(b) Figure 2. (a): CAD design of the sensor. (b): Assembled sensor with protective casing. (c): Calibration results of the sensor.

References

[1] J. Fastier-Wooller, H.-P. Phan, T. Dinh, T.-K. Nguyen, A. Cameron, A. Öchsner, D.V. Dao,

Novel Low-Cost Sensor for Human Bite Force Measurement, Sensors, 16 (8) 1244, 2016.

[2] D.V. Dao, T. Toriyama, J. Wells, S. Sugiyama, Six-degree of Freedom Micro Force-moment

Sensor for Application in Geophysics, Proceedings of the 15th IEEE International Conference on

Micro Electro Mechanical Systems 2002 (IEEE MEMS2002), pp.312-315, Las Vegas, Nevada, USA,

Jan.20-24, 2002.

350.4

350.6

350.8

351

351.2

351.4

351.6

0 50 100 150 200 250 300

Res

ista

nce

(o

hm

)

Force (N)

(c) Final Prototype - Resistance

Test 1 Test 2 Test 3 Test 4 Test 5 Average


Assessing the impact of integrating IMU wearable technology in an elite netball

training environment for performance analysis

Jonathan Shepherd *1,2, Christine Voge2, David Rowlands1, Daniel James1 1 Sport and Biomedical Engineering Laboratories (SABEL), Griffith University, Nathan Campus, 170 Kessels Road, Nathan QLD, Australia, 4111

2. Centre of Excellence for Applied Sport Science Research, Queensland Academy of Sport, Queensland Sport and Athletics Centre, Kessels Road,

Nathan QLD, Australia, 4111

*Correspondence: [email protected]; Tel.: +61-737-355-084

Keywords: Netball, IMU, Inertial Sensors, Wearables, Skill Acquisition

Introduction: Since its codification in 1901[1],netball has become increasingly popular with a

current playing population in excess of 20 million players [2]. The core aim of netball is for the

goal shooters to shoot the ball through a 3.05m high pole, with the team who scores the most goals

at the end of four 15 minute quarters winning the game. Subsequently, scoring has been deemed the

most important facet for competitive success of a netball team [2] with an impetus tracking

shooting percentages both in training and match environments. In high performance environments

shot tracking is generally manually recorded, a resource intensive process occurring both in training

and game environments. Recent advancements in technology, paralleled with an increased

professionalism within netball has led to new methods of performance analysis. Video analysis [3],

inertial sensor for workload intensity based classification [4], and GPS for load monitoring [5]have

been reported in literature as novel methods for performance analysis. The advent and increase in

popularity of crowd funding has led to a plethora of wearables based on inertial measurement units

(IMUs) being taken to market. One such device is the ShotTracker (ShotTracker, Kansas City,

United States of America), an IMU based basketball shooting unit, designed to automate the

tracking of basketball shooting percentages [6]. ShotTracker gives concurrent knowledge of results

in terms of shooting percentages, however there is no evidence in netball literature confirming the

effectiveness of this feedback in netball shooting.

Methods: Two elite female shooters (20yo, right handed) from an elite grade netball team were

chosen to participate in the study. The participants were the primary shooters for the same national

representative team, playing at an international netball competition which consisted of 6 games.

One player as the primary Goal Shooter (GS) and the other as the primary Goal Attack (GA). The

study was to assess whether utilising automated shot tracking in training would improve accuracy in

game. Three periods were examined, a control period of five weeks of state level based games with

no technology, an intervention period where shot tracking was used for five weeks, and finally the

international match period, consisting of 6 international matches. The intervention encompassed the

coaching staff using the ShotTracker system during specific shooting sessions. The coach, utilising

the iPhone application, had access to constant knowledge of shooting percentages during the

shooting session and delivering verbal cuing to the player when they deemed appropriate. Prior to

this study neither of the players or the teams coaching staff had used technology to automate shot

tracking. The ShotTracker system was set up as per the manufacturer’s instructions with the wrist

sensor, encased in a sweat band, mounted proximal to the distal radioulnar joint and a second, net

mounted sensor, clipped onto the net[6]. The beginning of the intervention period included a

baseline shooting test to validate the technology. This test encompassed a 10 minute shooting warm

up followed by 10 shots from a mid-range distance (2.45m from the goal post) at three locations,

directly in front of the post (0°), 69.7° to the left of centre, and 69.7° to the right of centre. On the

7th shot of each block, the coach put defensive pressure on the shooter with a pool noodle shown in

Figure 1. As an independent validation, the thirty shots for each shooter were filmed and the result


of the shot was recorded by a researcher blinded to the ShotTracker outputs. The research was

conducted under Griffith University Ethics (GU:2016/294).

