Combining Toddlers...?

passingofamerica

Inspired by Star Wars' AT-AT, using 'mechano' my robotics class partner and I have been trying to combine two toddlers. We were successful in physically combining them, but are now stuck on making the robots react to IR sensors in sync.

Our first attempt was to simply 'piggyback' the signal from the IR receivers in the front bot and have the wires plug in to the pins, Vss and Vdd ports in the back bot. While the front IRs were working correctly, the rear bot did not appear to receive the signal.

We then tried to use two individual IR receivers on each left and right (one for the back, and one for the front bot) however, that plan seems to have failed as well.

We are testing the IR signals using two IR Test Programs on individual computers (one for each bot) and we cannot seem to make this work. We are confident our wiring is correct (works well with LEDs), however, we are still stuck.

Any thoughts/suggestions?

Post Edited (passingofamerica) : 12/15/2008 3:02:30 PM GMT

Mike Green

How about a schematic showing how you have the two Bots connected?

GWJax

Also post the code as well so we can get a full picture of what your doing and a Picture of your bot may help as well.

Jax

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
If a robot has a screw then it must be romoved and hacked into..

Ugha

I've had good success with using one stamp as a master and the other as a slave... the master tells the slave to do an IR scan
via SEROUT and then uses SERIN to wait for the result.

Slave receives the message (It always sits waiting for instructions while its not actually doing anything) and then does the IR
scan, sending the IR results back to the master via serial.

Like the others, I'd be able to help you more by seeing how you have things wired up and coded.

erco

I'll be interested to learn how your FrankenToddler walks, or more specifically, how it turns. That tilt/stride mechanism seems to be optimized for a biped that rotates about its own axis, sort of a differential drive for walkers. It might struggle with four legs on the ground, or you may get a pleasant surprise and it may work great. If you get a chance, please post a Youtube video.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·"If you build it, they will come."

GWJax

If you attach a swivel joint between the body of the legs you can get it to make a good gant if programmed right.

Jax

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
If a robot has a screw then it must be romoved and hacked into..

Ken Gracey

I've programmed a set of Toddlers to dance. These guys are linked via our RF modules:

http://www.parallax.com/Resources/ApplicationsContests/Robotics/ToddlerTroupe/tabid/504/Default.aspx

Download the video to see what it's all about. Here's a photo of them:



Ken Gracey
Parallax, Inc.

Humanoido

Ken, that's an awesome project! The web page is
well put together with complete information and very
useful. But how did all those Toddlers become
black and blue? What's the story behind that?

humanoido

Ken Gracey

Hey humanoido,

The blue is the square dance caller, and has the RF transmitter. The black Toddlers are all receivers, or the square dance participants. Well, it's not really a square dance but a "hokey pokey" kind of party.

They're individually powder-coated. I tried that instead of anodization once to compare the finish. The quality of powder coating was fantastic, though it certainly introduces more issues relative to tolerance than anodizing.

Ken Gracey

passingofamerica

···· Thanks for the responses so far...

Okay I have some pictures and schematics...


Side View


Close Up


Wiring On Front Bot


Wiring On Back Bot


Schematics ...


Hopefully all of this is helpful.

' {$STAMP BS2}
' {$PBASIC 2.5}
'Testing only right IRs...
' -----[noparse][[/noparse] I/O Definitions ]-------------------------------------------------
rEmitter CON 10
rInput VAR IN11
' -----[noparse][[/noparse] Variables ]-------------------------------------------------------
rDetector VAR Bit ' detector output values.
' -----[noparse][[/noparse] Initialization ]--------------------------------------------------
OUTPUT rEmitter
' -----[noparse][[/noparse] Main Code ]-------------------------------------------------------
DO
  FREQOUT rEmitter, 1, 38500
  rDetector = rInput
  DEBUG " Right= ", BIN1 rDetector
  PAUSE 20
LOOP

This is all we have for code. Just wanted to make sure the IRs worked between both bots before we dove in. Our plan was to use the same program in both bots, have them run at the same time and hopefully react the same.

P.S.
····· The hokey bots are pretty cool. [noparse]:)[/noparse] Kinda creepy though....

Post Edited (passingofamerica) : 12/17/2008 2:29:53 PM GMT

GWJax

here, I made some changes to your code that should work better for you.

' {$STAMP BS2}
' {$PBASIC 2.5}
'Testing only right IRs...

'

---

[noparse][[/noparse] Variables ]

---

rDetector VAR Byte ' detector output values.
rOutput VAR Byte
rInput VAR Byte

'

---

[noparse][[/noparse] I/O Definitions ]

---

rEmitter CON 10
rInput = IN11

'

---

[noparse][[/noparse] Initialization ]

---

OUTPUT rEmitter

'

---

[noparse][[/noparse] Main Code ]

---

DO
rDetector = 0
FREQOUT rEmitter, 1, 38500
rOutput.LOWBIT = ~rInput
rDetector = rOutput
DEBUG " Right= ", BIN1 rDetector
PAUSE 20
LOOP


Jax

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
If a robot has a screw then it must be romoved and hacked into..

passingofamerica

UPDATE::
············ Because we only have until tommorow before our school's christmas holiday we decided to take a sure (but messy) route. We decided to use the toddlers pre-made IR pins for the set of IRs on the lead bot, while the board of the lead bot works as a sort of holding place for the back bot's·IRs.

Schems:



This works, okay, however, ...four fidgety IRs isn't·as precise as I wish

P.S.

·· Can someone explain the improvements Jax added?