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 "mutual update" training algorithm for fuzzy adaptive logic control/decision network (FALCON).

S Altug, H J Trussell, M Y Chow

    IEEE Transactions on Neural Networks
    |February 7, 2008
    PubMed
    Summary
    This summary is machine-generated.

    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

    Image analysis technique: a theoretical explanation.

    Applied optics·2010
    Same author

    Comparison of image restoration methods.

    Applied optics·2010
    Same author

    Scan-playback corrections.

    Applied optics·2010
    Same author

    Optimal nonnegative color scanning filters.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2008
    Same author

    Color device calibration: a mathematical formulation.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2008
    Same author

    Colorimetric restoration of digital images.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2008
    Same journal

    Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

    IEEE transactions on neural networks·2013
    Same journal

    Guest editorial: special section on white box nonlinear prediction models.

    IEEE transactions on neural networks·2011
    Same journal

    Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

    IEEE transactions on neural networks·2011
    Same journal

    Guest editorial: special section on data-based control, modeling, and optimization.

    IEEE transactions on neural networks·2011
    Same journal

    Neural network-based multiple robot simultaneous localization and mapping.

    IEEE transactions on neural networks·2011
    Same journal

    Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

    IEEE transactions on neural networks·2011
    See all related articles

    A new training method for the FALCON fuzzy/neural architecture improves accuracy and speed. This line search approach enhances motor fault detection by reducing steady-state error significantly.

    Area of Science:

    • Artificial Intelligence
    • Machine Learning
    • Fuzzy Systems

    Background:

    • The conventional FALCON (Fuzzy/Neural Architecture Learning and Control) training algorithm faces limitations in accuracy due to implicit parameter independence assumptions in fuzzy inference systems.
    • This study addresses these limitations by proposing a novel training scheme for fuzzy/neural architectures.

    Discussion:

    • The proposed training scheme utilizes line search methods, commonly employed in iterative optimization.
    • It features synchronous updates for parameters related to input/output space partitions and fuzzy rules, guided by the Armijo rule for determining update magnitude and direction.
    • This contrasts with the conventional algorithm's sequential or independent parameter updates.

    Key Insights:

    • The novel training algorithm significantly outperforms the conventional FALCON training in a motor fault detection case study.

    Related Experiment Videos

  • It achieved steady-state error reduction twice as fast as the conventional method.
  • Furthermore, the proposed method resulted in a tenfold decrease in steady-state error, indicating superior performance and convergence.

    • Outlook:

    • This line search-based training scheme offers a promising advancement for fuzzy/neural architectures, potentially applicable to a broader range of complex problems.
    • Further research could explore its efficacy in other dynamic systems and fault diagnosis applications.
    • Optimization of the Armijo rule parameters could lead to even faster convergence and higher accuracy.