Neural Network-Based State Estimation of Nonlinear Systems

Abstract

except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights.

Neural Control for Driving a Mobile Robot Integrating Stereo Vision Feedback

“…Remark 1 It is worth to notice that L i in page 3,4 refers to the number of higher-order connections and the dimension of the weight vector, which according to (2) are the same.…”

“…The RHONN (2) trained with the EKF-based algorithm (7) to identify the nonlinear plant (1) , ensures that the identification error (9) is semiglobally uniformly ultimately bounded (SGUUB); moreover, the RHONN weights remain bounded.…”

“…It is so extremely difficult to consider all the physical laws involved in the behavior of the system [1]. System identification, which has held the attention of researchers for several years, consists on determining a system mathematical model [2]. Neural networks have grown to be a well-established methodology, which allows solving very difficult problems in engineering, as exemplified by their applications to identification and control of general nonlinear and complex systems.…”

Observers for biological systems

“…Using the well-known approximation capabilities of neural networks, neural observers have emerged [4,22,23,24,25,26]; one of the main advantages of this kind of observers is its robustness to uncertainties, external disturbances, among others. In [27,28,29,30,31,32], neural observers are designed to estimate the state for continuous-time nonlinear systems. Although discrete-time observers are preferred for real time applications, the discrete-time case has not been exploited as the continuous one.…”

Fault detection in complex and distributed systems

“…The application of nonlinear models brought a qualitative change by improving the robustness compared to linear models. Examples of these nonlinear models include nonlinear state space models [152], neural networks [78,117,140], fuzzy-logic-based models [116,119].…”