Wednesday
Mar222017

Terrible Stupid Robots

My daughters school has a yearly science/art/culture project fair.  She comes up with ideas and we do them together.  One year we looked at the descrimination faced by green or purple skinned people.  One year we built a raspberry pi computer that was hooked up to fruit and worked like a capacitive music synth drum kit.  One year we tested the structural properties of eggs to the point of failure, which was about as messy as fruit being pounded on by grade school kids.  (FUTURE ME TELLS PAST ME:  You are really going to need that fruit beating touch sensor so that you can sense robot arm position but you will take a couple weeks to figure that out yet dummy!)

This year we are doing an Robot.  Yep, that's what my daughter wants, a robot - and she's been quite firm on the details of what the robot must do.  I'm in over my head, and not just because building a robot arm that requires thousands of ArcTan(Y/X)-(Pi - elbowAngle)/2 style calculations  to move isn't third grade math.  (FUTURE ME TELLS PAST ME:  thats inverse kinematics and you didn't "reinvent" it quite right on your first try dummy!)   And starting her soldering skills on a expensive robot is not the best idea either.  (FUTURE ME TELLS PAST ME:  Two girls did more than half the soldering and that worked great!  They rapidly mastered the skill with steady hands, keen eyes and patience)  I'm in over my head because of the state of electronic integration and interoperatability.  Servos PWM modulation is like FM raido.   Stepper Motors use AM radio style pulses.  Or maybe they are best compared to rotary telephone dialing and airport flag hand signals. 

I need both so stack on extra "HAT" boards.  Stepper motor sizes need to be matched to drivers and controling hardware in a way that seems arbitrary and over complicated.  And the motors generate enough electronic noise to cause odd twiches and normally have some mechanical offset.  Tessa's grand idea seems to therefore require machine vision, which I am not going to do.   (Future me tells past me:  And that touch sensor hat.)

Mathematica is normally my go-to choice for programming.   Its a strong choice for this project because Mathematica is free on the Raspberry Pi, our robot brain.   I already know how to program in mathematica and it has the needed morphological machine vision tools.   However, the motors are run from Python code, and Methematica and Python interoperate indirectly.  And my python skills are weak and rarely used.  (FUTURE ME TELLS PAST ME:  Python and Mathematica on the RPi lack simple workflow communication tools available on other platforms.  You are going to have to learn python and write quite a lot of code in the next month dummy!)

I'm increasingly uncertain we will have a fully working robot come exibit day.   The deeper in we go, the bigger and more patched together it gets. (FUTURE ME TELLS PAST ME:  it all worked out fine, its just a kids project fair and who needs sleep the night before the fair anyway dummy!)

Perhaps plug and play prototyping is a bit far fetched - but it seems like it could and should be real.  And it probably will be a thing given the direction arduio and the Raspberry Pi are taking robots.  It just can't happen soon enough.

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments

There are no comments for this journal entry. To create a new comment, use the form below.

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Post:
 
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>
« We Haz Robot | Main | Taxation Without Representation »