Journal Article10.1109/TCSII.2022.3231350
Finite-Time Convergent Modified Davidenko Method for Dynamic Nonlinear Equations
Yinyan Zhang,Guanggang Geng +1 more
- 01 Apr 2023
Vol. 70, pp 1630-1634
3
TL;DR: In this paper , a modified Davidenko method via normalization was proposed for solving DNEs with unknown derivatives, which converges in finite time regardless of additive noise, and theoretical guarantees and numerical examples of the proposed method are provided.
read more
Abstract: Neurodynamic methods, especially zeroing neural networks (ZNN), are widely used for solving nonlinear equations. Being simpler than the ZNN, the Davidenko method is efficient for static nonlinear equations. However, it has been shown that there are large lagging errors when the Davidenko method is applied to dynamic nonlinear equations (DNEs). In this brief, we address this issue by proposing a modified Davidenko method via normalization, making it suitable for solving DNEs with unknown derivatives. The proposed method converges in finite time for DNEs, regardless of additive noise. Both theoretical guarantees and numerical examples of the proposed method are provided.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Dynamic Neural Network Models for Time-Varying Problem Solving: A Survey on Model Structures
TL;DR: In this paper , a review of real-valued, complex-valued and noise-tolerant dynamic neural networks for solving various time-varying problems is presented, which discusses the finite-time convergence, fixed-time parameters, and noise tolerance properties of dynamic neural network models.
5
Discrete-time zeroing neural network with quintic error mode for time-dependent nonlinear equation and its application to robot arms
Naimeng Cang,Weidong Zhang,Weibing Li,Xuanxian Li +3 more
TL;DR: This paper proposes a discrete-time zeroing neural network (DTZNN) with quintic error mode for solving time-dependent nonlinear equations, demonstrating its validity and superiority through theoretical analysis and numerical results, with practical applications in robot arm simulations and experiments.
2
A Novel Fuzzy-type Zeroing Neural Network for Dynamic Matrix Solving and Its Applications
Lv Zhao,Xin Liu,Jie Jin +2 more
References
Noise-Tolerant ZNN Models for Solving Time-Varying Zero-Finding Problems: A Control-Theoretic Approach
TL;DR: Noise-tolerant zeroing neural network models derived from the proposed design formula demonstrate their advantages when applied to solving time-varying zero-finding problems in the presence of noises.
223
Comparison on Zhang neural dynamics and gradient-based neural dynamics for online solution of nonlinear time-varying equation
TL;DR: Computer-simulation results via power-sigmoid activation functions substantiate the theoretical analysis and efficacy of the ZND model for solving online nonlinear time-varying equations.
209
Activated Gradients for Deep Neural Networks.
TL;DR: In this article, a novel method by acting the gradient activation function (GAF) on the gradient is proposed to handle the saddle point problem, which enlarges the tiny gradients and restricts the large gradient.
Modified Primal-Dual Neural Networks for Motion Control of Redundant Manipulators With Dynamic Rejection of Harmonic Noises
Shuai Li,MengChu Zhou,Xin Luo +2 more
TL;DR: This work proposes a novel primal-dual neural network design that directly takes noise control into account and consistently and significantly outperforms the existing primal- dual neural solutions as well as feedforward neural one and adaptive neural one for redundancy resolution of manipulators.
170
Solving Time-Varying System of Nonlinear Equations by Finite-Time Recurrent Neural Networks With Application to Motion Tracking of Robot Manipulators
Lin Xiao,Zhijun Zhang,Shuai Li +2 more
TL;DR: Computer simulations based on a numerical example validate the preponderance of the proposed FTRNNs for time-varying SoNE, as compared to the recently proposed Zhang neural network and its improved version.
132