Louis LE LAY

MimicLite.

exploy. Exploy: EXport and dePLOY Reinforcement Learning policies

UFO. An open-source unsupervised RL framework for humanoid control with FB/TeCH training, robot-aware motion import, and real-world teleoperation.

esp-now. A connectionless Wi-Fi communication protocol

ScaleBFM. The official implementation of the paper "Scaling Behavior Foundation Model for Humanoid Robots"

bam. Identify and simulate extended friction models for servo-actuators

extreme-parkour. [ICRA 2024]: Train your parkour robot in less than 20 hours.

microban. Microban is an affordable, fully 3D-printable, and 100% open-source humanoid robot. Powered by a Raspberry Pi Zero 2W and 19 Dynamixel servomotors, it serves as an accessible DIY platform for learning, experimentation, and research in robotics.

mjlab_microban. Reinforcement Learning (RL) environments for the Microban platform, a 30cm open-source and affordable humanoid robot.

rohand_mujoco. Mujoco for ROH

lerobot-legged-zoo. LeRobot/Pollen legged robot models + MJLab training examples (no pretrained policies).

physics_based_sysid. Jupyter Notebook

elevation-mapping-realsense-d435i. Elevation Mapping with RealSense Camera d435i

wbc-mjlab. One policy. Any motion. Train universal whole-body tracking in mjlab—swap clips, not checkpoints.

pal_policy_deployer. A policy execution pipeline package that provides tools to deploy ONNX models

ros2_policy_execution_pipeline. A ros2 policy execution pipeline package that provides tools to deploy ONNX models

abc. ABC: Scalable Behavior Cloning with Open Data, Training, and Evaluation

roboto_origin. Roboto_origin Fully Open-Source DIY Humanoid Robot/萝博头原型机全开源手搓级人形机器人

wuji-mjlab. In-hand reorientation RL: PPO training and sim-to-real deployment for the Wuji Hand, built on mjlab.

Open_Duck_Mini. Making a mini version of the BDX droid. https://discord.gg/UtJZsgfQGe

mevius2. Practical Open-Source Quadruped Robot with Sheet Metal Welding and Multimodal Perception

mevius. A Quadruped Robot Easily Constructed through E-Commerce with Sheet Metal Welding and Machining

KISS-IMU. KISS-IMU: Self-supervised Inertial Odometry with Motion-balanced Learning and Uncertainty-aware Inference. @ ICRA'26 Award Finalist

onshape-to-robot. Converting Onshape assembly to robot definition (URDF, SDF, MuJoCo) through Onshape API

tdmpc2. Code for "TD-MPC2: Scalable, Robust World Models for Continuous Control"

mjlab_manipulation. A growing collection of manipulation tasks built with mjlab.

nano-world-model. A Minimalist, Batteries-included Repository for Advancing World Model Science.

AMP_mjlab. Python

UnrealRoboticsLab. A high-fidelity, open-source robotics simulator integrating Unreal Engine's photorealistic rendering with MuJoCo's precision physics.

MakeTrackingEasy. Python

KickCAT. A C++ open source EtherCAT master/slave stack

twist2_mjlab. Python

mjlab_playground. A collection of tasks built on mjlab

le-wm. Official code base for LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels

stable-worldmodel. A platform for reproducible world model research and evaluation

AME_Locomotion. This repository reproduces the Attention-Based Map Encoding (AME) method from the paper Attention-Based Map Encoding for Learning Generalized Legged Locomotion.

mujoco-motors. A database of motor specs for mujoco,mjax and mjlab

mjviser. Web-based MuJoCo viewer powered by Viser

kimolab. Isaac Lab API, powered by MuJoCo-Warp, for RL and robotics research.

SOMA-X. SOMA: Unifying Parametric Human Body Models

LATENT. Official implementation of Learning Athletic Humanoid Tennis Skills from Imperfect Human Motion Data

InstinctMJ. mjlab-native port of InstinctLab for humanoid RL and Project-Instinct workflows.

humanoid_skateboarding. [RSS 2026] "HUSKY: Humanoid Skateboarding System via Physics-Aware Whole-Body Control"

genesislab. A lightweight robotics RL framework built on Genesis with IsaacLab-aligned APIs, enabling fast task development and easy migration from existing IsaacLab environments.

rumi-mjlab. mjlab integration for our open source quadruped Rumi

coconuts. The CocoNuts project is an open-source humanoid robotics research platform. Its primary design philosophy is the minimization of leg inertia by concentrating mass near the root of the limbs, utilizing a combination of timing belts and parallel linkage mechanisms.

pal_mjlab. PAL Robotics' robots in mjlab

mjlab-skillkit. AI coding skill kit for IsaacLab migration and mjlab-native development

minimalist_compliance_control. Python

OmniXtreme. Python

open_robot_actuator_hardware.

