This article first introduces a sampled-data state estimator design method for continuous-time long short-term memory (LSTM) neural networks with irregularly sampled output. To this end, the structure of the LSTM is addressed to obtain its dynamic equation. As a result, the LSTM neural network is modeled as a continuous-time linear parameter-varying system that is dependent on the gate units. For this system, the sampled-data Luenberger-and Arcak-type state estimator design methods are presented in terms of linear matrix inequalities (LMIs) by using the properties of the gate units. Lastly, the proposed method not only provides a numerical example for analyzing absolute stability but also demonstrates it in practice by applying a pre-trained behavior generation model of a robot manipulator.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/TNNLS.2024.3359211 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Output sampling synchronization and state estimation in flux-charge domain memristive neural networks with leakage and time-varying delays.
Neural Netw
December 2024
Department of Mathematics, Periyar University, Salem 636011, India. Electronic address:
This paper theoretically explores the coexistence of synchronization and state estimation analysis through output sampling measures for a class of memristive neural networks operating within the flux-charge domain. These networks are subject to constant delayed responses in self-feedback loops and time-varying delayed responses incorporated into the activation functions. A contemporary output sampling controller is designed to discretize system dynamics based on available output measurements, which enhances control performance by minimizing update frequency, thus overcoming network bandwidth limitations and addressing network synchronization and state vector estimation.
View Article and Find Full Text PDFState estimation of networked nonlinear systems with aperiodic sampled delayed measurement.
ISA Trans
November 2024
School of Automation, Nanjing University of Science and Technology, Nanjing, 210094, China. Electronic address:
This paper aims to study the remote state estimation of networked nonlinear systems subject to aperiodic sampled delayed measurement. A novel sampled-data non-affine nonlinear observer (SNNO) is designed to address this issue. The designed observer is composed of two parts: the first part is a continuous-time observer, and the second part is an auxiliary variable design that provides continuous compensation of output estimation errors for the first part's state observer.
View Article and Find Full Text PDFMRI recovery with self-calibrated denoisers without fully-sampled data.
MAGMA
October 2024
Biomedical Engineering, Ohio State University, Columbus, OH, 43210, USA.
Objective: Acquiring fully sampled training data is challenging for many MRI applications. We present a self-supervised image reconstruction method, termed ReSiDe, capable of recovering images solely from undersampled data.
Materials And Methods: ReSiDe is inspired by plug-and-play (PnP) methods, but unlike traditional PnP approaches that utilize pre-trained denoisers, ReSiDe iteratively trains the denoiser on the image or images that are being reconstructed.
Article Synopsis
- The article explores a new approach to Model Predictive Control (MPC) called Periodic Event-Triggered MPC (PETMPC), aimed at managing nonlinear uncertain systems affected by changing disturbances without constantly generating new control sequences.
- The method incorporates a generalized proportional-integral observer to estimate unknown states and disturbances while using predictions from the forward Euler method to create future control inputs, which are stored for use between events.
- Through stability analysis and numerical simulations, the article demonstrates that the PETMPC method is effective, leading to fewer transmissions and calculations while ensuring system stability.
Sampled-Data Exponential Synchronization of Complex Dynamical Networks with Saturating Actuators.
Entropy (Basel)
September 2024
School of Science, Qingdao University of Technology, Qingdao 266520, China.
This paper investigates the problem of exponential synchronization control for complex dynamical systems (CDNs) with input saturation. Considering the effects of transmission delay, a memory sampled-data controller is designed. A modified two-sided looped functional is constructed that takes into account the entire sampling period, which includes both current state information and delayed state information.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!