Now that the 2004 season is comfortably past, I would like to bring up the new Robot Controller. It was one of the biggest changes from past years, and it was meant to be an improvement over the legacy system.
Last year, many teams which chose to operate autonomously had to create all sorts of kludges to achieve real-time or near real-time signal processing. People experimented with auxiliary processors, large custom circuits, etc. All of this was to overcome the limitations of the 2003 Issac32 RC. This year, almost all the electronics “firepower” was contained within the 2004 RC. The PIC is far more powerful than the BS2sx was, and I didn’t hear very much about custom circuitry this year. I only noticed a lot more software magic.
One thing I did note was the overall performance of robots this year versus last. There were no major differences that I heard about. The new controller didn’t seem to have enabled all new and wonderous autonomous systems or feedback loops. All I saw was last year’s abilities (sometimes less than that) repeated.
To all the programmers and electricians out there, do you think that the new controller was worth it?
I definitely liked programming in C a lot more than in PBASIC, so for that it was worth it.
But, as with any new technology, it needs some time to mature. Expect a few years of not-so-impressive robot programming (well, I’m hoping for just a year, which has already passed, but we’ll see), and then there will be a huge amount of improvements to robot programming, as teams finally understand how to work with the controls, and programmers are working with a new control system from the beginning (unlike learning the Issac32 and then relearning the 2004 RC).
I actually have talked to a current 6th grader, in an attempt to get him to learn some more C to start programming for the team next year, it’s people like that who will come up with the new and amazing things, because they’re in a different mindset. We’re all still somewhat used to what can’t be done I think, so it’s a bit hard for us to jump in and make amazing things, but not when you’re brought up on this new system.
The only thing I dread is IFI changing the system drastically in another few years. The only thing that should change is the addition of a USB --> Serial adapter on the RC itself, and that should only be for ease of use. See Elgin’s thread for more on that though.
So, bottom line, yes, the change was worth it, just give it some time, and we’ll see some very cool things (or so I hope :-)).
Im not a programmer and I do some electrical work but mostly mechanical/pnuematics and design. I do know that our programmer was ecstatic that he was programming in C. One thing he didnt like so much about it though was that he always had to make the project file then compile yatta yatta yatta. I also know though that he just finished with his second generation of this years control system and is making steps to a third generation. Other things that I saw from a drivers standpoint was that when we were low on power the bot didnt resetinsert cheering noice here. So from my point of view two thumbs up.
We’ve got something in the works, that will utilize a significant portion of the RCs processing power, and this is without any interrupts as all the encoder couting will be offloaded to custom circuits (the code is pretty efficient, too). I think Ian’s prophecy will be fulfilled within a year for sure.
I think the PICs are a great improvement from the BS2sx. The sheer speed alone makes them much more versatile as a robot controller, nevermind the flexibility of interrupts. I was born a BASIC programmer though, wrote my first BASIC programs when I was 3, so I could write essentially anything in PBASIC for the robot. Unfortunately, I never did much of anything with C, so this year was the first time I’ve ever seen/used it. From a novice programmer’s perspective, I think IFI did a pretty good job as far as publishing information and tutorials on programming. We did get about the same functionality out of this year’s controller as we did last year, but now that I have a feeling for C, I can certainly see it’s great advantages for the future, and can appreciate the amount of power the new controllers have. This year made a great transitional year for experienced teams, and the wealth of knowledge between here and Innovation First will certainly help rookie teams. I do expect great things from this controller in the future, from PID loops to much more complex autonomous modes.
I know I’ll be spending most of the summer playing with the edu-bot, trying out various snippets of code, and really getting a feel for what these things are capable of. I can’t wait to see what everyone collectively can make these things do next season!
It’s actually impressive that there were “no major differences”. With a change as big as this it could very easily have caused teams major problems. I agree with others that have posted, the really great things are yet to come. Everyone used this year to get back to where they were (us included) and maybe try a few new things. We had all kinds of grand plans but when it came down to it the added risk of trying something too new this year with all the other changes in the controller was too great. We will be working in the off season to perfect some of these ideas and systems.
The new controller did put the playing field a little off kilter. Rookies could come in with more prog expieriance than some senior teams. I wouldn’t say level though.