Results: The validation test showed consensus between the manually recorded and automated

shooting outcomes, with both indicating that 54/60 shots were made. During the intervention

period 7 sessions were tracked, with the ShotTracker automatically tracking 686 goals from 906

attempts at a mean shooting percentage of 75.72%. The shooting percentages for the state league

and international games are shown below.

Figure 1. (Left) Tabulated field goal percentages over the three periods. (Right) Image showing the 7th shot from the centre position.

Discussion: Although the technology was validated without error, the coaching staff reported an

error if there was a goal post collision, where the vibrations were sufficient enough to shake the net

mounted device. When this occurred the coaching staff manually adjusted the ShotTracker attempts

and goals. Shooting percentages, as shown in Figure 1, improved after the ShotTracker was used.

Interestingly this improvement happened irrespective to the elevation in competitive defensive

pressure inherent with the increased competitive standard of the international competition. It should

be noted that the authors don’t feel the improvements should be entirely attributed to the knowledge

of results provided by the automated tracking system. The nature of being tracked will invariability

increase practice pressure, increasing athlete accountability and ensuring shooting targets are

adhered to. Furthermore, both the athletes and the coach reported the enjoyment of utilising

technology, essentially gamifying the training environment. With higher practice it could be

reasoned that shooters will more likely engage in a higher frequency and duration of deliberate

practice, potentially enhancing skill gains. Furthermore, this study is limited by a small sample size

an inherent issue when there are only two primary shooters in an elite team.

Conclusion: The results indicate that the accurate automation of shot tracking in netball training is

possible using wearable IMU technology. Marginal performance gains, in relation to shooting

percentages, were noted however a direct correlation between this and automated shot tracking

cannot be made solely from this study.

References: [1] M. Treagus, “Playing Like Ladies: Basketball, Netball and Feminine Restraint,” Int. J. Hist. Sport, vol. 22, no.

1, pp. 88–105, Jan. 2005.

[2] A. Delextrat and M. Goss-Sampson, “Kinematic analysis of netball goal shooting: A comparison of junior and

senior players,” J. Sports Sci., vol. 28, no. 12, pp. 1299–1307, Oct. 2010.

[3] R. E. Jenkins, L. Morgan, and P. O’Donoghue, “A case study into the effectiveness of computerised match

analysis and motivational videos within the coaching of a league netball team,” Int. J. Perform. Anal. Sport, vol. 7, no.

2, pp. 59–80, May 2007.

[4] S. J. Cormack, R. L. Smith, M. M. Mooney, W. B. Young, and B. J. O’Brien, “Accelerometer load as a

measure of activity profile in different standards of netball match play,” Int. J. Sports Physiol. Perform., vol. 9, no. 2,

pp. 283–291, Mar. 2014.

[5] T. Higgins, G. A. Naughton, and D. Burgess, “Effects of wearing compression garments on physiological and

performance measures in a simulated game-specific circuit for netball,” J. Sci. Med. Sport, vol. 12, no. 1, pp. 223–226,

Jan. 2009.

[6] “Shottracker, Inc. ‘Basketball Net Which Detects Shots That Have Been Made Successfully’ in Patent

Application Approval Process (USPTO 20160096067),” Telecommunications Weekly, p. 238, 27-Apr-2016.


Designers Don’t Text - They Visualise: Integrating Coding into the University

Design Curriculum

James Novak1, a *, Jennifer Loy2,b 1, 2Griffith University, Gold Coast, Australia

[email protected], [email protected]

Keywords: Human-Machine Interaction, Human-Machine Interfaces, Visual Programming Language,

Coding, Industrial Design, Prototyping, Higher Education

Human-Machine Interaction (HMI) is ubiquitous; picking up a mobile phone and searching for a

nearby restaurant or sharing an image through social media is carried out without a second thought.

It is part of daily life. HMI has recently emerged as a research discipline in its own right as the

complexities of interactions, and their impact not only on products’ performance, but on human

development itself, is beginning to be more fully understood. "HMI, as a field of investigation, is

quite recent even if people have used machines for a long time. HMI attempts to rationalise relevant

attributes and categories that emerge from the use of (computerised) machines. Four main principles,

that is, safety, performance, comfort and aesthetics, drive this rationalisation along four human factors

lines of investigation: physical (that is, physiological and bio-mechanical), cognitive, social or

emotional."1 The challenge for designers, and indeed the education of design students, is making

sense of these broad technical and emotional requirements within a single unified product or service.

