Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

A multilayer recurrent neural network for solving continuous-time algebraic Riccati equations.

Jun Wang1, Guang Wu

  • 1Department of Mechanical and Automation Engineering, The Chinese University of Hong Kong, Shatin, Hong Kong

Neural Networks : the Official Journal of the International Neural Network Society
|March 29, 2003
PubMed
Summary

A novel multilayer recurrent neural network efficiently solves continuous-time algebraic matrix Riccati equations in real time. This approach also enables real-time synthesis of linear-quadratic control systems.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Ligand non-innocence enables HAT at dinuclear nickel complexes.

Dalton transactions (Cambridge, England : 2003)·2026
Same author

Repurposing primaquine diphosphate for imatinib-resistant chronic myeloid leukemia via targeting BCR-ABL and Wnt/β-catenin pathway.

iScience·2026
Same author

Synthesis of a Series of Ln(III) (Ln = La, Ce, Lu) Aryl Complexes and Analysis of Their Ln-L Bonding Using Multinuclear NMR Spectroscopy and DFT Calculations.

Inorganic chemistry·2026
Same author

A novel therapeutic monoclonal antibody targeting PZR demonstrates potent anti-metastatic efficacy in triple-negative breast cancer.

Journal of translational medicine·2026
Same author

Zoledronic Acid Inhibits the Growth of ER-Positive Breast Cancer Cells by Inducing Ferroptosis.

Biomolecules & therapeutics·2026
Same author

Remote ischaemic conditioning in patients with supratentorial intracerebral haemorrhage (RICH-2): a multicentre, randomised, sham-controlled phase 3 trial in China.

EClinicalMedicine·2026

Area of Science:

  • Control Systems Engineering
  • Computational Neuroscience
  • Applied Mathematics

Background:

  • Continuous-time algebraic matrix Riccati equations (CAREs) are fundamental in control theory and dynamical systems analysis.
  • Real-time solutions for CAREs are crucial for advanced control system synthesis and adaptive control applications.
  • Existing methods for solving CAREs may face limitations in real-time computational efficiency.

Purpose of the Study:

  • To propose a novel multilayer recurrent neural network (RNN) architecture for solving continuous-time algebraic matrix Riccati equations.
  • To demonstrate the capability of the proposed RNN for real-time synthesis of linear-quadratic (LQ) control systems.
  • To analytically investigate the stability of the RNN and the solvability of CAREs using this approach.

Main Methods:

Related Experiment Videos

  • Development of a four-layer, bidirectionally connected recurrent neural network.
  • Utilizing an array of neurons within each layer for parallel processing.
  • Analytical derivation of stability conditions for the RNN.
  • Empirical validation through three illustrative examples.

Main Results:

  • The proposed multilayer RNN effectively solves continuous-time algebraic matrix Riccati equations in real time.
  • The network successfully synthesizes linear-quadratic control systems with real-time performance.
  • Analytical results confirm the stability of the RNN and its capacity to solve CAREs.
  • Demonstrated practical applicability and operating characteristics through case studies.

Conclusions:

  • The multilayer recurrent neural network presents a viable and efficient solution for real-time CAREs.
  • This RNN-based method offers a powerful tool for real-time LQ control system design.
  • The study provides theoretical and practical evidence for the effectiveness of the proposed neural network architecture.