Isaac LAB for Flamingo

✨ New Features - Updated 🚀

✔️ Flamingo rev.0.1.4: Latest version of Flamingo added.
✔️ Flamingo Edu v1: Flamingo Edu version added.
✔️ Stack Environment: Observations can be stacked with arguments.
✔️ Constraint Manager: Constraints as Termination (CaT) method implementation added.
✔️ CoRL: Based on rsl_rl library, off-policy algorithms are implemented on off_policy_runner.

Sim2Real - ZeroShot Transfer

Isaac Lab Flamingo

Sim 2 Sim framework - Lab to MuJoCo

• Simulation to Simulation framework is available on sim2sim_onnx branch (Currently on migration update)
• You can simply inference trained policy (basically export as .onnx from isaac lab)

Setup

• This repo is tested on Ubuntu 20.04, and I recommend you to install 'local install'

1. Install Isaac Sim

https://isaac-sim.github.io/IsaacLab/main/source/setup/installation/binaries_installation.html

2. Install Isaac Lab

https://github.com/isaac-sim/IsaacLab

3. Install lab.flamingo package

i. clone repository

git clone https://github.com/jaykorea/Isaac-RL-Two-wheel-Legged-Bot

ii. install lab.flamingo pip package by running below command

• run it on 'lab.flamingo' root path

conda activate env_isaaclab # change to you conda env
pip install -e .

iii. Unzip assets(usd asset) on folder

• Since git does not correctly upload '.usd' file, you should manually unzip the usd files on assests folder

 path example: lab/flamingo/assets/data/Robots/Flamingo/flamingo_rev01_4_1/

Launch script

Train flamingo

• run it on 'lab.flamingo' root path

  python scripts/co_rl/train.py --task {task name} --algo ppo --num_envs 4096 --headless --num_policy_stacks {stack number on policy obs} --num_critic_stacks {stack number on critic obs}

Train example - track velocity

  python scripts/co_rl/train.py --task Isaac-Velocity-Flat-Flamingo-v1-ppo --algo ppo --num_envs 4096 --headless --num_policy_stacks 2 --num_critic_stacks 2

play flamingo

• run it on 'lab.flamingo' root path

  python scripts/co_rl/play.py --task {task name} --algo ppo --num_envs 64 --num_policy_stacks {stack number on policy obs} --num_critic_stacks {stack number on critic obs} --load_run {folder name} --plot False

play example - track velocity

  python scripts/co_rl/play.py --task Isaac-Velocity-Flat-Flamingo-Play-v1-ppo --algo ppo --num_envs 64 --num_policy_stacks {stack number on policy obs} --num_critic_stacks {stack number on critic obs} --load_run 2025-03-16_17-09-35 --plot False