mujoco-environments

A unified framework for robot learning

Tutorial on how to get started with MuJoCo Simulation Platform. MuJoCo stands for Multi-Joint dynamics with Contact.

Repository to accompany RSS 2018 paper on dexterous hand manipulation

Explorer is a PyTorch reinforcement learning framework for exploring new ideas.

OpenAI Gym environment solutions using Deep Reinforcement Learning.

Mujoco Gym environment for the control of quadruped robots

IsaacLab to Mujoco GO2 deploy, IsaacLab to Real world GO2 deploy

Multi-rotor Gym

Mujoco Model for UR5-Ridgeback-Robotiq Robot

PPO implementation for controlling a humanoid in Gymnasium's Mujoco environment, featuring customizable training scripts and multi-environment parallel training.

Multi-drone environment for RL with MuJoCo, with GPU vectorization, wind models, domain randomization, and curriculum learning

PPO implementation of Humanoid-v2 from Open-AI gym

Meta QLearning experiments to optimize robot walking patterns

Code of the article "Benchmarking the Sim-to-Real Gap in Cloth Manipulation"

Soft robotics in MuJoCo

A reinforcement learning playground inspired by the iconic robots from ARC Raiders!

Efficient Model-Based Deep Reinforcement Learning with Predictive Control: Developed a Model-Based RL algorithm using MPC, achieving convergence in 200 episodes (best case) and 1000 episodes on average, outperforming SAC/DQN (10,000+ episodes). Enhanced sample efficiency by 80-90% using learned dynamics and CEM for trajectory optimization.

End-to-end implementation of Action Chunking Transformers (ACT) for imitation learning in robot manipulation tasks. Trained and evaluated in MuJoCo simulation on a pick-and-place task using the ROBOTIS FFW humanoid robot.

Sparse environment for MuJoCo suite (v2 and v3)

Dockerfile for mujoco200 environment. Visualization via X server.