Neural networks have been generally deemed as important tools to handle kinds of online computing problems in recent decades, which plenty applications science and electronics fields. This paper proposes a novel recurrent neural network (RNN) the perturbed time-varying underdetermined linear system with double bound limits on residual errors state variables. Beyond that, bound-limited is converted into that consists nonlinear formulas through constructing nonnegative variable. Then,...

A typical recurrent neural network called zeroing (ZNN) was developed for time-varying problem-solving in a previous study. Many applications result linear equation and inequality systems that should be solved real time. This paper provides ZNN model determining the solution of systems. By introducing nonnegative slack variable, are transformed into mixed nonlinear system. The is established via definition an indefinite error function usage exponential decay formula. Theoretical results...

In this paper, for online solution of time-varying linear matrix inequality (LMI), such an LMI is first converted to a equation by introducing matrix, which each element greater than or equal zero. Then, employing Zhang 's neural dynamic method, special recurrent network termed (ZNN) proposed and investigated solving the as well LMI. Such ZNN model showed in explicit dynamics exploits time-derivative information coefficients. addition, theoretical analysis results are discussed presented...

In this paper, two simple-structure neural networks based on the error back-propagation (BP) algorithm (i.e., BP-type networks, BPNNs) are proposed, developed, and investigated for online generalized matrix inversion. Specifically, BPNN-L BPNN-R models proposed left right inversion, respectively. addition, same problem-solving task, discrete-time Hopfield-type (HNNs) developed in paper. Similar to classification of presented models, HNN-L HNN-R correspond Comparing BPNN weight-updating...

Recently, a new formula with noise-tolerant capability has been designed by Zhang et al., and the resultant dynamics (ZD) models have developed to solve different types of time-varying problems. Based on previous research, this paper presents investigates application such ZD design kinematic control redundant robot manipulators via nonlinear equations solving. Specifically, exploiting system involved in control, redundancy-resolution scheme is established. Such contains proportional,...

In this paper, a new path-planning scheme based on pseudoinverse-type formulation for redundant robot manipulators with the existence of noise is proposed and investigated. Such pseudoinverse-based (PPP) contains proportional, integral, derivative information desired Cartesian path (of end-effector), can thus be viewed as nonlinear proportional-integral-derivative (PID) controller manipulators. other words, PPP has PID characteristic in terms path. Theoretical results are given to show that...

The Zhang neural network (ZNN) has recently realized remarkable success in solving time-varying problems. Harmonic noise widely exists industrial applications and can severely affect the solution computed by ZNN models. This article attempts to solve aforementioned limitations providing first design with an inherent capability prohibit harmonic noise. Moreover, it opens new opportunities shift research on ZNNs ideal situations that theoretical consideration nonideal working environments. We...

Repetitive motion planning (RMP) is a crucial issue encountered in studies on redundant robot manipulators. Numerous RMP schemes have been established previous wherein simulations are assumed to be free of noise. However, noise ubiquitous and can severely affect the point causing failure. This paper attempts address limitations imposed by providing first scheme with inherent noise-suppression capability. The new for manipulators noisy environment proposed basis an equality criterion that...

As we know, harmonic noises widely exist in industrial fields and have a crucial impact on the computational accuracy of zeroing neural network (ZNN) model. For tackling this issue, by combining dynamics signals, two noise-tolerant ZNN (HNTZNN) models are designed for dynamic matrix pseudoinversion. In design HNTZNN models, an adaptive compensation term is adopted to eliminate influence noises, Li activation function introduced further improve convergence rate. The robustness proposed proved...

By following the inspirational work of McCulloch and Pitts [1], lots neural networks have been proposed, developed studied for scientific research engineering applications [2][18]. For instance, one classical network is Hopfield (HNN) which was proposed by in early 1980s [2]. Another based on error back-propagation (BP) algorithm, i.e., BP network, Rumelhart, McClelland others mid-1980s [3]. Generally speaking, according to nature connectivity, these can be classified into two categories:...

Recently, the approach based on recurrent neural network (RNN) has been considered a powerful alternative to mathematical problem solving. In this study, new discrete-time RNN (DTRNN) is proposed and investigated determine an exact solution of dynamic nonlinear equations. Specifically, resultant DTRNN model established for solving equations by utilizing Taylor-type difference rule. This then theoretically proven have O(τ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML"...

Repetitive motion planning (RMP) plays a remarkable role in the operation of robotic manipulators. In this article, RMP manipulators wherein high precision joint angle repeatability and end-effector is guaranteed investigated. particular, novel pseudoinverse-based (P-based) scheme designed proposed for by applying special difference rule to discretize existing with P-based formulation. Such theoretically analyzed proven simultaneously guarantee repetitive precision. Comparative simulation...

This paper proposes and investigates a weighted velocity acceleration minimization scheme to prevent the occurrence of high joint caused by minimum norm (MAN) in redundant robot manipulators. The proposed considers kinetic energy (MKE) MAN criterions via two weighting factors, thus guaranteeing final motion be near zero, which is acceptable for engineering applications. Joint physical constraints (i.e., angle limits, limits) are incorporated formulation scheme. reformulated as quadratic...