A Modified CMAC Algorithm Based on Credit Assignment
Neural Processing Letters
Design and analysis of direct-action CMAC PID controller
Neurocomputing
Credit assigned CMAC and its application to online learning robust controllers
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Learning convergence of CMAC technique
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
This paper presents the controller design of an XY parallel micropositioning stage aiming at a sub-micron accuracy for micro scale manipulation. Owning to the decoupled design of the mechanism, a simple single-input-single-output (SISO) PID controller is adopted for each axis. To compensate for the hysteresis arising from piezoelectric actuator, credit assigned CMAC (cerebellar model articulation controller) neural network with adjustable learning rate is employed into the PID control. Experimental results show that the hysteresis of the stage has been significantly reduced by the CMAC-based PID controller and the stage can achieve a sub-micron positioning accuracy, which demonstrate the effectiveness of the designed controller as well.