While I was busying programming the robot and learning about our arm design and how everything is working I seemed to miss one thing, my team has decided to not use the camera. I am not sure what our autonomous mode is going to be but the general consensus is that we should do nothing and just focus on the human controlled part of the tournament since we are a rookie team and autonomous is only 15 seconds. Personally I feel like this is a REALLY bad idea. Whats your feeling about this? If we should change it how should I approach the team?
In past years aid’s for autonomous weren’t used as much as first had wanted them to. White tracking lines were used by a small number of bots and hardly anyone actually used the infrared from last year. I think that the camera will be used more than past aids. Although there will be many teams that won’t use the camera, you should tell your team that it could be the difference between winning and loosing. If your willing to program it and put in all the extra work I don’t see why they should turn it down.
The great thing about autonomous mode is that it doesn’t require any additional effort from anyone else on your team, other than to find a place to mount the camera. (You could do that yourself, if you really had to, but I’ve noticed that people will usually offer to do it for you, rather than let the programmer touch power tools.) To avoid conflict with your team, finish your driver control code first. If you get this done quickly enough, you should have a couple of weeks to do nothing but experiment with the camera and come up with an autonomous program. If it works, great. If it doesn’t, you’ve lost nothing. My prediction is that teams that have an effective autonomous program will be more likely to win matches (since capping the center goal is like nine free points), to be picked for finals, and to win judges’ awards. Show your team this post, and then follow through with your end of the bargain. They will thank you in the end.
Frankly, the camera seems to my team as a waste os precious wieght. The tracker in the future will be used more frequently, and the autonomous mode will be used more and more extensively. Our robot design needs that 15 seconds for more important things. And I am not at liberty to be talking about it on here. wait till competition to find out. TRENTON.
If it really comes down to not using the camera, sit down with the drive team and figure out alternate strategies you could use. Trust me, if you think hard enough you will figure it out. There are many things you could do in this game.
It seems that I have heard far too many people that think that using the camera is the only useful thing that you can do in autonomous mode. Frankly, I applaud your teams decision not to focus resources on the camera. If you don’t have the people time to do it right it will just make other parts of you robot suffer. However, there are many other useful things that a robot can do in those 15 seconds. Just think about things that will gain you a strategic advantage when the RC mode starts, but don’t get in the way of your partner robots.
I just want to reinforce Matt’s comments, and point out that there are a lot of things that your robot can be doing during the autonomous period. If the teams feels that, for any of a number of reasons, using the camera is not something that they want to do this year, that is OK. However, don’t just sit there during the autonomous period. Design your robot such that when it starts in one of the two “pocket” starting areas, it just has to swing a simple little arm and knock down one of the two tetras hanging inside the corner goals. While this will only score one point, it is one point that you alliance would otherwise not have. Or make a simple arm to take the one “starting tetra” that your alliance will have and flip it on to the goal sitting next to you when you start the match. This can be done with just one rotational motion, and can give you the easiest three points you are likely to get in the match.
Potentially, 33 points can be scored by an alliance during the autonomous period (two vision tetras stacked on the center goal for three points each and each giving two more three-point bonus tetras in the corners, knocking down the two corner hanging tetras for one point each, and placing the “starting tetra” on the middle goal for three point and making a row with the bonus tetras for ten points, for a total of 33 points). I will be very (pleasantly) surprised if any alliance scores ALL of those points during a match. But it is probably reasonable to expect that 8-15 points will be scored on a fairly regular basis. Given that other threads in this forum have indicated the average expected scores will run in the 45-65 point range, that means that 20-25% of the points may be scored during the autonomous period. If these projections turn out to be true, then sitting there and not doing anything during the autonomous period while your opponent is scoring a like number of points will put you way behind in the match.
I’m about to suffer the wrath of the FIRST Gods by disagreeing with the esteemed Dave Lavery…
In general, I agree with what Dave outlined above. However, one must be careful to be honest in your communications with alliance team members in what you can do.
If your robot interferes with a partner’s robot during autonomous, you could prevent that robot from scoring OR cause damage to it or your robot (not a good thing).
You should strive to do something like knock down the Magnetic tetra or score with the starting tetra. you might even want to maneuver to the automatic loading station and grab a tetra from a known location to score (no vision required).
However, be honest with your partners. If your autonomous efforts are not proven reliable before you ship your robot, doing nothing may be the best strategy for the overall good of the alliance.
OK. I had better stop now… The dark clouds of FIRST blasphemy are starting to gather over my head! Sorry Dave… I’ll never do it again!
