Moving objects in x,y coordinates

Using Arduino Sensors
#1

Hi,

I have a challenge and I need your help.
I have a board (like a chess board) with X,Y coordinates. Say the board is 1x1 meters with 10 equal X sections and 10 equal Y sections (so each square on this board is 10x10cm).
I have a few objects on this board, for example, little toy cars red, blue, and green.
All cars are placed on the same initial square.
I want to ‘instruct’ the red car to move certain number of squares (right/left) and certain number of squares (up/down) so that the car goes to a new square on this board.

I want to figure out how do I instruct each car to move to a specific square without affecting other cars, and how do I recognize that a car reached the destination square on the board. How can I actually know that a car did not get stuck in the middle for instance? What technology would you recommend? Perhaps RFID? Cameras? Something else?

The objects cannot be hooked with wires to a 'central processor' so they got to communicate using some wireless connection.

What would you do if you needed to avoid 'obstacles' on the board (you know where those obstacles are on the board).

Many thanks for your help.

#2

what you are describing is an xy plotter; the core of a cnc engraver

start by looking though the Project Hub and on the internet for xy plotter and cnc engraver projects

pick the cars up with a magnet

#3

Think about the problem some more. Simulate it, in a spread sheet if that is all you have. If we give you the answer, it is not your challenge any more.

#4

Thank you for your response.
I already thought about this project, naturally, but could not figure out a way to implement this project. Using a 'plotter' like mechanism would not work because the cars need to be free moving (not hooked with metal arms, wires, etc). The other issue that I am having with this project is that I really need to move all cars and not wait till one car reached its destination before I move the other cars (think that all cars move at the same time each car to its destination and they do not collide with each other and/or collide with obstacles that I put on the board).
I am not asking for the 100% solution here just a bit of advice from somebody that built a similar project.

#5

How will the cars move (mechanically), and how do you tell them to do so?

#6

That is the question I raised. Can I use Arduino for this at all?

#7

Menashay:
That is the question I raised. Can I use Arduino for this at all?

Sure, an Arduino is just a microcontroller that happens to have an abstraction layer that hides the complexity of the underlying hardware. You can bypass the abstraction layer if you want, but for this kind of project you probably don't need to do that.

Pick something that has at least 8k of RAM otherwise you are going to run into memory problems with running multiple sensors and comms systems. Each vehicle will need a compass so it knows which way it is pointing and an RFID reader to query RFID targets embedded in the board. There are many options for wireless communication.

The rest of the project is just a giant software monster as you are creating algorithms to control self driving cars in an environment where the road map is constantly changing. To preserve your sanity and minimise cost, first create a simulator on a desktop computer. Get this perfected before spending money on hardware.

#8

Maybe a RFID Tag reader in each car, and 100 tags. Or 100 readers and x tags! (Parallax do them).
Each car has a bluetooth receiver.
Main arduino controller knows where each car is.
Main Arduino controller controls each car via bluetooth.

(Edit) As mentioned - yes it sounds expensive.

Steve

#9

Thank you for your suggestions. The digital compass looks interesting. I was not even aware it existed.
RFID I will need to experiment with this and see how accurate the reader is.
I think that with the above info I can start.

#10

High Menashay,
You need to think about it some more... I guess it is sort of proof of concept, not an actual physical system you have to build. Do you intend for each car to have it's own intelligence, once given the xy coordinate for where it has to move? Do you want the individual cars to communicate to each other? Or probably easier, an over-riding controller that assesses the situation and moves the cars itself, or sends out individual instructions to each car, the cars being 'dumb' just driving where told.

I think the individual intelligence, with communication would be the richest field to explore. The solution would not depend on number of cars, or number of squares, unless a car on every square.
So, rough out a few algorithms to start with, write a list of what functions are needed, then develop each function. You are miles away from thinking about the type of microprocessor required. Does cost come into this, is it a one off, or going to sell in thousands. Is it intended as a game for three year olds, or a tool for traffic engineers? Work on a list of requirements, to stop you getting bogged down in the detail.

But, it seems, from your first post that you only want to tell each car to move a square at a time, from a central controller. That is a bit boring :o . How do you intend giving this controller it's instructions - eg, a list of cars and coordinates where you want each car to end up, it finds where cars are now, and navigates them to the destination, or will you tell it each individual move for each car? The problem is fairly standard, being similar to pcb auto-routing, and 'the traveling salesman'.

If this is a project set by someone else, e.g. a college tutor, then it is possible that there is an expectation that you will apply knowledge you have most likely acquired. If it is something that you would just like to explore, then look into pcb routing methods, traveling salesman, Monte-carlo, and the like wrt programing, (not car rallies or gambling).

Best wishes

#11