David, I am actually in the middle of revamping the segway-rmp-ros-pkg to include support for both usb and serial communication, and in the process I am changing a lot about the way the node works. If I remember correctly there rmp100 series have different constants (for determining wheel speed, integrated position, etc...) because it has different wheel sizes and such to the rmp200, I am including an option in my new node to support both the rmp200/400 and the rmp50/100, but that doesn't exist just yet.Your new libsegwayrmp library sounds great. I'm also mainly using linux so having it working with the default linux VCP interface is definitely a bonus.
I literally just finished the underlying segway library (http://williamjwoodall.com/2011/06/21/my-segway-rmp-library-libsegwayrmp/) and rewrote most of the ROS node today, it should be ready by the end of the week.
Yes. I've learnt my lesson about having the segway on blocks. The shocked look on my face when the segway took off would probably have made one of those funniest home video type shows :). I (stupidly) didn't look at the source and wasn't expecting the IRI test program to move the robot. I guess I just thought it would print information to the screen to the effect that it had established the connection and everything was looking good.Blocks are _always_ a good idea, the Segway RMP's are very powerful machines and always command some respect. I think the IRI test program should warn you, but I know there are warnings in the comments of the source of the test program, not that I would expect anyone to look there first.
Ok. I'm sure I had no other nodes running so the wheels spinning was not the expected behaviour :).The ROS node shouldn't move unless messages are being published on the /cmd_vel topic, but I have not explicitly tested the RMP100 so this might be new behavior. If you can wait, the new packages will be out soon, or I am happy to try and help you debug this version more tomorrow.
Good night,_______________________________________________ ros-users mailing list ros-users@code.ros.org https://code.ros.org/mailman/listinfo/ros-users
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
William Woodall
Graduate Software Engineering
Auburn University
w@auburn.edu
wjwwood@gmail.comwilliamjwoodall.com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
On Thu, Jun 23, 2011 at 2:10 AM, David Rajaratnam <daver_ml@gemarex.com.au> wrote:
Hi all,
I'm trying to get the segway_rmp stack
(http://www.ros.org/wiki/segway_rmp) working and was wondering if there
was anyone else using these drivers (which wrap the IRI drivers) to
connect to a segway rmp100.
I haven't found too much information on the difference between the rmp
100 and 200 so was assuming/hoping that it would work the same - my
reasoning was that someone else on the team got the segway working under
player using some general rmp player driver.
Anyway, when I ran the iri's 'test_segway_rmp_200' program it sent the
segway hurtling into the desk. To protect myself (and everyone else)
from my stupidity the segway is now on blocks. I next ran the
segway_rmp200 ros node and the node looked like it started up ok with an
info message saying something like "Segway Ready". However, it had a
similar result as before where the wheels were sent spinning wildly till
some sort of overload failsafe kicked in.
So I was wondering if anyone else is trying to do the same - or am I
doing something obviously stupid.
Cheers,
Dave
_______________________________________________
ros-users mailing list
ros-users@code.ros.org
https://code.ros.org/mailman/listinfo/ros-users