Line/magnet follower

julianE

Senior Member
Edmunds, fascinating project. I have a very crazy idea based on magnetism. Adhere cassette tape to the board and have a cassette tape head on the bottom of the vehicle reading information from the tape. You would have to record data on the tape beforehand. The idea is not unlike credit card readers that read the credit card magnetic stripe. Of course, the vehicle would need to move at a steady speed but that should not be too difficult.
 

techElder

Well-known member
I have a very crazy idea based on magnetism. Adhere cassette tape to the board and have a cassette tape head on the bottom of the vehicle reading information from the tape.
... and there would have to be direct contact between the head and the tape
... and the tape would eventually be demagnetized in the fields from the vehicle
... and the head would probably be a substantial part of the vehicle volume

However, speaking of "crazy idea", Edmund could possibly use a stack of small magnets with soft iron spacers to create his "message". Each message would change the sequence of the materials in the stack.

Controlling the field from the magnets would compress the message and detection area needed, and would make a system that can read the message quicker.

Sort of a mechanical strip of magnetic tape. :D
 

premelec

Senior Member
I think there are hall 'gear tooth' sense units which detect changes in flux when put near ferromagnetic materials. Basically a hall sensor with integrated magnet. Note that an iron wire can be considered a one tooth gear...
[ for robotic instructions:
:-D]
 
Last edited:

edmunds

Senior Member
Thanks for all the ideas, guys!

I have been battling mechanical problems with steering to exhaustion, but it seems to be sorted now.

Here is the first attempt - no real PID tuning, nothing. Looks promising to me :).


Now one has to build a test layout.

Cheers,

Edmunds
 
Last edited:

edmunds

Senior Member
I think there are hall 'gear tooth' sense units which detect changes in flux when put near ferromagnetic materials. Basically a hall sensor with integrated magnet. Note that an iron wire can be considered a one tooth gear...
[ for robotic instructions:
:-D]
I was aware of the sensors, but did not know how they work. Now, knowing how they work (if I understand it correctly) they need uniform and moving past the sensor 'tooth' of some kind to generate the sinusoidal signal used for detection. I cannot imagine, how could I generate that for lateral position of the car on the road. Anyhow, idea can be useful for some other things in the future. Thanks for sharing.

Edmunds
 

edmunds

Senior Member
Dear all,

The code below [did not fit in the post] works with both magnet polarities. I have not implemented the 'message' yet, but it is a mere technicality of counting the magnet transitions after an S magnet 'start condition'. In the meanwhile, I need to build the track and improve the code to use fewer variables and to be more efficient/readable. Any ideas on how to macro-nize or subroutin-ize this are welcome.

Thank you for your time,

Edmunds
 

Attachments

premelec

Senior Member
I didn't have in mind the gear tooth device being used on a gear tooth - but to show sideways deviation from an iron wire... which would be less regular than a rotating gear but should indicate deviation OK.
 

edmunds

Senior Member
Dear all,

You should see a smile on my face :). I don't have time, nor energy tonight for more than one turn, but this scores as a proof of concept in my book.


Cheers,

Edmunds
 

edmunds

Senior Member
It actually does work quite a bit faster, too. The distance between 2x2mm cylinder magnets is 10mm. Approximately the same for straights and the turn.


Edmunds
 
Top