"@Aurelien
Yaw torque problem looks like a Gazebo version issue (I'm using Gazebo 6.5):
https://github.com/ethz-asl/rotors_simulator/commit/71c1ddef1ff4234..."

"@Markus  I don't know if this is the same problem you are having but I was using Alex Buyval's version of rotor_simulator and found and issue with that.  The motor plugin used for the rotors only applies torque to the rotor link and not the base…"

"@Markus Well, no luck yet.  It's acting as if the physics simulation is ignoring the rotors' angular momentum. The rotors' have a non-zero moment of inertia defined, though, so I can't see why that would be the case."

"I've been playing with this (the ArduCopter SITL Gazebo plugin) and am experiencing something weird with the yaw control. In modes where overriding the RC channel 4 is supposed to rotate the vehicle, nothing happens unless RC 4 is set to an extreme…"

"I'm already working from a clean and up to date version of your master branch.

How are you slowing down SITL sim? I added the following to Firmware/Tools/worlds/iris.world:
    <physics type="ode">       …"

"@PX4
How exact is slowing the PX4 SITL sim accomplished? When I slow down the physics in my gazebo world file, the copter loses control (and the SITL keeps pumping out heartbeats at the same rate).

This lead me to believe that slower than realtime…"

"@Andrew_Tridel and whoever else: Does the SITL backend always run in real time or can I clock it slower (i.e. if I slow down the real_time_update_rate in the gazebo physics so that the sim time runs slower than real time, will the SITL sim use the…"

"This step lock is a nice feature.  Is the step length configurable, though?  I'm trying to emulate a sensor that needs a simulation update faster than 2.5ms. Also, does this same mechanism control the ROS simulation time?
The step lock mechanism…"