Nope, Mike, you are exactly correct. In my earlier post, it would have been more accurate of me to say that the ALLIANCE should not sit and do nothing during the autonomous period. Based on the capabilities of the other robots in the alliance, having your robot sit and count NO-OPS for 15 seconds may be the exact right thing to do.
However, you should also be prepared for the inverse case and the occasional alliance pairings where your robot will be the strongest member of the alliance. In such situations, making the strategic decision during the build period to not implement any type of autonomous capability will come back to haunt you. Even if you are not going to use the camera system, it seems that you would always want to at least build in one of the simpler capabilities. Having the option to use (or not use) those capabilities during the autonomous period would always be to your advantage.
Our strategy this year in designing a robot has been sort of like Dave Lavery said. Don’t ask yourself “why do we need this” ask yourself “why not have this?” We tried to keep as many options open as possible this year. We don’t want to be restricted in game play by robot design/functionality.
My team is probably heading towards the same conclusion. We’re already short-staffed as far as people who know what they’re doing. So getting the really cool stuff up and running will probably take longer than we have.
Last year we had one of the most effective auto modes around. We had at least four different ways to get on the platform. We also had other autonomous programs we never used. Some of our programs had a delay so we could let our partner go across our path without interference. All were dead reckoning. One of them we call “Mars Rock”. Guess what it does? That’s right, the robot sits there waiting for Opportunity to come along.
If nothing else, drive forward into your opponents zone and park in a strategic place. While some robots will be able to deal with a robot parked in front of them, most probably will not. If you are playing defense then at least you are better positioned than if you stayed put. This requires only a minimal effort on the part of the programmer. No other hardware is involved if you use dead reckoning.
There are ways to score tetras without vision. One of our mentors has a complete distrust of all sensors, so we have come up with several “fall-back” strategies on what to do if we can’t get vision to work for us. Some of them can be accomplished with dead reckoning alone.
The bottom line is that there are only so many robot seconds out there on the field. Just sitting there doing nothing during autonomous wastes 4% of them. That means your alliance is going to have to work at least 8% harder to catch up during the rest of the match.
sometimes not having an autonomous mode may be very helpful. in 2003 (Stack attack) we did not have a very good autonomous mode. during one match the casters on our robot were not positioned right and during autonomous mode our robot ran against a wall and fell on its side. We were dead for the rest of the match. I would say a team should consider some sort of autonomous mode that at least moves your robot down the field so you don’t need to wast the teleoperated mode time.
although it would seem in practical, couldn’t a team potentially take one of the magnetic tetras off, and then stack it? This would give a maximum autonomous mode point value of 37
Well I hate to be a pessimist (ok … maybe I don’t hate it ) but I will be very (pleasantly) surprised if more than one percent of all robots will be able to cap reliably during autonomous mode. And if you can’t cap during automonous then using the camera becomes much less important. Although finding and picking up a vision tetra during automonous will at least have you loaded and prepared to score immediately once the driver takes control.
WOW … I have not seen or heard that 45-65 number … Again I will be very surprised to see the average over 50. And not to hijack this thread, but I also feel that because of the low scores the penalties are very dis-proportionate to the scoring potential this year. Play clean … penalties will KILL you!
Our team was tossing around the idea of not using the camera either, but we figured it could help us get over to the automatic or manual loading zone so we’re using it. Which is good because otherwise I would have gotten stuck with writing autonomous scripts…joy… :o
Last year we didn’t focus on autonomous and just let our mentors from RIT mess around with it. It wasn’t until the championship we tried to get up on the platform during autonomous, but we actually ended up breaking one of our drive chains. Point it you can do it w/o autonomous, and autonomous DOES open you up to some damage with other robots (collisions during dueling autonomous modes aren’t uncommon,) but overall you SHOULD focus on the human element and once you have a strategy down for that THEN focus on autonomous mode.
Last year we were a rookie team, and we really did not see that much advantage to autonomous. When we actually went to the competition and saw the number of robots that actually had functional autonomous (that knocked the ball off the tee) we concluded that FIRST had not intended it to be that difficult. But when we got picked by another high placed rookie team for the finals, they sent one of their mentors over to help, and he showed us how to write a quick and simple program to drive forward and block another robot from knocking a ball off.
The Moral of the story is that Although the program didnt work perfectly, it still got us halfway down the field, which later allowed us to score 15-20 points we otherwise would not have.
However I do strongly recomend that unless you are on a finals aliance where you know your partners, you should definetly have an off switch for your autonomous program, so you dont risk interfering with your aliance members robots.