SLSpy provides a Python-based framework to design and simulate model-based control systems, especially for system level synthesis (SLS) methods. The details of the framework are described in the paper
Shih-Hao Tseng and Jing Shuang (Lisa) Li, ``SLSpy: Python-Based System-Level Controller Synthesis Framework,'' 2020.
The synthesis workflow follows the structure in the paper
Shih-Hao Tseng and James Anderson, ``Deployment Architectures for Cyber-Physical Control Systems,'' 2019. (also in Proc. IEEE ACC, 2020)
A system with both Python 3 and Python 2.7 may encounter installation error. In that case, please make Python 3 your default choice by appropriate aliasing.
SLSpy might still work for Python 2.7 (or higher Python 2 versions), but one should ensure CVXPY is version 1.1 or higher.
First install the necessary packages by
sudo make setup
then install SLSpy by typing
sudo make install
which has been tested under Fedora 30 and Ubuntu 18.04.
For Windows Subsystem for Linux and macOS, the user might need to install without sudo after the setup by
make install
To get started, please refer to the examples in the folder ``examples.''
For more detailed explanations about the basic framework structure, please refer to
Shih-Hao Tseng and Jing Shuang (Lisa) Li, ``SLSpy: Python-Based System-Level Controller Synthesis Framework,'' 2020.
System Level Synthesis (FIR version), c.f.\ James Anderson, John C. Doyle, Steven H. Low, and Nikolai Matni,\ ``System Level Synthesis,'' Annual Reviews in Control, 2019.
Yuh-Shyang Wang, Nikolai Matni, and John C. Doyle,\ ``A System Level Approach to Controller Synthesis,'' IEEE Trans. Autom. Control, 2019
Input-Output Parameterization (FIR version), c.f.\ Luca Furieri, Yang Zheng, Antonis Papachristodoulou, and Maryam Kamgarpour,\ ``An Input-Output Parameterization of Stabilizing Controllers: Amidst Youla and System Level Synthesis,'' IEEE Control Systems Letters, 2019
Currently, SLSpy can handle a linear time-invariant (LTI) system by synthesizing finite impulse response (FIR) controller using SLS/IOP. It is possible to deal with linear time-variant (LTV) system under SLSpy framework by introducing customized system model, controller model, and synthesizer.
The way SLSpy works is to express transfer matrices as a list. For example, an FIR transfer matrix Φ with the z-transform
Φ = Σ_{t=0}^T Φ[t] z^{-t}
is expressed by
[ Φ[0], Φ[1], ..., Φ[T] ]
in the codes.