CLOT. official code for paper CLOT: Closed-Loop Global Motion Tracking for Whole-Body Humanoid Teleoperation

urdf-visualizer. A VSCode extension for visualizing the URDF file and xacro file.

simtoolreal. Official implementation of SimToolReal: An Object-Centric Policy for Zero-Shot Dexterous Tool Manipulation

decentralized_multibiped_controller. Official code repository for decMBC paper (CoRL 2024)

on-policy. This is the official implementation of Multi-Agent PPO (MAPPO).

relic. Supplementary code for “Versatile Loco-Manipulation through Flexible Interlimb Coordination”

marladona-isaac-lab. External extenstion template based on Isaac Lab

x1-mjlab. Python

tron1-rl-mjlab. Python

mjlab_jump_style. Python

in-hand-rotation-mjlab. Sim-to-real RL for in-hand cube rotation with the LEAP Hand, built on Mjlab.

freemocap. Free Motion Capture for Everyone 💀✨

perceptive_mpc. Code for "Perceptive Model Predictive Control for Continuous Mobile Manipulation"

solidworks_urdf_exporter. SolidWorks to URDF Exporter

awesome-humanoid-robot-learning. A Paper List for Humanoid Robot Learning.

GVHMR. Code for "GVHMR: World-Grounded Human Motion Recovery via Gravity-View Coordinates", Siggraph Asia 2024, TPAMI 2026

TWIST2. [arXiv 2025] TWIST2: Scalable, Portable, and Holistic Humanoid Data Collection System

robotic_world_model_lite. Lightweight implementation of Robotic World Model (RWM) and Uncertainty-Aware Robotic World Model (RWM-U)

scenesmith. Code for "SceneSmith: Agentic Generation of Simulation-Ready Indoor Scenes", ICML 2026 Spotlight

ASAP. [RSS 2025] "ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills"

mjlab. Isaac Lab API, powered by MuJoCo-Warp, for RL and robotics research.

cnpy. library to read/write .npy and .npz files in C/C++

smplx. SMPL-X

crl-humanoid-ros. A ROS 2 framework for humanoid robot simulation and control, developed by the Computational Robotics Lab (CRL) at ETH Zurich.

robotic_world_model. Repository for our papers: Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Robotics and Uncertainty-Aware Robotic World Model Makes Offline Model-Based Reinforcement Learning Work on Real Robots

fpo-control. Flow Policy Gradients for Robot Control

bipedal-robot-learning-collection. Collection of high-quality robo learning papers for bipedal robots.

legged_template_controller. DO NOT FORK, CLICK ON "Use this template"

mjlab-homierl. Fork of mjlab which supports unitree H1/G1 & robotiq_2f85, personal implementation of the locomotion policy training in "HOMIE: Humanoid Loco-Manipulation with Isomorphic Exoskeleton Cockpit [RSS2025]"

anymal_c_velocity. Integrating a custom robot (ANYmal C) with mjlab's velocity task

unitree_rl_mjlab. This is a repository for reinforcement learning implementation for Unitree robots, based on Mujoco.

spider. A general physic-based retargeting framework.

awesome-physical-ai. A curated list of academic papers and resources on Physical AI — focusing on Vision-Language-Action (VLA) models, world models, embodied ai, and robotic foundation models.

claude-sneakpeek. Get a parallel build of Claude code that unlocks feature-flagged capabilities like swarm mode.

sage. Framework for measuring sim-to-real gaps in robot joint motions. Supports different humanoids with physics simulation, real hardware data collection, and statistical analysis.

Contact-Aided-Invariant-EKF. Example code for contact-aided invariant extended Kalman filtering.

invariant-ekf. C++ library to implement invariant extended Kalman filtering for aided inertial navigation.

asimov-mjlab. Asimov Humanoid Locomotion

genie_sim. Simulation Platform from AgiBot

pyroki. A Modular Toolkit for Robot Kinematic Optimization

rsl_rl_rwm. Algorithm implementation for our paper: Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Robotics

urdf-usd-converter. A URDF to OpenUSD Data Converter

mjlab_myosuite. Interface of myosuite with https://github.com/mujocolab/mjlab

Isaac-Sim-Physical-consistency-plugin. Isaac Sim Physical consistency plugin

ProtoMotions. ProtoMotions is a GPU-accelerated simulation and learning framework for training physically simulated digital humans and humanoid robots.

SoloParkour. Python

IsaacLab-Go2. Python

h1v2-Isaac. Python

holosoma. Python

cleanrl. High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)

legged_perceptive. C++

legged_control. NMPC, WBC, state estimation, and sim2real framework for legged robots based on OCS2 and ros-controls

mujoco_vendor. ROS 2 vendor package for MuJoCo simulator

edge-policy-inference. Quick and Dirty Version

mujoco_ros2_control. ROS2 control package for Mujoco simulation physics simulator, including plugins for various sensors, RGB-D cameras, and lidar.

