A comprehensive framework for reinforcement learning in robotics, which allows users to train their robots in both simulated and real-world environments.

RealROS is an open-source Python framework that seamlessly integrates with ROS (Robot Operating System) to create real-world robotics environments tailored for reinforcement learning (RL) applications. This modular framework simplifies RL development, enabling real-time training with physical robots

Explore the capabilities of RealROS and MultiROS in training robots for real-world tasks. This repository showcases real-world training and Gazebo simulation-based training for a reach task based on the ReactorX 200 robot manipulator.

This repository contains the support materials for the UniROS framework