Robust precise trajectory tracking of hybrid stepper motor using adaptive critic-based neuro-fuzzy controller

Abstract

In this paper, an adaptive critic-based neuro-fuzzy controller (ACNFC) is developed for robust and high-precision speed trajectory tracking of a hybrid stepper motor (HSM). The proposed model-free controller uses the critic-based learning and backpropagation of errors for adaptive tuning of the consequence part of the fuzzy inference rule. This makes the ACNFC reconfigurable and robust in high-precision tracking applications, such as robot-assisted surgery, involving with parametric uncertainties and environmental disturbances. To investigate the performance and robustness of the ACNFC, HSM system is simulated under various conditions in MATLAB/Simulink. These operating conditions consider mechanical parameter variations, load disturbance, noise impact, and sudden fault occurrence. To verify the effectiveness of the proposed controller, test results are compared with the results obtained by optimized-PI and brain emotional learning-based intelligent controllers. Simulation results confirm the effective performance of the ACNFC for adaptive and precise speed response as well as dealing with nonlinearity and uncertainty in realistic applications.