ampyc 0.0.3

Python implementation of various advanced MPC algorithms

AMPyC

GitHub | PyPI | Issues | Changelog

ampyc -- Advanced Model Predictive Control in Python

General Python package for control theory research, including some reference implementations of various advanced model predictive control (MPC) algorithms.

Features:

• Implements dynamical systems and control interfaces to allow seamless interactions
• Provides abstract base classes to allow custom implementation of any type of dynamical system and controller
• Reference implementations of many advanced MPC algorithms; for a full list of implemented algorithms see below
• Global parameter management for easy experiment setup and management
• Various utility tools for set computations, polytope manipulation, and plotting
• Lecture-style notes and notebook tutorials explaining advanced predictive control concepts

Installation

ampyc requires Python 3.10 or higher. Just use pip for Python 3 to install ampyc and its dependencies:

    python3 -m pip install ampyc

Local (editable) installation

1. Clone this repository using

    git clone [e-mail]:IntelligentControlSystems/ampyc.git

2. Install all dependencies (preferably in a virtual environment) using

    python3 -m pip install -r requirements.txt

3. Install ampyc in editable mode for development. Navigate to this top-level folder and run

    pip install -e .

Getting Started

To get started with the ampyc package, run the tutorial notebook, which provides an introduction to all parts of the package.

For specific control algorithms implemented in ampyc, run the associated notebook in the notebook folder.

Implemented Control Algorithms

Year	Authors	Method/Paper	AMPyC
-	-	Linear Model Predictive Control	code
-	-	Nonlinear Model Predictive Control	code
2001	Chisci et al.	Systems with persistent disturbances: predictive control with restricted constraints	code
2005	Mayne et al.	Robust model predictive control of constrained linear systems with bounded disturbances	code
2013	Bayer et al.	Discrete-time incremental ISS: A framework for robust NMPC	code
2016	Kouvaritakis & Cannon	Stochastic constraint-tightening Model Predictive Control	code
2017	Agrawal & Sreenath	Discrete control barrier functions for safety-critical control of discrete systems with application to bipedal robot navigation	code
2018	Wabersich & Zeilinger	Scalable synthesis of safety certificates from data with application to learning-based control	code
2018	Wabersich & Zeilinger	Linear model predictive safety certification for learning-based control	code
2018	Hewing & Zeilinger	Stochastic Model Predictive Control for Linear Systems Using Probabilistic Reachable Sets	code
2020	Hewing et al.	Recursively feasible stochastic model predictive control using indirect feedback	code

Cite this Package & Developers

If you find this package/repository helpful, please cite our work:

@software{ampyc,
  title  = {AMPyC: Advanced Model Predictive Control in Python},
  author = {Sieber, Jerome and Didier, Alexandre and Rickenbach, Rahel and Zeilinger, Melanie},
  url    = {https://github.com/IntelligentControlSystems/ampyc},
  month  = jun,
  year   = {2025}
}

Principal Developers

  Jerome Sieber     |   Alex Didier     |   Mike Zhang     |   Rahel Rickenbach