Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Robot Showcase (http://www.chiefdelphi.com/forums/forumdisplay.php?f=58)
-   -   pic: Mini-Arm - Controller (http://www.chiefdelphi.com/forums/showthread.php?t=26067)

Rich Wong 26-02-2004 13:52

Re: pic: Mini-Arm - Controller
 
Quote:

Originally Posted by nuggetsyl
that is how we control the arm that way it makes it easier for the driver. Trying to drive this arm with a joystick would be toooooooooo much.

Just this Control component has a lot of award potential!

Good luck!
:)

Bharat Nain 26-02-2004 17:11

Re: pic: Mini-Arm - Controller
 
Quote:

Originally Posted by mightywombat
Pots are most likely used. We did the same thing last year and it works sooo nice, as long as you have the time to tune the code to react correctly to the movement in the model. I am curious. How did you guys do it? Is it as simple as finding the difference between the actual and the requested pot values, dividing by a constant and adding the signed int to 127? I'm still working on code for a model, right now we are using joysticks and probably will for competition but its a project of mine.... How do you do it?


We used a Axis control code. It wasnt too easy to program this thing. Basically we used the difference between the pots and set the speed accordingly. Also this thing has been programmed with a lot of safety measures. It also has a timer set, so incase we get stopped by anything, it multiplies the speed forcing movement. Ask if anything else :D

Adams High Man 26-02-2004 21:44

Re: pic: Mini-Arm - Controller
 
Yea, i was origionally going to make a mini-arm to control ours, but then i decided to write inverse kinematics code! Unfortionatly, we ran out of time to actually TEST our robot, but it looks very hopeful. This way, we can control our 3 joint arm using 1 joystick. We can tell it to go anywhere in x-y space.

It was interesting to write, because i had to make my own integer trig library.

Alaina 27-02-2004 16:23

Re: pic: Mini-Arm - Controller
 
814 did the same thing last year with their stacking arm.
We won leadership in controls at Sacramento for it.

KenWittlief 27-02-2004 16:56

Re: pic: Mini-Arm - Controller
 
so whos car is now missing its headlight switch and knob? it looks like its off a VW.

Aignam 27-02-2004 18:09

Re: pic: Mini-Arm - Controller
 
Quote:

Originally Posted by KenWittlief
so whos car is now missing its headlight switch and knob? it looks like its off a VW.

Shhhhhh. Big Mike hasn't noticed yet...

eugenebrooks 28-02-2004 01:15

Re: pic: Mini-Arm - Controller
 
Quote:

Originally Posted by mightywombat
How did you guys do it? Is it as simple as finding the difference between the actual and the requested pot values, dividing by a constant and adding the signed int to 127? I'm still working on code for a model, right now we are using joysticks and probably will for competition but its a project of mine.... How do you do it?

We control a single rotating arm with a similar setup. The drive
to the motor is derived from the difference of the values returned
by the pots. This produces a torque on the motor that is proportional
to the error. The result, if you increase the gain to get good precision,
is a harmonic oscillator. If you run in to that, you add a damping force
derived from speed of the arm, calculated by reading its position on two
(or more) successive radio packet cycles between the OI and the RC.
The arm then behaves as a damped harmonic oscillator. With well
chosen damping it won't oscillate at all. Use quality pots that do a good
job of maintaining continuity as they are turned.

Mike Schroeder 28-02-2004 01:51

Re: pic: Mini-Arm - Controller
 
Quote:

Originally Posted by Aignam
Shhhhhh. Big Mike hasn't noticed yet...

oh i have noticed, its just that my car is so pathetic that needing a pair of pliers to turn my headlights on doesnt change much ;) (good thing this is a joke or else i would be pretty mad my cars headlight knob was gone)

Aignam 28-02-2004 08:36

Re: pic: Mini-Arm - Controller
 
Quote:

Originally Posted by eugenebrooks
We control a single rotating arm with a similar setup. The drive
to the motor is derived from the difference of the values returned
by the pots. This produces a torque on the motor that is proportional
to the error. The result, if you increase the gain to get good precision,
is a harmonic oscillator. If you run in to that, you add a damping force
derived from speed of the arm, calculated by reading its position on two
(or more) successive radio packet cycles between the OI and the RC.
The arm then behaves as a damped harmonic oscillator. With well
chosen damping it won't oscillate at all. Use quality pots that do a good
job of maintaining continuity as they are turned.

Funny story about this. The team was at the Bristol-Myers Squibb shop, and I kept hearing the programmers on our team and the BMS tradesmen talking about "pots", and I couldn't count the times I heard them make references to the "pots", since we were programming the arm. It took me most of the day to figure out that "pots" were potentiometers. Then everything started making sense. Alas, clarity.

steven114 28-02-2004 10:34

Re: pic: Mini-Arm - Controller
 
You wouldn't believe how many pot jokes were flying around while we were making our arm controllers... :p

Bharat Nain 28-02-2004 10:36

Re: pic: Mini-Arm - Controller
 
Sometimes, we even forgot which pot was hooked up to which part, and we used to upload the code and wonder what went wrong. Thank God for the "Kill Switch" :ahh:

Swan217 28-02-2004 10:44

Re: pic: Mini-Arm - Controller
 
Quote:

Originally Posted by steven114
You wouldn't believe how many pot jokes were flying around while we were making our arm controllers... :p

Oh we'd believe it - I'm sure that it does on almost every team that uses pots. It gets much worse when you break one of them and someone accuses you of "smoking the pot" :rolleyes:


Quote:

Originally Posted by Big Mike
oh i have noticed, its just that my car is so pathetic that needing a pair of pliers to turn my headlights on doesnt change much (good thing this is a joke or else i would be pretty mad my cars headlight knob was gone)

Don't feel bad - my blinker stopped working the other day. It still lights, but it doesn't "blink." It's quite interesting having to changing lanes or attempting to complete a Michigan Left turn while actuating a turnsignal by hand!!!

Note: Yes, I do even make the "blink" sound myself too.

steven114 28-02-2004 10:47

Re: pic: Mini-Arm - Controller
 
Yeah, we actually did smoke one of them. I accidentally hooked the wires up backwards, so that ground and power were next to each other (that's what I get for not having the right colors of the thin wire I was using!)
When I turned it all the way to one side, it started to glow red and smoke came pouring out. We quickly pulled the power :)

Amazingly, it still worked for a while, then it crapped out.

Joe Ross 28-02-2004 11:09

Re: pic: Mini-Arm - Controller
 
I think team 308 was the first to do an arm about this, all the way back in 2000. Here is a thread about it: http://www.chiefdelphi.com/forums/sh...ad.php?t=10706 Unfortunately, I can't seem to be able to dig up any pictures of it.

eugenebrooks 28-02-2004 19:16

Re: pic: Mini-Arm - Controller
 
Quote:

Originally Posted by Aignam
Funny story about this. The team was at the Bristol-Myers Squibb shop, and I kept hearing the programmers on our team and the BMS tradesmen talking about "pots", and I couldn't count the times I heard them make references to the "pots", since we were programming the arm. It took me most of the day to figure out that "pots" were potentiometers. Then everything started making sense. Alas, clarity.

A potentiometer has been called a pot since the early days of radio.
I have been asked for the "thingamajig" for the arm measurement,
to which I replied the proper way to ask for the part is: "Dr. Brooks,
I would like you to give me the pot." The teenagers in the shop thought
it was very funny. The nine dollar Bournes pots from digikey, very
good for continuity as you turn them, are, of course, smooth pots. In
our shop everyone was careful wiring them, so no one smoked the pot.


All times are GMT -5. The time now is 03:42.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi