An Adaptive Learning Rate for RBFNN Using Time-Domain Feedback Analysis

Radial basis function neural networks are used in a variety of applications such as pattern recognition, nonlinear identification, control and time series prediction. In this paper, the learning algorithm of radial basis function neural networks is analyzed in a feedback structure. The robustness of the learning algorithm is discussed in the presence of uncertainties that might be due to noisy perturbations at the input or to modeling mismatch. An intelligent adaptation rule is developed for the learning rate of RBFNN which gives faster convergence via an estimate of error energy while giving guarantee to the l 2 stability governed by the upper bounding via small gain theorem. Simulation results are presented to support our theoretical development.

[1]  Zhang Jingjing,et al.  Variable Step Size LMS Algorithm , 2012 .

[2]  Daniel E. Rivera,et al.  Internal Model Control , 2010, Encyclopedia of Machine Learning.

[3]  Bernard Widrow,et al.  Adaptive Inverse Control: A Signal Processing Approach , 2007 .

[4]  Ranjan Ganguli,et al.  Filter design using radial basis function neural network and genetic algorithm for improved operational health monitoring , 2006, Appl. Soft Comput..

[5]  Torbjörn Wigren,et al.  Recursive prediction error identification using the nonlinear wiener model , 1993, Autom..

[6]  Ali H. Sayed,et al.  A time-domain feedback analysis of filtered-error adaptive gradient algorithms , 1996, IEEE Trans. Signal Process..

[7]  P. Olver.  Nonlinear Systems , 2013 .

[8]  Syed Saad Azhar Ali,et al.  Hammerstein Model Identification Using Radial Basis Functions Neural Networks , 2001, ICANN.

[9]  Ranjan Ganguli,et al.  Structural damage detection in a helicopter rotor blade using radial basis function neural networks , 2003 .

[10]  James C. Hung,et al.  Internal Model Control , 2011, Control and Mechatronics.

[11]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[12]  Narasimhan Sundararajan,et al.  A fully complex-valued radial basis function classifier for real-valued classification problems , 2012, Neurocomputing.

[13]  Fouad M. AL-Sunni,et al.  U-Model Based 2DoF Multi-Variable IMC for Improved Input-Disturbance Rejection: A Case Study on a 2-Link Robot Manipulator , 2011 .

[14]  Ranjan Ganguli,et al.  Rotorcraft parameter estimation using radial basis function neural network , 2010, Appl. Math. Comput..

[15]  Muhammad Shafiq,et al.  U-model based learning feedforward control of MIMO nonlinear systems , 2010 .

[16]  Stanley H. Johnson,et al.  Use of Hammerstein Models in Identification of Nonlinear Systems , 1991 .

[17]  Carlos E. Garcia,et al.  Internal model control. A unifying review and some new results , 1982 .

[18]  R. Zouhal,et al.  Adaptive control of multivariable Hammerstein's nonlinear model , 1993, Proceedings of IEEE Systems Man and Cybernetics Conference - SMC.

[19]  W. Marsden.  I and J , 2012 .

[20]  Sundaram Suresh,et al.  Sequential Projection-Based Metacognitive Learning in a Radial Basis Function Network for Classification Problems , 2013, IEEE Transactions on Neural Networks and Learning Systems.

[21]  Sundaram Suresh,et al.  An on-line learning neural controller for helicopters performing highly nonlinear maneuvers , 2012, Appl. Soft Comput..