Noise-tolerant neural algorithm for online solving time-varying full-rank matrix Moore-Penrose inverse problems: A control-theoretic approach

Abstract In this paper, zeroing neural network models are redesigned and analyzed from a control-theoretical framework for online solving time-varying full-rank Moore-Penrose inversions. To solve time-varying full-rank Moore-Penrose inverse problems with different noises in real time, some modified zeroing neural network models are developed, analyzed and investigated from the perspective of control. Furthermore, the proposed zeroing neural network models globally converge to the theoretical solution of the full-rank Moore-Penrose inverse problem without noises, and exponentially converge to the exact solution in the presence of noises, which are demonstrated theoretically. Moreover, in comparison with existing models, numerical simulations are provided to substantiate the feasibility and superiority of the proposed modified neural network for online solving time-varying full-rank Moore-Penrose problems with inherent tolerance to noises. In addition, the numerical results infer that different activation functions can be applied to accelerate the convergence speed of the zeroing neural network model. Finally, the proposed zeroing neural network models are applied to the motion generation of redundant robot manipulators, which illustrates its high efficiency and robustness.

[1]  Zhongbo Sun,et al.  A new trust region–sequential quadratic programming approach for nonlinear systems based on nonlinear model predictive control , 2019 .

[2]  Yunong Zhang,et al.  Design and experimentation of acceleration-level drift-free scheme aided by two recurrent neural networks , 2013 .

[3]  Ping-Cheng Yeh,et al.  An Interpretation of the Moore-Penrose Generalized Inverse of a Singular Fisher Information Matrix , 2011, IEEE Transactions on Signal Processing.

[4]  António M. Lopes,et al.  A fractional perspective on the trajectory control of redundant and hyper-redundant robot manipulators , 2017 .

[5]  Shuai Li,et al.  Noise-Tolerant ZNN Models for Solving Time-Varying Zero-Finding Problems: A Control-Theoretic Approach , 2017, IEEE Transactions on Automatic Control.

[6]  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.

[7]  Yunong Zhang,et al.  Zhang Functions and Various Models , 2015 .

[8]  Dongsheng Guo,et al.  Novel Discrete-Time Zhang Neural Network for Time-Varying Matrix Inversion , 2017, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[9]  Jun Wang,et al.  Global exponential stability of recurrent neural networks for synthesizing linear feedback control systems via pole assignment , 2002, IEEE Trans. Neural Networks.

[10]  Vasilios N. Katsikis,et al.  Modified discrete iterations for computing the inverse and pseudoinverse of the time-varying matrix , 2018, Neurocomputing.

[11]  Heinz H. Bauschke,et al.  Fitzpatrick Functions and Continuous Linear Monotone Operators , 2007, SIAM J. Optim..

[12]  Shuai Li,et al.  Cooperative Motion Generation in a Distributed Network of Redundant Robot Manipulators With Noises , 2018, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[13]  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.

[14]  Blake Hannaford,et al.  A Model-Based Recurrent Neural Network With Randomness for Efficient Control With Applications , 2019, IEEE Transactions on Industrial Informatics.

[15]  Sitian Qin,et al.  A new one-layer recurrent neural network for nonsmooth pseudoconvex optimization , 2013, Neurocomputing.

[16]  Zhijun Zhang,et al.  A Varying-Parameter Convergent-Differential Neural Network for Solving Joint-Angular-Drift Problems of Redundant Robot Manipulators , 2018, IEEE/ASME Transactions on Mechatronics.

[17]  Zhijun Zhang,et al.  Two Hybrid Multiobjective Motion Planning Schemes Synthesized by Recurrent Neural Networks for Wheeled Mobile Robot Manipulators , 2021, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[18]  Xiang Chen,et al.  A Visual Distance Approach for Multicamera Deployment With Coverage Optimization , 2018, IEEE/ASME Transactions on Mechatronics.

[19]  Yantao Tian,et al.  A novel projected Fletcher‐Reeves conjugate gradient approach for finite‐time optimal robust controller of linear constraints optimization problem: Application to bipedal walking robots , 2018 .

[20]  Shuai Li,et al.  Manipulability Optimization of Redundant Manipulators Using Dynamic Neural Networks , 2017, IEEE Transactions on Industrial Electronics.

[21]  Shuai Li,et al.  On Generalized RMP Scheme for Redundant Robot Manipulators Aided With Dynamic Neural Networks and Nonconvex Bound Constraints , 2019, IEEE Transactions on Industrial Informatics.

[22]  Dongsheng Guo,et al.  New Pseudoinverse-Based Path-Planning Scheme With PID Characteristic for Redundant Robot Manipulators in the Presence of Noise , 2018, IEEE Transactions on Control Systems Technology.

[23]  Lin Xiao,et al.  Noise-Tolerant and Finite-Time Convergent ZNN Models for Dynamic Matrix Moore–Penrose Inversion , 2020, IEEE Transactions on Industrial Informatics.

[24]  Peiyao Shen,et al.  Complete and Time-Optimal Path-Constrained Trajectory Planning With Torque and Velocity Constraints: Theory and Applications , 2018, IEEE/ASME Transactions on Mechatronics.

[25]  Karim Djouani,et al.  Toward Lower Limbs Functional Rehabilitation Through a Knee-Joint Exoskeleton , 2017, IEEE Transactions on Control Systems Technology.

[26]  Shuai Li,et al.  Dynamic neural networks aided distributed cooperative control of manipulators capable of different performance indices , 2018, Neurocomputing.

[27]  Rachel W Jackson,et al.  Human-in-the-loop optimization of exoskeleton assistance during walking , 2017, Science.

[28]  Siyuan Chen,et al.  Compatible Convex–Nonconvex Constrained QP-Based Dual Neural Networks for Motion Planning of Redundant Robot Manipulators , 2019, IEEE Transactions on Control Systems Technology.

[29]  Punyaphol Horata,et al.  Robust extreme learning machine , 2013, Neurocomputing.

[30]  Ning Tan,et al.  Zhang neural network solving for time-varying full-rank matrix Moore–Penrose inverse , 2010, Computing.

[31]  Shuzhi Sam Ge,et al.  Adaptive Neural Network Control of Robotic Manipulators , 1999, World Scientific Series in Robotics and Intelligent Systems.

[32]  Yueming Hu,et al.  Robustness Analysis and Robotic Application of Combined Function Activated RNN for Time-Varying Matrix Pseudo Inversion , 2019, IEEE Access.

[33]  Lin Xiao,et al.  New Varying-Parameter ZNN Models With Finite-Time Convergence and Noise Suppression for Time-Varying Matrix Moore–Penrose Inversion , 2020, IEEE Transactions on Neural Networks and Learning Systems.

[34]  Yantao Tian,et al.  A superlinear convergence feasible sequential quadratic programming algorithm for bipedal dynamic walking robot via discrete mechanics and optimal control , 2016 .

[35]  Shuai Li,et al.  Modified ZNN for Time-Varying Quadratic Programming With Inherent Tolerance to Noises and Its Application to Kinematic Redundancy Resolution of Robot Manipulators , 2016, IEEE Transactions on Industrial Electronics.

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

[37]  Walter Fichter,et al.  Fast Exact Redistributed Pseudoinverse Method for Linear Actuation Systems , 2019, IEEE Transactions on Control Systems Technology.