This project is about an autonomous vehicle, based on a modified toy RC car, that can drive along a "road" without any manual interaction required.
To this end, the car's remote control is modified so it can be attached to a microcontroller, that receives commands from a Python program running on a laptop. The camera, mounted on the top of the car, streams its view wirelessly to a neural net on the laptop, that decides what steering commands are the most appropriate at every time step/frame.
This is a very cool project. I should get into Raspberry Pi as it is a real computer and more amenable to machine learning but I just have too many hobbies and I don’t think I will work with vision (though that would be fun) so will just stick with Arduino. I like neural nets too and want to work more with them but just may not happen.
As an aside, here is an interesting video (scaremongering stuff) https://www.youtube.com/watch?v=m1QKaZUKgK4 but at about 58 minutes they say the neural net was tracking the researchers - would like to know more about what that’s all about.
Thanks for your reply, it’s always nice to hear from you. I noticed that detailing the post takes up more space than I expected and will take a bit.
Regarding the scaremongering stuff: I think automation can be scaring in many ways, the real issue is the hybris of people, that lead to employ such systems assuming they are safe to use and “better” than anything else. Apart of that I am very much for any progress than can be made, it is fascinating, it just needs mature people who don’t confuse the fascination with fail-safeness.
The main purpose is to learn something about neural nets and machine learning in a practical project. You could do it in a simulator, but it’s more fun to see a real car (well toy car, but in the real world) drive
