Variability on the NRP

#1

Dear NRP community/developers,

I have a question regarding the sources of variability that seem to be present on the platform,

To clarify, this is when I observe the variability: repeatedly running the same experiment in which I feed my robot the same motor signals (so no spikes/brain involved) result in my robot completing pretty different trajectories.

Now I am curious where this variability comes from, does the asynchronous operation of the CLE contribute to this? Or is it perhaps due to the physics engine? (I am using the default ODE). I also think this is an important point for roboticists that are perhaps used to deterministic simulations.

happy with any thoughts on this,

Kind regards,

Alexander

NRP simulation appears non-deterministic?
#2

Hi Alexander,

Could you send us the models and the experiment files that you use so that we can also test on our side and see what causes this variability?

Regards,
Manos

#3

Hi Alexander,

Simulation determinism/variability with the Neurorobotics platform has been investigated extensively and we know different reasons which explain why a simulation run by the Neurorobotics platform is not deterministic.

First, the current implementation of our Closed Loop Engine’s control loop is not deterministic by design: the world simulation step (Gazebo) is threaded and runs in parallel with the execution of transfer functions. But we also have a deterministic version of the control loop (see $HBP/CLE/hbp_nrp_cle/hbp_nrp_cle/cle/DeterministicClosedLoopEngine.py) and you should be able to use it in the future with a simple change in the experiment configuration.

Second, the asynchronous nature of ROS and its current integration with transfer functions make them non-deterministic: there is no guarantee that the same ROS message will be available at the exact same time by a transfer function, from one run to the other.

Eventually, Gazebo has its own source of non-determinism as its publishing rate of images is not deterministic. A patch is available (see https://bitbucket.org/osrf/gazebo/issues/2293/non-deterministic-image-render-frame), but it is not part of the NRP yet.

For each of the aforementioned issue, there is a solution and an epic has been written to address them all. The pace at which we will tackle non-determinism in NRP depends on our priority sort and users’ requests.

Best regards,
Luc