Guide Bar

Our goal was to make a system for farmer which wil allow them to know where they are in there fields, the “smartest” path to do, and how much of the field they still need to parcours.

It’s stuff that already exist in real life but our goal was to make it as cheap as possible(< 200€). Of course we didn’t included features like linking to other component of the tractor because we where newbies 3rd year students.

My part

I was in charge of port the app on the raspberry. Because we were using Qt it was possible to a less technical member to us the feature Qt Design to build the graphics.

My work was to make the link between the differents bricks, the GUI, the PathBuilder, the Bluetooth with the system.

Also, because i was already curious about this kind of stuffs, i’m also responsible for the bad decision in term of material use and protocol choosen (i was a newbie).

Structure

GPS

• An arduino mega
• A GPS module
• An IMU
• A Bluetooth module
• Battery

Rasp

• Just a rasp plug to the cigar-lighter

Why does it failed ?

It was more a gestion problem, in our school it’s mandatory to do a semester abroad, but this time it was forced to do it in third year and this project was to big for a semester. I stayed in france because of a VISA problem so i decided to take the lead of the project the second semester.

But i’m not a leader and it was too hard to me guide the other and still work on my stuffs.

Changes

However, more technical, if i had to do it again today i will also use different materials. Bluetooth consume to much energy for the need and a C1101 is a cheap but working solution.

Also we used a smartphone portable battery in USB connected to the rasp, today i would use something more efficient.

And if i had to choose my role i would prefere to work on the embedded system on the roof. Even if Qt is interesting i prefer very low level stuffs now.