[feature request] Sawyer w/ reinforcement learning (openai gym)

[MartinaRuocco]

Hi, I hope someone is taking care of this project :)

Can you please provide a tutorial like this one that explains how to setup ROS Kinetic to run an Openai gym algorithm using Sawyer?

I have evidence that suggests that launching ./intera.sh modifies somehow the settings, therefore the classic tutorials are quite ineffective in this case.