PS2 Controller on an Axon tutorial.

[dunk]

hey guys,
i finally got round to playing with the shinny Axon Admin sent me.

i found the whole process of getting it working to be fairly painless.
anyone with some microcontroller experience should get a "hello world" working within an evening.
the board is well designed and practically laid out. i'm afraid i don't have anything to compare it to though as i've always made my own development boards.

as a test project i set about interfacing a Sony PS2 controller and displaying the output to my laptop.

could someone please take a read through the following and see if it makes sense?
http://www.societyofrobots.com/member_tutorials/node/200


thanks,

dunk.

[airman00]

any video demo?

[Admin]

Wow that was quick!

Ok three requests:
1) video demo!
2) a good pic of the Axon with a PS2 controller to use as a thumbnail on the Axon examples page
3) in the code:

Code:

  // main program loop:

change to:

Code:

  // main program loop
  // ADD YOUR CODE HERE
  // use data0, data1, data2, etc. after rprint as your control input variables

Also, out of curiousity, why does communication need to open and close in that loop and not just remain open?

Perhaps the PS2 controller initialization should be made as a separate function (like gameByte()) so that it can be called? I'm just thinking of ways it can be modded so that it can be easily hacked into someone else's project. I understood it on first glance, but others new to robotics might have to think hard

[dunk]

there.
added picky, YouTube clip and the suggested readability improvements to the code.

Also, out of curiousity, why does communication need to open and close in that loop and not just remain open?

that's just the way the PS2 packet works. the packet will start when the Attention line goes low and finishes when it goes high (whether all requested data has been sent or not).

or do you mean "why are there multiple calls to the function gameByte ?"?
gameByte is called once per byte. one byte is sent to the controller and one bye is received from it.
after the 3 byte header gameByte is called once more per data byte.

Perhaps the PS2 controller initialization should be made as a separate function (like gameByte()) so that it can be called?

yea. i was thinking that myself but could not be bothered rewriting it just now. maybe in the future.
if anyone else feels the need to adopt this project then feel free to make it more user friendly.


dunk.

[arixrobotics]

Hi Dunk,

I noticed you said in the video that the firmware only works with original Sony PS controllers.

I've done a similar project before, interfacing PS controller to a PIC. But what I got was opposite of yours. Mine worked great with cheap China made controllers, but not as good with the original Sony ones. All the digital buttons work ok, but not the analog joysticks. They didn't give out the expected values for the joysticks.

Any idea why this could have happened?


Arif

[dunk]

I noticed you said in the video that the firmware only works with original Sony PS controllers.

yea. i'm fairly sure it could be made to work with both with a bit of timing adjustment.
the Axon code is very similar to my original code for the ATmega8 at 8MHz which works with both original Sony controllers and cheapo imports. http://mrdunk.googlepages.com/ps2controlleronanavr
the only real difference is the timing. the cheapo controller will probably need additional delays inserted somewhere at 16MHz.

if anyone gets a non Sony controller working at 16MHz let me know the exact delays involved and i'll adjust the tutorial.

All the digital buttons work ok, but not the analog joysticks. They didn't give out the expected values for the joysticks.

sounds like you are not sending the right control codes to put the controller in analogue mode.


dunk.

[awally88]

Good job Dunk,

I was looking at interfacing a PS Controller on the Axon after I finish working the Wii nunchuck but you've beaten me too it! After I finish on the Wii nunchuck I think I'm gonna go old-school and get some Super Nintendo controllers working for robots.

Keep up the good work
A

[SmAsH]

this was done a year ago and was resurrected by a spammer