Abstract
Artificial intelligent control methods using fuzzy control, neural networks, and genetic algorithm have recently been recognized as important algorithms that can improve the performances of electronic devices. These methods use their algorithms for solving problems of devices of nonlinear condition. An adaptive control algorithm adjusts the nonlinear system with various parameters for demand performance. In this chapter, we propose a feedback control method using the adaptive neuro-fuzzy inference system (ANFIS), which is a hybrid algorithm that combines fuzzy control, neural network control and adaptive control, for an outer rotor spherical actuator. The proposed algorithm has the combined advantage of expert knowledge of the fuzzy inference system (FIS), learning capabilities of the neural networks (NN) for control of a nonlinear system and adaptabilities of adaptive control for adjusting various parameters. An example application is actuator drives which require adaptability and robustness. This control method using ANFIS is expected to produce more accurate rotor positions compared to other control methods. In order to verify this, experiments using a prototype of an outer rotor spherical actuator fitted with rotary encoders are conducted using a dSPACE controller with MATLAB/Simulink, and we compare the accuracy of the ANFIS control method with the traditional PID control method for feedback control of the outer rotor spherical actuator.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call