New robot controller next year?

[Dave Flowerday]

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.)

I know Mike already responded to this point, but I want to emphasize something: what we did with our robot was NOT done entirely with the built-in processor. It wouldn't have been possible. We had to add our own external microprocessor to do it. Think about that.

Honestly, a better processor will help out other teams much more than it will help us. Most of our team's mentors are software/computer engineers. That makes us very well equipped to deal with limited computing resources of the current system - like being able to add our own external processor this year. For a lot of teams, that's not a realistic option, which means those teams will gain much more freedom with a new processor than we will.

[Matt Leese]

Originally posted by 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.

The current Basic Stamp based microcontroller fully supports signed arithmetic (well, at least for addition subtraction; I believe it also works for multiplication and division). Two's-compliment numbers work perfectly (the real question would be why they wouldn't). However, the only thing that doesn't support negative numbers are the input and output devices. So this is probably much more a problem of the additional hardware than the Basic Stamp directly.

The real stupidity with the Basic Stamp math is its lack of order-of-operations.

Matt

[Adam Y.]

The real stupidity with the Basic Stamp math is its lack of order-of-operations.

Actually I think most processors handle math that way from left to right.

[Sachiel7]

Ladies and Gentleman, I have an announcement to make!
Upon emailing IFI asking for some information on the new control system, I received this message in response.

I think it should answer several questions:

----- Original Message -----
From: Tom Watson
To: sachiel7@comcast.net
Sent: Monday, September 15, 2003 4:22 PM
Subject: Re: New System



Sachiel7(I changed my name),

We will have updates on our website beginning on 01 October. Until then we can not give out any details.


Innovation First

So, We'll know in 16 days the specs of the new system.
I think I'll make a countdown timer for it.

Also, I have received word from a few sources that claim that the new language will be C based, and from word coming from other people, the pointers are highly suggesting this. I'm just glad I got C under my belt this summer .
A note to wildstang - I wasn't trying to be mean or anything I just wanted to note that even though the system didn't live up to your expectations, it got you the national trophy, with some help. Plus, I was a bit confused as to why you were using trig on a processor like this (???) From the matches I saw your bot had a skid/crab drive with the 2 "flaps" and a pretty good solid auto. What components did you have that required that level of calculation? Just curious. Also, yes my main source for claiming that IFI was made for FIRST was the remarks at the Awards Ceremony that Dean/Woody made. I saw it live, and I remembered reading something about that somewhere.
Anywho, I'm looking forward to this new system. Hopefully it will be a bit easier to use. I just through though, all the creations made for the STAMP, like the load of extra tools teams have developed, is all going down the drain.
My main suspicion that first is making the jump also is with the predicted 1500 teams, which is the largest jump to date. Could this also mean the field size/shape might finally be changed?
I guess we'll find out.

[Rickertsen2]

YES!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Fianlly!!!!!!!! I really can't wait. Its like christmas or something.

[Sachiel7]

Christmas for some...
The Apocalypse for Others.
(Not us )

[Matt Leese]

Originally posted by Adam Y.
Actually I think most processors handle math that way from left to right.

Most processors only do one operation at a time. Given that the Basic Stamp is programmed in PBasic, this means that you program it such that it does multiple operations at once. Because of this, it can't do order of operations. Most programming languages will execute things in the correct order. The Basic Stamp will not.

Matt

[Ian W.]

Originally posted by Rickertsen2
YES!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Fianlly!!!!!!!! I really can't wait. Its like christmas or something.

What about chanukah? Or do I have to get Adam Sandler in here to make me feel happy?

[Rickertsen2]

lol u know what i mean. No hard feelings.

[Dave Scheck]

Originally posted by Sachiel7
A note to wildstang -Plus, I was a bit confused as to why you were using trig on a processor like this (???) From the matches I saw your bot had a skid/crab drive with the 2 "flaps" and a pretty good solid auto. What components did you have that required that level of calculation? Just curious.

Our autonomous system, StangPS, required the use of trig. We used the offboard processor to calculate our current position on the field at all times. The communication between the RC and the processor was at a minimum due to the lag we experienced. Every loop we passed the x,y, and angle to the field (theta) across. The RC then used that information along with the point on the field it knew it wanted to get to in order to calculate the direction to drive. That's where the trig comes in. Take a look at the presentation that we put together at http://www2.wildstang.org/2003/video/StangPS

As Dave and Mike have mentioned, we really pushed the stamp to its limits. By upgrading the processor, other teams may have the opportunity to attempt something similar completely within the RC.

