Install and Use Isaac Sim and Isaac Lab

Install Isaac Sim

If you haven't already done so, install Isaac Sim first.

Clone the Isaac Lab repository into your workspace

Clone Isaac Lab from the NVIDIA GitHub repository.

git clone --recursive https://github.com/isaac-sim/IsaacLab
cd IsaacLab

Create a symbolic link to the Isaac Sim installation

Be sure that you have already installed Isaac Sim from Isaac Sim before running the following command.

echo "ISAACSIM_PATH=$ISAACSIM_PATH"

Create a symbolic link to the Isaac Sim installation directory.

ln -sfn "${ISAACSIM_PATH}" "${PWD}/_isaac_sim"
ls -l "${PWD}/_isaac_sim/python.sh"

Install Isaac Lab

./isaaclab.sh --install

Run Isaac Lab and Validate Humanoid Reinforcement Learning Training

Launch Isaac Lab using the provided Python executable. You can run the training in one of the following modes:

Option 1: Headless Mode (Recommended for Faster Training)

Runs without visualization and outputs logs directly to the terminal.

export LD_PRELOAD="$LD_PRELOAD:/lib/aarch64-linux-gnu/libgomp.so.1"
./isaaclab.sh -p scripts/reinforcement_learning/rsl_rl/train.py --task=Isaac-Velocity-Rough-H1-v0 --headless

Option 2: Visualization Enabled

Runs with real-time visualization in Isaac Sim, allowing you to monitor the training process interactively.

export LD_PRELOAD="$LD_PRELOAD:/lib/aarch64-linux-gnu/libgomp.so.1"
./isaaclab.sh -p scripts/reinforcement_learning/rsl_rl/train.py --task=Isaac-Velocity-Rough-H1-v0