Traditionally, industrial designers develop the physical form and function of a product, and engineers

and information technology (IT) specialists develop the internal electronics that connect the product

to the internet or power electromechanical movements or sensors. Whilst all parties may meet the

requirements of the design brief, a traditional, isolated approach to the development of each element

can lead to fractured product outcomes. This is because disjointed elements can be merged together

at the end of the design process, rather than developed simultaneously in a collaborative fashion. This

can lead to human-machine interactions where the use of the product and the electronics of that

product that do not effectively work together to create a seamless experience for the user, fully

designed to be in sync with daily human activities. However, this is changing, as industrial design

students are increasingly being “re-coded”2 to be able to simultaneously develop the hardware and

the software for HMI products. This is possible because of the emergence of Visual Programming

Languages (VPL’s) that can replace traditional text-based coding methods. A VPL can be likened to

a two-dimensional representational diagram with visual blocks and linking connectors rather than

strings of text.3

Griffith University introduced a core course for industrial design students called Human Machine

Interfaces in 2015 to respond to the new opportunities to better engage the students in the development

of electronics-based products. In this course, accessible, open tools, such as Arduino

microcontrollers, allow students to rapidly generate creative, advanced prototypes. These can be

completed to working model stage within the space of two six-week projects. While it was previously

possible to engage the students with introductory learning about electrics and electronics within that

time frame, there was little space for students to engage with the design opportunities and the

outcomes tended to be predictable and low level in design terms. VPL’s allow for a different way of

working that enables design-led development, rather than outcomes limited by conventional

electronics.

Grasshopper VPL has been chosen as the main mode of programming and interacting with the

Arduinos. In the Griffith University example, the first three weeks of the course teaches both the

more common text-based coding method as well as Grasshopper to allow students to make their own


comparisons and understand some of the logic behind programming. At the conclusion of the first

three weeks of the 2015 course, an anonymous survey was completed by the industrial design students

about their experience and understanding of coding with both methods, with the results published by

Novak and Loy2 clearly indicating a preference for the Grasshopper VPL. In fact given a hypothetical

design brief, one-hundred percent of respondents indicated they would prototype using Grasshopper

over a text-based tool. This is in line with similar research conducted by Celani and Vaz4 who had

success with Architecture students and VPL’s, indicating that the inherent visual skills of creative

designers allows them to more rapidly and easily adopt VPL’s rather than traditional coding.

What is significant about such research is that new approaches to programming mean that design

education can maintain its focus on user-centered principles, rather than being derailed by a

requirement to train designers to code using text-based systems in isolation from the design process.

By using VPL’s, design students now have the capacity to rapidly prototype both the physical and

electronic aspects of their design. This means they can be iteratively developed as holistic solutions

to a design brief, and represents a paradigm shift from the previous process where designers were

dependent on the technical expertise and perceived limitations of technicians from traditional

disciplines. As a result, rather than focusing only on the physical design and showing minimal

understanding of the technical systems to make their product work, industrial designers have the

potential to produce design-driven outcomes where the electronics is integral to the design thinking

embedded in the product, rather than integrated into it.

As a tool for empowering designers, Visual Programming Languages are proving to be effective

within the industrial design course at Griffith University, and student feedback to the opportunities

they provide has been positive. Whilst text-based coding will likely remain the best method for

programming large interconnected systems and complex software, industrial designers who are not

necessarily interested in becoming programmers are for the first time able to genuinely integrate

electronics into their design process. The role of the industrial designer is broad, encompassing

ergonomics, user experience, aesthetics, form, manufacturing constraints and much more. The

complexities of programming languages restrict the thinking and creativity of a designer when

responding to a complex, human-centred design problem. VPL’s are changing the game. They are

providing designers with workflows more akin to other visual practices already common to the design

process, allowing them to remain creative and design-led, whilst engaging more fully with the

interaction design elements of their product. Designers are beginning to see “code as a kind of

material just as a potter sees clay.”5

References

[1] G.A. Boy, Introduction: A Human-Centered Design Approach, in: G.A. Boy (Ed), Handbook of

Human-Machine Interaction: A Human-Centered Design Approach, Ashgate Publishing Limited,

England, 2011, pp. 1-20.

[2] J. Novak, J. Loy, Recoding Product Design Education: Visual Coding for Human Machine

Interfaces, Proceedings of DESTECH Conference (2016).

[3] C. Reas, C. McWilliams, LUST, Form + Code in Design, Art, and Architecture, Princeton

Architectural Press, New York, 2010.

[4] G. Celani, C. Vaz, CAD Scripting And Visual Programming Languages For Implementing

Computational Design Concepts: A Comparison From A Pedagogical Point Of View, International

Journal of Architectural Computing, 10(1), 2012, pp. 121-138.

