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

Data-Driven Finite-Horizon Approximate Optimal Control for Discrete-Time Nonlinear Systems Using Iterative HDP

Chaoxu Mu, Ding Wang, Haibo He

    IEEE Transactions on Cybernetics
    |October 14, 2017
    PubMed
    Summary

    This study introduces a data-driven optimal control method for nonlinear systems using adaptive dynamic programming (ADP). The approach iteratively learns control policies without needing a system model, ensuring stability and demonstrating effectiveness through simulations.

    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

    SACI framework-based fixed-time learning control for nonlinear systems with asymmetric constraints.

    Neural networks : the official journal of the International Neural Network Society·2026
    Same author

    Distributed Inertial k-Winners-Take-All Neural Network Based on Quadratic Optimization Problems.

    IEEE transactions on neural networks and learning systems·2026
    Same author

    Influence factors and model selection for conflict risk in different car-following behaviors: Insights from automated and human-driven vehicles.

    Accident; analysis and prevention·2025
    Same author

    Nesterov Accelerated Gradient Tracking With Adam for Distributed Online Optimization.

    IEEE transactions on neural networks and learning systems·2025
    Same author

    Observer-Based Event-Triggered Fault-Tolerant Synchronization for Memristive Neural Networks Subject to Multiple Failures.

    IEEE transactions on neural networks and learning systems·2025
    Same author

    Finite time dynamic analysis of memristor-based fuzzy NNs with inertial term: Nonreduced-order approach.

    Neural networks : the official journal of the International Neural Network Society·2025

    Area of Science:

    • Control Theory
    • Machine Learning
    • Nonlinear Systems

    Background:

    • Optimal control for discrete-time nonlinear affine systems is challenging.
    • Existing methods often require precise system models.
    • Data-driven approaches offer a promising alternative.

    Purpose of the Study:

    • To develop a data-based finite-horizon optimal control strategy.
    • To apply iterative adaptive dynamic programming (ADP) for solving the Hamilton-Jacobi-Bellman equation.
    • To enable control without prior system function knowledge.

    Main Methods:

    • Utilizing heuristic dynamic programming (HDP) with a model network for initial control.
    • Employing an action network for approximate optimal control law.
    • Using a critic network to approximate the optimal cost function.

    Related Experiment Videos

  • Analyzing convergence of the ADP algorithm and stability of weight estimation errors.

    • Main Results:

    • The iterative ADP algorithm demonstrates convergence.
    • Weight estimation errors remain stable within the HDP structure.
    • The proposed method successfully controls nonlinear systems using only data.
    • Simulation examples validate the theoretical findings.

    Conclusions:

    • The data-based finite-horizon optimal control approach is effective for discrete-time nonlinear affine systems.
    • The heuristic dynamic programming implementation allows control without system models.
    • The method offers a robust and stable solution for complex control problems.