Proposing and Validation of a New Four-Point Finite-Difference Formula With Manipulator Application

In this paper, a four-point one-step-ahead finite-difference formula is presented, which obtains higher computational precision in approximating the first-order derivative. Then, the formula is used for the discretization of the continuous-time Zhang neural network (CTZNN), and it can greatly overcome the limitation of the conventional formulas in CTZNN discretization. Based on this formula, a new-type discrete-time Zhang neural network (DTZNN) model is proposed and investigated for time-variant matrix pseudoinversion. Numerical experiments further validate the feasibility, effectiveness, and superiority of the proposed new-type DTZNN model for solving the time-variant matrix pseudoinversion. Moreover, the proposed new-type DTZNN model is applied to the control of a robot manipulator. Physical experiment performed on a four-link planar robot manipulator is presented to demonstrate physical realizability and effectiveness of the proposed new-type DTZNN model.

[1]  Jinde Cao,et al.  A new neural network for solving quadratic programming problems with equality and inequality constraints , 2014, Math. Comput. Simul..

[2]  Shuai Li,et al.  A dynamic neural network approach for solving nonlinear inequalities defined on a graph and its application to distributed, routing-free, range-free localization of WSNs , 2013, Neurocomputing.

[3]  Yunong Zhang,et al.  Repetitive Motion Planning and Control of Redundant Robot Manipulators , 2013 .

[4]  Anneli Folkesson,et al.  Numerical methods for engineers , 2007 .

[5]  Giuseppe Acciani,et al.  Application of neural networks in optical inspection and classification of solder joints in surface mount technology , 2006, IEEE Transactions on Industrial Informatics.

[6]  Jun Li,et al.  A time‐varying coefficient‐based manipulability‐maximizing scheme for motion control of redundant robots subject to varying joint‐velocity limits , 2013 .

[7]  Ju H. Park,et al.  A study on H∞ state estimation of static neural networks with time-varying delays , 2014, Appl. Math. Comput..

[8]  Oscar Camacho Nieto,et al.  Proportional derivative fuzzy control supplied with second order sliding mode differentiation , 2014, Eng. Appl. Artif. Intell..

[9]  Shuai Li,et al.  Distributed Task Allocation of Multiple Robots: A Control Perspective , 2018, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[10]  Dongsheng Guo,et al.  Zhang neural network, Getz-Marsden dynamic system, and discrete-time algorithms for time-varying matrix inversion with application to robots' kinematic control , 2012, Neurocomputing.

[11]  J. Marsden,et al.  Dynamic inversion of nonlinear maps with applications to nonlinear control and robotics , 1995 .

[12]  Ke Chen,et al.  Performance Analysis of Gradient Neural Network Exploited for Online Time-Varying Matrix Inversion , 2009, IEEE Transactions on Automatic Control.

[13]  Long Jin,et al.  Taylor $O(h^{3})$ Discretization of ZNN Models for Dynamic Equality-Constrained Quadratic Programming With Application to Manipulators , 2016, IEEE Transactions on Neural Networks and Learning Systems.

[14]  M. Combrinck.  Analysis of numerical differentiation methods applied to time domain electromagnetic (TDEM) geophysical data in the S-layer differential transform , 2009, Comput. Geosci..

[15]  J. S. C. Prentice.  Truncation and roundoff errors in three-point approximations of first and second derivatives , 2011, Appl. Math. Comput..

[16]  Yoichi Hori,et al.  An Adaptive Speed Sensorless Induction Motor Drive With Artificial Neural Network for Stability Enhancement , 2012, IEEE Transactions on Industrial Informatics.

[17]  I. R. Khan,et al.  Closed-form expressions for the finite approximations of first and higher derivatives based on Taylor series , 1999 .

[18]  Long Jin,et al.  Taylor-type 1-step-ahead numerical differentiation rule for first-order derivative approximation and ZNN discretization , 2015, J. Comput. Appl. Math..

[19]  Dongsheng Guo,et al.  Novel Recurrent Neural Network for Time-Varying Problems Solving [Research Frontier] , 2012, IEEE Computational Intelligence Magazine.

[20]  Katta G. Murty,et al.  Nonlinear Programming Theory and Algorithms , 2007, Technometrics.

[21]  Panagiotis Angelikopoulos,et al.  NDL-v2.0: A new version of the numerical differentiation library for parallel architectures , 2014, Comput. Phys. Commun..

[22]  Leonardo Franco,et al.  FPGA Implementation of the C-Mantec Neural Network Constructive Algorithm , 2014, IEEE Transactions on Industrial Informatics.

[23]  Long Jin,et al.  Discrete-time Zhang neural network of O(τ3) pattern for time-varying matrix pseudoinversion with application to manipulator motion generation , 2014, Neurocomputing.

[24]  Shuai Li,et al.  Decentralized kinematic control of a class of collaborative redundant manipulators via recurrent neural networks , 2012, Neurocomputing.

[25]  Yunong Zhang,et al.  From different ZFs to different ZNN models accelerated via Li activation functions to finite-time convergence for time-varying matrix pseudoinversion , 2014, Neurocomputing.

[26]  R. Rakkiyappan,et al.  Passivity and passification of memristor-based complex-valued recurrent neural networks with interval time-varying delays , 2014, Neurocomputing.

[27]  Tsuneo Yoshikawa,et al.  Manipulability of Robotic Mechanisms , 1985 .

[28]  Vinod Khadkikar,et al.  Application of Artificial Neural Networks for Shunt Active Power Filter Control , 2014, IEEE Transactions on Industrial Informatics.