OpenArm Isaac Lab

Overview

This repository provides simulation and learning environments for the OpenArm robotic platform, built on NVIDIA Isaac Sim and Isaac Lab. It enables research and development in reinforcement learning (RL), imitation learning (IL), teleoperation, and sim-to-real transfer for both unimanual (single-arm) and bimanual (dual-arm) robotic systems.

What this repo offers

• Isaac Sim models for OpenArm robots.
• Isaac Lab training environments for RL tasks (reach, lift a cube, open a drawer).
• Imitation learning, teleoperation interfaces, and sim-to-sim / sim-to-real transfer pipelines are currently under development and will be available soon.

This repository has been tested with:

• Ubuntu 22.04
• Isaac Sim v5.1.0
• Isaac Lab v2.3.0
• Python 3.11

---

Index

• OpenArm Isaac Lab
  • Overview
    • What this repo offers
  • Index
  • Installation Guide
    • (Option 1) Docker installation (linux only)
    • (Option 2) Local installation
  • Reinforcement Learning (RL)
    • Training Model
    • Replay Trained Model
    • Analyze logs
  • Sim2sim
  • Sim2Real Deployment using OpenArm
  • Related links
  • License
  • Code of Conduct

Installation Guide

(Option 1) Docker installation (linux only)

1. Pull the minimal Isaac Lab container

docker pull nvcr.io/nvidia/isaac-lab:2.3.0

2. Create container

xhost +
docker run --name isaac-lab --entrypoint bash -it --gpus all --rm -e "ACCEPT_EULA=Y" --network=host \
   -e "PRIVACY_CONSENT=Y" \
   -e DISPLAY \
   -v $HOME/.Xauthority:/root/.Xauthority \
   -v ~/docker/isaac-sim/cache/kit:/isaac-sim/kit/cache:rw \
   -v ~/docker/isaac-sim/cache/ov:/root/.cache/ov:rw \
   -v ~/docker/isaac-sim/cache/pip:/root/.cache/pip:rw \
   -v ~/docker/isaac-sim/cache/glcache:/root/.cache/nvidia/GLCache:rw \
   -v ~/docker/isaac-sim/cache/computecache:/root/.nv/ComputeCache:rw \
   -v ~/docker/isaac-sim/logs:/root/.nvidia-omniverse/logs:rw \
   -v ~/docker/isaac-sim/data:/root/.local/share/ov/data:rw \
   -v ~/docker/isaac-sim/documents:/root/Documents:rw \
   nvcr.io/nvidia/isaac-lab:2.3.0

3. Clone git at your HOME directory

cd /workspace
git clone [email protected]:enactic/openarm_isaac_lab.git

4. Install python package with

cd openarm_isaac_lab
python -m pip install -e source/openarm

5. With this command, you can verify that OpenArm package has been properly installed and check all the environments where it can be executed.

python ./scripts/tools/list_envs.py

(Option 2) Local installation

It is assumed that you have created a virtual environment named env_isaaclab using miniconda or anaconda and will be working within that environment.

1. Clone git at your HOME directory

cd ~
git clone [email protected]:enactic/openarm_isaac_lab.git

2. Activate your virtual env which contains Isaac Lab package

conda activate env_isaaclab

3. Install python package with

cd openarm_isaac_lab
python -m pip install -e source/openarm

4. With this command, you can verify that OpenArm package has been properly installed and check all the environments where it can be executed.

python ./scripts/tools/list_envs.py

Reinforcement Learning (RL)

You can run different tasks and policies.

Replace <TASK_NAME> and <POLICY_NAME> with one of the following available tasks:

Task Description	Task Name	Policy Name	Demo
Reach target position	Isaac-Reach-OpenArm-v0	rsl_rl, rl_games, skrl
Lift a cube	Isaac-Lift-Cube-OpenArm-v0	rsl_rl, rl_games, skrl
Open a cabinet's drawer	Isaac-Open-Drawer-OpenArm-v0	rsl_rl, rl_games, skrl
Reach target position (Bimanual)	Isaac-Reach-OpenArm-Bi-v0	rsl_rl, rl_games, skrl

Training Model

python ./scripts/reinforcement_learning/<POLICY_NAME>/train.py --task <TASK_NAME> --headless

Replay Trained Model

python ./scripts/reinforcement_learning/<POLICY_NAME>/play.py --task <TASK_NAME> --num_envs 64

Analyze logs

python -m tensorboard.main --logdir=logs

And open the google and go to http://localhost:6006/

Sim2sim

Coming soon...

Sim2Real Deployment using OpenArm

Coming soon...

Related links

• Read the documentation
• Join the community on Discord
• Contact us through [email protected]

License

Apache License 2.0

Copyright 2025 Enactic, Inc.

Code of Conduct

All participation in the OpenArm project is governed by our Code of Conduct.