[Robotgroup] Divide and Conquer!

[Acy Stapp]

If you want to go crazy with distributed control you could reverse engineer
the AX-12+ firmware and replace it with your own. Each has an Atmega8 with
8K flash and 1K RAM. That could be a whole project in and of itself before
you get to actually working on your robot :)

The AX-12+ TTL bus is probably not set up to handle multiple masters well,
as the AX-12s are designed to be slaves. If you went this route you'd
probably want to switch to some protocol that supports multiple bus masters.


If anyone has questions about the AX-12+ servos or other Robotis/Dynamixel
gear I'm becoming very familiar with it. Ask away. They are pretty easy to
interface.

On 9/13/07, Gray Mack <gray_mack at yahoo.com> wrote:
>
> From my perspective, other than fewer wires and more
> modular design I don't know if it buys you much
> anymore but its a neat concept.
> Processors like the propeller can allow you to run 8
> tasks in one chip. It can also control up to 32
> servos. (Now-days servo controllers can cost less than
> one servo).
> The AX-12 servos have lots of logic and self
> protection built in and chain together on a 3 wire
> power-ground-data network but at $45 each that add$ up
> quick for hexapods. But they are likely easier to
> drive from one central processor.
>
> Whether you are time-slicing tasks or running on
> separate hardware, smoothly accomplishing movement,
> sensing, etc is still going to take lots of software
> to get it done.
> The cost of a bunch of PIC MCUs and support components
> and circuit boards may end up being more than one
> hefty enough processor and a servo controller and you
> will need a solid inter-ic-communication link before
> you get very far with multi-MCUs. Be wary of brownouts
> and motor noise.
>
> The software can be modular in design whether its all
> running on one processor or split out to many.
>
> So maybe its more about whats neatest to you and what
> gets your development going faster, but I would like
> to hear other opinions on this.
>
> -Gray
>
>
-- 
Acy Stapp

"When I'm working on a problem, I never think about beauty. I think only how
to solve the problem. But when I have finished, if the solution is not
beautiful, I know it is wrong." -- R. Buckminster Fuller (1895 - 1983)