Feedback control of outer rotor spherical actuator using adaptive neuro-fuzzy inference system

Adaptive neuro-fuzzy inference systems (ANFIS) has the advantage of expert knowledge of the fuzzy inference system (FIS) and learning capabilities of the neural networks (NN) for control of a nonlinear system. The membership function parameters are tuned using a combination of the least squares estimation and back propagation. So, a control method using ANFIS will produce more accurate results compared to other control methods. In this paper, we propose a feedback control method using the adaptive neuro-fuzzy inference system (ANFIS) for the outer rotor spherical actuator. Outer rotor spherical actuators are a type of multi-DOF actuator that can produce a high torque. In order to verify the ANFIS, experiments using a prototype of the outer rotor spherical actuator are conducted using a dSPACE controller with MATLAB/Simulink.