isaac-sim/IsaacLab

Unified framework for robot learning built on NVIDIA Isaac Sim

What it solves

Isaac Lab simplifies and unifies robotics research workflows, specifically for reinforcement learning, imitation learning, and motion planning. It addresses the challenge of sim-to-real transfer by providing a high-performance simulation environment that balances speed and accuracy.

How it works

Built on NVIDIA Isaac Sim, the framework uses GPU acceleration to run complex physics and sensor simulations rapidly. It supports various physics backends and renderers, allowing simulations to run locally or be distributed across the cloud for large-scale data generation and training.

Who it’s for

It is designed for robotics researchers and developers focusing on robot learning, including those working with manipulators, quadrupeds, and humanoids.

Highlights

  • Diverse Robot Library: Includes over 16 commonly available models across different robot types.
  • Extensive Environments: Over 30 ready-to-train environments compatible with popular RL frameworks like RSL RL, SKRL, RL Games, and Stable Baselines.
  • Advanced Sensor Simulation: Features RTX-based cameras (RGB, depth, segmentation), LIDAR, IMU, and contact sensors.
  • Broad Physics Support: Handles rigid bodies, articulated systems, and deformable objects.

Related

  • Project
    umfieldrobotics/OceanSim

    A GPU-accelerated underwater robot perception simulation framework that uses physics-based rendering to model visual and acoustic sensors and reduce the simulation-to-real gap.

  • Project
    NVlabs/ProtoMotions

    A GPU-accelerated simulation and learning framework for training physically simulated digital humans and humanoid robots, supporting large-scale motion learning and sim-to-real deployment.

  • Project
    mani-skill/ManiSkill

    An open-source robot simulation and training framework that uses GPU parallelization to enable high-throughput data collection and training for manipulation skills.

  • Project
    google-deepmind/mujoco_playground

    A GPU-accelerated suite of robot learning environments built with MuJoCo MJX, designed for efficient reinforcement learning and sim-to-real research.

  • Project
    nvidia-isaac/WBC-AGILE

    A reinforcement learning framework based on NVIDIA Isaac Lab for training humanoid loco-manipulation policies with validated sim-to-real transfer capabilities.