Also like you I am trying to figure out the best mode of self locating within a environment.  If I can get my bot somewhat accurate then I dont think I will be to much of a problem.  My reasoning is this, if its locomotion is not perfect my grid will become obsolete after x time has passed. basically it will forget in a sense but if it can uptake enough information to off set this "forgetting" then I think it will be ok.  But the hard part is getting the bot somewhat accurate lol.

Now this is my question to the folks with more hardware experience.  Can I use 2 or 3 IR receivers on my bot to triangulate its position from a IR emitting base?  I was also pondering the idea of using the parts out of a mouse to get feedback of the bots actual movement but this idea also lacks accuracy but should be quasi easy to implement.


edit actually with some rough calculations with the 16mhz clock rate on the arduino and the speed of light the ir receivers would have to be placed ~19 meters apart on the bot.  Unless it is possible to detect the amount of noise in the IR signal or strength of the signal to calculate location but I dont know if that is possible due to my lack in hardware knowledge

oddly enough Im working on pretty much the same project as you.  Im pretty new to computational interactions with the physical world  ie I dont have much experience with uC and sensors.  But I am a forth year specializing in Computing science so I really do not think the software aspect is going to be that difficult.  Im using solar power to charge my lipos.  My plan was to implement a grid like you but each cord will have three values, one with the average amount of light, the probability of obstruction, and the number of times i have been there.  When the bot needs to "feed" (go to a source with light)  it will search the map for the highest average concentration of light and head that direction.  It will use the probability of obstruction as a weighted edge and will predetermine the shortest(<% of a obstruction) path to its destination. The varible storing the number of times the bot has been to that cord will be used for calculating both the average light and obstruction % as well as removing that cord from the graph(memory) which I am hoping will allow the useless parts, cords it doesnt travel to often, to be over written.  This would save memory for more valuable parts of the environment to be used.

Im getting my pc back at the end of the week and see if it works in a windows 7 enviroment

here is the output with verbose


System wide configuration file is
"/Applications/arduino.app/Contents/Resources/Java/hardware/tools/avr/etc/acrdude.conf"

user configuration file is "/users/XXXXXXXXX/.avrduderc"
user configuration file does not exist or is not a regular file, skipping

         Using Port            : /dev/tty.usbmodem0000103D1
         Using Programmer      : stk500v1
         Overriding Baud Rate  : 115200
avrdude: Send: 0 [30]   [20]
avrdude: Send: 0 [30]   [20]
avrdude: Send: 0 [30]   [20]
avrdude: Recv: . [86]
avrdude: stk500_getsync(): not in sync: resp=0x86
avrdude: Send: Q [51]   [20]
avrdude: ser_recv(): programmer is not responding
avrdude: stk500_recv(): programmer is not responding



does the usb cable length matter? mines about 4 feet

so I did screen /dev/tty.usbmodem..... and pressed reset a few times and it printed "a"$$ ???a??  &??a?&?? "?`?a("&0$?a0& ??""?`?" so I think its communication one way or my baud is off

Any ideas anyone. Im gonna try uploading onto it today using a pc hopefully its something wrong with my lap top but I have a sneaking suspicion that its not the computer causing the issue

No I have already read through that post and was unable to resolve my problem with that solution. But this morning I used screen /dev/tty.usbmodem..... and it wrote 5 or 6 random ascii characters and thats all. I think it is connecting because the tx and rx leds do flash a bit and my mac does seem to see the arduino.

anyone?

Specs
Macbook 10.4.11
new Arduino Uno atmel 328


I just got it and followed the getting started instrustion on this site(mind you i have nothing but the usb plugged in). Ive tried this with a several of the example sketches all with the same out come.

plug in the usb
open arduino 0021
open a example
compile/verify
upload

while Im uploading the Tx and Rx do flash a little bit (the rx flashes much more the the Tx)

then it says

"avrdude: stk500_getsync(): not in sync: resp=0x86"
a few seconds pass then
"avrdud: stk500_recv(): programmer is not responding"



I think this issue is often asked and im sorry I did search and none of the solutions seemed to work. ps does it matter which usb it is plugged into? also yes in tools its set to uno and tty.usbmodem

Thank you for your time and help


pps I also tried it on another macbook with exactly the same problem. And if this helps the L led flashes 4 times then stutters the flashes 4 times and stutters over and over when it is just plugged in

I have searched the internet for a week now and cant find the solution or even someone with the exact same problem as me and it is kind getting ridiculous is there anyone at arduino that can atleast tell me if it is my laptop or the arduino uno that is causing the issue

http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1287015838
here is my info. sounds like the op and I have close to the same problem

tired using a dc adapter while uploading and sitll got the same issue. Im really considering just sending it back

But then why isnt it uploading anything. my tx and rx flash but I get a avrude error

when I go into my network preferences it says "arduino uno is configured, but is not connected"

Sounds like the exact problem I have But im using a mac.