[5] J. Sauter, Preface, in: R. Klanten, S. Ehmann, V. Hanschke (Eds), A Touch of Code: Interactive

Installations and Experiences, Gestalten, Berlin, 2011


Modelling resistance exercise lifting technique with inertial sensors: Knee

extension during deadlifts

Sam Gleadhill1, 2, a *, Daniel James2, b and James Lee1, c 1 Physiolytics Laboratory, Charles Darwin University, Australia.

2 SABEL, Griffith University (Nathan campus), Australia.

a [email protected], b [email protected], c [email protected]

Keywords: Wearables, Resistance exercise, Inertial sensor, Technique, Knee extension.

Introduction

Biomechanical models provide insight into the constructs of resistance exercise, allowing health

professionals to ensure lifting technique is safe and efficient. Defined from past kinematic research,

the deadlift has been modelled with inertial sensors to have three separate phases during the lift [1].

These phases include the lift off phase (LO), knee passing phase (KP), and the lift completion phase

(LC) [2]. Analysing biomechanical models with wearable technology may provide tools for health

professionals to identify injury risks and provide accurate feedback to clients. Inertial sensors have

been used to accurately classify resistance exercise technique into categories based on expertise with

greater accuracy than professional observation [3]. The aim of this research was to model deadlift

technique with inertial sensors to interpret trends of movement patterns.

Methods

11 Volunteers performed unweighted conventional deadlifts for two sets and five repetitions per set

(Charles Darwin University Ethics Committee clearance: H14046). Lifting technique was instructed

by a qualified strength and conditioning coach. One set was performed with simultaneous knee, hip

and spine movement and one set was performed with knee extension occurring as the primary initiator

of movement. Volunteers were monitored with tri-axial inertial sensors (SABEL sense) located on

the skin at the Cervical (C7), Thoracic (T12) and Sacral (S1) spinal landmarks [4]. A five Hertz low

pass filter was applied and the concentric phase of each lift was time normalised (N=100). Students

T tests were used to compare the difference of gyroscope Y axis rate of change group means, due to

this data representing spinal movement in the sagittal plane. Group means were modelled to cross

reference video playback for interpreting real world trends.

Results

A null hypothesis of no significant technique rate of change difference was rejected for the

relationship at the sacral spine (Table 1).



Table 1. Knee extension rate of change difference

Variables Correlation P value

DL UW C7 vs. KneeExt C7 0.525 P>0.05

DL UW T12 vs. KneeExt T12 0.592 P>0.05

DL UW S1 vs. KneeExt S1 0.169 P=0.014 Legend: DL = Deadlift; UW = Unweighted; KneeExt = Deadlift with

knee extension occurring first; C7 = Inertial sensor at cervical spine; T12

= Inertial sensor at thoracic spine; S1 = Inertial sensor at sacral spine.

Figure 1. Sacral spine (S1) degrees per second model comparison between typical unweighted deadlifts (DL UW S1)

and deadlifts with knee extension as the initiator of movement.

Discussion

Results suggest a significant measurable difference between the two lifting trials if measured from

the sacral spine. This may be due to isolated or greater knee extension as the initiator of movement

leading to larger hip displacement whilst the spine remains neutral [1]. As defined from past research

for hip and knee rate of change, spinal rate of change models can be classified into the same three

phases (Figure 1) [2].

Figure 1 demonstrates the clear differences of the three phases between trials with simultaneous

movement and isolated knee extension, specifically the KP phase. Therefore, it was concluded that

inertial sensors can model the three phases of a deadlift from S1 and may be able to identify if knee

extension is occurring as the primary mover. This may have significant impacts for health

professionals feedback to clients for skill acquisition in amateurs, and safety or equilibrium for spinal

and knee loading. Various other techniques and resistance exercises may be interpreted with similar

methods to assist health professionals feedback. Resistance exercise risk factors to injury which

inertial sensors may be capable of monitoring include joint angles, posture, speed of movement,

displacement, fatigue, and power. Future research is needed to develop dynamic biomechanical

models of resistance exercise which can be monitored in real time with user friendly automated

inertial sensor method designs. Such research could have a large impact on numerous societal areas

0

10

20

30

40

50

60

70

80

0 10 20 30 40 50 60 70 80 90 100

°/sec

Time points N = 100

DL UW S1

Knee ext S1

KP

LO KP LC

LO KP LC


including workplace health and safety, recreational exercise, health and rehabilitation, physical

activity classification, and professional sport.

References

[1] M. Hales, Improving the deadlift: Understanding biomechanical constraints and physiological

adaptations to resistance exercise, Strength & Conditioning Journal. 32(4) 2010 44-51.

