[ros-users] call for testing: camera1394 in C-turtle

[Andrew Harris]

Hi -

   So after some more tinkering, the problems I have been having with
my Point Grey Chameleon on the beagleboard xm don't have anything to
do with the camera.  The problems seem to be the beagleboard xm and
it's USB interface.  I am getting messages like:

"[11646.000000] usb 1-2: device descriptor read/64, error -71"

   which I thought were due to the camera, but I have plugged in
another device and I get the same messages.  It seems like keyboards
and mice and other simple things work fine, but "more complex" things
for some value of complex seem not to work on the beagleboard xm.

   The problem is somewhat intermittent -- I've got at least two
bagfiles that show camera1394 working OK on the beagleboard xm with
the Point Grey Chameleon doing IIDC over USB.  But I am certainly not
going to use this set up for an autonomous robot at this point!

regards,
-andrew

On Tue, Nov 23, 2010 at 1:28 AM, Andrew Harris <andrew.unit at gmail.com> wrote:
> On Mon, Nov 22, 2010 at 11:31 PM, Eric Perko <wisesage5001 at gmail.com> wrote:
>> There is a wiki page that talks about some of the alignment issues that have
>> come up with ROS on ARM.
>> http://www.ros.org/wiki/eros/alignment (take a look at the linked ticket as
>> well). Also take a look at this
>> thread: http://code.ros.org/lurker/thread/20101122.232925.f0d885ce.en.html and
>> perhaps try the -msoft-float option and see if that changes anything.
>
> Hi Eric,
>
>   Thanks for the suggestion.  I applied Josh's patch, and that seems
> to have fixed the problem within dynamic_reconfigure.  I think
> camera1394 might be working now -- I'm getting some intermittent
> results because I think my camera may be occasionally exceeding the
> power limits of the USB host bus on the beagleboard.  I am going to
> try connecting the camera to a powered hub and see if I can get more
> reliable results.  Thanks to you both again for your help.
>
> regards,
> -andrew
>