Using RBFs in a CMAC to prevent parameter drift in adaptive control

A radial Basis Function Network (RBFN) works well as a nonlinear approximator in direct adaptive control, as long as the number of inputs is low. A Cerebellar Model Arithmetic Computer (CMAC) indexes basis functions efficiently and can handle many inputs, but is prone to adaptive-parameter drift and subsequent bursting. This paper proposes using overlapping RBFs inside a CMAC structure. Specifically the RBFs associated with past and future (predicted) CMAC cells on a CMAC layer are activated along with the currently indexed cell's RBF on that layer. The novel neural network structure achieves the computational efficiency of the CMAC, yet can avoid drift when RBF widths are wide enough. Simulation results with a pendulum compare the performance and robustness of CMAC, RBF, and the proposed RBFCMAC in both the disturbance-free case and with sinusoidal disturbance.

[1]  Chih-Min Lin,et al.  Adaptive Dynamic Sliding-Mode Fuzzy CMAC for Voice Coil Motor Using Asymmetric Gaussian Membership Function , 2014, IEEE Transactions on Industrial Electronics.

[2]  Chris J. B. Macnab,et al.  Preventing bursting in adaptive control using an introspective neural network algorithm , 2014, Neurocomputing.

[3]  Ella M. Atkins,et al.  Road Risk Modeling and Cloud-Aided Safety-Based Route Planning , 2016, IEEE Transactions on Cybernetics.

[4]  Fu-Chuang Chen,et al.  Practical stability issues in CMAC neural network control systems , 1996, IEEE Trans. Control. Syst. Technol..

[5]  Haoyong Yu,et al.  Hybrid feedback feedforward: An efficient design of adaptive neural network control , 2016, Neural Networks.

[6]  Ella M. Atkins,et al.  Cloud aided safety-based route planning , 2014, 2014 IEEE International Conference on Systems, Man, and Cybernetics (SMC).

[7]  Chris J. B. Macnab,et al.  Robust adaptive control of a quadrotor helicopter , 2011 .

[8]  Xiao-Hua Yu,et al.  An adaptive filtering approach for electrocardiogram (ECG) signal noise reduction using neural networks , 2013, Neurocomputing.

[9]  L. G. Kraft,et al.  Real-time vibration control using CMAC neural networks with weight smoothing , 2000, Proceedings of the 2000 American Control Conference. ACC (IEEE Cat. No.00CH36334).

[10]  Chih-Min Lin,et al.  Adaptive recurrent cerebellar model articulation controller for linear ultrasonic motor with optimal learning rates , 2007, Neurocomputing.

[11]  Wen Yu,et al.  Stable adaptive compensation with fuzzy CMAC for an overhead crane , 2011, Inf. Sci..

[12]  Haralambos Sarimveis,et al.  Radial Basis Function Network Training Using a Nonsymmetric Partition of the Input Space and Particle Swarm Optimization , 2013, IEEE Transactions on Neural Networks and Learning Systems.

[13]  Ming-Yang Cheng,et al.  Development of a Recurrent Fuzzy CMAC With Adjustable Input Space Quantization and Self-Tuning Learning Rate for Control of a Dual-Axis Piezoelectric Actuated Micromotion Stage , 2013, IEEE Transactions on Industrial Electronics.

[14]  Abdullah Toprak,et al.  Reduction in impulse noise in digital images through a new adaptive artificial neural network model , 2014, Neural Computing and Applications.

[15]  Charles W. Anderson,et al.  Comparison of CMACs and radial basis functions for local function approximators in reinforcement learning , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[16]  Chun-Shin Lin,et al.  CMAC with General Basis Functions , 1996, Neural Networks.

[17]  Jing Hu,et al.  Removal of EOG and EMG artifacts from EEG using combination of functional link neural network and adaptive neural fuzzy inference system , 2015, Neurocomputing.

[18]  James S. Albus,et al.  Data Storage in the Cerebellar Model Articulation Controller (CMAC) , 1975 .

[19]  Robert M. Sanner,et al.  Gaussian Networks for Direct Adaptive Control , 1991, 1991 American Control Conference.

[20]  Chih-Hong Lin,et al.  Adaptive recurrent Chebyshev neural network control for PM synchronous motor servo‐drive electric scooter with V‐belt continuously variable transmission , 2015 .

[21]  Dan Wang,et al.  Adaptive neural network control for a class of uncertain nonlinear systems in pure-feedback form , 2002, Autom..

[22]  Frank L. Lewis,et al.  Robust backstepping control of nonlinear systems using neural networks , 2000, IEEE Trans. Syst. Man Cybern. Part A.

[23]  K. Narendra,et al.  A new adaptive law for robust adaptation without persistent excitation , 1987 .

[24]  Bo Yang,et al.  Robust Hybrid Control Based on PD and Novel CMAC With Improved Architecture and Learning Scheme for Electric Load Simulator , 2014, IEEE Transactions on Industrial Electronics.

[25]  Bo Egardt,et al.  Stability of Adaptive Controllers , 1979 .

[26]  James S. Albus,et al.  New Approach to Manipulator Control: The Cerebellar Model Articulation Controller (CMAC)1 , 1975 .