[2] M. Hales, B.F. Johnson, J.T. Johnson, Kinematic analysis of the powerlifting style squat and the

conventional deadlift during competition: is there a cross-over effect between lifts?, The Journal of

Strength & Conditioning Research. 23(9) 2009 2574-2580.

[3] R. Adelsberger, G. Troster, Experts lift differently: Classification of weight-lifting athletes, 2013

IEEE International Conference on Body Sensor Networks (BSN). 2013 1-6.

[4] S. Gleadhill, J.B. Lee, D.A. James, The development and validation of using inertial sensors to

monitor postural change in resistance exercise, Journal of biomechanics. 49(7) 2016 1259-1263.


27

Progress in Metal Additive Manufacturing towards a Wide Implementation in

Dentistry

Leonhard Hitzler1,a *, Frank Alifui-Segbaya2,b, Wayne Hall1,c and Andreas Öchsner1,d

1 Griffith University, Griffith School of Engineering, Southport 4222, Australia 2 Griffith University, Griffith School of Dentistry and Oral Health, Southport 4222, Australia

[email protected], [email protected], [email protected], [email protected]

Keywords: Selective Laser Melting, Laser Polishing, Anisotropy, Biocompatibility, Dental Alloys

In recent years, the availability of high power lasers has enabled the full melting (FM) of raw

materials, as in the case of selective laser melting (SLM), and hence, has overcome the necessity of

a binder [1]. Thus, the FM allows the fabrication of parts with alloys suitable for dental

applications. SLM eliminates many physical stages, required in the time-consuming investment

casting process, by offering speed of the manufacturing process for bespoke metallic devices, such

as denture frameworks and implant abutments [2-5]. The manufacturing process is solely based on

dental and medical CAD data, which are designed on virtual anatomical patterns received from

high-resolution dental scanners and CT, CBCT, MRI, photogrammetry, or surface scan data [6,7].

Cobalt-chromium (Co-Cr) and titanium alloys remain by far the most widely used alloys for

affordable prosthodontics treatment. Documented studies show SLM fabricated parts possess

adequate mechanical properties for clinical use [8-10]. In addition, available data indicates superior

properties, in comparison to cast, for both tensile strength and fracture toughness [4,11-13]. These

results show that a direct substitution of investment casting with SLM in the dental area is

uncritical. In the assessment of biocompatibility, corrosion data, which forms an excellent adjunct

to cytotoxicity studies, show that SLM fabricated samples have lower ion emission rates than cast

ones (Fig. 1) [2,4]. Cytotoxicity studies also proved them safe, non-irritant and nontoxic on oral

tissues and the body as a whole [14,15].

However, the design and dimensioning of components can be very challenging, as the layer-wise

generation process results in anisotropic material properties, which are highly volatile to the

manufacturing conditions [16]. For components designed close to the material limits, the

dimensioning state has to account for the anisotropic properties, in both, the elastic and elasto-

plastic range, including the stress-concentration effects of localized, inherent imperfections [17].

Currently, the surface roughness of ‘as-built’ SLM parts appears of insufficient quality for dental

applications [18]. It is worth stating that the particular surface characteristics, and consequently

their roughness, depend on the alignment and the positioning in the built space [19]. The choices

available in post processing technologies to overcome the challenges of insufficient surface quality,

while still maintaining the geometric flexibility in the built parts, are limited. On a positive note,

recent studies have advanced from a proof of concept stage to polish the surface of SLM

components with a laser; a process that could eliminate manual polishing altogether [20,21].


Figure 1: Elemental release of Co-Cr-Mo in artificial saliva after 42 days in accordance with ISO 10271 on: CP = cast

and polished; CEB = cast and electro-brightened; SP = SLM-fabricated and polished; SEB = SLM-fabricated and

electro-brightened; adopted from [4]

Conclusion

The SLM process, although high-priced to acquire, offers increased speed and flexibility in dental

technology. Likewise, the SLM process, being controlled digitally, offers a standard method for the

production of dental devices, which is likely to be much closer to the manufacturer’s specifications

than investment casting that is fraught with fabrication steps and operator variations. Nevertheless,

further studies are required to understand the parameters that influence the physico-mechanical

properties of SLM fabricated components.

References

[1] P. Rochus, J.Y. Plesseria, M. Van Elsen, J.P. Kruth, R. Carrus, T. Dormal, New applications of

rapid prototyping and rapid manufacturing (RP/RM) technologies for space instrumentation. Acta

Astronaut 61 (2007) 352-359.

[2] F. Alifui-Segbaya, P. Foley, R.J. Williams, The corrosive effects of artificial saliva on cast and

rapid manufacture-produced cobalt chromium alloys. Rapid Prototyping Journal 19 (2013) 95-99.

