robo-gym -- An Open Source Toolkit for Distributed Deep Reinforcement Learning on Real and Simulated Robots

Applying Deep Reinforcement Learning (DRL) to complex tasks in the field of\nrobotics has proven to be very successful in the recent years. However, most of\nthe publications focus either on applying it to a task in simulation or to a\ntask in a real world setup. Although there are great examples of combining the\ntwo worlds with the help of transfer learning, it often requires a lot of\nadditional work and fine-tuning to make the setup work effectively. In order to\nincrease the use of DRL with real robots and reduce the gap between simulation\nand real world robotics, we propose an open source toolkit: robo-gym. We\ndemonstrate a unified setup for simulation and real environments which enables\na seamless transfer from training in simulation to application on the robot. We\nshowcase the capabilities and the effectiveness of the framework with two real\nworld applications featuring industrial robots: a mobile robot and a robot arm.\nThe distributed capabilities of the framework enable several advantages like\nusing distributed algorithms, separating the workload of simulation and\ntraining on different physical machines as well as enabling the future\nopportunity to train in simulation and real world at the same time. Finally we\noffer an overview and comparison of robo-gym with other frequently used\nstate-of-the-art DRL frameworks.\n