Reinforcement learning environment for UR5e arms with MuJoCo 3 — SAC training for reach, pick-and-place, and symmetric multi-arm cooperative tasks. Includes a ROS 2 policy node for Gazebo deployment.
python reinforcement-learning robotics deep-reinforcement-learning universal-robots gazebo manipulation robot-arm sac gymnasium robot-learning ros2 mujoco robotiq-gripper pick-and-place ur5e multi-arm sim-to-real stable-baselines3 contact-rich
-
Updated
Apr 29, 2026 - Python