[3] F. Alifui-Segbaya, J. Lewis, D. Eggbeer, R.J. Williams, In vitro corrosion analyses of heat treated

cobalt-chromium alloys manufactured by direct metal laser sintering. Rapid Prototyping Journal 21

(2015) 111-116.

[4] F. Alifui-Segbaya (2011) 'In vitro' tensile tests and corrosion analyses of a rapid manufacture-

produced cobalt-chromium alloy compared to cast cobalt-chromium alloys. Thesis, Cardiff

Metropolitan University, Cardiff

[5] F. Alifui-Segbaya, R.J. Williams, The corrosion of electrobrightened and polished cast and rapid

manufacture-produced cobalt chromium alloys. The Dental Technician Journal 64 (2011) 10-11.

[6] R. Van Noort, Introduction to Dental Materials, 4 edn. Elsevier, Sydney, 2013.

0

1

2

3

4

CP CEB SP SEB

Ele

men

tal

rele

ase

(μ

g/L

)Co

Cr

Mo

Cast SLM


[7] Medical Models (2015) PDR Cardiff Metropolitan University. http://pdronline.co.uk/surgical-

and-prosthetic-design/medical-models. Accessed 18 October 2015

[8] F. Alifui-Segbaya, J. Evans, D. Eggbeer, R. George, Clinical relevance of laser-sintered cobalt-

chromium alloys for prosthodontic treatment: a review. Proc. of the Intl. Conf. on Progress in

Additive Manufacturing, pp 115-120, (2014)

[9] D. Jevremovic, T. Puskar, B. Kosec, D. Vukelic, I. Budak, S. Aleksandrovic, D. Egbeer, R.

Williams, The analysis of the mechanical properties of F75 Co-Cr alloy for use in selective laser

melting (SLM) manufacturing of removable partial dentures (RPD). Metalurgija 51 (2012) 171-174.

[10] A.R. Lapcevic, D.P. Jevremovic, T.M. Puskar, R.J. Williams, D. Eggbeer, Comparative analysis

of structure and hardness of cast and direct metal laser sintering produced Co-Cr alloys used for dental

devices. Rapid Prototyping J 22 (2016).

[11] Y. Kajima, A. Takaichi, T. Nakamoto, T. Kimura, Y. Yogo, M. Ashida, H. Doi, N. Nomura, H.

Takahashi, T. Hanawa, N. Wakabayashi, Fatigue strength of Co–Cr–Mo alloy clasps prepared by

selective laser melting. Journal of the Mechanical Behavior of Biomedical Materials 59 (2016) 446-

458.

[12] A. Takaichi, Suyalatu, T. Nakamoto, N. Joko, N. Nomura, Y. Tsutsumi, S. Migita, H. Doi, S.

Kurosu, A. Chiba, N. Wakabayashi, Y. Igarashi, T. Hanawa, Microstructures and mechanical

properties of Co–29Cr–6Mo alloy fabricated by selective laser melting process for dental

applications. Journal of the Mechanical Behavior of Biomedical Materials 21 (2013) 67-76.

[13] L. Hitzler, J. Hirsch, J. Schanz, B. Heine, M. Merkel, W. Hall, A. Öchsner, Fracture toughness

of selective laser melted AlSi10Mg. P I Mech Eng L: J Mat SUBMITTED (2017).

[14] D. Jevremovic, V. Kojic, G. Bogdanovic, T. Puskar, D. Eggbeer, D. Thomas, R. Williams, A

selective laser melted Co-Cr alloy used for the rapid manufacture of removable partial denture

frameworks - initial screening of biocompatibility. J Serb Chem Soc 76 (2011) 43-52.

[15] T. Puskar, A. Lapcevic, S. Arandjelovic, S. Radulovic, I. Budak, D. Vukelic, D. Jevremovic,

Comparative study of cytotoxicity of direct metal laser sintered and cast Co-Cr-Mo dental alloy.

Metalurgija 54 (2015) 481-484.

[16] L. Hitzler, C. Janousch, J. Schanz, M. Merkel, B. Heine, F. Mack, W. Hall, A. Öchsner, Direction

and location dependency of selective laser melted AlSi10Mg specimens. J Mater Process Technol

242 (2017) IN PRESS.

[17] A. Öchsner, Continuum Damage and Fracture Mechanics. Springer, Singapur, 2016.

[18] L. Hitzler, C. Janousch, J. Schanz, M. Merkel, F. Mack, A. Öchsner, Non-destructive evaluation

of AlSi10Mg prismatic samples generated by Selective Laser Melting: Influence of manufacturing

conditions. Mat-wiss u Werkstofftech 47 (2016) 564-581.

