ros_controllers icon indicating copy to clipboard operation
ros_controllers copied to clipboard

Published 20 hours ago • verified publisher icon ros-controls

Full reset of diff_drive_controller in rostests

Open bmagyar opened this issue 5 years ago • 1 comments

@martiniil do you think it would be a viable and stable approach to start/stop or unload/reload the controller in the test fixture DiffDriveControllerTest to properly reset the controller state?

This discussion was split from #283 and the new test affected by this was merged in #469 . Now we need to make sure it actually runs in a meaningful way ;)

bmagyar avatar Mar 15 '20 08:03 bmagyar

The problem I observed was not the controller state but the state of the robot Diffbot, that is used in the tests.

Diffbot has a start/stop functionality. Stop triggeres Nan in position/velocity. Start should lead back to defined values, but it doesn't work. I am not sure which scenario of a real robot this start/stop function corresponds to. Do you have an explanation? I see two possibilities

  • Always start at 0.0 pos/vel
  • Start at last defined position

martiniil avatar Mar 16 '20 07:03 martiniil