Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Programming (http://www.chiefdelphi.com/forums/forumdisplay.php?f=51)
-   -   Any Team Using A Timer? (http://www.chiefdelphi.com/forums/showthread.php?t=90556)

SM987 11-04-2011 12:56

Re: Any Team Using A Timer?
 
Quote:

Originally Posted by BMWilson (Post 1051684)
We designed a sensor to detect the tower's on/off sequence. It initially began as a tower illumiantion and color sensor, but the brightness of the field lighting pretty much monkeywrenched that idea. There are six transitions of light to dark (you can count them if you disagree). After the sixth transition a digital high is sent to the cRio to tell it to deploy the minibot. Works like a charm!

We tried that as well, but had issues when we wound up at the pole after 15 seconds, so we just "hid" our sensor inside our alignment arms and tuned the voltages for red/blue.

jvriezen 11-04-2011 14:11

Re: Any Team Using A Timer?
 
Our bot deployed when both button is pressed and clock <= 10 seconds. We worried about timer differences, but in practice, where we deployed in most all of 20 matches at Lake Superior, we were never called for being early. Our MB was a few inches behind the frame perimeter so we had a fraction of a second of grace there. However, we used pretty fast surgical tubing deployment drive.

In at least one case, where we and our finals alliance partner (2169) were both deploying, our slower bot deployed via timer beat the noticeably faster MINIBOT deployed via human only.

2169 adopted our timer based deployment for their next regional, and I watched them via video streaming in the elims and don't recall any early deployments there, but I can't say for sure.

We tossed around the idea of setting the deployment at something less than 10 seconds to account for clock difference, but we never implemented and it never cost us a penalty. I think we won nearly all (if not all) MB races where we got the bot on the pole (which was most of them) This was due to a combination of timer deploy, fast deploy and reasonably fast bot (gear boxes with Tetrix wheels)

Maybe we were lucky about 17 or 18 consecutive times or so, maybe it was reliable. Who knows.

John Vriezen
Team 2530 "Inconceivable"
Mentor, Drive coach, Inspector

tr6scott 12-04-2011 06:16

Re: Any Team Using A Timer?
 
We use a timer to block inadvertent deployments before 15 seconds left. After 15 seconds, the drivers have to hit 2 buttons to deploy. If the drivers hold the buttons for 2 seconds, it will bypass the timer altogether.

The drivers have deployed early once. Kettering, and the first time the minibot climbed the pole.

DinerKid 12-04-2011 06:46

Re: Any Team Using A Timer?
 
Quote:

Originally Posted by tr6scott (Post 1051832)
We use a timer to block inadvertent deployments before 15 seconds left. After 15 seconds, the drivers have to hit 2 buttons to deploy. If the drivers hold the buttons for 2 seconds, it will bypass the timer altogether.

The drivers have deployed early once. Kettering, and the first time the minibot climbed the pole.

At 15 seconds the finale noise sounds and i must say it seems out of place to me (should be at 10 IMO) but even if you still rely on a driver to deploy the mini bot having a way of making sure that they get through the rest of the match (up to say 12 seconds remaining) without deploying would be a good idea.

My opinion has always been that it should be made the way it was described above where a driver must hit a button and the timer must be <= to 10 seconds. And you would definitely want a manual override in case the timer messed up somehow (10pts for last is better than 0 points for not being able to deploy)

~DK

aechmtwash11 12-04-2011 08:30

Re: Any Team Using A Timer?
 
just wanting to clarify....

So are you all saying the Game time feedback that is available in the FRC default framework code (labview) is not accurate???? Doesn't it just pull from FMS?

Owen Meaker 12-04-2011 12:01

Re: Any Team Using A Timer?
 
I made the code prevent deployment until the 110 second mark. Factoring in the roughly 0.5-1 second deployment time, there is little chance of deploying early, even if the driver clock is off a little. One of the main advantage is that the driver can hold the trigger down, and it will immediately deploy when the 110 second mark is hit. So once they are in position, all they have to do is hold down the trigger and wait, rather than trying to time a button press to a timer.

tr6scott 12-04-2011 12:42

Re: Any Team Using A Timer?
 