[19] L. Hitzler, J. Hirsch, M. Merkel, W. Hall, A. Öchsner, Position dependent surface quality in

Selective Laser Melting. Mat-wiss u Werkstofftech IN PRESS (2017).

[20] J. Schanz, M. Hofele, L. Hitzler, M. Merkel, H. Riegel Laser polishing of additive manufactured

AlSi10Mg parts with an oscillating laser beam. In: Machining, Joining and Modifications of

Advanced Materials. Springer, Singapore, 2016, pp 159-169.

[21] J. Schanz, M. Hofele, S. Ruck, T. Schubert, L. Hitzler, G. Schneider, M. Merkel, H. Riegel,

Metallurgical Investigations of Laser Remelted Additively Manufactured AlSi10Mg Parts. Mat-wiss

u Werkstofftech IN PRESS (2017).


Using Inverse Finite Element Simulation to Characterise Soft Tissues by MSC Marc/Mentat

Zia Javanbakhta *, Andreas Öchsner b

School of Engineering, Gold Coast Campus, Griffith University, Brisbane, Queensland, Australia

[email protected], [email protected]

* Corresponding author

Keywords: Inverse finite element analysis, Depth-Sensing Indentation Test, Thin Film.

Extended Abstract. The study of materials is connected to a framework comprising of four inseparable

elements: processing, mechanical properties, characterization, and theory [1]. Changing the process

path of manufacturing results in altered mechanical properties which require a theory to be justified.

Meanwhile, characterization provides the insight to the multi-scale structure of the material.

Generally, characterization of classical materials such as metals, ceramics, and polymers can be

readily done by conventional materials testing methods such as the tensile test, the torsion test, and

multi-axial tests among others.

Although monolithic materials cover a wide range of applications, generally non-homogeneous

anisotropic materials are created to perform under many critical cases. The non-uniformity of these

rather complex materials can also be characterized, to some extent, using the classical tests. A higher

level of complexity can be found in the hierarchical structure of soft tissue which consists of

interconnected multi-level structures. Approaching non-conventional materials with classical

material tests makes the characterization process cumbersome, if not impossible. This demands for

new methods of characterisation.

A very attractive alternative to the classical experiments is the depth-sensing indentation test which

is preformed, in its very primitive form, as an indentation hardness test. Although the original depth-

sensing tests were carried out using the Berkovich indenters, other axisymmetric indenters can also

be used [2]. Furthermore, the beauty of the indentation test is the fact that it can be implemented in

macro-, micro- and nano-scale levels which makes it an appropriate tool for multi-scale applications.

More recently, a new type of indenter, i.e. the lateral cylindrical indenter, was proposed to characterise

structured material, e.g. a porous structure. In addition, the possibility of using AFM to conduct

experiments using the mentioned particular cantilever is considered [3, 4]. Nevertheless, the nonlinear

nature of such contact problems limits the derivation of a closed-form analytical solution almost

impossible for many cases. Approaching the problem numerically also raises complexity because of

the required iterations for numerically solving nonlinear simultaneous system of equations [5, 6].

The widely known numerical method for the solution of partial differential equations is the finite

difference method which is now an alternative to the mainly used numerical method, i.e. the finite

element method [7]. Finally, the finite element method, is selected as the computational tool of this

investigation since analytical solutions are not available for such a multi-parametric contact problem

that is faced in the indentation of materials.

Generally in a finite element simulation, most of the parameters regarding the FE model are known

and the user is looking for the results of the simulation, e.g. deformations or stresses. In contrast, in

the inverse finite element method, an ordinary FE model is coupled with an optimisation algorithm.

The goal is to find the optimal target parameters of the model which corresponds to the correct

behaviour of the model. These target parameters are incorporated into a user-defined objective

function which is used to measure the degree of optimality of the target parameters (see Fig. 1).

Target parameters are either physical quantities such as the geometry of the model in a topology

optimisation problems, or material parameters in parameter estimation problems. In any case,

choosing the target parameters must be done carefully [8].


For an inverse FE solution, extra programming, or at least scripting, will be required in addition to a

typical commercial FE package. Nevertheless, such level of complexity demands a powerful package

to handle the simulations. The MSC Marc/Mentat semi-open commercial package is selected to carry

out the analyses. More flexibility is obtained by incorporating the user-defined FORTRAN

programming language code into the software [9].

Although Marc/Mentat provides many facilities for nonlinear simulations by default, there a few

parameters which cannot be directly calculated—especially in contact problems. For instance, the

contact area cannot be calculated by the software. The solution for such problems is presented in [7]

where a rather extensive package of ready-to-use subroutines (more than 50) has been made available

for users. The result is a great versatility in terms of defining customized material properties,

boundary conditions, contact properties, and elements which makes the validation of the experimental

results possible.

