A Python based deployment repository for executing
policies trained with LFMC-Gym
.
It can be used with both
PyBullet and
RaisimPy.
Switching between the two engines is extremely simple.
Project website: https://ori-drs.github.io/lfmc/
Training repository: https://github.com/ori-drs/lfmc_gym
Deployment (C++): https://github.com/ori-drs/lfmc_cval
Preprint: https://arxiv.org/abs/2209.14887
@article{lfmc-locomotion,
title = {Learning Low-Frequency Motion Control for Robust and Dynamic Robot Locomotion},
author = {Gangapurwala, Siddhant and Campanaro, Luigi and Havoutis, Ioannis},
url = {https://arxiv.org/abs/2209.14887},
publisher = {arXiv},
year = {2022},
doi = {10.48550/ARXIV.2209.14887},
}
This is optional and is
only applicable if you would like to use RaisimPy.
Ensure that you have
compiled it and that $PYTHONPATH
contains the
RaisimPy install location.
export PYTHONPATH=$PYTHONPATH:<raisim-install-directory>/lib
To clone lfmc_pyval
, use the following command. The Python
version does not depend on other repositories.
git clone [email protected]:ori-drs/lfmc_pyval.git
A setup.py
script is provided with the Python
package which handles all the necessary dependencies.
It is recommended that you use a
virtual environment.
If python3-venv
is not already installed, use the following command.
sudo apt-get install python3-venv
Now create and source a virtual environment.
cd lfmc_pyval
python3 -m venv venv
source venv/bin/activate
The dependencies can then be installed by
pip install -e .
This will install numpy
, torch
, pybullet
, and
pyyaml
along with their dependencies.
To run the included example, just use the following command.
python scripts/command_tracking.py
This will execute the locomotion policy in PyBullet. To switch to
RaisimPy, simply edit the first line in the simulation
configuration file. The file is called simulation.yaml
and can be found at $LFMC_PYVAL_PATH/configuration/simulation.yaml
.
By default, this is what Line 1 reads.
engine: pybullet # Options: [raisim, pybullet]
To switch to Raisim, change it to
engine: raisim # Options: [raisim, pybullet]
Assuming RaisimUnity
has been launched for visualization,
you can execute the command_tracking.py
script just as before.
The example will now utilize the Raisim engine.
python scripts/command_tracking.py
└── common # Utilities used by modules throughout the project
├── paths.py # Utility to handle project related paths
├── networks.py # Contains MLP and GRU classes based on PyTorch
└── configuration # Configuration files
├── simulation.yaml # Simulation configuration parameters
└── motion_control # Locomotion controller related files
├── controller.py # Controller interface class
└── resources # Assets used in the project
├── models # Robot URDFs
├── parameters # Network parameters for locomotion and actuation
└── scripts # Python scripts
├── command_tracking.py # Example script for executing command tracking policy
└── simulation # Simution utilities
├── actuation.py # Contains the actuator network class for ANYmal C
├── pybullet.py # Interface utility for PyBullet
├── raisim.py # Interface utility for RaisimPy
├── wrapper.py # Wrapper to switch between PyBullet and RaisimPy
└── setup.py # Installs LFMC-PyVal and dependencies