Adaptive recurrent cerebellar model articulation controller for linear ultrasonic motor with optimal learning rates

Authors:
Ya-Fu Peng;Chih-Min Lin
Affiliations:
Department of Electrical Engineering, Ching-Yun University, Chung-Li, Tao-Yuan, 320, Taiwan, Republic of China;Department of Electrical Engineering, Yuan-Ze University, Chung-Li, Tao-Yuan, 320, Taiwan, Republic of China
Venue:
Neurocomputing
Year:
2007

Citing 6
Cited 3

CMAC with general basis functions

Neural Networks
Optimal design of CMAC neural-network controller for robotmanipulators

IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
Smooth trajectory tracking of three-link robot: a self-organizingCMAC approach

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Generalizing CMAC architecture and training

IEEE Transactions on Neural Networks
Discrete-time CMAC NN control of feedback linearizable nonlinear systems under a persistence of excitation

IEEE Transactions on Neural Networks
High-order MS CMAC neural network

IEEE Transactions on Neural Networks

CMAC-based neuro-fuzzy approach for complex system modeling

Neurocomputing
Standalone CMAC control system with online learning ability

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Learning in fully recurrent neural networks by approaching tangent planes to constraint surfaces

Neural Networks

Quantified Score

Hi-index	0.01

Visualization

Abstract

In this study, an adaptive recurrent cerebellar model articulation controller (ARCMAC) is investigated for the motion control of linear ultrasonic motor (LUSM). The proposed ARCMAC has superior capability to the conventional cerebellar model articulation controller in efficient learning mechanism and dynamic response. The dynamic gradient descent method is adopted to online adjust the ARCMAC parameters. Moreover, the analytical method based on a Lyapunov function is proposed to determine the learning-rates of ARCMAC so that the stability of the system can be guaranteed. Furthermore, the variable optimal learning-rates are derived to achieve the fastest convergence of tracking error. Finally, the effectiveness of the proposed control system is verified by the experiments of LUSM motion control. Experimental results show that high-precision tracking response can be achieved by using the proposed ARCMAC.