Nashua Robot Builders Website
Oct/Nov 96

Site Navigation

Nashua Robot Builders October-November '96 Newsletter

Hi all,

This is a two-month newsletter to cover October and November's meetings.

October saw the re-run of our line-following contest our line-following contest. Bascially a robot has to follow a crooked strip of black tape along a white sheet of Tyvek house wrap to beat the clock. Two robots re-entered and both won.

Jay Francis and I competed with his modifed toy buldozer and my modifed tank. The sensors really made the difference from last time we ran the contest, we were both using Sharp IR detectors and modulated IR light so there were no false detects due to sunlight, wrinkles in the Tyvek or gaps in the tape. I used the "analog hack" of the Sharp IR sensors ( http://www.catwoman.mil.ufl.edu ) to get a signal proportional to the ammount of IR we were getting back from the sheet/tape. The "analog hack" is really just a plain modulated IR detector from Radio Shack with a tiny wire soldered to a spot inside, then use an LM324 op amp to buffer the signal for later processing.


Basically:

To Do the analog hack.

Attach +5 to pin 2 of the IR sensor, GND to pin 3 of the sensor

Connect the wire from the Sharp analog hack to pin 3 of the op-amp

Connect pins 1 and 2 of the op amp together

Use pin 1 as output (buffered signal)

Attach +5 Volt power to pin 4, Gnd to pin 11

Ignore the other pins.

Use a wire-wrap socket to hold the IC and stick-tape it to the IR sensor, then wrap them up with tape for a firm hold.

This gives you the analog hack plus a voltage buffer so you can plug the signal into some other analog input (like a 6.270 analog input) without degrading the signal due to overloading the weak signal from the analog hack.

Then we got into "Gee Wally, what do we do with the signal now that we've got it?" Well.....

I built three of the above sensors and stuck them all together on the front of the bot with modulated IR LEDs shining right down in front of them. Then, I added some short heat shrink tube over the sensors so they'd see just the reflection from the line and not the LEDs. This gets you a signal from at least one sensor whenever the line is underneath them and no signal when its not. Now you
KNOW if you're over the line or lost.

With this in hand I basically told the robot to steer straight if middle was darkest, left if right darkest, right if left darkest, and reverse with a twist if lost.

There you go gang, like they used to say: Garbage In Garbage Out and with good sensors, you get good stuff IN.

On to November!

Today we had really good turnout. Quentin Lewis, Matt Cross, Ron Beck and Jay Francis were all there and we had a really good talk about robotics, computers, where the industry is heading, who the leaders are, where we as professionals are heading and what to next with the club. We also talked alot about the 68HC11 processor and all the neat stuff you can do with it.

Matt was so interested in HC11 I loaned him a CPU, reset chip, MAX232 and a copy of PCBug11 to build his own prototype like I did. With all that its possible to get a simple HC11 system up and running in a good afternoon. Add another afternoon for external memory.

Ron brought a unique viewpoint as a previous company owner as to where things are going and how to run a good company, thanks Ron!


N E W C O N T E S T !!!


Our next contest is a "find the light" game. We'll build a 40kHz IR beacon and robots will have to find it. See the attached rules for details!

Date is 2nd Sat of March, 97. Be there!


 

[Homepage] [Club Purpose] [Directions] [Newsletters] [Std Platform] [Contests] [Members] [Other Links] [Contact Us] [Events]