New robot controller next year?

[MichalSkiba]

If FIRST wants to move towards either a longer autonomous period or splice it into a different section of the heat were dead-reckonning would be impractical, then a better processor would definately be needed. If a mock GPS system, multiple real-time IR, gyro,... sensors, and intra-alliance communication during the autonomous mode is a realistic goal for either this comming year or the one following, then I'd suggest using an ARM Thumb microcontroller, and programming in C.

The ARM Thumb uC has more then enough processing power and speed to comprihand the demands of keen programmers for years to come. Its core has DSP features such as barrel shifters and single cycle multiplication, to speed up signal interpretation. It was a plethora of General Purpose IOs and on-board peripherals that would allow the designs to be expandable (allowing for the addition of an NTSC/H.263 codec for feeding video streams).

C is fast, powerful and relatively easy to use. Additionaly, theres a better chance that a software mentor would be able to help out with C then with pBASIC. Teams won't have to program low-level hardware settings like registers if a kernel (and RTOS?) is provided by FIRST (most likely burned into a ROM). The kernel could be something like uCLinux for example. The kernel could include a block that would give FIRST admin privliages [FIRST = root], such as the ability to cut power to the robot (like at the end of a heat). Headers with libraries and various global functions could be provided by FIRST, or made public by various teams, to further ease programming.

I strongly beilive that there is enough time within the 6 week timeframe to learn fundamental C and to program the robot. If you already know one programming language, such as pBASIC for example, switching to another would definately be easier. Additionally, even more so then linux, there are huge talent pools and reasources avalible on the internet and in books, incase anyone has any questions about programming in C; far more then for pBASIC.

So what about teams that don't have enough people to dedicate to programming? Well, too bad. Autonomous was a critical element of last year's game, and I don't see a reason for why its importance should change this year. Therefore, select a person to do the programming.

Most of all, FIRST provides the oppertunity and environment for students to learn and excel. This should be extended to programming (and not limited by crappy hardware).


That's my $0.02 PLN


*S*K*I*B*A*

[Rickertsen2]

That would be great. there really really isn't even a need for a rtos if they stick with the current approach of 2 processors. (a user configurable and a hard coded one). But then again, from a hardware standpiont, an rtos would be more economical because it involves less components. I really have the feeling though, that all we will see is like a change to a jstamp, BasicX or something else like that.



-------------------------------------------------------------------
pretend this is the start of a new post.

Originally posted by FreeBSDboy
Some food for thought.

I have programmed autonomous robots on motorola HC processors for some of my classes when I was a junior in college. It is not a trivial step up from pbasic.

It's not so much learning the C language that would give the most problems, but rather I'd say it would be the seting up of and writing to registers and using interrupts that would be the most challenging aspect. There are a lot of difficult things that the basic stamps / pbasic make very easy to acomplish.

An example would be PWM output on the HC series. You need to have a good understanding of the real time clock and how real time interrupts work in order to get it to function properly. And it is much easier to accidently destroy your motor controlers and servos by screwing up the PWM programming- to the extent that it would be handy to have use of an oscilliscope for diagnostic purposes.

While I am sure that many of the veterans and the more tech savy teams would greatly appriciate such added control over the control system (I know I wouldn't mind it), you have to keep in mind that it also has to be simple enough for rookie teams and beginners to be able to pick up reasonably quickly, which pbasic allows.

So just some food for thought. Not everyone would be appriciative of a control system with a large learning curve. I'm confident that First and IFI will take all this into account if/when they decide to tackle the issue. And I would hypothesize that any changes to the control system would take the form of expanded capabilities without loosing sight of these issues.

Just my perspective. I hope I could shed some light on the topic.

[edited for minor gramatical errors - whoops!]

Thats why they offload the PWM and other duties such as enable/disable to another processor, so your code is runnign in a "sandbox". Also i found the lack of interrupts to be a limitation rather that a good thing. As far a setting up and writing to registers, FIRST could provide example code an libraries to do all the grunt work for the unexperienced teams.

