Hey everyone, I’m working on a project for my undergrad course and had this idea for a ‘Smart Traffic Control System using V2X.’ The concept is to send traffic signals and speed limits directly to vehicles at an intersection.
I’m planning to simulate a four-way intersection with four cars, one on each road. I’ve attached a picture of the setup. The cars will be controlled by Arduinos, and I’ll have a Raspberry Pi with a camera positioned above the intersection to track the vehicles. The cars will communicate their intended directions to the Raspberry Pi using NRF24L01 modules. The Pi will process this info, assign priorities and speeds to the cars, and send it back to them to ensure they move through the intersection smoothly and without collisions.
So, I’m wondering: Is this idea too complicated or maybe unnecessary? Is it realistic for me to pull this off? I’d love to hear any suggestions or improvements you might have. Thanks in advance!
As a project to teach you micro controllers, imagine recognition, real time control systems and programming it think it's an excellent idea. As an exercise in the reality of driving and driver behaviour not so much.
Following with interest although I don't have any particular advice to offer.
Interesting. I'd say ok for a simple exercise ( small scale ), for a second step ( a large scale exercise ) that includes more than 1 intersections the system should address 'the right cars at the right intersection'.
Also the system could verify if each car is following the 'recommendations' and warn the other vehicles if not.
In my opinion this is a very large project. I would start with something smaller.
Some time ago I have written a traffic simulation. A city map with several intersections and lot of "cars" with several tasks. Each car drives to its own target without causing crashes and according to the possible ways, stopping at traffic lights ...
create your full story and calculcate how long your pupils will need to do each task.
Sum it together and calculate how long it will take.
Does this fit in your time constraints? A year?
When you have your time estimation, pick out the smallest "stand alone" thing to start with. Start with the "Minium Viable Product" which draws your pupils attention to your project.
And now I suggest you read again the link I have provided in my previous post.
first level ( single intersection super simplified )
1.1) develop pc/raspberry 'system' program
1.2) develop arduino 'car' program
Both should be super simplified tasks ( no rf communication needed at this level, is enough a 'serial/textual' communication of some kind ), cars are already 'identified' ( right car is 'R1', left is 'L1'... ) define a series of task each program should do and define a basic set of commands to interact between the two. Also defining a 'test procedure' to verify that everything is working as expected
2.1) convert the pc program to parspberry ( if not already developed on a raspberry )
2.2) develop rf communication ( simple but... )
2.3) develop a method to identify cars ( not that simple at all )
...
second level ( second year? )
...
P.S.
also consider that driving on the left side is very dangerous ; - )
And a very good plan. Are the car builders aware they are going to build something that cannot be tested until the whole project is completed? Or are you going to design test facilities so the cars can be tested in steps as they are being built?
