40x1 based micro mouse

[beny1949]

hello all,
got a few pics and some information about my micromouse i have been working on and asking questions about on here for the last few weeks. thought i owed you all a little look at it. i will try and keep the site reguarly updated....

for the pictures, there is a link in the left hand menu.

Ben

<A href='http://www.ben-howes.co.uk' Target=_Blank>External Web Link</a>

 

 

[BeanieBots]

Looks good. Would be nice to see a section to view the circuit diagram and another for code.
Why do you need a 40X1?
What are you going to do with all that I/O?
Maybe a 28X1 would be a better choice.

 

[beny1949]

i need all the ADC! there are 6 LDRS detecting the walls (3 each side on the stalks) and there will also be a further on on the front when i make it.

Ideally i would have access to even more adc... are there any chips you guys would recomend as a good solution?

Also, another reason for the X1 is the fact that i can have it running at much fast speeds, as well as with more lines of code, which really matters when you are trying to solve a maze which has 256 cells and many different paths to the middle.

This needs consideration of not only the route with the least cells to pass through, but the ones which have more striaght lines, which saves on turning time as well as rudcing the risk of a dogy turn.

<b> hence the need for the shiny new brain compliments of Rev-Ed. </b>

Ben

&#160;

Edited by - beny1949 on 06/05/2007 22:49:49

 

[BeanieBots]

So use an analogue multiplexer and have as many ADC inputs as you want.
I2C memory for storage and if required, several 28X1s talking to each other over I2C.
Probably still a smaller footprint than a 40X1. Just a thought as memory and speed are AFIK the same for 28X1 and 40X1.

 

[beny1949]

fair enough

i just thought that as the 40x1 is what 12p more or something stupid like that that i may as well go with it! especially as im using a 28/40 proto board.

I just put a handfull of new pics in the picture section if you interested.

about this analogue multiplexer....

Ben

 

 

[BeanieBots]

Only a suggestion. Your robot, your choice. It's just that bigger is not always better. Several 'smaller' ones will give more processing power and memory than 1 'large' one but with a few more complications than having everything in one package.

Anyway, good luck with it and please keep us up to date with progress. Always good to see how a project progresses. Even if it's a total failure (which I'm sure it won't be) let us know what went wrong.
You say you are using LDRs for sensors. It would be interesting to 'see' (pardon the pun) how well they perform.

 

[MORA99]

about analog multiplexor
http://www.rev-ed.co.uk/picaxe/forum/topic.asp?topic_id=6839

 

[Michael 2727]

You can make a LF Mouse from an 08M.
Slurp borrowed this from me to tease the Hawaiians ~ ;o)
<A href='http://www.cfht.hawaii.edu/~robotics/club/viewtopic.php?p=833#833' Target=_Blank>External Web Link</a>
Currently it's in the shop for a refit, not
happy with the performance, LDRs, MOSFETs
but it does work.
By the way Dick Smith's have a LT Mouse Kit for
under $20 bucks, mine should arrive tomorrow,
be interesting to see what you get for $20.00
these days.

Beny you could try pairing up opposing LDRs
in a voltage divider arrangement to save 3
inputs.


Edited by - Michael 2727 on 07/05/2007 09:37:13

 

[beny1949]

the multiplexers look good! I think i will give it a miss for this year, as i am rather straped for cash and i would like to keep it simple.

So far i have spent about &#163;10 on it, on the LCD and the PICAXE, everything else was reused.

&#160;

Edited by - beny1949 on 07/05/2007 10:38:22

 

[bgrabowski]

A good source of information for micromouse builders is the forum of micromouseonline.com, which is every bit as helpful as this one and has the advantage of being contributed to by many of the top UK competitors.

It has links to excellent technical articles for micromouse builders at all levels of proficiency. It explains difficult concepts like the maze solving algorithm and covers chassis design, motor selection and sensors in some detail.

 

[beny1949]

thanks for that bgrabowski.

i am assuming that running at 4Mhtz will be fast enough for maze solving logs?

Ben

 

 

[Bloody-orc]

Good work man.

Very good choice of sensors (LDRs) and LED color. But why so long boon for them? why do you need &quot;half a meter&quot; for it? But very nice indeed. (your camera could use some manual focus and macro though )

 

[beny1949]

thanks!

the &quot;half a meter&quot;... was that just a guess from the pics? or have I put that somewhere? if so sorry! the length of the sensor bar is 20.5 CM.

 

 

[Bloody-orc]

well I just looked the pictures. Line sensors are normally placed under the robot or in front of it. somehow robot builders try to avoid hammer-shark look with their line followers )

 

[beny1949]

not line follower... maze solver!!!

<A href='http://micromouse.cannock.ac.uk/index.htm' Target=_Blank>External Web Link</a>

this is the real deal!


just got an auto calibrate set of subs up, which work really well, but not in bright sunlight. ill have to make it some shades to add to the 'hammer head effect'!

Ben


 

 

[Bloody-orc]

ok now I'm lost...
Pardon me; I'm not familiar with mazerunners and their concept...
But those red LED sensors are aimed to the floor right? to detect what? a line maybe?

Anyway what ever the reason for those sensors they are(for me) oddly positioned. As mentioned earlier: it's a hammer-shark and IMHO that's not a bad thing to name a bot now is it


Anyway what ever the bot was constructed for I wish you luck with it and if it is going to some competition, then I wish you good luck in there too. nice bot (although odd-looking to my eyes). But extraordinary things are usually better than the usual plane box things . Daring to differ is something I admire. I do it a bit too often sometimes...

 

[beny1949]

lol, good philosophy though

anywho, the sensors look down on 5cm high walls. have a look at the link. the mice in the vid are the world class ones, not my school boy first attempt.

i know i have no chance in winning, not this year, but i can build on it for next year.

Ben