here - Workspace - Imperial College London

powerfuelSoftware and s/w Development

Nov 9, 2013 (5 years and 8 days ago)

300 views

INTRODUCTION
DESIGN DEVELOPMENT
& IMPLEMENTATION
TESTING &
VALIDATION
COMPLICATIONS
ARDUINO DEVELOPMENT
CONCLUSION &
FUTURE WORK
Author: Suraj Pathak Imperial College London,
Supervisors: Dr. Southgate, Department of Bioengineering
Prof. Bull
Introduction
Providing feedback to athletes can
considerably improve performance
[1]. Information derived from
technology enables athletes to
improve technique, performance and
reduce injury.
Project Aims
The aim of the project was to develop
a data acquisition platform using the
iPod touch, which would provide
feedback to Paralympic athletes.
Proposed Idea
After the initial research phase , it was
decided that the project would be
focussed on providing feedback to javelin
throwers.

Accelerometers on board the iPod touch
can only detect ± 2g at a sampling
frequency of 100Hz.

Javelin throwers generate much larger
accelerations, well over ± 20g.
The following system was
designed to overcome the
problem that was identified:
The following system specification was developed based on initial research:
- To develop the software application on an iPod platform for the iOS
mobile operating system.
- To process the acceleration data and display it in a useful way to the user in
order to provide feedback to Paralympic javelin throwers.
- To accurately measure and record tri-axial acceleration up to ± 20g
- To enclose the system in durable and ergonomic case.


SPECIFICATION
Arduino UNO
Serial to TTL
adapter
LIS331HH
Accelerometer
Power Supply
(Battery Pack)
iPod Touch
Redpark Serial Cable
Multi -core cable
A Serial Peripheral Interface was chosen as the method
for data communication between the sensor and the
Arduino board. The Arduino was programmed so that it
only sends data to the iPod touch if it receives a byte from
the iPod touch. The Arduino then stops transmitting data if
it receives another byte.
An enclosure was chosen and
modified for the sensor and the
electronics that were developed
as a part of this system.
ENCLOSURE
A native iOS application was
developed using the iOS software
development kit (XCode
development environment) and the
Objective-C programming
language. The following diagrams
summarise all the views that were
created.
During experimentation it was found that the
accelerations observed by the
accelerometers were not entirely accurate.
They had to be adjusted by the following
multiplication factors: 1.073, 1.097 and
1.200 for the X, Y, and Z-Axes
respectively.
iOS DEVELOPMENT
VALIDATION
To validate the accelerometers, an accelerometer was
attached to a spring mass oscillation system.
Experimental peak acceleration values were then
compared to the theoretical values derived from a
model.
M
M
k
C
It was vital to test the system that was developed practically. This meant
recording the accelerations that were generated during a throw.
4 volunteers threw 5 trials for each distance, the results are discussed
below.
Soft: 40%
Capacity
Medium: 60%
Capacity
Hard: 80%
Capacity
100cm 650cm
1050cm
- The results suggest that there is a positive correlation between
the distance and capacity of the throw, and the resultant
peak and mean accelerations obtained from the sensor.
Axis

Correlation Coefficient

X
-
Axis

0.91

Y
-
Axis

0.96

Z
-
axis

0.94


Results:
- These results also show that the application developed will
successfully provide relevant feedback concerning the
performance of the throw.

Results:
When comparing the achievements of
this project to the design brief that was
initially stated, it can be seen that all the
objectives and aims were completed
successfully. However, there are many
improvements that could be made:
- Implementation of filters to make the
data less noisy.
- To add the functionality that allows a
pdf report to be generated for each
throw.
- System miniaturisation and
incorporating the use of a wireless
system to transmit data.
- Use of other sensors to calculate more
performance linked parameters.
- More thorough testing/ validation of the
accelerometers used (only 4% of the
total measurable accelerations
were validated).
- Applications to other sports.

OFFSET ERRORS
TESTING
ATHLETE MONITORING
ON AN IPOD PLATFORM
PROJECT AIMS
Figure 1: Acceleration time graphs for three able bodied athletes for different javelin masses. Data was obtained from strain-gauge
accelerometers. The red line indicates the limits of what the LIS331DL accelerometer on board the iPod touch can measure [2].
Figure 3: The main enclosure for the Arduino
and associated components.
Figure 4: The enclosure developed for
the accelerometer.
Figure 5: The
experimental set-up
Figure 6: The model used to describe the spring
mass oscillation system
REFERNCES
[1] Motor control and learning: a behavioural emphasis. By Richard
A. Schmidt, Timothy Donald Lee,
[2] Vladimir M. Zatsiorsky (2000), Biomechanics in Sport:
Performance enhancement and injury prevention. Part 4:
Throwing and hitting, Chapters 18-22
ACKNOWLEDGEMENTS
Dr Southgate, Prof Bull, Mr Holloway & Reva Vaze
SYSTEM DESIGN
Figure 2: The proposed design of the system.