Let's Make Robots!

Location, location, location!

Hello everyone!

Now, I'm sure I can't be the first with this problem. A quick scan of the forums didn't turn up much, so I'll ask outright. I have a semi-autonomous robot project (won't bore with details) and I absolutely NEED to have a coordinate system (X, Y, and the direction the robot is facing). This can't involve any GPS or expensive digital compass sensors, and it needs to be pretty accurate. And because this might be asked, I'm running a twin gearbox Tamiya motor and two rubber wheels. It's a pretty small setup. Someone here has to have run into this before...what are your suggestions or solutions?

Thanks in advance! 

- P 

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

I've had this thought going through the back of my mind for a month or so now. Haven't got very far (well, anywere at all) on it because of other projects. I'm about to jump back into that project again though, so your post is pretty nicely timed ;)

OddBot's post (as you may notice) looks like it'll help a lot. I've also found this book (pdf, free), but managed to not get started on it until today (right after my son brought home a nasty flu bug from school... arg). One of the thoughts I was having was to just periodically check distance to a wall from two angles. If readings at X and -X sensor positions are the same, the bot is facing a wall. There should be some math to derive a rotational position if the two readings are not the same. This would be putting a lot of faith in the squareness of my house though...

Anyway, thanks for bringing it up, and thanks for that other link you just posted (off to read it now too).

I'm going to have to have this done within two weeks (school deadline) so I'll be posting everything as soon as possible...hopefully this will spark from inspiration from more people here! I'll be taking a look at that book you posted there, but I'll probably leave the reading until tomorrow hehe.


- P 

You could also use wavefront. Keep in mind you would need encoders to make sure you move 1 gril location at a time and can make precise 90 or 45 degree turns.

I'll be using techniques similar to that...this has all got to be done with my own hand, part of which is the decision algorithm. 

Thanks for the link. It's given me some ideas on how to start planning routes.

- P 

If you want a cheap compass then two of these sensors mounted 90 degrees to each other and connected via an op-amp to your analog inputs should work.

For those like me who want it simpler, there are these that I iam using:

ZX-Magnetic - Magnetic Field sensing board

Hooks right up to an analoge input, cable even included :)





I LOVE your creative, cheap, ad hoc, lowtech solutions :) Does this really work? If so it's just what I've been looking for...

All info regarding details would be appreciated. How to set it up? What are the drawbacks/limitations? How precise are they? Is the op amp really necesary (I saw on the specs that the output is 2.25-2.75V which shoud be enough to give meaningful data)?

Anyway thanks for the tip ;)


Really?! Holy crap man! I've been lamenting the ~$60 price tag of cheap compasses lately. I didn't know these could do that. (you have no idea how stoked I am about this)(damn, this is the highlight of my week, that's sad)

This may have just solved a few more problems...for anyone else interested in how this works, read the fourth paragraph of this website.


Thanks very much! I'll be ordering those shortly, once I come up with a proper schematic, which I'll post here too.

- P