References

[1] Meyers, M. A. and Chawla, K. K., Mechanical behavior of materials, 2nd ed. Cambridge, New

York: Cambridge University Press, 2009.

[2] Fischer-Cripps, A. C., Nanoindentation. New York, NY: Springer New York, 2011.

[3] Argatov, I. and Mishuris, G., Contact mechanics of articular cartilage layers: Asymptotic models.

Advanced structured materials volume 50. Cham: Springer, 2015.

[4] Argatov, I. I., Mishuris, G. S., and Paukshto, M. V., Cylindrical lateral depth-sensing indentation

testing of thin anisotropic elastic films, European Journal of Mechanics - A/Solids, vol. 49, pp.

299–307, 2015.

[5] Öchsner, A. and Merkel, M., One-Dimensional Finite Elements: An Introduction to the FE

Method. Berlin, Heidelberg: Springer, 2013.

[6] Öchsner, A., Elasto-plasticity of frame structure elements: Modeling and simulation of rods and

beams. Berlin: Springer, 2014.

[7] Öchsner, A., Computational statics and dynamics: An introduction based on the finite element

method. Singapore: Springer, 2016.

[8] Kauer, M., Inverse Finite Element Characterization of Soft Tissues with Aspiration Experiments,

PhD, Swiss Federal Institute of Technology, 2001.

[9] MSC Software Corporation, Marc 2015: User's Guide. Newport Beach, California: MSC

Software Corporation, 2015.

[10] Javanbakht, Z. and Öchsner, A., Advance finite element simulation with MSC Marc: Application

of User Subroutines. Cham: Springer International, 2017.

Figure 1: Flowchart of an

inverse FE problem. Adapted

and updated from [8]


Fused deposition modeling of 3D parts using acrylonitrile butadiene styrene

(ABS) polymer

Maksym Rybachuk1,2a*, Charlène Alice Mauger3,4b, Thomas Fiedler5c and

Andreas Öchsner1,5d 1 School of Engineering, Griffith University, Engineering Drive, Southport QLD 4222, AUSTRALIA

2 Queensland Micro- and Nanotechnology Centre, Griffith University, Nathan QLD 4111, AUSTRALIA

3 Télécom Physique Strasbourg, Université de Strasbourg, 300 boulevard Sébastien Brant, 67412 Illkirch

CEDEX, FRANCE

4 Faculty of Medical and Health Sciences, University of Auckland, 85 Park Road, Grafton, Auckland 1023,

NEW ZEALAND

5 School of Engineering, University of Newcastle, Callaghan University Drive, Callaghan NSW 2308,

AUSTRALIA

[email protected], [email protected], [email protected], [email protected]

Keywords: Acrylonitrile butadiene styrene polymer, disordered solids, mechanical properties of solids,

mechanical properties anisotropy, fused deposition modeling

Extended Abstract. Anisotropic mechanical properties of parts fabricated using acrylonitrile

butadiene styrene (ABS) polymer relative to part built orientation employing fused deposition

modeling process is reported. ABSplus-P430 polymer was used to investigate the effects of infill

orientation on parts mechanical properties under tensile and compression loading. The results

revealed that infill orientation strongly affects tensile properties of fabricated ABS samples, namely

the values for Young’s modulus were found to be ranging from ~1.5 to ~2.1 GPa, ultimate tensile

strength from ~12.0 to ~22.0 MPa, yield strength from ~1.0 to ~21.0 MPa and elongation-at-break

from ~0.2 to ~4.8 % for different infill orientations. Samples with infill orientation aligned to the

vertical (i.e. Z-) axis were found to display the highest values relative to all other infill orientations

studied. Mechanical properties anisotropy was lower for parts under compressive loading, that is the

Young’s modulus, ultimate compressive and yield strength were weakly correlated with infill

orientation apart from samples those built orientation was aligned at 45° to the vertical Z-axis. The

latter samples displayed inferior mechanical properties under all compressive tests. The effects of

sample gauge thickness on tensile properties and ABS sample micro- and bulk- hardness with respect

to infill orientation are also discussed [1].

References

[1] M. Rybachuk, C. Mauger, M. Fielder and A. Ochsner “Anisotropic mechanical properties of

fused deposition modeled parts fabricated using acrylonitrile butadiene styrene (ABS) polymer",

Journal of Polymer Engineering (2016); in press



mailto:[email protected],%[email protected]

mailto:[email protected],%[email protected]

engineering research theme 3 retreat...iii preface welcome to this special edition focused on human...

Documents