Closed GoogleCodeExporter closed 8 years ago
John,
It should set both CW and CCW slopes. Here is what happens: it writes 2 bytes
(1st byte - CW slope, 2nd byte - CCW slope) starting from register 28, so that
1st byte goes to register 28 (CW Compliance Slope) and 2nd byte goes to
register 29 (CCW Compliance Slope).
Are you saying that when using some other driver and setting compliance slopes
to the same value the behavior is different?
Original comment by areb...@gmail.com
on 29 Jun 2011 at 9:02
Okay, my apologies I see that now. Is there a way that I can easily read out
the values of the registers so that I can verify they're being set correctly?
It could be in my head, but it appears that the behavior differs between when I
use your driver vs. some of our matlab code that directly drives the servos.
Original comment by haro...@gmail.com
on 29 Jun 2011 at 9:38
No worries. You could write a simple python script that just connects to the
serial bus, reads the compliance values before and after you set them to
something else. Look at [1] and [2] for an example.
[1]
http://ua-ros-pkg.googlecode.com/svn/trunk/arrg/ua_drivers/ax12_driver_core/scri
pts/info_dump.py
[2]
http://ua-ros-pkg.googlecode.com/svn/trunk/arrg/ua_drivers/ax12_driver_core/scri
pts/set_servo_config.py
Original comment by areb...@gmail.com
on 29 Jun 2011 at 9:47
John,
I would also suggest to migrate to dynamixel_motor [1] stack, that I have
released a few days ago. Any bug fixes and new features will go there. Ubuntu
packages are available for ROS Diamondback.
[1]
http://code.google.com/p/ua-ros-pkg/source/browse/stacks/dynamixel_motor/trunk
Original comment by areb...@gmail.com
on 29 Jun 2011 at 10:22
Okay, I've migrated over to the dynamixel_motor stack. If I encounter any
problems (or verifiable bugs) I'll submit a new report. Thanks.
Original comment by haro...@gmail.com
on 30 Jun 2011 at 2:07
Great! So are compliance settings set properly?
Original comment by areb...@gmail.com
on 30 Jun 2011 at 2:51
Original comment by areb...@gmail.com
on 12 Jul 2011 at 4:53
Original issue reported on code.google.com by
haro...@gmail.com
on 29 Jun 2011 at 8:43