URDF2MJCF. Convert URDF robot definitions to MuJoCo environments

librepods. AirPods liberated from Apple's ecosystem.

spider. A general physic-based retargeting framework.

zrm. Zenoh ROS-like Middleware

sage. Framework for measuring sim-to-real gaps in robot joint motions. Supports different humanoids with physics simulation, real hardware data collection, and statistical analysis.

space_robotics_bench. Robot Learning Beyond Earth

isaac_ros_pose_estimation. Deep learned, NVIDIA-accelerated 3D object pose estimation

FoundationPose. [CVPR 2024 Highlight] FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects

viser. Web-based 3D visualization in Python

BDX-R-MjLab. mjlab reinforcement learning for the BDX-R robot

mujoco_ros2_simulation. This project has been moved to the ros-controls organization, see there for updates: https://github.com/ros-controls/mujoco_ros2_simulation

pace-sim2real. PACE: A systematic approach for sim-to-real transfer of legged robots, identifying actuator and joint dynamics with standard joint encoders.

constraints-as-terminations. Integrating Constraints in PPO (using Isaac Gym or Isaac Lab)

usd2mjcf. This is a Python tool for converting USD models to MuJoCo MJCF format.

best-of-robot-simulators. A Best-of-list of Robot Simulators, re-generated weekly on Wednesdays

torchax. torchax is a PyTorch frontend for JAX. It gives JAX the ability to author JAX programs using familiar PyTorch syntax. It also provides JAX-Pytorch interoperability, meaning, one can mix JAX & Pytorch syntax together when authoring ML programs, and run it in every hardware JAX can run.

diffusion-policy. Implementation of Diffusion Policy, Toyota Research's supposed breakthrough in leveraging DDPMs for learning policies for real-world Robotics

barkour_robot. Barkour Robot: Agile Quadruped Robots by Google DeepMind

g1_spinkick_example. Train a Unitree G1 humanoid to perform a double spin kick using mjlab

TWIST. [CoRL 2025] TWIST: Teleoperated Whole-Body Imitation System

DeepMimic. Motion imitation with deep reinforcement learning.

MimicKit. A lightweight suite of motion imitation methods for training controllers.

mink. Python inverse kinematics based on MuJoCo.

pixidock_template. Template for hermetic, reproducible, and neat Python environments; with some extra envs for roboticists

pymujoco. Python bindings for open source MuJoCo

embodiment-scaling-laws. Implementation of the paper on Embodiment Scaling Laws in Robot Locomotion (CoRL 2025)

PARC. PARC: Physics-based Augmentation with Reinforcement Learning for Character Controllers

uv. An extremely fast Python package and project manager, written in Rust.

BDX-R-Description. CAD, URDF & USD Files for the BDX-R Robot

Robot-Trains-Robot. [CoRL 2025] Real-world RL. Official implementation of "Robot Trains Robot: Automatic Real-World Policy Adaptation and Learning for Humanoids"

MikuMotionTools. Tool to bridge Blender animation and physics-based robotic simulation

GMR. [ICRA 2026] GMR: General Motion Retargeting. Retarget human motions into diverse humanoid robots in real time on CPU. Retargeter for TWIST.

ubisoft-laforge-animation-dataset. Ubisoft La Forge Animation Dataset

pinocchio. A fast and flexible implementation of Rigid Body Dynamics algorithms and their analytical derivatives

teleop. Turns your phone or VR headset into a robot arm teleoperation device by leveraging WebXR

walk-these-ways. Sim-to-real RL training and deployment tools for the Unitree Go1 robot.

actuator_net. Python

madrona_mjx. C++

so101_deploy. Deployment code for so101 based on lerobot interface

Indy7_RL_sim2real. This repo provides sim2real code for Neuromeka Indy7

cassie-mujoco-sim. A simulation library for Agility Robotics' Cassie robot using MuJoCo

amp-rsl-rl. 🔁 AMP-RSL-RL: Adversarial Motion Priors for robotic RL (PPO + motion imitation)

mevita. MEVITA: Open-Source Bipedal Robot Assembled from E-Commerce Components via Sheet Metal Welding

brax_training_viewer. Brax Viewer is a real-time, interactive web viewer for monitoring reinforcement learning (RL) policies

unitree_rl_lab. This is a repository for reinforcement learning implementation for Unitree robots, based on IsaacLab.

genesis_ros. A light-weight, pythonic ros2 package to connect the genesis simulator and ROS2

