Update: I've also noticed that under some pid setting a set point of zero still results in an error value for a continuous joint on a wheel. This should not be the case as it should already be at a neutral gravity state on flat ground.

Update: As I play with this more I see that this error value can be over or under the stated set point this seems to be a pid related issue. However I do not understand why being that both left and right motors have the same pid setting why the difference between the left and right error values.

I am having a problem or miss understanding of the Joint Controller State. I send the controller two velocities for a right and left wheel each +.05 meters per sec. The Joint Controller State message show the following. I am treating the values as velocities in meters per sec.

process_value -> pv
er -> error
sp -> set_point

[ INFO] [1380852360.412367809, 4.000000000]: pv1=[-0.023923]  pv2=[0.013752]
[ INFO] [1380852360.412429679, 4.000000000]: er1=[0.073923]   er2=[0.036248] 
[ INFO] [1380852360.412461154, 4.000000000]: sp1=[0.050000]   sp2=[0.050000]

First why are the errors different, the wheels are exactly the same and running on a flat surface in gazebo.

Second why is the error on joint-1 greater then the velocity sent and why did the actual process value go negative.

I thought the error is that part of the set point (original velocity) that was not executed. so pv1 should be 0 I would guess.