Quote:

Originally Posted by DinerKid (Post 1051839)
At 15 seconds the finale noise sounds and i must say it seems out of place to me (should be at 10 IMO) but even if you still rely on a driver to deploy the mini bot having a way of making sure that they get through the rest of the match (up to say 12 seconds remaining) without deploying would be a good idea.
~DK

The cut off point was debated, and because the cRIO is not in sync with the field management, the decision was to "open the window" around the 10 seconds to not interfere with deployment.

After the first premature launch, it was debated again. :) The drivers have learned. Our deployment releases with a servo lock, this action takes .5 second, before the minibot moves. The drivers like the ability to deploy at 10.5 seconds if it is going to be a close match.

ChrisH 12-04-2011 15:22

Re: Any Team Using A Timer?
 
Quote:

Originally Posted by santosh (Post 1051662)
Has anybody done any tests this year to determine whether or not this years matches are running a little long like people were saying earlier in this thread? or has anyone had issues with deploying slightly early/late if they were using an automated system?

I have it from a reliabe source in FIRST HQ that this issue has been addressed and corrected in the current version of FMS.

Joe Ross 12-04-2011 15:39

Re: Any Team Using A Timer?
 
Quote:

Originally Posted by aechmtwash11 (Post 1051866)
just wanting to clarify....

So are you all saying the Game time feedback that is available in the FRC default framework code (labview) is not accurate???? Doesn't it just pull from FMS?

FMS does not publish the time. The LabVIEW framework looks at the state transitions (telop, autonomous, disabled) and uses the cRIO timer to measure the elapsed time in each state.

Last year, the FMS timer was Accurate but not Precise. Hopefully, that's now be rectified.

Eric222 12-04-2011 20:00

Re: Any Team Using A Timer?
 
I would advise you against using a timer. At Boston, the Driver Station and our robot lost communication for about 10 seconds (Due to a bad wire connection, we learned later on), and we had a timer programmed in for firing the minibot. The robot luckily reconnected, but the end result was that the minibot did not deploy when we tried to launch it in the end of the match. We basically threw the timer out of the program after that match. So, I would listen to everyone else when they say that it is probably best to leave minibot deployment up to the human controller. Take the chance of complete failure to deploy out of the picture.

Ziv 12-04-2011 21:56

Re: Any Team Using A Timer?
 
Team 125 also used a timer at Boston. The fraction of a second head start our minibot had against team 78's equally fast minibot was crucial in the finals. Given that the FMS's time seems well behaved—we were never penalized for early deployment—I would at least give the timer careful consideration. (However, as is the case with any automated function, there should be a manual override that can deal with cases like the one Eric described above.)

Chris Fultz 12-04-2011 22:36

Re: Any Team Using A Timer?
 
Quote:

Originally Posted by Eric222 (Post 1052151)
So, I would listen to everyone else when they say that it is probably best to leave minibot deployment up to the human controller. Take the chance of complete failure to deploy out of the picture.

It is smart to make automated systems, but it is always wise to have a manual override.

dwodrich 13-04-2011 08:27

Re: Any Team Using A Timer?
 
So what value are teams using? Elapsed time is what the framwork provides, so 15+120-10 = 125 as the deployment time or does the elapsed time start over at the start of teleopt like the big screen? Should we use 125 or 110?

plnyyanks 13-04-2011 09:20

Re: Any Team Using A Timer?
 
Quote:

Originally Posted by dwodrich (Post 1052347)
So what value are teams using? Elapsed time is what the framwork provides, so 15+120-10 = 125 as the deployment time or does the elapsed time start over at the start of teleopt like the big screen? Should we use 125 or 110?

we only allowed out minibot to be deployed after 110 seconds of teleop had elapsed. We never got penalized for an early deployment

Blackphantom91 13-04-2011 16:32

Re: Any Team Using A Timer?
 
A team at a regional used an I phone timer to do the time (which is not legal) I would agree with most above do it of visual and practice way, The refs will deactivate quick if you pass that cylinder the minibot.


All times are GMT -5. The time now is 23:31.

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