BDX-R-IsaacLab. Isaac Lab reinforcement learning for the BDX-R robot

wandb. The AI developer platform. Use Weights & Biases to train and fine-tune models, and manage models from experimentation to production.

tyro. CLI interfaces & config objects, from types

mjlab. Isaac Lab API, powered by MuJoCo-Warp, for RL and robotics research

TienKung-Lab. Tien Kung-Lab: Direct IsaacLab Workflow for Legged Robots

TienKung_URDF. Python

whole_body_tracking. Python

gpt-oss. gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

mujoco_mpc. Real-time behaviour synthesis with MuJoCo, using Predictive Control

PlotJuggler. The Time Series Visualization Tool that you deserve.

nav-suite. Navigation Suite for IsaacLab

human2humanoid. [IROS 2024] Learning Human-to-Humanoid Real-Time Whole-Body Teleoperation. [CoRL 2024] OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learning

isaac_so_arm101. Isaac Lab external project for SO-ARM100/101 arm robot.

HOVER. HOVER

genesis-world. Simulation platform for general-purpose robotics & embodied AI learning.

Gymnasium. A standard API for single-agent reinforcement learning environments, with popular reference environments and related utilities (formerly Gym)

skrl. Modular Reinforcement Learning (RL) library (implemented in PyTorch, JAX, and NVIDIA Warp) with support for Gymnasium/Gym, NVIDIA Isaac Lab, MuJoCo Playground and other environments

isaac-go2-ros2. Unitree Go2 simulation platform for testing navigation, decision-making and autonomous tasks. (NVIDIA Isaac/ROS2)

ros1_bridge_sandbox. Minimal packages to demonstrate usage of ros1_bridge with custom message types

cyclo_lab. This repository provides tutorials for reinforcement learning and imitation learning using ROBOTIS robots, and supports Sim2Real functionality for deploying the learned policies on real robots.

Lightwheel_Kitchen. A complete kitchen scene designed by Lightwheel for simulation and interaction in Isaac Sim.

mujoco_kangaroo_sim2sim. Minimal package to run locomotion policies on a Kangaroo robot in MuJoCo

tiago_simulation. Python

mjcf2usd. A powerful Isaac Sim extension for converting MuJoCo XML Configuration Format (MJCF) files to Universal Scene Description (USD) format

ros2_so_arm. ROS 2 packages for SO-ARM100 and SO-ARM101 arms.

mjcf_urdf_simple_converter. Minimal script to convert MuJoCo MJCF robot model files to URDF files usable in ROS

isaaclab_kangaroo. Isaac Lab locomotion tasks for the Kangaroo robot

so_arm_100_hardware. a ROS 2 Control hardware interface plugin for the SO100-arm low-cost 5DoF robotic manipulator. This interface supports both direct serial communication with the physical robot and simulation via ROS topics.

IsaacSim. NVIDIA Isaac Sim™ is an open-source application on NVIDIA Omniverse for developing, simulating, and testing AI-driven robots in realistic virtual environments.

rl_sar. Simulation verification and physical deployment of robot reinforcement learning algorithms, suitable for quadruped robots, wheeled robots, and humanoid robots. "sar" represents "simulation and real"

mujoco_menagerie. A collection of high-quality models for the MuJoCo physics engine, curated by Google DeepMind.

SO-100-arm. Standard Open Arm 100 - Version 1.3

kangaroo_robot. The perfect robot to learn about walking locomotion.

newton. An open-source, GPU-accelerated physics simulation engine built upon NVIDIA Warp, specifically targeting roboticists and simulation researchers.

IsaacLabTutorial. Tutorial project for the accompanying doc

SO-ARM100. Standard Open Arm 100

ksim-zbot. Policies for Z-Bot

kos-sim. Simulation backend for KOS

bdx_nao_rl_isaaclab. Python

kinova_isaaclab_sim2real. Codebase for Kinova Gen3 training in Isaac Lab, sim2real transfer and ROS deployment.

loco-mujoco. Imitation learning benchmark focusing on complex locomotion tasks using MuJoCo.

ros2_kortex. ROS2 driver for the Gen3 Kinova robot arm

WheeledLab. Environments, assets, workflow for open-source mobile robotics, integrated with IsaacLab.

velo2cam_calibration. Automatic Extrinsic Calibration Method for LiDAR and Camera Sensor Setups. ROS Package.

Leatherback_tutorial. Python

orbit-surgical. ORBIT-Surgical: An Open-Simulation Framework for Learning Surgical Augmented Dexterity

ros2_SimRealRobotControl. This repository provides ready-to-use ROS2 (Humble) packages to execute simple programs and sequences and control different Industrial and Collaborative Robots using ROS 2.

mujoco_warp. GPU-optimized version of the MuJoCo physics simulator, designed for NVIDIA hardware.