Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/arbit3rr/nn-control

Control Methods for Dynamic Systems based on Neural Networks

https://github.com/arbit3rr/nn-control

Science Score: 13.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ○
    .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (5.5%) to scientific vocabulary

Keywords

actor-critic-algorithm control-systems neural-network nn-controller pytorch reinforcement-learning rl
Last synced: 5 months ago · JSON representation

Repository

Control Methods for Dynamic Systems based on Neural Networks

Basic Info
  • Host: GitHub
  • Owner: arbit3rr
  • License: gpl-3.0
  • Language: Jupyter Notebook
  • Default Branch: main
  • Homepage:
  • Size: 1.35 MB
Statistics
  • Stars: 2
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Topics
actor-critic-algorithm control-systems neural-network nn-controller pytorch reinforcement-learning rl
Created about 2 years ago · Last pushed over 1 year ago
Metadata Files
Readme License

README.md

NN-based Control Methods for Dynamic Systems

Reference Tracking Neural Network Controller for Dynamic Systems

The controller utilizes Neural Networks to control a nonlinear dynamic system by tracking a given reference signal. The main goal is to minimize the error between the system output and the desired reference trajectory.

Block Diagram

diagram

Components Description

  1. Reference Model:

    • The reference model is defined by the transfer function: math G_m(s) = \frac{K}{\frac{1}{\omega_n^2}s^2 + \frac{2\xi}{\omega_n}s + 1}
    • Generates the desired reference signal.
    • Uses the reference input which is given by: math r(k) = \sin\left(\frac{2\pi k}{25}\right) + \sin\left(\frac{2\pi k}{10}\right)

  2. Dynamic System:

    • The nonlinear dynamic system is represented by: math y(k+1) = \frac{y(k) y(k-1) u(k) + u^3(k) + 0.5 y(k-1)}{1 + y^2(k) + y^2(k-1)}

  3. NN Controller:

    • Adjusts the control input to minimize the tracking error.
    • Utilizes gradients and parameters of the RBF NN model to update the control signal.

Result

NN-based Control

Reference

[1] Slema, S., Errachdi, A., & Benrejeb, M. (2018, March). A radial basis function neural network model reference adaptive controller for nonlinear systems. In 2018 15th International Multi-Conference on Systems, Signals & Devices (SSD) (pp. 958-964). IEEE.

Actor-Critic-Identifier DRL Framework for Pendulum Balancing Problem

DRL Pendulum

Owner

  • Name: Amirhossein Heydarian Ardakani
  • Login: arbit3rr
  • Kind: user

GitHub Events

Total
  • Watch event: 1
Last Year
  • Watch event: 1

Dependencies

requirements.txt pypi
  • gym ==0.25.2