Detecting plants rows with Pixy

[bvasseu]

I’m currently working on a projet in order to build a mini robot able to go through 2 sugar beets rows. I fist wanted to do it with classic sensors (ultrasonic, potentiometers and co…) because of the limited power of an arduino board.
Then I discovered the CMUcam5 Pixy and I think it can be a good solution to detect my rows and follow the middle path.
Do you guys think that I would be able to detect such a complex form with that technology ?
Taking in account the huge variation of color that there can be within the field ? (I put a pic of an average sugar beet field so that you can know how it looks)

Thanks in advance

[Edward]

Hello Baptiste,
I’m not a machine vision expert, but I would guess that you will need some custom vision algorithms to navigate in an environment like this. Pixy can detect objects based on color, which might help for part of your solution, but I don’t see how it will do everything

Looks like a cool project though!

Edward

[briz]

Wondering if you had any results?