[Adam Y.]

and keep in mind, though not everyone knows C, NOBODY knows PBASIC

Actually PBASIC is easy enough to learn on your own and that it is fairly easy enough to understand without much programming background.

[Sachiel7]

Ok, a few more thoughts here...
First off, to all the teams complaining that the Basic stamp stinks, isn't fast enough, doesn't give you a huge range of options, then let me say this: duh.
Do you honestly expect FIRST to supply teams with a fast system with plenty of memory and a typical programming language?
No! It is part of the FIRST challenge that you work with things that don't always do exactly what you want, when you want them to. When you get a job working as an engineer, do you think that every project you work on will have no problems, and be performed in an ideal environment? That you will have the latest, biggest, and baddest gizmos at your disposal???
Tell me if you find such a career, because I certainly haven't found one this way yet.
FIRST will supply a slower processor with limited memory and a language that is not-too-typical as a challenge. You have to learn the hardware, the language, and how to work with it. You should design your robot to function well with it, not rely on FIRST to give you the best.
(Also, a note to Wildstang, you complain about the speed of the processor, yet even with a highly significant delay, you still managed to take Nats. Think about that.)
Also, note that (as I stated before) IFI was created for FIRST. And they more than likely aren't going to change their entire control system for a new processor and memory. It is known that FIRST will include auto mode for at least 2 more years. IFI reps around the regionals have confirmed this several times. Why, when they just developed a system to function autonomously, with the multi-slot programming, would they just scrap all that?
It is my belief that IFI will stick to their control system for at least 2-3 more years before even considering a change.
When I previously posted about C, I meant that C and Basic are the main two different choices of languages out there. Since the system is currently basic-based, it would be asking alot to shift to C. That would likely require the change of processors for more compatability, and as I just said, I doubt that will happen.
One clue we may receive will lie within the eduBot, if we receive them this year, and I believe from all the positive feedback that we probably will receive them again.
If FIRST is going to switch languages/processors, then they will perform this switch on the eduBot. Note that the eduBot documents on IFI's site have been updated since competition, and yes, they still use the same control system as before.

[dez250]

i am not going to go out on a screaming and yelling tangent here, and i will make this short so everyone will read this.

IFI was not created with FIRST, or in any way completely for FIRST. They were a control system that took over when the Motorola controllers were discontinued. IFI was around before they were involved with FIRST, and if you have taken a look much at their site you would might figure that out. Also FIRST has just helped them as a company succeed, and grow more. I would suggest some of you to take a look at what you type here and what you say, as this thread may be the rumor mill, but it still reflects upon your team and yourself what you say, and this board will leave lasting impressions on many if not all of us!

~Mike

[FotoPlasma]

Originally posted by dez250
. . . IFI was not created with FIRST, or in any way completely for FIRST. They were a control system that took over when the Motorola controllers were discontinued. IFI was around before they were involved with FIRST, and if you have taken a look much at their site you would might figure that out. . . .

~Mike

I seem to recall that, during the awards ceremony (which ran concurrently with the final matches) at Championships, in Houston, 2003, Woodie Flowers (possibly someone else) specifically stated that Innovation First was formed by professionals who were involved in FIRST, at one point, and decided to split off, and work mainly on providing FIRST with control system equipment.

If I'm wrong, please correct me, as soon as possible.

[Ricky Q.]

Once again Jim, you are correcto, as soon as I saw this post I pulled up the video of the Founder's Award off SOAP's website, http://www.soap108.com/2003/movies/cmp/ , and watched it, and low and behold its Dean Kamen saying

This company started as a couple of guys on a team in the second year of this competition, and one day they decided it was better to do FIRST everyday than just once in a while, so they quit their jobs and decided to start helping us. They started helping us build better motor controls.....their new company grew as we grew, they started building complete systems, to run robots, fields, power, and then they decided they'd help us with EDU robots...

And so on, so IFI did start out as some guys on a FIRST team.

