Set the paused argument in the gazebo empty_world.launch to False

[JorisWeeda]

Description

In order to properly launch the moveit package the simulation time must be active otherwise the movegroups will not be initialized. If the movegroups are not initialised the init of the gait selection is not finished causing the state machine to be stuck waiting for the gait selection which will not be finished. Therefore the paused argument must be false. This will not cause any weird behavior of the simulation because the physics are loaded after the state machine is initialized.

Changes

• Set paused argument to false

[RutgerVanBeek]

I definitively approve this.

[Roelemans]

Perhaps make a paused argument that is set to $(arg not balanced) in march.launch?

[JorisWeeda]

Perhaps make a paused argument that is set to $(arg not balanced) in march.launch?

We could add an argument but Rutger and I tested the normal simulation launch and the argument change did not degrade the functionalities of the simulation so maybe not necessary?

[Roelemans]

The simulation is started paused to avoid the issue of feet flapping after startup (though you may have noticed that this does not always work). I am not sure how big that problem currently is. I do have a preference for starting it paused when possible, because I don't wanna look into the flappin feet again 🙃

[RutgerVanBeek]

The simulation is started paused to avoid the issue of feet flapping after startup (though you may have noticed that this does not always work). I am not sure how big that problem currently is. I do have a preference for starting it paused when possible, because I don't wanna look into the flappin feet again

The physics are only turned on when the state machine is finished initializing, right?
This just make the gazebo publish on /clock.

[Roelemans]

EDIT: Just realised that the "unpause physics" at the end of the state machine unpauses the simulation. But that only works if it is started paused, which is what you are turning off here. i.e. if you set paused to false as you propose in this PR, the physics will start unpaused and the unpausee at the end of the state machine will do nothing.