View Full Version : Step 1 for new game: Longer Auto/Hybrid
02-02-2008, 08:03 PM
I'm wishy/washy on the Hybrid period this year, but I'd like to see a longer autonomous period.
Adding encoders and all that is expensive- our team might not be able to afford it, but still... perhaps some canned routines that beginning coders could build upon (such as the kamikaze ultrasonic wall avoider) would be great.
I believe truly that a 30 second autonomous period would encourage and increase the value for this first period. Right now, teams that know how to do it will continue to do it. Teams that don't continue to struggle.
It's all about the code reuse.
And 15 more seconds might make it much more interesting...
02-03-2008, 10:35 AM
While I agree that having a 30 second period would make it more valuable, and thus require more teams to try it, I would currently be dead set against it. I don't think the majority of teams are really have the engineering/machining capabilities to finish with time left over for programming autonomous, and it’s very hard (if not impossible) to program with a field, or large enough area with carpet to make into a semi-real field. Driving on concrete or snow 30 feet and then turning does not give you the same results as industrial carpet. My guess is that instead of having 5-10% of teams trying you would get 10-25%, but those extra teams would all be paying for it in how good their robot was in the teleported period instead.
02-10-2008, 03:42 PM
That's why I'm torn. We are starting out this year with an all-freshman cast of programmers. They're starting out at the ground floor and... some of them are taking to it better than others. I don't know how much to start other than to give them a java book to get started.
Code reuse is great- but tis just getting to the point that you can teach it to someone else.... and then it will build.
I still would like to build an ultrasonic runabout. We'll see this year we've got a good prototype that it might be possible.
02-10-2008, 05:06 PM
There was a thread on this about a year ago here (http://www.chiefdelphi.com/forums/showthread.php?t=56090). Most sides have been argued to death, but there's still room for conversation.
Echoing on what I what I mentioned in the old thread was that for some teams, it'll still be a robot beauty contest. Hopefully with the implementation of the Robocoach this year, there won't be as many dead auto/hybrid matches as last year, but it'll still happen. From the stands, this can be the most boring part, especially to those new to FIRST.
Although I do agree with the importance of software rising in FIRST, I do disagree with extending autonomous. Making the autonomous challenges harder and worth more will inspire better software, not just allowing more time, or making it easy to score. 15 more seconds just widens the gap between the have and the have-nots in terms of software capabilities.
I have a feeling that the results of this game will be significant in how the GDC plans out the auto or hybrid periods of the upcoming games, especially in terms of robot-human interaction or other communication agents. We'll just have to see eventful those first 15 seconds are this year...
This is my first year programming, and it was so overwhelmed at first. We spent a large amount of time on Hybrid mode, and I agree that if it was 15 seconds longer we probably wouldn't have enough time to program it all. While it would give you oppurtunities to do more, I don't know if it would be worth it.
02-11-2008, 09:21 AM
It can't be that hard: Drive?Straight!Turn?Left!!
02-11-2008, 10:13 AM
It can't be that hard: Drive?Straight!Turn?Left!!
It's actually pretty hard to make your robot drive exactly straight for ~40 feet, especially if you're 2-3 degrees off of "straight" when you start out. And for this game, how hard do you turn left? Do you turn left with a 0 or 5 foot turn radius? There are a HUGE amount of things to consider if you want to have a successful hybrid mode. Going over the first line is a start, but it doesn't take 30 seconds to do it. Rookie teams should be able to setup a simple couple of lines that pseudo-says "After 5 seconds into the match (so I don't crash into my alliance), go forward for 5 seconds." Even at 7 ft/second, on a full battery that's at least 35 feet which is enough to cross the first line.
I'm still an advocate of the 15-second autonomous period. It's hard enough to build a good teleoperated robot for some teams. Telling those teams they have to work even HARDER is just plain mean.
If the glamour bots this year really want to step up their own hybrid mode, they should either hurdle or knock the ball off of both their own and their opponents' overpasses. That in itself is pretty hard and requires some sort of extra guidance or tracking system with some extra speed too.
02-20-2008, 03:42 PM
As a rookie team, we didn't even have much of a discussion about the hybrid mode (let alone autonomous). Yes, we have a robocoach and yes we can drive and turn using it but we ran out of time to consider doing much else. And - YES, we have a robot that COULD have done so much more but no time to program it properly. I'm not sure whether or not 15 vs. 30 seconds amounts to a ton more programming or not - seems like most of it should be autonomous anyway so it could just run longer. That's a pretty simplistic view however.
In any case, our motivation as a rookie team was to build a working robot that would not break down. i think we have succeeded in this. Next year, if we keep a similar drive train then we will have more time to program. That seems to be a bigger advantage for veteran teams than just the programming knowhow (unless you don't really have a programmer on the team).
well - that's my thoughts on it.
vBulletin® v3.6.4, Copyright ©2000-2013, Jelsoft Enterprises Ltd.