[Mike Soukup]

Originally posted by dez250
IFI was not created with FIRST, or in any way completely for FIRST. They were a control system that took over when the Motorola controllers were discontinued. IFI was around before they were involved with FIRST, and if you have taken a look much at their site you would might figure that out. Also FIRST has just helped them as a company succeed, and grow more. I would suggest some of you to take a look at what you type here and what you say, as this thread may be the rumor mill, but it still reflects upon your team and yourself what you say, and this board will leave lasting impressions on many if not all of us!

You're right, what you type here reflects on you & your team, so enough with the know-it-all, condescending attitude. Even if you were right about the origins of IFI (and you aren't, it was founded by some engineers from FIRST teams as Jim & Ricky pointed out), don't assume that others haven't looked at their website or done some research into the origins of IFI.

Originally posted by Sachiel7
(Also, a note to Wildstang, you complain about the speed of the processor, yet even with a highly significant delay, you still managed to take Nats. Think about that.)

Us complaining about the speed of the processor & us winning nats are really independent. We were surprised that we got as much mileage out of the control system as we did and we enjoyed the challenge. We were pushing the limits of the current controller. We were dangerously close to the upper bound on program loop length, a few hundred or so more lines of code (I didn't actually calculate how close we were) and we wouldn't have executed our loop fast enough and the RC would have reset. Sure we could have offloaded more processing to another off board controller, but we had electrostatic problems with our existing processor & didn't need more problems. Besides, it's almost unrealistic to require teams without mentors with CompE & CS backgrounds to use an off board processor if they want to do anything exciting with autonomous mode. We've already heard some grumbling from teams who think it's unfair to use an off board processor just because their teams don't have experienced mentors. If FIRST wants a majority of teams to progress beyond dead reckoning in autonomous mode they have to give us a processor with more power. Many teams in 2003 could have come up with awesome autonomous programs, but they were limited by the speed & memory constraints of the current controller so they stuck with dead or rotation counting. And if FIRST wants the more experienced teams to continue to push the envelope of what's possible in autonomous mode they'd better give us a controller that's capable of accomplishing more.

Originally posted by Sachiel7
Why, when they just developed a system to function autonomously, with the multi-slot programming, would they just scrap all that?

This isn't true. The 2003 controller was pretty much the same as all previous controllers. The old ones had multi-slot programming, we actually used 4 slots in 2002 (and 6 in 2003). And when we loaded our 2003 code onto previous year's RCs - I believe we used the 2001 controller on our prototype bot - we didn't have any problems, even in autonomous mode.

Mike

[Doug Leppard]

This was my FIRST year with FIRST. But I have been doing autonomous computers as a hobby for years.

The FIRST controller was in adequate not just because it was slow and had no memory to speak of, but it couldn't process interrupts, or look at an IO port more than once a cycle which means you did not have direct connection to the ports. It was ok if it was just an remote controller under radio command. But if you wanted to do auto mode a new controller is needed.

We did a modified external computer to do simple wheel encoders, which gave us better control of the robot in autonomous mode.

The kids I worked with had no problem understanding the concepts and implementing them.

What I understand is that FIRST is coming out with a controller soon and it will be radical change from previous years. The only way to do auto mode well is that you need a different controller. So I do hope that the controller will be released soon so we can train our kids now.

[dez250]

i was not trying to sound like a know-it-all... i was trying to convey a point to you all... your all running wild with things you may say and things you may here, you all are just going around saying what you may like to say, and what i did was try to show a point to you all... The last post i said in this thread was to see how quick some of you may jump at someone else's throat, and to try and show you that maybe for one time you should step back, take a rest and read all of what some others have to say before you put your 2 cents in. i see mike read through my whole post, and the start of it all with IFI was a b/s to see how many of you would just stop reading in the middle. Mike from 111 read through and what he pointed out is what i wish everyone could have seen, and not the inaccurate points of my post. Mike, the Student he is to FIRST, and the mentor he is to Wildstang, read through and showed that what you post on here comes to bight you back, and reflects upon your team and/or any organizations you may be involved with. This may be the rumor mill, but there is still a genuine state of respect in here, please read what you post, before you say it. For sometime or another you will meet someone from here at a competition or in real everyday life and see that this is more then computer to computer, but face to face.