[Mike Soukup]

Originally posted by Sachiel7
A note to wildstang - I wasn't trying to be mean or anything I just wanted to note that even though the system didn't live up to your expectations, it got you the national trophy, with some help. Plus, I was a bit confused as to why you were using trig on a processor like this (???) From the matches I saw your bot had a skid/crab drive with the 2 "flaps" and a pretty good solid auto. What components did you have that required that level of calculation?

It's quite alright, your original post probably came across the wrong way. The problem isn't that the controller didn't live up to our expectations - we've been dealing with it for a few years so we knew what to expect - the problem is that FIRST is asking us to do more with it than ever before. The controller came out when the most complex programs teams were writing were simple feedback loops for controlling arms & crab wheels, now teams are asked to write programs that drive the entire robot without operator control. Fortunately FIRST & IFI realized that in order for the competition to progress beyond simple autonomous programs, they need to give us a better controller.

We used trig functions (actually an arctan lookup table) to find the angle that our robot needed to travel in order to reach its destination. We knew the robot's coordinates (got that from the off board processor) & the target coordinates so we simply took ((delta y) / (delta x)) and used our arctan subroutine to get the angle of travel relative to the field. Then we had to take into account our robot's angle relative to the field and get the angle to point the crabs, but that was just a subtraction. After that we had to adjust one side of our crab wheels in order to correct for any skew in the robot's orientation (what we called theta correction). Then that data was passed to our normal crab control subroutines to actually turn the wheels to the proper angle. I almost forgot that we also passed data to the wing control subroutines so we could control the wings in autonomous mode. All of that was done each loop in autonomous mode.

Thanks for the info about the status of the new controller. You may want to edit out your email address from the post if you don't want it public.

And on a side note, I sure hope IFI builds a lot of extra controllers since it's common for teams to build a second robot for use after the build deadline. Either that or FIRST lets us keep our controller instead of shipping it with the robot.

Mike

[Sachiel7]

Hey thanks for letting me know more about what you did last year.
I tried to make something like wilddraw this past year but ran out of Time. I'm working on a newer one now, and I'm having it generate rough code in C, in case that is the language we use.
My drive system I designed will function as Crab\Skid\Car and a few others, so we may agree on using it if there aren't any obvious issue with it being used in the game after it is revealed. That would really cut down out crunch.
My email is available to anyone who wants it on chiefdelphi, so feel free to give me a ring.

[Wetzel]

From a FIRST email blast.
"The rumors are true; a new control system is coming. The new RC has a user programmable Microchip PIC" controller operating at 10 MIPS, gives direct access to I/O, has interrupts, and is programmable in C. Please do not contact Innovation First requesting information as more details will be available at www.InnovationFirst.com in the next few weeks when the product ships.

Last year all teams received the FIRST EDUrobotics kit to assist in preparation for the FIRST Robotics Competition season. Based on the overwhelming response and positive feedback from teams for survey results, FIRST will again offer the FIRST EDUrobotics Kit to the incoming rookie and returning teams that did not compete last year. After these teams register for an event, and upon receipt of a $500 deposit at FIRST, they will receive the kit that will include the new Innovation First Control System. All deposits are applied toward the $5,000 registration fee; however, once product is shipped the deposit is non-refundable. In order to allow everyone to build familiarity with the system, all teams that received the FIRST EDUrobotics kit last year will receive the new control system when they register. On-line instructional materials will be available on our website at www.usfirst.org. Both items should start shipping approximately October 15, so register early.

In order to provide the new robot controllers and other new parts in the kit this coming year and stay within budget constraints, it is necessary for FIRST to make some changes to the kit of parts. Battery chargers, joysticks and compressors will be supplied to teams in their first and second year. Veteran teams should have two sets of these parts and can use them on their 2004 robot. We hope you find some of the new parts exciting and challenging."


Whos excited?

Wetzel

[Mike Yan]

It was stated the the new micro was a PIC operating at 10MIPS.
Does anyone know what the clock speed is?

PICs run at a quarter of whatever clock speed it gets, so I was just wondering.

[FotoPlasma]

Originally posted by Mike Yan
It was stated the the new micro was a PIC operating at 10MIPS.
Does anyone know what the clock speed is?

PICs run at a quarter of whatever clock speed it gets, so I was just wondering.

Mike, you just answered your own question. If Fcyc is Fosc/4, and Fcyc is 10M, Fosc is 40MHz.