Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Extra Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=68)
-   -   pic: Uber3.0 - prototyped (http://www.chiefdelphi.com/forums/showthread.php?t=98281)

kajeevan 15-11-2011 07:24

pic: Uber3.0 - prototyped
 

sgreco 15-11-2011 07:25

Re: pic: Uber3.0 - prototyped
 
That is one sweet looking frame.

Want to give us some specs?

kajeevan 15-11-2011 08:04

Re: pic: Uber3.0 - prototyped
 
The cad and specs are both in this link.

http://www.chiefdelphi.com/media/photos/36993

Hawiian Cadder 15-11-2011 23:24

Re: pic: Uber3.0 - prototyped
 
We completed a prototype this fall geared similarly to this, I completely see why you chose a faster speed, with a reasonable amount of practice it is very controllable and highly useful.

Andrew Lawrence 15-11-2011 23:31

Re: pic: Uber3.0 - prototyped
 
Quote:

Originally Posted by Hawiian Cadder (Post 1085278)
We completed a prototype this fall geared similarly to this, I completely see why you chose a faster speed, with a reasonable amount of practice it is very controllable and highly useful.

I'm a bit mechanical dumb, so sorry if this is obvious, but how do you achieve the speed of 16 ft/s? I see it's through CIMple Boxes, but is that it?

Thanks!

P.S: After reading the CAD paper again, it says it goes through CIMplex Boxes, not CIMple Boxes. What are CIMplex Boxes? Are they like CIMple Boxes?

Chris is me 15-11-2011 23:41

Re: pic: Uber3.0 - prototyped
 
16 FPS single speed sounds brutal. Not enough room for error there... A 12 feet per second skid might have that.

Did you not account for frictional losses? If you didn't, then your drive is very similar to 1503's this year.

kajeevan 16-11-2011 00:40

Re: pic: Uber3.0 - prototyped
 
Off the gearbox I'm running 32:38 for a total ratio of about 5.55:1. The wheel well layout is inspired by 1503 but instead of 35 chain I went 25 and threw in bolt action tensioners to pull the end wheels to take up any slack. Also I decided dead shafts opposed to live. Also I'm not worried about to much speed, we enjoy going fast, plus we have the option to throw in an electronic shifter as we have in past years.

P.S. My mistake it is CIMple not cimplex.

Brandon Holley 16-11-2011 08:49

Re: pic: Uber3.0 - prototyped
 
Quote:

Originally Posted by Chris is me (Post 1085283)
16 FPS single speed sounds brutal. Not enough room for error there... A 12 feet per second skid might have that.

Did you not account for frictional losses? If you didn't, then your drive is very similar to 1503's this year.

In high gear this past year, we were moving faster than this (actual speed). It required some pretty sophisticated control software that 254 helped to develop. Once we had this setup, it made driving at that speed much simpler.

I would not recommend trying to drive that fast without some software support. I think a simple 2 stick, or 1 stick drive setup would be quite difficult to handle. However, every driver is different, so YMMV.

-Brando

JamesCH95 16-11-2011 10:04

Re: pic: Uber3.0 - prototyped
 
Quote:

Originally Posted by Brandon Holley (Post 1085312)
In high gear this past year, we were moving faster than this (actual speed). It required some pretty sophisticated control software that 254 helped to develop. Once we had this setup, it made driving at that speed much simpler.

I would not recommend trying to drive that fast without some software support. I think a simple 2 stick, or 1 stick drive setup would be quite difficult to handle. However, every driver is different, so YMMV.

-Brando

Care to elaborate about what it did? Balance wheel speeds using encoder feedback?

Brandon Holley 16-11-2011 13:51

Re: pic: Uber3.0 - prototyped
 
Quote:

Originally Posted by JamesCH95 (Post 1085318)
Care to elaborate about what it did? Balance wheel speeds using encoder feedback?

We utilized both encoders in the drive system, and a gyro.

We basically just elaborated on Cheesy Drive. 2 stick drive system- 1 stick is essentially "throttle" the other stick is essentially "steering". There was a button on each stick. One button for shifting, the other button for "quick turn". Quick turn essentially turned the steering stick into a single stick drive system for turning in place.

As we traveled at a faster rate of speed, the gains in the turning loop would adjust so that dumping the steering stick all the way to one side or the other would create a smooth, controlled turn. This occurred even at 16+ ft/s top speed when we traveled back and forth across the field. Additionally, even at low speeds, the robot could only turn in place using the quick turn button, otherwise it would drive in a small arc.

I did not do the actual coding of it, so I'm speaking more from a theoretical/design point of view, but that's essentially what it did.

Driving without this setup at top speed was quite difficult. We noticed right away as our driver began driving just the drive system in week 3 or so. He practiced driving the full field and then dipping into/out of the human player lanes over and over again. We continued to tune the system until he felt it was solid, and we were satisfied with the results.

-Brando

JamesCH95 17-11-2011 11:11

Re: pic: Uber3.0 - prototyped
 
So, if I'm understanding you correctly, turning rate limits were controlled based upon the speed of the robot.

That does seem like it would make the robot easier to control. I also like the idea of a "quick turn" button too. I assume that made the robot to a very fast about-face?

Brandon Holley 17-11-2011 14:01

Re: pic: Uber3.0 - prototyped
 
Quote:

Originally Posted by JamesCH95 (Post 1085506)
So, if I'm understanding you correctly, turning rate limits were controlled based upon the speed of the robot.

That does seem like it would make the robot easier to control. I also like the idea of a "quick turn" button too. I assume that made the robot to a very fast about-face?

That is correct. As the robot moved faster, the turning became less "sensitive". It essentially allowed the driver to still use coarse controls to maneuver at high speeds.

Quick turn would change the control scheme to allow turning in place. If you held the button down and moved the stick left and right, you essentially created the control scheme you get from moving one stick forward and the other backward in a standard 2 stick setup.

-Brando

AdamHeard 17-11-2011 14:09

Re: pic: Uber3.0 - prototyped
 
Quote:

Originally Posted by Brandon Holley (Post 1085532)
That is correct. As the robot moved faster, the turning became less "sensitive". It essentially allowed the driver to still use coarse controls to maneuver at high speeds.

Quick turn would change the control scheme to allow turning in place. If you held the button down and moved the stick left and right, you essentially created the control scheme you get from moving one stick forward and the other backward in a standard 2 stick setup.

-Brando

Just to clarify, this means turning in place is exclusive to quick-turn, and in all other modes turning is in arcs?

Brandon Holley 17-11-2011 14:23

Re: pic: Uber3.0 - prototyped
 
Quote:

Originally Posted by AdamHeard (Post 1085537)
Just to clarify, this means turning in place is exclusive to quick-turn, and in all other modes turning is in arcs?

That is correct.


All times are GMT -5. The time now is 09:14.

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