~Dez

[FotoPlasma]

Originally posted by dez250
....

~Dez

Uh. I didn't jump down your throat. I corrected a technical inaccuracy. If you have any kind of valid complaint which you want to talk about, IM me.

<edit>
Heh. By the way, nice little disclaimer in your sig.
</edit>

[Ian W.]

Originally posted by dez250
i am not going to go out on a screaming and yelling tangent here, and i will make this short so everyone will read this.

Originally posted by dez250
....

~Dez

That's too long, I don't want to read it...

To be completely serious though, You can't expect to post something that has obviously incorrect info here and be surprised when everyone tries to "jump down your throat", if that's what you want to call it.

Just let this drop, because this is the stupidest thing I've seen in a while. These forums have many people, like Jim and Ricky who will always correct you, just because that's what they do. I like when the correct me, because it's either because I was stupid and didn't look it up, or I learned something new. If you don't like that people are correcting you, either look stuff up so everything posted is true, don't post fake information, or just leave. No one is making you stay here, and you're obviously sad because people are "jumping down your throat", but I can't help you, and I'm not sure that anyone else can.

Now, back on topic, about those controllers for next year. Anyone know when we're being told what they look like or what language they'll run?

<EDIT>
Typo.
</EDIT>

[Sachiel7]

All I'm saying is this:

-The division of IFI that produces the First components is heavily involved with FIRST.

-I agree that we need a system upgrade for more auto-heavy tasks. But my point was that it is part of the Challenge to use what is given to you, and if your complaining that its not good enough, you're not rising to the challenge.

-IFI has confirmed an entirely new control system for 2004. THIS IS NOT A RUMOR. I and at least 4 local teams have contacted IFI directly and each time they have confirmed that we are receiving an entirely new system next year. My prediction? I'm guessing they're going to use a BasicX. They'll probably stick with a BASIC based system, or something simple because every single (predicted 1000) team will have less than 2 months to learn this new language. Many of you big teams out there say "scoff, we have 20 programmers, and at least one of them knows C, or Java, or something or other" and you don't take into consideration that there could be well over 500 other teams out there that have no clue how to code in C or Java and would have a big issue on their hands with this type of shift.
I'm also curious as to whether or not IFI will change what type of PWM/Relay we use. I'm hoping that the Max amperage will be increased again, and that we may receive an even beefier drive motor than the CIM next year. Possibly a larger battery. I don't know, but I just emailed IFI asking what they could tell me, if anything, about the new system. FIRST claimed that information would be posted about this a while back, but I haven't seen a change yet.
This also raises quite a large bump in our workshop schedule. A major change in the system (electronics included) would ruin the fact that we have been planning our control workshops to take place only weeks away from now.

[TechnocratiK]

The RCs we're using could definitely use improvement One of the biggest faults of these microcontrollers is their inability to properly calculate with negative numbers. Sure, floating-point values would be a nice frill, but when working with a gyro, it'd be nice for the value of 1 / -1 not to equal 0.
Secondly, a new programming language would definitely be nice. Personally, I favour C++, mainly because it provides a good balance between complexity and efficiency. Nevertheless, I'd be tolerant of practically any language, as long as it allowed for user-defined functions.
Finally, more variables is a must. Robot controller programs aren't that big if you use some tricks, but being limited to 26 unsigned variables is torture.
As for the OIs, I really don't forsee FIRST using a Playstation controller plug, mainly because of the limitations of the device. Sure, digital inputs all around would be nice, but PSX controllers aren't the solution, not when you need a 3 way switch or the likes.

[Rickertsen2]

[i]Why, when they just developed a system to function autonomously, with the multi-slot programming, would they just scrap all that?
[/b]

IFI didn't actually develope the multi-slot thing in 2003 or ever. That was a result of Basic Stamp software updates, which have nothing to do with IFI. As far as auto mode goes, all they did was make a minor update to the firmware on the RCs, and some minor hardware and firmware changes to the field controllers.