Skip to content

pyvrft - Virtual Reference Feedback Tuning

pyvrft is a Python toolbox for designing feedback controllers from input/output data using Virtual Reference Feedback Tuning (VRFT).

The package supports SISO and MIMO controller design with standard least-squares estimation and instrumental variables.

Install

pip install pyvrft

Quick Start

import vrft

p = vrft.design(u, y, y, Td, C, L)

where u and y are input/output data, Td is the reference model, C is the controller structure, and L is the VRFT pre-filter.

Signals are represented as NumPy matrices shaped (N, n), where N is the number of samples and n is the number of inputs or outputs.

Features

  • SISO and MIMO VRFT controller design
  • Least-squares implementation
  • Instrumental-variable implementation through a second output data set
  • Filtering utilities for discrete-time transfer functions
  • Stable inversion utilities for reference-model inversion
  • CSV helper for loading input/output data

Documentation

  • Installation: install from PyPI or from a development checkout.
  • Quick Start: design a first SISO controller with VRFT.
  • Basic Concepts: signal shapes, transfer-function conventions, MIMO lists, and controller structures.
  • Examples: runnable SISO, MIMO, and CSV examples from the repository.
  • API Reference: public functions re-exported by vrft.
  • Citation: citation information and DOI.

Citation

The archived software release is available at https://doi.org/10.5281/zenodo.20602578. See Citation.