It's better to ask such question on the mailing list.
There is two way to communicate with the e-puck BTcom protocol.
 In binary mode (when command a negative)
 In ascii mode (when command are positive)
In binary mode (one use for the camera datas) you can stack the commands, and 
then gets the answer when you send à "0". So probably its what it is missing in 
your protocol.
Also I recommend to use only binary commands... (the same commands as in ascii 
but negative) 

To better understand the protocol have a look on its code...

Bonani Michael

Tél :  + 41 21 693 78 69
Fax : + 41 21 693 78 07
E-mail : michael.bon...@epfl.ch
MEB 3 30
Station 9
1015 Lausanne

 > -----Original Message-----
 > From: www-data [mailto:www-d...@lsa1pc65.epfl.ch] On Behalf Of Stephan
 > Zibner
 > Sent: Tuesday, July 27, 2010 9:06 AM
 > To: Bonani Michael
 > Subject: epuck: Using the E-Puck camera with Matlab
 > This is an enquiry e-mail via http://www.e-puck.org from:
 > Stephan Zibner <stephan.zib...@ini.rub.de>
 > Dear Michael Bonani
 > I'm an employee of the Institut für Neuroinformatik in Bochum, Germany and
 > our work group bought three sets of E-Puck robots a year ago. We are very
 > interested in using the built-in camera of the E-Puck from within Matlab
 > and did spend quite a while on searching for a working solution. Attempts
 > on getting camera images with the provided ePic Matlab interface failed
 > because the program was not able to establish the serial connection to the
 > robot (the required toolboxes are installed). Using our own C-code wrapped
 > in mex files allowed us to control the robot with all the given letter
 > commands (like seen by sending 'H' for help) but this set of commands does
 > not comprise getting a camera image. Sending -'I' as seen in the ePic code
 > did not return the expected current resolution, followed by image data.
 > Only zeros are returned instead. What are we doing wrong? Is this maybe a
 > known issue? Our test platforms are Ubuntu 32bit and 64bit together with
 > Matlab 7.9.0 (R2009b) and a bluetooth connection via rfcomm. ePic worked
 > fine with the same Matlab version on the latest Mac OSX.
 > I would appreciate it, if you could provide us with a solution for the
 > given problem or point us to someone who can help with this.
 > Best regards
 >    Stephan Zibner, M.Sc.
 >    Institut für Neuroinformatik
 >    Ruhr-Universität Bochum
 >    44780 Bochum, Germany

Attachment: BTcom.c
Description: Binary data

E-puck-user mailing list

Reply via email to