Python library for simulating dynamics and ensemble model predictive control.
The code in this repository was prepared to implement the methodology described in
- C. Folkestad, D. Pastor, J. Burdick, "Ensemble Model Predictive Control", in Proc. Conf on Decision and Control Control, (submitted) 2020
The simulation framework of this repository is adapted from the Learning and Control Core Library.
Set up virtual environment
python3 -m venv .venv
Activate virtual environment
source .venv/bin/activate
Upgrade package installer for Python
pip install --upgrade pip
Install requirements
pip3 install -r requirements.txt
Create conda environment
conda create --name ensemblempc
conda activate ensemblempcconda install matplotlib numpy pyqtgraph
pip install torch cvxpy % not available in conda
To run the code, run one of the examples in
core/examples
Run the example scripts as a module with the root folder of repository as the working directory. For example, in a Python 3 environment run
python -m core.examples.1d_drone_landing
To visualize use
python -m core.examples.1d_pyqtgraph