Fourth Year Design Project – LiMar
Our forth-year design project aimed to improve how line markings on sports fields are painted. Currently, at the start of
the year, the field maintence team spends about 10-20 manhours/field to measure, mark and then paint the field lines. Then
every week, the team spends about 0.5-1 manhours/field to paint over the previous lines to ensure they stay visible and
aren't fading. This is done weekly or sometimes bi-weekly depending on the condition of the painted lines. This is to
avoid the huge time commitment of painting the field from scratch if the lines were allowed to fade completely. This is
waste of paint, time, and labour. Especially as this weekly painting is continued even during weeks when no-one is using
the field.
Solution
Our proposed solution to this problem was to create an attachment for the line marker cart. This attachment would assist
the operator in easily and more accurately paint the lines. It would serve to reduce the need for the large time investment
to start from scratch and thus eliminate the need for weekly re-painting. This attachment would make use of RFID tags
buried within the ground to localize itself and provide feedback to the operator. With the addition of other sensors such as
speed measurement and a camera, additional feedback can be provided to help the operator.
I worked on all aspects of the mechanical design using SolidWorks and contributed to the programming of the graphical user interface in Python. The sections below highlight the various aspects of design associated with the project.
I worked on all aspects of the mechanical design using SolidWorks and contributed to the programming of the graphical user interface in Python. The sections below highlight the various aspects of design associated with the project.
Display Mount
To aid the operator, feedback would be given to them through the display (a 7IP-CAPLCD Display). As this display had its
electronics exposed it needed a protective casing, in addition to a way to mount it to the cart. The first iteration had a
combined design that served both as an enclosure and a mount. This is pictured below:
However, it was soon found that while this solution was sufficient, it was not ergonomic and would hurt the operator's neck during operation. As well, it would make it difficult for the operator to view both the screen and the field causing unwanted inaccuracies due to focus issues. As such a redesign was required. This involved use a prebuilt tablet mount and designing only an enclosure for the display. This is shown below:
However, it was soon found that while this solution was sufficient, it was not ergonomic and would hurt the operator's neck during operation. As well, it would make it difficult for the operator to view both the screen and the field causing unwanted inaccuracies due to focus issues. As such a redesign was required. This involved use a prebuilt tablet mount and designing only an enclosure for the display. This is shown below:
RFID Reader Mount
This system made use of a Radio Frequency Identification (RFID) technology to assist the operator in painting straight lines.
RFID tags embedded within the ground underneath the lines would function as markers which can be used to determine the
current position and provide feedback. Thus, the RFID reader needed to be mounted near the ground and on the side where the
paint spray was located. It also needed to hold the battery which would provide power to the reader and Arduino. The CAD and
the final 3D printed design is shown below:
RFID Tag Casing
As the RFID tags would need to be buried in the ground, a protective casing was designed. This casing made use of SolidWork's
snap hook and groove feature to allow for a 3D printed part that could easily and securely be put together.
GUI
The graphical user interface was created to provide information to the user. The interface was designed in python using the
TKinter library. A live feed from the camera was included within the GUI to ensure the operator has a clear view of the field
as well as the provided feedback. A mock-up of the GUI can be seen below:
Results and Next Steps
When assembled onto the cart, this system was able to give some rough feedback to the operator on whether they were following
the line or not. However, there are many improvements that can be made. These are as follows:
- An attempt was made at an algorithm that could detect pre-existing lines and use that data in addition to the RFID data to create a virtual line for the operator to follow. This however, did not work due to the large amount of noise due to the grassy environment. More research can be put into this aspect.
- The range of the RFID tags were quite limited especially when under dirt. More research can be done to either amplify the signal or find a better RFID tag/reader.
- Better integration of the electronics as currently two different batteries were used. One for the reader, and another for the display screen and the Jetson Nano.
Photos 🖼
