Repeatability of real world training experiments: A case study

We present a case study of reinforcement learning on a real robot that lear ns how to back up a trailer and discuss the lessons learned about the impor tance of proper experimental procedure and design. We identify areas of par ticular concern to the experimental robotics community at large. In particu lar, we address concerns pertinent to robotics simulation research, impleme nting learning algorithms on real robotic hardware, and the difficulties in volved with transferring research between the two.