differential_wheel_vel not zero on standby ?

[mherrb]

Hi,

we're using a RMP440LE platform with our own control software, but the issue should be visible on the default OCU demo too.
When the robot is idle (either in STANDBY mode or in TRACTOR mode with zero linear and angular speeds commands sent periodiacally) the mean reported differential_wheel_vel is not zero (it's always around 0.001 rad/s). Moreover, looking at the left and right wheels velocities and computing the difference ourselves leads to a comparable mean value.

Is this expected ? Is there anything that can be done to avoid this bias ?

[sdnalloh]

The RMP uses its motors to hold position. In Tractor Mode (with zero motion being commanded) or in Standby Mode the RMP's wheels will not rotate, even when pushed or on a steep slope.

The very small velocities you are seeing are a result of the RMP holding position. The wheels are fighting external forces (like gravity) and are also fighting each other. If you put your hand on one of the tires while the RMP is is Standby Mode, you will feel a very slight vibration: this is the motor rocking back and forth to hold position.

These small values (+/- 0.002) are normal and expected.

-Greg