Bioloid I/O interface

[pepperm]

Hi jaho434

I've looked at Limor's code and it looks like is pretty sound. I don't think it is compileable yet though and there isn't anything really executable to test in it. I believe Limor is working on it now.

Mark

[jaho434]

Cool.

Nathan has given me a whole lot of his AVR code that he has used before. There's a lot of library functions that could be useful.

I'll have a closer look at them over the weekend and post the useful stuff.

My Bioloid should be here by the weekend! w00t!

EDIT: Anywhere on here where I can upload the files to?

EDIT AGAIN: Right, found the Manage File Uploads link

Nathan's code can be found here

It has nothing to do with this current project, but should provide some insight or useful code for us. He has given me a brief explanation of some of it, but I will go through it in more detail on the weekend.

Limor: Aparrently ASCOR.c is a good place to start. You will see that an interrupt is generated any time a signal is received on the serial line and then it is processed. I've only had a quick look at it, so if you can make more of it, you're doing better than me!

Hope it helps.


Julian

[limor]

hi juho434,

thanks! Briefly looked at the code and there are things there that can be useful for the Atmega8 sensor board project. I will be looking out on the web for some objective-driven high-level libraries hopefully in c++.. that are MCU agnostic. something that will help my code look like this:

Code: Select all

RS485 bus(PF4, PF5, PE6, PE7); // 4-wire SUART constructor
ADC a2d(PE1); // A2D constructor

s.init(_1MBPS); // set serail lines at 1mbps
while(1) {
  p = s.wait_for_master_to_call_me(175); // 175 - my ID
  p.val1 = a2d.get(); // get a single A/D value
  s.send_packet_to_master(p);
}


I'm waiting for peperm's board and will spend the weekend making it do useful things.

[jaho434]

The ADC.c file in the include directory of that code has some A/D conversion functions. Again, I haven't had a good look at it...... will do so this weekend.

Maybe Nathan could give us a brief description of the useful code in there?

[limor]

I received pepperm's magic sensor board today.
This must be the shortest hardware/software development cycle in history.

i'll be coding away this weekend.
Thanks!

[pepperm]

Limor

Excellent you got the board. Sorry the connectors are a bit make shift but I can't find the right ones in any of the shops nearby.

Keep me updated as to the coding.

Mark

[Pev]

Blimey this is impressive. Hats off to you Guru types its well beyond me.

Pev

[rebuildit]

I was wondering how this project has progressed so far? I havebeen reading and searching and I would love to be able to add sensors to the Bioloid.

By the way this is one great site!

[pepperm]

Hi

Thanks for asking rebuildit. Currently, we are trying to get the processors code to work. I believe Limor is developing it in "C" but if I get any time I will give it a go in BASCOM-AVR.

pepperm

[rebuildit]

Thanks pepperm.

Me and my sons will be getting a bioloid for christmas. I am looking forward to what this hoby can and will bring!

[Joe]

Just putting in my $0.02 of encouragement, as I too would very much like to see this. I don't even have a Bioloid yet, but I sure want to get one, and I want to see Bioloid succeed... things like this will certainly help.

Go, pepprm! Go, Limor! Go go go!

Best,
- Joe

[limor]

Finally here's some code that works as expected on Pepperm's board.

All it does is expect a packet from the CM5 and return a packet.
It also changes the led status when this happens.
It ignores packets that are not destined for him (arbitrary number 69 was elected).

Next step is to connect an analog gyro or accelerometer to the board and respond to something like a "READ 1 1" command.

If you want to have a look at the code for the Atmega8, here it is:
http://robosavvy.com/Builders/limor/Pep ... -works.zip

[pepperm]

Fantastic Limor. Well done.

Mark

[pepperm]

Just a thought. Instead of designing a PCB for the production of the Bioloid I/O, how about using a board that is already in production. So far I have fould 2 that could be suitable.

One is the megabitty Mega8 board info here:
http://www.junun.org/MarkIII/Info.jsp?item=57

and the other is the Open Servo board V1.1 info here:
http://www.openservo.com/moin.cgi/Hardware

We would maybe need a daughter board for the 1 wire interface but other than that these boards have things like LEDs, lots of I/O and motor drivers available. I think the megabitty board is best. They are pretty cheap too and have already been developed and manufactured. They have bootloaders too and use the Mega8 processor that we are already developing for.

What are everyones thoughts?

Mark

[Pev]

Mark,

It would certainly simplify things if we only had to worry about the daughter board.

I too like the megabitty board, looks nice and compact and the daughter board could be made to the same form factor and stacked with the processor board.

Pev