First point in ABB trajectory takes a long time to execute

[GoogleCodeExporter]

The ABB server side(rapid) driver(abb_common) sometimes takes a long time to 
execute the first trajectory point.  This occurs because the joint trajectory 
action node (industrial_robot_client) tolerance for a finished motion is 0.01 
radians.  When the joint trajectory action node reports the motion as finished, 
the last reported joint position is taken as the beginning of the next motion.  
On subsequent motions, the initial point differs from the true end of the last 
motion by more than 0.002 radians (or 0.1 degrees) as hard coded in the ABB 
server.  Because the move duration is set to zero for the first point, the 
robot interprets the first motion as a really slow move (there's probably more 
explanation but I don't know what it is).

A workaround to this is to override the joint trajectory action in the launch 
file as follows:

<node pkg="industrial_robot_client" type="joint_trajectory_action" 
name="joint_trajectory_action_server" output="screen">
        <param name="constraints/goal_threshold" value="0.001"/>
</node>

Original issue reported on code.google.com by shaun.ed...@gmail.com on 23 Feb 2013 at 1:41

[GoogleCodeExporter]

A possible fix to this problem is to simply implement the robot status message 
that reports the motion state.  The joint trajectory action is already set up 
to to use the robot status to determine the end of a move by default (when this 
message isn't published the method looks for a close enough distance).

Original comment by shaun.ed...@gmail.com on 23 Feb 2013 at 1:44

• Added labels: Stack-industrial_core

[GoogleCodeExporter]

That's a good short-term solution.

I think the "better" fix long-term is to correct how we calculate the motion 
durations.  See issue 38 for more details.

Original comment by jz...@swri.org on 25 Feb 2013 at 4:01