go2RL

Setting up the environment

1. Create a workspace folder

2. Clone this repository inside the workspace folder

3. Create a virtual environment

python3 -m venv <name of virtual environment>

4. Activate the virtual environment

source <name of virtual environment>/bin/activate

5. Install necessary libraries

pip install -r requirements.txt

6. Inside the workspace folder, clone the rsl_rl repository found here https://github.com/leggedrobotics/rsl_rl/tree/main

7. Install the python package

cd rsl_rl && git checkout v1.0.2 && pip install -e .
pip install tensorboard

8. Go back to the root of the workspace

cd ..

9. Clone the genesis repository found here https://github.com/Genesis-Embodied-AI/Genesis/tree/main

10. Navigate to this go2RL directory

cd go2RL/

11. Run the training script

python go2_train.py

12. Run the evaluation script

python go2_eval.py